An Ecology of High-Altitude Infancy: A Biocultural Perspective (Cambridge Studies in Medical Anthropology)

This page intentionally left blank An Ecology of High-Altitude Infancy Andrea Wiley investigates the ecological, his...

Author: Andrea S. Wiley

14 downloads 680 Views 2MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

This page intentionally left blank

An Ecology of High-Altitude Infancy

Andrea Wiley investigates the ecological, historical, and sociocultural factors that contribute to the peculiar pattern of infant mortality in Ladakh, a high-altitude region in the western Himalaya of India. Ladakhi newborns are extremely small at birth, smaller than those in other high-altitude populations, smaller still than those in sea-level regions. Factors such as hypoxia, dietary patterns, the burden of women’s work, gender, infectious diseases, seasonality, and use of local health resources all affect a newborn’s birthweight and raise the likelihood of infant mortality. An Ecology of High-Altitude Infancy is unique in that it makes use of the methods of human biology but strongly emphasizes the ethnographic context that gives human biological measures their meaning. It is an example of a new genre of anthropological work: “ethnographic human biology.” Andrea S. Wiley is Associate Professor of Anthropology at James Madison University. She has published in Medical Anthropology Quarterly, Social Science & Medicine, American Journal of Human Biology, Journal of Biosocial Science, and High Altitude Medicine and Biology, among others. Dr. Wiley has traveled extensively in South and Southeast Asia, Europe, and North America.

Cambridge Studies in Medical Anthropology Editor ALAN HARWOOD University of Massachusetts, Boston Editorial Board WILLIAM DRESSLER University of Alabama RONALD FRANKENBERG Brunel University, UK MARY JO GOOD Harvard University SHARON KAUFMAN University of California, San Francisco SHIRLEY LINDENBAUM City University of New York MARGARET LOCK McGill University CATHERINE PANTER-BRICK University of Durham, UK

Medical anthropology is the fastest growing specialist area within anthropology, in both North America and Europe. Beginning as an applied ﬁeld serving public health specialists, medical anthropology now provides a signiﬁcant forum for many of the most urgent debates in anthropology and the humanities. It includes the study of medical institutions and health care in a variety of rich and poor societies, the investigation of the cultural construction of illness, and the analysis of ideas about the body, birth, maturity, aging, and death. This series includes theoretically innovative monographs and state-ofthe-art collections of essays on current issues. 1 Lynn M. Morgan, Community Participation in Health: The Politics of Primary Care in Costa Rica 2 Thomas J. Csordas (ed.), Embodiment and Experience: The Existential Ground of Culture and Health 3 Paul Brodwin, Medicine and Morality in Haiti: The Contest for Healing Power 4 Susan Reynolds Whyte, Questioning Misfortune: The Pragmatics of Uncertainty in Eastern Uganda 5 Margaret Lock and Patricia Kaufert, Pragmatic Women and Body Politics 6 Vincanne Adams, Doctors for Democracy 7 Elizabeth Hsu, The Transmission of Chinese Medicine Series information continues after the index.

An Ecology of High-Altitude Infancy A Biocultural Perspective Andrea S. Wiley James Madison University

cambridge university press Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Cambridge University Press The Edinburgh Building, Cambridge cb2 2ru, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521830003 © Cambridge University Press 2004 This publication is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published in print format isbn-13 isbn-10

978-0-511-18695-0 eBook (EBL) 0-511-18695-9 eBook (EBL)

isbn-13 isbn-10

978-0-521-83000-3 hardback 0-521-83000-1 hardback

isbn-13 isbn-10

978-0-521-53682-0 paperback 0-521-53682-0 paperback

Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

For Aidan and Emil

Contents

List of Figures, Photographs, and Tables Acknowledgments Abbreviations Glossary of Ladakhi Words

page x xiii xix xxi

1

Introduction

2

Challenges of High-Altitude Living

25

3

Contextualizing Reproductive Health Research in Ladakh

39

4

Big Mountains, Small Babies

71

5

An Ecology of Infancy in Ladakh

105

6

Comparative Perspectives on Reproductive Health in Ladakh

146

Toward Relevant Research: Adaptation and Policy Perspectives on Maternal-Infant Health in Ladakh

178

7

Notes References Index

1

205 209 227

ix

Figures, Photographs, and Tables

Figures 1.1 1.2 2.1 2.2 3.1 3.2 4.1 5.1 5.2 5.3 5.4 5.5 5.6 5.7 6.1 6.2 7.1

Biocultural adaptive dynamics The proximate determinants of child mortality The relationship between birthweight and altitude Birthweight, gestational age, and neonatal mortality risk Ladakh in the context of South Asia Geography of Ladakh Birthweight distribution Chen’s model of the proximate determinants of infant mortality Birthweight and mortality in the ﬁrst year of life Patterns of infant and child hospital deaths Causes of pediatric hospital deaths Monthly variation in hospital stillbirths, neonatal deaths, and births Pediatric hospital admissions by age and sex Seasonal variation in admissions and diagnosis Birthweight and altitude: A summary of the evidence Birthweight and mortality in Ladakh and the Andes Proximate determinants of infant mortality in Ladakh

page 9 20 28 33 40 43 84 107 113 119 120 121 122 124 147 153 180

Photographs 1 2 3 4

x

The wide ﬂood plain of the Indus River, where much of the population of Ladakh lives and cultivates crops Leh, the district capital of eastern Ladakh Plowing with the help of dzos in June outside of Leh Grinding barley into tsampa, using a water-driven mill

42 44 48 54

Figures, Photographs, and Tables

5 6 7

The Sonam Narboo Memorial Hospital, the only public hospital in Leh District and the site of this study Mothers carry their bundled-up infants and toddlers in baskets on their backs Ache Tsewang with the produce from her garden for sale along the gutters in the main Leh bazaar

xi

59 130 188

All photographs by Andrea S. Wiley Tables 4.1 Sample characteristics relative to census data for Leh District 4.2 Maternal and neonatal characteristics of the sample 4.3 Ladakhi women and infants compared with other general populations 4.4 Regression equations for predicting birthweight 4.5 Dietary practices and activity during pregnancy 4.6 Comparison of maternal and neonatal parameters among Ladakhis, Tibetans, and migrants from low altitudes in Ladakh 5.1 Coefﬁcients and signiﬁcance of anthropometric indices of newborns for predicting early mortality 5.2 Neonatal and infant mortality rates from reproductive histories, 1990 and 1995 5.3 Age and sex patterns in hospital admission diagnoses 6.1 Comparison of neonatal and maternal characteristics in Ladakh and the Andes 6.2 Neonatal, infant, and child mortality in the highland Andes and Ladakh 6.3 Comparison of neonatal and maternal characteristics among high-altitude Himalayan populations 6.4 Infant mortality rates from the Himalaya 6.5 Infant mortality statistics from North India 7.1 International infant mortality rates (IMRs) and per capita wealth

78 82 85 87 89

97 112 117 123 148 152 158 162 172 196

Acknowledgments

The research on which this book is based and the actual writing of the manuscript collectively span more than a decade. Inevitably, there have been changes in my perspective and substantial contributions to the relevant literature that have forced me to abandon previously held assertions and working hypotheses. At the same time, there have been some enduring themes and motivations for remaining engaged in this project. One result is that the list of those who have inﬂuenced and facilitated this work has grown ever longer. It is a pleasure to acknowledge the intellectual, ﬁnancial, and personal inspirations of this work. I was initially drawn to anthropology because of its commitment to holism. Caught between interests in the social and biological sciences, but not sufﬁciently compelled by any of the existing disciplines I explored in college, I chose an interdisciplinary major that required an introductory course in biological anthropology. I was hooked in short order; this was exactly what I was after – a venue for exploring humans as both social and biological without having to sacriﬁce one for the other. Ultimately I chose to pursue what was then the “emerging” ﬁeld of medical anthropology, as it held the promise of an approach that was intrinsically biocultural – how could one not think of health as both biologically based and socially situated? Though anthropology’s uniqueness is holism, in reality the ﬁeld’s commitment to this foundation waxes and wanes, and even its most ardent supporters realize that while easy enough to hold up as an ideal, it is difﬁcult to put into practice. The subdisciplines require specialization in theory, research methods, and data, and within the context of ﬁeldwork not all relevant questions can be asked or answered. Further, although the term “holism” implies consideration of the “whole,” it also refers to the interdependence and relationships among different facets of culture and biology. Thus, although some of the research questions and methods described in this work are quite focused (measuring fatfolds on newborns certainly seems to represent a pretty narrow slice of reality), I have always xiii

xiv

Acknowledgments

tried to frame the speciﬁc by linking it to the larger socio- and natural ecological context. This book is intended as a celebration of my initial discovery of anthropological holism. I consider it to be a new genre, what I term “ethnographic human biology” or “biocultural ethnography.” While some human biologists and probably more ethnographers are likely to take issue with these labels and each may ﬁnd gaps in the biological or ethnographic details, I want to emphasize that both have contexts that are often not fully appreciated. Ethnography, especially one that deals with health issues, might proﬁt from considering the details and evolutionary history of human biological makeup. Likewise, human biology can never be fully understood without considering the myriad social and cultural forces that ultimately impinge on biological function. To some, these statements may seem tired and worn at this point, but I hope that this book represents an effort to put them into practice, to reafﬁrm that holism is neither empty rhetoric nor an impossible ideal. My commitment to biocultural work has endured, but the nature of my relationship to the topic of maternal and infant health and the data presented in this book has changed a great deal as a result of my own experiences with reproductive health. When the research was done I had no children of my own; while writing the book, I gave birth to two sons, now aged two and ﬁve years. What had been a population-based study, focused on average birthweight and the rates and probabilities of infant mortality in a remote mountainous area of India (Ladakh), became a more emotionally laden topic that was difﬁcult to engage in sometimes while writing. I can no longer think about birth, nursing, infant health, and infant death with the same kind of purely intellectual interest coupled with emotional detachment that I was able to maintain throughout the research and early analysis. Although the data are primarily quantitative, I found myself forced to consider each hard labor, each tiny infant, and each early death a unique and deeply painful experience for these mothers, their babies, and their families. As I note in the text, emotional responses to infants and infant deaths are quite different in Ladakh than in the American cultural context, but I cannot help feeling anxious and depressed when I think about the fragility of Ladakhi newborns and the precariousness of their early lives as well as their mothers’ anxiety over their survival. This changed emotional responsiveness is undoubtedly recognizable by other parents, especially those with small children, but it also relates to a more speciﬁc experience with my second son, who was hospitalized at aged three months for a prolonged period with RSV, respiratory syncitial virus. RSV, which I discuss in more detail in Chapter 6, is a particularly

Acknowledgments

xv

insidious, and life-threatening viral infection for infants, as their bronchioles are small. RSV damages these tissues and causes inﬂammation sufﬁcient to dramatically impede their oxygen uptake. My son had either a particularly severe form or an underlying propensity for inﬂammation of the airways, probably both, as he was rehospitalized a month later with non-RSV bronchiolitis, and again with RSV when he was just over one year old. The symptoms associated with RSV include respiratory distress (rapid breathing, gasping for air, wheezing) and may require exogenous oxygen for the infant to maintain adequate oxygen saturation. While the experience of watching my child struggle with breathing has been difﬁcult enough, when I think of this in the context of Ladakh, it is even more horrifying, as there it occurs against the backdrop of an environment deﬁcient in oxygen (hypoxia) and in a context in which health care resources are not adequate to provide infants with supplemental oxygen while they ﬁght off the disease. Even a minor lower respiratory tract infection in this context can precipitate respiratory distress; in the hospital in Ladakh where I worked, respiratory distress and bronchial pneumonia were the most common diagnoses for infant admission to the hospital as well as the most common causes of infant mortality. In light of my work in Ladakh I am ever more profoundly grateful to have had ready access to the health care resources that kept my son alive. At the same time, I am nothing short of outraged by the comparison. My son needed supplemental oxygen, some generic inhaled medicines, and basic monitoring while he slowly recovered; these my local community hospital easily provided and insurance, thankfully, covered. And yet this is well beyond what any Ladakhi baby could hope for in the best of the local public biomedical facilities. As I think back to the young infants who died in the hospital in Ladakh, I now see my son among them. Sadly, although the chronicity of his condition remains to be seen, his pulmonary problems preclude a return to Ladakh with my family for the foreseeable future. On a lighter note, my own experience has also been quite at odds with that of Ladakhi mothers and infants. My ﬁrst son weighed in at nine pounds, ten ounces, a weight unheard of among Ladakhi newborns, none of whom weighed more than eight pounds, ﬁve ounces in my study of 168 newborns! At over the 90th percentile, he was large by American standards, but monstrous by Ladakhi standards, which I had come to adopt by the end of the research. It further indicates that anthropology requires not only the appreciation of cultural relativism, but biological relativism as well. So, ﬁrst I must acknowledge my great fortune in having these wondrous children who have provided me with profound insights into reproduction and its risks and rewards, which have enriched this work

xvi

Acknowledgments

tremendously. Likewise my husband, Richard Lippke, has steadfastly supported this project in part by helping me balance parenting, scholarship, and subsistence. My sister, Marcia Wiley, visited Ladakh three times while I was there, and her own research on water quality issues in Leh enhanced the analysis of child morbidity patterns. Her perennial enthusiasm and adventurous zeal kept my spirits up when they began to ﬂag. With her I ventured well beyond the Indus valley where I was able to learn more about women and children who lived far from the hospital and experience some of the dangers and beauty of foot travel among villages. Although my parents never made it to Ladakh, their love of the outdoors and quest for challenging experiences introduced me to mountains at a young age, and they never questioned my decision to become an anthropologist or to strike out on my own to the farthest reaches of the Himalaya. And although our paths diverged many years ago, I must thank Brian Holmes for introducing me to South Asia and its many pleasures. E. A. Hammel, in the department of demography at the University of California, Berkeley, was an attentive advisor and mentor and provided me with a congenial environment in which to complete my dissertation. The other readers of my dissertation – Fred Dunn, Janet King, and James Anderson – are also to be credited with helping me reﬁne my ideas and analysis. I am indebted to a long line of scholars in human adaptability studies and biocultural approaches to human biology; of them I would like to speciﬁcally thank Mike Little, whom I was fortunate to work with at Binghamton University (SUNY), and also Brooke Thomas and Rebecca Huss-Ashmore, with whom I have not worked, but whose work I have always admired and which has profoundly inﬂuenced mine. Sol Katz was my original mentor at the University of Pennsylvania. His creative spark, particularly his ideas about food and biocultural evolution, continue to inspire my own research. Colleagues Ivy Pike, Nancy Winterbauer, Leslie Carlin, Alex Brewis, Catherine Panter-Brick, Debbie Blackwell, and many others have encouraged me and engaged in insightful discussion about some of the issues raised in this book. I have been inspired by the work of Helena Norberg-Hodge and others at LEDeG and by their dedication to the well-being of Ladakhis and a sustainable future for Ladakh. They have encouraged me to think beyond the narrow scope of this research project to the social and biological health of Ladakhis and their unique environment. Alan Harwood and three anonymous and conscientious reviewers provided an incredibly constructive set of comments and suggestions, which have improved the book greatly. Lapses, inaccuracies, or other shortcomings, while I hope they are minimal, are entirely attributable to me.

Acknowledgments

xvii

I was fortunate to have received funding from a variety of sources in support of this work. The original research was supported by a Population Council dissertation research grant, a National Science Foundation (NSF) dissertation improvement grant, and a small Lowie grant from the department of anthropology at the University of California, Berkeley. The follow-up research in 1996 and the initial writing of the book were sponsored by a Wenner-Gren Hunt Postdoctoral Fellowship, and I also received a Faculty Summer Research Grant from James Madison University to continue the writing of the book. I was granted course release time in spring 2000. In Ladakh, I am grateful to the Lden family, especially both Aches Tsewang Drolma and their children, for their hospitality, care, and support. They were truly a second family to me, enduring my faltering spoken Ladakhi and poor gardening skills. I feel lucky to have been included in their lives. My sincerest thanks and appreciation go to Dr. Tsering Ladhol, who allowed me to undertake this project in her department and taught me the basics of obstetrics. I have a profound admiration of and respect for her humility and dedication to the well-being of women and children in Ladakh. Without her, this project would not only have been impossible, but her interest in it continued to stimulate my research efforts. I would also like to acknowledge the help of Dr. Stanzin Thundup and other doctors at the Sonam Narboo Memorial Hospital for sharing their knowledge and materials with me. Likewise, this project would not have gone so well without enthusiastic assistance of the labor room nurses and the students in training there who went out of their way to help. And ﬁnally, most humble thanks and boundless appreciation to the women of Ladakh who participated in this study. They patiently endured my questions during a particularly intimate, draining, and critical moment in their lives. And of course to their fragile newborns, who endured my seemingly endless measurements, some with placidity, some with fury. I mourn those who died and celebrate those who survived and are now teenagers! The depth and complexity of their experiences are so much more profound than is portrayed in this account; I only hope that I have not done them any injustice. As I write these ﬁnal words, conﬂict between India and Pakistan over the status of Kashmir continues, with both countries amassing troops along their borders, which extend along the northern and western ends of Ladakh. The scale of this threat is potentially quite profound because both nations have nuclear capability. Ladakh is not part of the disputed area, so its largely Buddhist population is not at the heart of the conﬂict, but, nonetheless, the large military presence there means that it is involved in these hostilities. Fortunately, massive offensives have been

xviii

Acknowledgments

avoided to date, but ﬁghting between anti-government forces and the Indian military within Kashmir contributes to internal dangers, as well as to political and economic instability. Many countries have warned their citizens not to travel to this region, and tourism, which had become one of the mainstays of the Ladakhi economy, is plummeting. Predicated in part on the outcome of these ongoing conﬂicts, the future of Ladakh is unpredictable and may very well be quite different from the situation described in this book.

Abbreviations

ANM

Auxiliary nurse midwife – a biomedically trained nurse assistant

EGA

Estimated gestational age – the age of the newborn, in this study estimated from the mother’s fundal height at admission for labor; also measured from date of last menstrual period

IMR

Infant mortality rate (birth–1 year) – calculated from the number of infant deaths, usually expressed per 1000 births or infants at risk

LBW

Low birthweight – birthweights <2500 grams

NMR

Neonatal mortality rate (birth–28 days) – calculated from the number of deaths of infants less than one month old, usually expressed per 1000 births or infants at risk

PEM

Protein-energy malnutrition

PI

Ponderal Index (weight (g)∗100/length3 )

PIH

Pregnancy-induced hypertension – blood pressure readings above 140/90 during pregnancy

PNMR

Postneonatal mortality rate (1–12 months) – calculated from the number of deaths to infants one month to twelve months of age, usually expressed per 1000 births or infants at risk

RSV

Respiratory syncitial virus – a lower respiratory tract infection of the bronchioles that is extremely dangerous for young, especially premature, infants insofar as it reduces oxygen uptake

SGA

Small for gestational age – usually <2500 grams, or <10th percentile in weight

TBA

Traditional birth attendant xix

Glossary of Ladakhi Words

When possible, spellings of Ladakhi words were taken from NorbergHodge and Palden (1991) or follow conventions in wide usage for Ladakhi terms. amchi

practitioner of Tibetan medicine

ane

respected female Buddhist nun

balang

Ladakhi cow

chang

local beer brewed from barley

chodkhang

household temple

chomo

Buddhist nun

dzo

yak-cow hybrid (dzomo if female)

gonpa

Buddhist monastery

gos

a heavy woolen or velvet coat worn over underclothes by Ladakhi women

gur-gur cha

Ladakhi tea, with salt and butter

khambir

homemade ﬂat bread made of barley or wheat ﬂour

khang-chen

“big house”; main house within an extended household compound

khang-chung

“small house”; one of the smaller houses within a household compound

kolak

a common preparation of tsampa mixed with gur-gur cha

lama

Tibetan Buddhist monk

lhaba

Ladakhi shaman or oracle (lhamo if female) xxi

xxii

Glossary of Ladakhi Words

lhu

underground spirit, especially related to water sources

marzan

barley ﬂour mixed with butter and sometimes sugar

mik-ya

evil eye

nyingka

heart; “heart-stomach”

oma

milk

onpo

astrologer

phaspun

household mutual-aid network

rdun

seven; also a celebration of the birth of a child (held ∼1–2 months after birth)

tsampa

ground roasted barley ﬂour

tukpa

a thick soup made of vegetables and or meat and barley or wheat noodles

yak

large Himalayan bovine species (dimo if female)

zhing

ﬁeld

An Ecology of High-Altitude Infancy A Biocultural Perspective

1

Introduction

Reproduction is a fundamental problem for all organisms. Animal species exhibit a dazzling array of strategies to produce offspring, which often require large inputs of energy and are associated with major risks to their survival. Among mammals, females incur more direct costs of reproduction insofar as they carry (literally) the burden of embryonic and fetal growth and nourishment of the young through lactation. Thus the health of a female mammal directly affects reproductive outcome, both in terms of her fertility and the survival or death of her offspring. Human females experience these costs as a function of their mammalian heritage, but the variety of environments they inhabit generates substantial differences in the reproductive risks that women face. Their abilities to reduce these risks are important determinants of individual and population differences in maternal and child health and survival. Given the centrality of reproduction to the lives of organisms, including humans, it is useful to know what constitute the optimal conditions for reproduction. Certainly, adequate nutrition, absence of infection, monitoring, and judicious intervention are among the factors that enhance pregnancy outcome. In biomedicine, complex technologies are routinely used to increase the odds of conception and a healthy newborn. In-vitro fertilization, fetal genetic screening, and cesarean sections are now routine procedures, and ever more complicated procedures such as fetal surgery are on the horizon. But these are primarily, though not exclusively, practices found in wealthy countries, where today most women have relatively few births, later in life, and “invest” each of them with intense meaning and resources. Conditions for reproduction for populations in much of the world are at stark variance from those in wealthy countries; many women experience dramatic constraints on their ability to create or gain access to the conditions necessary for optimal fetal development and infant/child survival. Impoverished living conditions, social inequalities, heavy physical workloads, infectious disease, and nutritional deﬁciencies all compromise a process that is inherently fraught with demands on a woman’s resources. 1

2

An Ecology of High-Altitude Infancy

Anthropologists are interested in reproduction insofar as it represents a universal aspect of the human experience and also because the myriad ways in which different cultural and ecological contexts shape the reproductive process contribute to cross-cultural variation. All societies attempt to control reproduction and its outcome in some way, be it simply through the universal institution of marriage, which restricts access to and timing of what constitutes locally appropriate reproduction; complex prescriptions or proscriptions (prohibitions) for behavior (especially diet) during pregnancy; or child-rearing practices that enhance the well-being of some offspring at the expense of others. These social mandates occur within the parameters of ecological contexts and basic human reproductive biology, which shape the outer limits of reproductive possibilities. Collectively these factors contribute to a tremendous range of reproductive experiences in human populations. Salient aspects of this variation in the reproductive process include women’s health, both before and during pregnancy; the health or survival of their offspring; and the personal and cultural meanings of reproduction. These factors ultimately have longterm effects on a population, by shaping its age distribution and growth rate, which in turn impact the environment in which the population lives. The case to be considered here is reproduction under a unique constellation of natural ecological and sociocultural conditions (what can be termed “socioecology”) that constitute a set of challenges to successful, healthy reproduction. The geographic locale is a little-known region called Ladakh, which in the past constituted an independent kingdom in the remote Himalaya, and which is now a district in the hotly contested state of Jammu-Kashmir in India. It forms the western end of the transHimalayan (Tibetan) plateau on the northern side of the great Himalaya range. Ladakh is a high-altitude desert area where the average altitude of human habitation is over 3000 meters. It is in the rain shadow of the Himalaya, which effectively blocks the summer monsoon rain clouds that amass to the south, resulting in average annual rainfall of only 10 to 15 centimeters (Mani 1981). In the winter, bitterly cold winds blow across the steppes of northern and central Asia into the region. Ecologically and culturally, Ladakh has many afﬁnities with the better-known regions of Tibet and Tibetan areas of Nepal – similar languages, Buddhist religious practices and beliefs, marriage and family structures, diet and subsistence agricultural practices, among others, all link peoples in this broad Himalayan band. This blend of culture and ecology impacts the reproductive process in this region, but it does so differentially; individuals and populations living here vary in meaningful ways that may lead to dramatically different reproductive outcomes.

Introduction

3

Based on studies in other high-altitude areas of the world, especially the Andes of South America, there is reason to believe that the ecological stresses of high-altitude life impinge on human biology and reproduction. High-altitude areas (deﬁned as >2500 m) are characterized by numerous stresses, including cold, ultraviolet (UV) radiation, harsh terrain, and probably most important, hypoxia, the decreased partial pressure of oxygen that occurs at higher elevations (Pawson and Jest 1978). With increasing altitude, oxygen density decreases steadily so that at 3000 meters (10,000 ft), the partial pressure of oxygen is less than 70% of sea-level pressure. Above 5000 meters, oxygen pressure is sufﬁciently low that permanent human settlements cannot be maintained. Oxygen plays a key role in physiological functioning across animals, including cellular respiration and growth, so it comes as no surprise that humans should be profoundly affected by a high-altitude context. Shortness of breath, rapid heart rate, and in worst-case scenarios, headache, nausea, ﬂuid buildup in the lungs or cerebrum, or even death are among the responses an individual may have to hypoxia (Frisancho 1993). One of the interesting features of the hypoxic conditions of high altitude is that unlike other ecological features such as temperature that are easily buffered by material innovations (e.g., clothes, shelters), the experience of hypoxia cannot be modiﬁed by technologies on a population level. Certainly there are now technological solutions, such as the use of oxygen tanks by high-altitude climbers, but these never evolved in an indigenous high-altitude setting, and are, for all practical purposes, unsustainable in a large population. At the same time, however, an individual’s biological resources may alter the experience of hypoxia, and these resources are shaped by culture-speciﬁc behaviors and social organizations that inﬂuence nutritional status or a genetic inheritance that allows for more efﬁcient oxygen utilization, among others (Leonard 1989; Moore et al. 1998). With regard to the effects of hypoxia on reproduction, women traveling to high-altitude areas have anecdotally reported disruptions in menstrual activity. During the colonial period, the Spanish noted a very high rate of reproductive failure among their populations trying to establish a permanent presence in the city of Potosi, at 4000 meters in the Andes. No Spanish child was born who survived childhood until ﬁfty-three years after the city was founded; most died at birth or within a short time after birth (de la Calancha 1639, quoted in Clegg 1978). As a result, Spanish women often descended to lower altitudes for pregnancy and birth and remained there until the child was at least one year old. At the same time, the Spanish noted that the indigenous population did not seem to experience these problems. A similar situation has been described in

4

An Ecology of High-Altitude Infancy

contemporary Tibet, which has been colonized by increasing numbers of Han Chinese from low altitudes over the past few decades (Moore et al. 2001a). So, residence at high altitude is widely associated with signiﬁcant reproductive problems among migrants to high altitude. Migrant women recognize these risks and often descend to lower altitudes for pregnancy and for some time after birth. A proximate mechanism that mediates the relationship between hypoxia and birth outcome has been identiﬁed. Numerous studies from New World mountainous areas have demonstrated that the average birthweight of newborns decreases as altitude increases, indicating a possible link between oxygen shortage and reduced delivery of essential nutrients to the fetus (Beall 1976; Grahn and Kratchman 1963; Haas 1980; Haas et al. 1977; Haas et al. 1980; Lichty et al. 1957; McClung 1969; Unger et al. 1988; Yip 1987). Oxygen deprivation in utero could result in impaired fetal growth and compromised birth outcomes, which, in turn, may contribute to high early mortality, although some authors have suggested that they do not (Beall 1981; Wilcox 1993). The negative effect of hypoxia on fetal growth is particularly severe among newcomers to high altitude but is evident to a lesser extent among indigenous populations. Thus, a population’s history in a high-altitude environment appears to affect their reproductive success and the relative severity of its impact, suggesting evidence of adaptive processes at work among longer-term inhabitants (Haas 1980; Moore et al. 1998). Furthermore, diversity in cultural behavior, social organization, and other local ecological conditions is likely to contribute to variation in the effects of mountain life on reproduction. Reproduction in Ladakh occurs in an ecological matrix characterized by hypoxia and dramatic seasonal differences in temperature that constrain subsistence. Most Ladakhis practice subsistence agriculture based on various strains of barley, wheat, legumes, potatoes, and a variety of other root vegetables, and many families keep some animals for dairy products and help in plowing and threshing. Agricultural production is labor- and seasonally intensive as there is no mechanization, and both men and women work hard at agricultural tasks, especially between April and October (Osmaston 1994). Women in Ladakh experience substantial energy burdens as they contribute to agricultural production for their households, and these energetic demands are exacerbated by the hypoxic conditions under which they must work. Aerobic work in a hypoxic environment is exceptionally arduous, even for people who are well acclimatized, and it places high demands on energy-producing, oxygendependent metabolic pathways in a context where oxygen is not abundant (Kashiwazaki et al. 1995; Leonard et al. 1995). A Ladakhi woman’s

Introduction

5

reproductive status often has little effect on her work patterns; most pregnant women labor throughout their pregnancies. Thus their energetic resources and oxygen transport systems experience double and triple demands during this stage. On the other hand, there are widely articulated cultural ideologies suggesting that pregnant women should not work hard, and indeed should be particularly well provided with extra highly valued foods such as meats and dairy products. The extent to which women may lay claim to these beneﬁts depends on the availability of resources and the level of demand for their labor within their household. Both are inﬂuenced by season and household dynamics, among other factors. Ladakhi women, particularly Buddhist women, enjoy relatively high social status compared to women in other parts of Hindu or Muslim South Asia, and women’s status is generally thought to be correlated with positive health indices (Thaddeus and Maine 1994). Women freely engage in interactions with unrelated men and are active vendors in the vegetable bazaar in the main town. They engage in valued productive work and control the distribution of food in the household and the proﬁts and produce from large household gardens. Moreover, Ladakh is one of the few places where polyandry was traditionally practiced, and some authors have idealized this marital form as an institution that fosters the empowerment of Ladakhi women (Norberg-Hodge 1991; Rizvi 1983). There are thus reasons to believe that aspects of Ladakhi social and cultural life highly value the productive and reproductive capacities of women, and that this valuation might translate into beneﬁts to reproductive health and well-being. However, it is difﬁcult to get an accurate sense of health parameters in Ladakh because of the lack of systematic data reporting. Nutritional research done suggests that Ladakhis subsist on an adequate base of calories and protein (Attenborough et al. 1994; Osmaston 1994; Palriwala 1988), while there are scattered references to some vitamin and mineral deﬁciencies (Attenborough et al. 1994; Meyer 1981; Stobdan 1990). There are two village studies indicating that infant and child mortality rates are high (Attenborough 1994; Elford 1994). Maternal health and health differentials between males and females have not been examined, so existing sources do not tell us much about the reproductive health situation. It is important to note, however, that high rates of infant and child mortality represent a loss of maternal biological resources and a need for higher fertility, which requires further resources, already strained by workload demands. The research presented in this book documents that reproductive health is compromised among a large sample (n = 168) of mothers and infants in eastern Ladakh. Most babies were very small at birth; the

6

An Ecology of High-Altitude Infancy

median (and mean) birthweight (2764 g) was only slightly over the World Health Organization cutoff for infants at increased risk of mortality (<2500 g). Indeed, these small infants in the sample had very poor survival prospects. But the pattern of mortality was quite distinctive; while infant mortality rates were on the order of 20%, deaths in the ﬁrst month of life accounted for over 75% of deaths in the ﬁrst year of life. After the ﬁrst critical month, mortality rates declined to relatively low levels. Birthweight was a very strong predictor of mortality in the ﬁrst month of life, and this implies that the determinants of birthweight are at the roots of high mortality. Since birthweight is largely a function of maternal characteristics, constraints on women’s health status are profoundly implicated in the analysis of these very high rates of early infant death. I suggest that this dismal reproductive health situation results from the competing demands of production and reproduction, a situation not unique to Ladakh, but a problem common to subsistence societies throughout the world. Each woman experiences this conﬂict somewhat differently as a function of different natural and socioecologies. In the Ladakhi context, women’s engagement in subsistence work interacts with the demands placed on them by reproduction and the metabolic constraints of hypoxia. A focus on reproductive health can give us important insights into numerous aspects of Ladakhi social and cultural life and ecology as well as into the lived experience of women and children. Women’s health is closely tied to that of their offspring, especially early in life, and thus has implications for infant survival. Infant mortality rates, in combination with fertility rates, determine the age structure and growth rate of the Ladakhi population; demographic processes are crucial aspects of Ladakhi subsistence under what are generally quite marginal and stressful ecological conditions. Expectations about the likelihood of success or failure of the reproductive process condition emotional responses to pregnancy, birth, fetal loss, or infant death. Child-care patterns are affected by maternal workloads and also by expectations about the roles and status of women and children in Ladakhi society. Common reproductive health issues also make their way into local healing traditions. Ladakh is home to numerous medical traditions, ranging from shamanism to Tibetan medicine, to the more recent importation of allopathy (Western or biomedicine). In sum, while this study focuses on a very narrow window of time in a woman’s reproductive life and in a child’s life, it provides insights into multiple aspects of life in contemporary Ladakh. The reproductive process is, somewhat obviously, the foundation of a population’s existence.

Introduction

7

A Biocultural Perspective on Health An ideal approach to reproductive health in Ladakh is a biocultural theoretical perspective, a uniquely anthropological form of inquiry that considers a full range of biological and sociocultural factors to be related to health outcomes. Health status is usually described by biological indices of individuals (e.g., growth, immune responsiveness, blood pressure, hormone status), and the trends in these indices in a population (rates of diseases, average growth proﬁles, etc.). These biological data are then considered as outcomes of interactions between individuals and their environments. In other words, the body is in a dynamic relationship with the environment and reﬂects the stresses and resources in that environment, both in the past (over the course of the lifetime of an individual, or over generations) and in the present. Such a perspective draws heavily on evolutionary theory, which views biological and behavioral outcomes in relation to these environmental characteristics and emphasizes the ways that organisms respond to them to enhance their survival and reproduction. Such traits (be they biological or behavioral) are deﬁned as “adaptive.” Biocultural perspectives have their origins in the human adaptability studies by biological anthropologists in the 1960s and 1970s that were designed to broaden understanding of human biological variation under a variety of ecological conditions. This body of research uncovered both genetic adaptations to different environmental contexts and evidence for adaptability (also known as plasticity), the physiological alterations generated by exposure to particular environmental stresses over the course of an individual’s life cycle (Harrison 1997). Since an individual’s ability to respond adequately to a stressor impacts his or her biological function, it was inevitable that many adaptability studies began to consider the health implications of human-environment interactions, and that the concept of adaptability became a core concept in the emerging ﬁeld of medical anthropology in the 1970s. Anthropologists using the concept of adaptability have increasingly realized that biological responses to environmental stresses are mediated in signiﬁcant ways by behavior, which for humans is often culturally patterned and socially situated. Cultural traditions, technologies, and social conditions can both exacerbate or ameliorate natural ecological stresses. Hence the emergence of a biocultural paradigm within human adaptability studies and medical anthropology. Although traditionally the biocultural approach has been linked to evolutionary analyses emphasizing functional adaptation to the natural environment, more recently some have argued that a change of emphasis

8

An Ecology of High-Altitude Infancy

is required to understand how the body is affected by contemporary social environments, especially those characterized by poverty (Goodman and Leatherman 1998b). The important environmental stresses facing many human populations are seen to be of largely social – as opposed to natural – origin, and derive from the intersection of global and local political economies that more often than not produce conditions that severely threaten health. Human adaptive capacities may be inadequate to respond to the multiple stresses deriving from poverty, and hence the focus shifts away from adaptation, to identifying and describing the stresses in the environment, the ways that the body responds to such stresses, and the short- and long-term consequences of these responses for both the body and the social group (Goodman and Leatherman 1998b; Thomas 1998). This shift in focus is an important complement to analyses of adaptive outcomes and provides additional power to the biocultural paradigm. It is crucial to identify the sources of stressors in an environment, their historical roots, and the means by which they are sustained. At the same time, the links between stressors and biological outcomes need to be speciﬁed. In some contexts, adaptive outcomes are evident; in others, stressors may be so severe or numerous that the body’s adaptive capacity is overwhelmed, with negative health consequences. This synthetic approach should provide insight into the complex determinants of a population’s health status, and to do so demands investigation into history, ecological, social and cultural dynamics, and biology. Figure 1.1 provides an abstract of this biocultural model, demonstrating the ways that inequalities and other constraints on access to resources require biological and behavioral responses that ultimately impact human biology and health. Four core concepts of the biocultural enterprise frame this study of reproductive health in Ladakh: the population/the body, environment, health, and adaptation. They are described below.1 The Population and the Body Anthropologists are generally concerned with humans as members of speciﬁc populations. Individuals arrange themselves in myriad ways for different purposes, and the researcher’s interests and local contingencies usually determine the group to be studied. A group is most often bounded by geographical space, and its members share a loosely deﬁned “culture.” It may be a household, clan, community, or nation-state. It can also be considered a gene pool if most reproduction occurs within the group. Thus a population can be both biologically described (by common

Introduction

9

Environmental Stressors

Biological Impact

Psychological Stressors

Behavioral Adjustment

Biological Adjustment

Essential Resources Inequalities - Gender - Class - Race

Factors Affecting Access to Essential Resources

Figure 1.1. Biocultural adaptive dynamics (Redrawn with permission from Thomas, R. 1998. The evolution of human adaptability paradigms: Toward a biology of poverty. In Goodman, Alan H., and Leatherman, Thomas L. (eds.), Building a New Biocultural Synthesis: Political-Economic Perspectives on Human Biology. Ann Arbor: University of Michigan Press.).

descent and relatedness) as well as socially deﬁned. Populations can be large or small, although larger units tend to exhibit more heterogeneity in biology and behavior than do small groups. In general, we are interested in the frequency of physiological or behavioral characteristics of individuals as a way of describing a group but also to understand and account for variation within a group. Biocultural analyses are concerned with biological outcomes that are measurable at the level of the individual – that is, the body. The body is best thought of as a socially situated and culturally inﬂuenced biological entity that responds to forces in the environment that challenge its ability to function. Sociocultural phenomena can affect the body by shaping the behavioral decisions that individuals make and by determining the resources available to support the body’s needs. In turn, the body affects sociocultural life most importantly by having needs that must be met

10

An Ecology of High-Altitude Infancy

for survival. These requirements pose limits to the kinds of sociocultural systems that can be maintained. The body has a history that affects its ability to function in a given environment; that is, it is an evolved entity. A body is not a random accumulation of cells and organs nor a perfectly designed machine; it is a form that has been shaped by a history of life within a particular lineage in a particular environment (Nesse and Williams 1994). That history was shaped by the forces of evolution that include, most importantly, natural selection, but also mutation, genetic drift, and gene ﬂow. Natural selection, the contribution of Charles Darwin and Alfred Wallace to evolutionary theory in the nineteenth century, refers to the idea that the environment in which organisms live poses certain threats to survival, and individuals who are able to avoid these threats or overcome them survive and reproduce at a greater rate, and thus pass on their traits (in the form of genes) at a greater rate than do other individuals. Over time, these characteristics are found in more and more individuals in the population. These traits are referred to as adaptations, or adaptive traits, because they enhance reproductive success, which is measured as the product of survival through the reproductive years and the number of offspring produced. Since health is closely related to survival and reproductive outcomes, it is appropriate to investigate the health of an individual as part of evolutionary inquiry. Bodies are the product of their evolutionary history, but they are also the product of their life history in a given environment. As noted before, human biology is plastic; it is capable of changing in response to conditions encountered over the course of the life span. The environment in which growth and development occurs is particularly important, as it can leave indelible marks on an individual’s biology. Stresses encountered while an individual is growing up may manifest as reduced growth or compromised physiological or psychological function, and they can affect the way an adult responds to current stresses. The other forces of evolution shape the body as well and have health implications. Mutation simply refers to random changes (“mistakes”) made in the formation of egg and sperm cells. It is the source of variation on which natural selection can act, since natural selection can only “select” from an existing pool of individuals with variable traits, and thus without variation, natural selection cannot occur. Mutations are also responsible for many inherited genetic diseases (e.g., Tay-Sachs disease, sickle-cell anemia, cystic ﬁbrosis), although these have likely reached relatively high frequencies in some populations as a function of natural selection (cf. Desowitz 1981; Diamond 1991; Gabriel et al. 1994). Some environmental conditions (e.g., UV radiation, pollutants of various kinds) can

Introduction

11

stimulate mutanogenesis and increase the frequency of genetic diseases. Gene ﬂow, also known as migration, can bring new characteristics into a population or remove variation from a population. This is particularly crucial at present, as the high rate of global migration over the past two centuries has brought many people to novel places. Individuals who live in environments different from those of their ancestors may ﬁnd themselves with traits that are not well suited to their current environment. Last, genetic drift is an evolutionary process that affects only small populations. It refers to random events having disproportionately large effects on the diversity of traits within a population and to the historical impacts of relatively low genetic diversity. Health Health must of course be a central concept in medical anthropology, but, as with many key concepts, there is a range of deﬁnitions. Here health is deﬁned as an individual’s ability to respond effectively to the challenges of his or her environment, which may include infectious agents, food scarcity, extremes of temperature, psychological or social stress, or any environmental conditions that compromise biological (including reproductive) function (noxious chemicals, mutagens, etc.). Individuals have mechanisms to keep physiological function within homeostatic parameters that can be altered somewhat during stress (allostasis), so that the body can maintain functionality without excessive costs at least for short periods of time. If these challenges become chronic, or if an individual faces several at once, the body’s ability to cope may be seriously compromised, and disease results. Thus, at any given time an individual’s health status lies somewhere between the poles of basic survival and optimal function (Armelagos et al. 1978). This is a very biological deﬁnition of health, and there is certainly a more subjective aspect to health as well. A person may not exhibit any evidence of physiological impairment yet say that he “doesn’t feel well”; or someone may assert that she “feels ﬁne” yet display evidence of physiological malfunction. Medical anthropologists thus make a distinction between the terms “disease” (a biological index) and “illness” (a subjective index of health). The broader concept of “well-being,” which includes both a subjective and biological component, may be more inclusive than a strictly biological deﬁnition of health. When viewed from this perspective, health is seen as a relative state, and those characteristics that describe health will vary by environmental context. There is unlikely to be one deﬁnition of “optimal” health for all humans. Descriptors of “health” or “normal human function” derived from

12

An Ecology of High-Altitude Infancy

Western industrialized populations are not necessarily appropriate standards against which to judge the health status of other populations living in different contexts. For example, very high levels of hemoglobin might be seen as pathological among sea-level populations, but at high altitude this same phenomenon could be adaptive in the context of low levels of oxygen. At the same time, we need to recognize that health categories are not entirely relative and that dismal living conditions preclude the attainment of any kind of “optimal health” for much of the world’s population. This is glaringly evident in the vast disparities in life expectancy2 among countries – with the range between ﬁfty-two in subSaharan African countries to over eighty in northern Europe and Japan. Differences in life expectancy mean that years of life are lost in some populations because of much greater disease burdens, and that these burdens fall disproportionately on the young. The Environment Interactions between the body and the environment are crucial to health, and here we deﬁne the environment to include abiotic (climate, geology, altitude, etc.), biotic (plant and animal life), social (other individuals, groups and communities and their organization), and cultural aspects (ideologies, worldviews, technologies, etc.). The abiotic features set the broad parameters within which all the others exist and interact. Key aspects of the biotic and abiotic environment, from the perspective of human biological function or health, are certain threats or various stresses that affect the body (e.g., cold or hypoxic stress, infectious disease, toxins, predation, treacherous landscapes). At the same time, the environment may provide resources for enhancing health (e.g., materials for shelter or protection, antimicrobial compounds, alternative food sources). Similarly, aspects of the sociocultural environment, or socioecology, can act both as stressors and resources. Often these act by inﬂuencing access to, or use of resources, and access to resources (food, shelter, clean water, health care) is certainly among the most important determinants of health (Huss-Ashmore and Thomas 1997). For example, Wilkinson (1996) has convincingly demonstrated that health is negatively affected by social stratiﬁcation, as countries with more egalitarian social and economic structures have better health indices. Stratiﬁcation increases stress across the social hierarchy and also creates differentials in access to resources (Baer et al. 1986). Social cohesion and democratic participation in decision making tend to be associated with better health. Cultural understandings about how the body works (or how male, female, and children’s bodies differ), religious dietary proscriptions, the causes of disease

Introduction

13

(e.g., microbes versus witchcraft), and social ideologies such as racism and sexism can all inﬂuence disease vulnerabilities and access to adequate treatment. In sum, variation in environmental conditions is key to understanding health differences among and between populations, and the qualities of an environment, especially those that affect access to key resources, are primary determinants of health and well-being. Adaptation The concept of adaptation is fundamental to evolutionary theory and to biocultural studies in anthropology, although there has been considerable discussion of what exactly constitutes evidence for, and the meaning of, adaptation. As noted earlier, a very basic deﬁnition of adaptation is a trait (biological or behavioral) that enhances reproductive success, and the process that increases the frequency of these traits is called natural selection. Populations have evolutionary histories during which they have undergone the continual process of natural selection over generations in their environments. Therefore it is reasonable to expect that their members have traits that are adaptive – that enhance, rather than detract from, their well-being. However, it is important to recognize that populations are not likely to be exclusively characterized by traits that are adaptive in the current environment if it has in fact been subject to numerous modiﬁcations over historical time or if the population has only recently settled there. Under these conditions, we are more likely to observe the adaptive process rather than well-established adaptive outcomes. Hence the study of adaptation is, in part, a historical project. For humans, as a species with tremendous behavioral ﬂexibility and innovative ability, adaptations are often encoded in culturally speciﬁed behaviors that attempt to buffer the effect of environmental stresses on the body. Adaptation is not simply about passively responding to whatever challenges come up; it is proactive as well. One can adapt to as well as adapt something to meet one’s needs. Thomas et al. (1979) outlined seven possible modes of adaptation: avoidance, modiﬁcation, buffering, distribution, resistance, conformity, and change. One can act to avoid a stress, or if it is unavoidable, conform to it (as long as this is possible while maintaining internal homeostasis), or change in response to it (e.g., natural selection or adaptability). One can attempt to modify or redistribute the stressor, or buffer the body from it by using various technologies. Resistance is an attempt at adaptation, in the sense of manipulating the social organization to enhance health or reduce morbidity or mortality. All of these strategies can be assessed by their efﬁcacy (how well do they solve the problem) and efﬁciency (what other costs do they produce). The net

14

An Ecology of High-Altitude Infancy

effects of the efﬁciency and effectiveness of a response should enhance health and well-being. Depending on the characteristics of the stress itself and the resources (material and biological) available to an individual or group, one or more of these adaptive responses may be engaged to deal with a challenge to well-being. Common goals or ideologies (i.e., culture) will inﬂuence how individuals experience the environment and prioritize their responses to stresses. Cultural traditions have a history, rendering them adaptive, neutral, or potentially maladaptive under current environmental conditions. Generally speaking, these traditions provide the parameters within which decisions are made; some traditions are quite entrenched and may preclude adaptive change; others may be more ﬂexible and responsive to new conditions (Huss-Ashmore and Thomas 1988). Culture – like biology – is not static, but if stable environmental conditions prevail for long periods of time, one should expect to see little change in either. Some human populations live under such conditions; many others live in environments that are novel, highly variable, or unpredictable (e.g., rapidly urbanizing environments). As new threats to health emerge or as old threats diminish, different biological and behavioral strategies are called for. It is important to realize, however, that the ability to adapt is not unlimited, but can be overwhelmed by multiple stresses, particularly severe acute stresses, or by exposure to chronic stress. Disease, illness, debility, and ultimately death may result. Behavioral or physiological adaptations are made in response to or to prevent sickness, a state in which the body cannot maintain adequate function. Documentation of this process is complicated by the fact that environmental stresses are not experienced one at a time. Instead, several stresses are often experienced at once and are responded to with available resources. Thus, responses to one stressor can easily result in negative effects with regard to another, just as responses that are adaptive in the short term can become problematic in the long term (Mazess 1975). The outcome of such actions is not likely to be a “perfect ﬁt” between the individual and the environment. Adaptation is compromise; every response or preventive mechanism comes at some cost, and sometimes multiple stresses are acting simultaneously. Given the heterogeneity of resources available, challenges to, and priorities of an individual or group, adaptation in one realm is likely to have repercussions in other domains (it may be effective but inefﬁcient). For example, many physiological responses to infectious disease can be considered adaptive, but they are biologically costly. Fever is a common response to infection and is effective in terms of raising the temperature of the body so as to reduce the growth rate of bacteria, but this requires substantial energetic resources and over time, may be detrimental to body tissues (Nesse and Williams 1994).

Introduction

15

How do we know whether a biological or behavioral change is adaptive? That is, how do we measure adaptation? Indices of health are used as measures of adaptation, since they are linked – either directly or indirectly – to survival and/or reproduction (Little 1989). These are often quite standard biological measures such as hemoglobin status, infectious disease status or parasite load, child growth, blood pressure, reproductive outcomes such as miscarriage, or birthweight. All of these signal something about biological well-being that has signiﬁcance for survival chances and overall reproductive success. Sometimes it is possible to measure survival and reproduction directly by looking at rates of death or birth. Ideally these measures can be linked together in a prospective study – that is, by assessing individual health characteristics and following that individual over longer periods of time to assess the person’s fertility and survival (or age at death). Prospective studies are difﬁcult to do, since they require monitoring of individuals over long periods of time. So, more often than not, health measures are used as proxies for adaptation if there is a demonstrable or theoretical link to survival or reproduction. Biocultural Synthesis: High-Altitude Examples Biocultural research grew out of human adaptability studies, many of which were carried out in high-altitude areas of the Andes. Although early studies emphasized hypoxia as the main stress there, more recent work has focused on the social sources of stress and how they compromise health, including a person’s ability to respond to natural stresses such as hypoxia. Leonard and colleagues (1989) have shown that nutritional stress, which varies by socioeconomic status, alters the extent to which a child’s growth and development will be affected by hypoxia. Other researchers have emphasized the historical roots of social stresses, which have created the current political economic conditions that enhance the well-being of some groups at the expense of the health of others. In Peru, Carey (1990) has shown how individuals in households that have been increasingly marginalized since the colonial period suffer from inadequate food supplies, resulting in malnutrition, infectious diseases, and lost days of work. Households respond to these very tangible threats to their well-being by forming cooperative support systems, which seem to be associated with better health of their members. Some households are so socially and economically impoverished as to be unable to participate in such networks, and their health status is reduced as a consequence. In another part of the same study, Leatherman (1998, 1996) has shown that illnesses (particularly chronic conditions) reduce net household productivity and income and further exacerbate poverty, resulting in a downward spiral of social and biological well-being.

16

An Ecology of High-Altitude Infancy

In an earlier study of iodine deﬁciency and social stratiﬁcation in Ecuador, Greene (1977, 1980) demonstrated that a local ecological stress of low iodine soils (common in mountain environments) resulted in a high frequency of iodine deﬁciency, which manifests as goiter (enlargement of the thyroid gland), or more profoundly, cretinism (severe mental retardation). These very serious problems were most common in lower socioeconomic groups and served to perpetuate their poverty. Greene further showed how dominant groups in Ecuador had been able to take advantage of this situation by employing cretins for onerous and low-paying tasks, while the households of such individuals became increasingly economically and socially marginalized, with ensuing negative health outcomes. Huss-Ashmore and Thomas (1988) emphasized the problem of adapting to increasing unpredictability of resources in highland areas of Lesotho (South Africa) and Peru. Uncertainty derives in part from the seasonality inherent in these environments but also from the inhabitants’ increased incorporation into national and regional cash economies whose stability depends on forces beyond local control. This unpredictability results in changing household dynamics, which shift the burden of production to women, as men more frequently engage in migrant labor. The local human-environmental system has been profoundly changed by these conditions, such that previously evolved adaptive responses no longer work, and the authors have demonstrated negative impacts on child health. These studies each demonstrate the utility of a biocultural perspective. They also point to the necessity of investigating variation in health status within populations, whereas previous human adaptability studies had focused on differences between populations that could be attributed to ecological differences. The study of intrapopulation variation illustrates the role of the intervening social factors that impinge on health from the perspectives of both exacerbated stresses and adaptive responses. This research also points to the household as an important institution that mediates larger ecological and social stresses to individuals. It becomes apparent too that local and global histories are important to a fuller understanding of current environmental conditions, health status, and adaptive responses (cf. Goodman and Leatherman 1998a; Ulijaszek and Huss-Ashmore 1997). A Biocultural Perspective on Reproductive Health in Ladakh This book weaves together several lines of inquiry within a comparative, biocultural perspective on reproductive health in Ladakh. Reproductive

Introduction

17

health is particularly amenable to an evolutionary perspective since this aspect of health affects the two dimensions of ﬁtness – fertility (the production of a live birth) and early mortality. I focus on the health of pregnant women and infants and the biological index that links them: birthweight. Birthweight reﬂects the resources that a mother was able to channel to the fetus and is thus related to her health status, and, as a measure of a newborn’s biological resources, it is a very good predictor of the probability of an infant’s survival within a population (Chen 1983). The central questions that frame this project include these: How does the adaptive process play out as mothers and infants struggle to navigate the very complex and ﬂuid natural ecology and socioecology that prevails in contemporary Ladakh? How can we trace the pathways of inﬂuence that ultimately result in reproductive success or failure – infant survival or death? How do the experiences of Ladakhi mothers and infants compare with those in other situations? The answers to these questions contribute in signiﬁcant ways to a variety of existing anthropological projects in reproductive ecology, high-altitude adaptation, and the ethnography of Ladakh. Historically, the environmental factors affecting conception and fertility (production of a successful birth) have been studied quite separately from those affecting infant survival, and the label “reproductive ecology” has been used to describe this body of work (Ellison 1994). Reproductive ecology has been particularly concerned with how reproductive parameters represent adaptations to conditions such as food availability, seasonality in diet and workloads, and hypoxia, among others (cf. Vitzthum et al. 2000a). However, because successful reproduction is a process that does not end with birth, I propose that we need to include an ecology of infancy to generate a broader picture of the ecology of the reproductive process. Furthermore, by developing the concept of an ecology of infancy, this work responds to the call for an “anthropology of childhood,” a topic that has not been sufﬁciently explored, as recently lamented by PanterBrick (1998) and Scheper-Hughes and Sargent (1998). The focus on mothers and infants in the Himalaya also redresses a notable bias in the long tradition of research on adult men in high-altitude environments, particularly in the Andes. This is not to say that some extremely valuable work has not been done on the reproductive process in this context, but Man in the Andes (Baker and Little 1976) has long been the standard for describing high-altitude adaptation. Hence, by refocusing attention on women and children in the Himalaya we broaden our understanding of how this ecological context affects human biology. A comparative perspective is particularly useful in this regard. While the Ladakhi population faces the ecological stresses common to many

18

An Ecology of High-Altitude Infancy

temperate latitude high-altitude environments, the social conditions and cultural traditions of Ladakh may ameliorate or exacerbate these stresses, resulting in unique health outcomes for Ladakhi mothers and infants. We can relate the Ladakhi experience to that of other high-altitude populations, including those in the Andes, that share the hypoxic but not the sociocultural or historical context, and to other Himalayan populations that have similar ecological and sociocultural experiences, but in the context of different historical processes. Likewise, we can compare Ladakh to the larger South Asian culture area south of the Himalaya; as part of India, Ladakh shares some of that country’s public infrastructure and has been inﬂuenced by Indian social and cultural trends, but ecologically the region is radically different from the plains of North India. Our understanding of Ladakh has been limited by the lack of such a comparative perspective. Heretofore, writing on Ladakh has been fairly particularistic with only local frames of reference. While not taking anything away from the valuable work done so far, there has been a tendency to idealize and emphasize the uniqueness of traditional lifeways and to treat them as evidence of “optimal” organization and function in the context of dramatically limited natural resources, and further, to treat the population as a homogeneous group, undifferentiated in meaningful ways that affect health and well-being (Norberg-Hodge 1991; Osmaston et al. 1994). Among the important social institutions that inﬂuence health and contribute to variation in health in Ladakh is the household. The household mediates or ﬁlters environmental stressors for those whose lives are largely circumscribed by activities in and around it. In many societies, including Ladakh, these are primarily, though certainly not exclusively, women and children. Several researchers have pointed to the household as the key institution in the “production of health” (cf. Berman et al. 1994) and the crucial locus of child health in particular. Harkness and Super (1994) have referred to it as the “developmental niche” of a child. In Ladakh, a child’s experience of hypoxic stress can be mediated by the household environment, especially aspects related to heating and ventilation. Domestic pollution can compromise oxygen uptake by increasing levels of carbon monoxide or airborne particulate matter. Respiratory pathogens circulate more densely in small, crowded households and further compromise oxygen uptake. Households with abundant resources may have larger and more adequately heated and ventilated houses; they may also have access to more food, and calories are an important resource for adequate response to hypoxia. Within the household, food is distributed to members in particular ways; not everyone necessarily has access to

Introduction

19

sufﬁcient food, even within households whose resources are more than adequate. The problematization of the household in Ladakh has focused on this institution as the locus of both production and reproduction. Balancing these two fundamental yet often conﬂicting tasks has been routinely identiﬁed as the major problem facing Ladakhi households and the entire population, and polyandry is often analyzed from that perspective (see Chapter 3). The literature suggests, then, that the household is a good place to look for adaptive compromises, and this conﬂict plays out in ways that have signiﬁcance for maternal and infant health. The multiple ways in which the household acts as the proximate or “localized” environment that inﬂuences the health of women and children need to be speciﬁed. While the household emerges as the most important “immediate” environment of women and children, it does not exist in isolation from its larger social, cultural, and ecological milieu. It is useful to think of individuals being nested within wider spheres of inﬂuence that expand in space and time. This type of analysis is referred to as “proximate determinants analysis,” which attempts to locate the immediate (proximate) factors inﬂuencing birth or death, then the factors that inﬂuence the proximate factors, and so on. This way the pathways of inﬂuence from an environment to an individual’s likelihood of having an offspring, dying, or getting sick can be traced and analyzed (Bongaarts 1978; Chen 1983; Millard 1994; Wood 1994). Millard (1994) has drawn up a model of the determinants of child health by nesting proximate determinants (infection and undernutrition) within a household context, and the household within the regional, national, and international context (Figure 1.2). To date, none of the proximate determinant models has incorporated a historical component that would indicate the chronological sequence of events that led to the current set of risk factors, or that would demonstrate the role that population history has had in shaping people’s vulnerability to different risks. I hope to provide such a model as part of this work and extend the utility of proximate determinants analysis to the study of reproductive health, and to bring the household to the fore as an important locus of the ecology of infancy in Ladakh. Ethnographic Human Biology This project is a template for what I conceive of as “ethnographic human biology” or “biocultural ethnography,” a new anthropological genre that details the larger social, cultural, and ecological context of human biology. Human biological data are the core of the book, but they are situated within analysis of the ecological, historical, and sociocultural context of

20

An Ecology of High-Altitude Infancy

Ultimate Tier

Economic Factors International Ties to Donors, Lending Institutions, Global Economy Social & Cultural Systems Political Systems Administration of Agricultural & Other Programs

Natural Ecology

Settlement Patterns, Sanitation, Water Supply, Animal Husbandry Practices

Household Division of Labor In Agricultural and Other Work Household Decision-making about Disposal of Income in Kind and in Cash

Household Food Security: Food Production by Household s Access to Foods from Outside Households

Proximate Tier

Intermediate Tier

‘

Distribution of Food within the Household Food Preparation Child Care Practices Inadequate Diets of Children (frequency of feeding, meal composition, seasonal varieties, weaning practices)

Exposure of Children to Pathogens

Malnutrition

Diarrhea

Lower Respiratory Tract Infection

Child Mortality

Other Factors Outside This Model

Figure 1.2. The proximate determinants of child mortality (Redrawn from Millard, Ann V. 1994. A causal model of high rates of child mortality. Social Science & Medicine 38: 253–268, with permission from Elsevier Science.).

Introduction

21

Ladakh, other high-altitude areas, and South Asia. An overarching evolutionary perspective shapes the interpretation of the interaction between biology and environment. What becomes evident from the analysis is that reproductive health in Ladakh is compromised by the combined effects of the natural and sociocultural environment, which have acted in a speciﬁc historical way on the Ladakhi population. Reproductive health outcomes represent compromises – the net effect of several stresses interacting with a naturally and socially circumscribed set of resources. Chapter 2 will describe the ways the ecological stresses of high altitude, especially hypoxia, impact on human biology. Much of the research in this domain has been conducted in the Andes, but there have also been a number of recent important ﬁndings from the Himalaya. Particular attention will be paid to the relationship between hypoxia and reproductive health, and how responses to hypoxia are mediated by other ecological stresses as well as social and cultural processes. Chapter 3 sets forth both the broad outlines of Ladakhi ecology, society, and culture with speciﬁc attention to those likely to impact on reproductive health, and the context of doing research on reproductive health in Ladakh in the 1990s. The evolutionary and social history of the Ladakhi population are key to understanding its current conﬁguration and characteristics. Infrastructural development sponsored by the Indian government, increasing military activity, and tourism have wrought profound changes on people’s lives, including their reproductive lives. The household will be presented as a key social institution that mediates the relationship between the natural and social environment and reproductive health. Traditions of health and medicine, in both their formal textual forms and nontextual forms, will be described as well as the current understanding of the epidemiology of Ladakh. The research on which this book is based has been constrained and biased in various ways, as is all ﬁeldwork. In this chapter I discuss these issues and how they affected the research, along with the day-to-day aspects of doing research in the labor room of the Leh hospital. Reﬂexivity about ﬁeldwork is notably absent from most human biology, though it is now a core component of contemporary ethnography. In reﬂecting on my ﬁeld experience I hope to convey the reality of this kind of research, thereby making the process more transparent to the reader and also helpful to students anticipating ﬁeld research in human biology or medical anthropology. The protocol for the prospective study of birth outcome and the sample characteristics are laid out in Chapter 4. The results and analysis of two sets of speciﬁc research questions are presented: (1) What are the characteristics of a large sample of Ladakhi newborns? How do they compare

22

An Ecology of High-Altitude Infancy

with those of other populations? (2) What factors are important determinants of birthweight and other neonatal characteristics in this population? How do these and other sociocultural factors contribute to intrapopulation variation in birth outcome? The relevant determinants of maternal health and birth outcome are described here in a quantitative fashion for a sample of 168 women and newborns. This analysis is ﬂeshed out by a more qualitative examination of the context in which pregnancy takes place. Here the productive activities of women, which are shaped by their household context, come to the fore to inﬂuence the different experiences of mothers and newborns in the study. The condition of the neonate has important implications for his or her survival, and Chapter 5 outlines how neonatal characteristics affect survival in the ﬁrst month and year of life. Data from the prospective study, along with analysis of reproductive histories of a larger cohort of women and pediatric admissions data, are used to address the following speciﬁc research questions: (1) What is the relationship between newborn characteristics, especially birthweight, and neonatal mortality (death in the ﬁrst month)? How do other known risk factors interact with birthweight and inﬂuence early mortality? Since early mortality is so closely related to birth outcome, which in turn is a function of the pregnancy, how can we trace the determinants of neonatal mortality through maternal behavior and household ecology? (2) What are the determinants of infant mortality (death in the ﬁrst year) in this population? What is the relative importance of birthweight for overall infant mortality, and how does it interact with environmental causes of death such as infectious diseases? (3) What are common child-care patterns, and how do they relate to household and gender dynamics, socialization goals, and infant physical and emotional health, and mortality? (4) Why and how do women utilize different local health resources for their infants, and how does this vary by the infant’s gender? (5) How did the political, economic, and social turbulence speciﬁc to this area in the early 1990s inﬂuence women’s reproductive strategies or the health of their new infants? One of the key ﬁndings presented in this chapter is that neonatal size was strongly predictive of early mortality, and mortality was focused during the neonatal period. This is a very peculiar pattern of infant mortality and can only be understood by this analysis of infant ecology. The household constitutes the most relevant “microecology” of the infant, as it ﬁlters sociocultural, ecological, and historical conditions unique to Ladakh and frames the experiences of children and their caretakers. While Chapters 4 and 5 detail the experience of mothers and infants in this study, this will be put into a more global, comparative perspective in Chapter 6 to provide broader insight into the following questions

Introduction

23

concerning biocultural adaptation in Ladakh: (1) How does the study of neonatal anthropometric characteristics such as birthweight, their causes and consequences, help us assess the adaptive status of the population of Ladakh? Insofar as birthweight both reﬂects the availability of maternal resources for fetal growth and is a good predictor of mortality in the neonatal period, it has important implications for the reproductive success of this population. In what ways have social institutions and cultural traditions acted to buffer ecological stress, or impose additional stresses on health and reproduction? (2) How can we interpret these ﬁndings from Ladakh relative to those in other high-altitude areas, including the Andes and Himalaya, and those from the lowlands of India? Comparison with the Andes controls for altitude but allows for examination of other factors that vary (evolutionary history, recent history, culture). Comparison with other Himalayan populations controls for altitude and culture to some extent and highlights local factors. The comparison with reproductive indices from lowland India, which is both ecologically and culturally quite distinct from Ladakh, allows us to detect regional patterns and assess some important infrastructural and social inﬂuences. The book concludes with Chapter 7, which takes a further look at the short-term and long-term implications of the reproductive health situation. How does a high rate of infant death interact with local rates of fertility to affect overall population dynamics, and how does this relate to population-environment interactions and the adaptive process? As the process of radical social change continues apace in Ladakh, how is reproductive health affected? How do Ladakhi women and infants continue to try to cope with their unique circumstances? This chapter also considers the relevance of the biocultural perspective and the results of this study for health intervention. Such information should hopefully lead to the design of appropriate interventions, if called for, and more effectively target infants and mothers at risk with sensitivity to maternal and familial desires for ideal family size and composition, and the constraints that characterize this high-altitude environment. Common interventions to reduce infant mortality, for example, might be useless in Ladakh, because of the unique constellation of factors that contribute to infant death. Medical anthropologists have been, quite correctly, critical of health interventions that are not locally appropriate, or that are merely “band-aid” responses that do not deal with the fundamental roots of disease. But this should not lead to the conclusion that intervention is useless or unwarranted; a 10 to 15% death rate for Ladakhi infants in the ﬁrst month of life cannot be viewed as if it is simply an intrinsic part of the “otherness” of Ladakh (cf. Farmer 1999). It is most certainly of crucial importance to the well-being of thousands of Ladakhi

24

An Ecology of High-Altitude Infancy

mothers, infants, and their families, and it is incumbent on anthropologists to contribute productively to the dialogue on how quality of life might be improved for all populations. Finally, it is worth noting that the biocultural analysis of reproductive health in Ladakh laid out in this book is but one particular story that can be told. A biocultural perspective is extremely valuable in uncovering and organizing the various proximate and ultimate sociocultural, ecological, and historical determinants of, and pathways of inﬂuence on, reproductive outcomes. In many ways, this analysis has currency among Ladakhi women as well: many of the variables that the analysis suggested were important to reproductive outcome are the very ones that Ladakhi women themselves articulated to me, and to each other. However, the holistic approach of biocultural research may lead one to believe that all of the relevant aspects of reproduction have been determined. But the reproductive process, encompassing both birth and all too frequently premature death, is a multilayered experience, and there is no one “right” or “complete” analysis. It would be easy to imagine an entirely different story – one with the theme of karma and suffering, for example – that would illuminate the phenomenological experience of birth and mothering. In any event, it is the power of anthropological thinking that allows for these multiple explanatory frameworks to provide different kinds of insights into this universal aspect of human experience in the very speciﬁc cultural and ecological context of Ladakh.

2

Challenges of High-Altitude Living

Human populations living at high altitude must ﬁnd ways to adapt to a unique complex of ecological stresses. Hypoxia is the best studied of these, but it often occurs in conjunction with high levels of short-wave (UV) radiation, dramatic seasonal or diurnal variation in temperature resulting in periods of intense cold, and harsh and barren terrain. Mountainous regions of the world vary greatly in their geographies, but they are linked by this package of ecological features. These have both direct and indirect effects on human biology insofar as they constrain the possibilities for food production and inﬂuence the distribution of disease-causing microorganisms, among others. Further, these ecological factors act as stressors – singly, additively, or interactively – on individuals and populations, and ultimately they can have effects on people’s health and wellbeing. The elevation of 2500 meters is often used as the cutoff to distinguish “high” from “low” altitude, but this is only a rough indication of when most people begin to experience the physiological effects of hypoxia and where other aspects of high-altitude ecology become increasingly apparent (Pawson and Jest 1978). It is estimated that over 140 million people live in the mountainous areas of the world, the majority in the Andes and Himalaya. There are also substantial populations in the East African highlands, the European Alps, and the Rockies of the United States (Moore et al. 1998). These areas have been inhabited by humans for widely varying lengths of time. Some authors contend that populations of the Tibetan plateau have the greatest antiquity, citing evidence of upper Paleolithic (ca. 25,000 bp) settlement in the area (Moore et al. 1998). Others contest this date and the likelihood of a direct link with contemporary Tibetan populations (Beall 2000). There are excellent archaeological data from the Andes indicating continuous occupation for the past 10,000 years (MacNeish and Berger 1970). Extant populations in the Rockies have only a few generations of antiquity at most, as there is ongoing migration to high elevations. The antiquity of a population living at high altitude is likely to be a strong inﬂuence on its adaptive status; populations with relatively 25

26

An Ecology of High-Altitude Infancy

long histories of high-altitude residence are expected to be characterized by more efﬁcient genetic adaptations to stresses such as hypoxia. The genetic characteristics of high-altitude populations have been shaped not only by natural selection but also by other evolutionary forces such as genetic drift, since they may be small, localized, and fairly isolated; gene ﬂow through migration; and mutation, possibly increased by exposure to UV radiation. Cultural characteristics have been shaped by each population’s unique history and also by the ecological conditions of high-altitude life, which set the parameters for human habitation and subsistence. High Altitude: Laboratory for Human Adaptability How humans adapt to the challenges of a high-altitude environment has come to be a key example of human adaptation described in introductory biological anthropology textbooks. There is a long history of anthropological studies of high-altitude biological adaptation, especially in South America. Many of these studies were directed by Paul Baker at Pennsylvania State University and continue to be carried out by some of his former students.1 Topics of investigation have included fertility, mortality, disease, cardiovascular and hematological characteristics, growth and development, and nutrition, all aimed at assessing the quality and extent of human adaptation to high altitude, particularly to hypoxia. Two older volumes review the conclusions of this research as they existed in the late 1970s (Baker 1978; Baker and Little 1976). The material has been updated in Frisancho (1993) and subsequent review articles (Beall 2000; Moore et al. 1998). Not surprisingly, hypoxia has been found to have its most signiﬁcant effects on cardiovascular and hematological characteristics. Increased ventilation, hemoglobin production and capillarization, and enlargements in lung volume are among the responses to hypoxia, and all act to increase oxygen transport to tissues in a context where oxygen is not abundant (Frisancho 1993). These are all inducible (acclimatization) responses; that is, they are changes that individuals can make over the course of their lifetime when exposed to hypoxia. Although some responses, such as lung volume, are sensitive to the timing and duration of hypoxia exposure, other responses that individuals of any age can engage when chronically exposed to hypoxia include increased ventilation, hemoglobin, and capillary formation. High-altitude natives also make such modiﬁcations. These responses appear to be sufﬁcient to maintain adequate cardiovascular function, insofar as they are both efﬁcient and effective. In the Andes, carefully controlled migration studies designed to ascertain the

Challenges of High-Altitude Living

27

role of genetic and acclimatization responses in high-altitude adaptation have shown that acclimatization accounts for much of the variation in hypoxia response (Frisancho 1993; Haas et al. 1980; Harrison 1966). At the same time, such studies have shown that there may be generalized genetic inputs to high-altitude adaptation, but speciﬁc genes affecting speciﬁc proteins have not been identiﬁed. The extent to which the differences seen between populations living at different altitudes are due to pure altitude effects is often difﬁcult to assess accurately due to the confounding effects of other differences between populations living at high versus low altitudes. Because of their unique population histories and local combinations of social and ecological stresses, there is no reason to expect that all high-altitude populations will be characterized by similar adaptations. The current state of high-altitude adaptation studies suggests that there is variability in the kinds and modes of responses to hypoxia exhibited across human populations. For example, it appears that the pathways by which Andeans and Tibetans achieve oxygenation differ, and these demonstrate the range of options for responding to hypoxia. Reports from Tibet have identiﬁed the presence of major genes inﬂuencing oxygen saturation, hemoglobin concentration, resting ventilation, and ventilatory response to hypoxia (Beall et al. 1994), while evidence of similar genes has not been found among Andean populations (Beall et al. 1999). The Tibetan research provides the strongest evidence thus far for one or more genetic adaptations to hypoxia in a human population, and they occur in a population that may have the greatest antiquity in a highaltitude environment. Andeans rely more on increases in lung volume, hemoglobin, and oxygen saturation, while Tibetans rely on increased ventilation and ventilatory responsiveness, but less so on increases in lung volume, hemoglobin, or oxygen saturation (Beall 2000; Beall et al. 1997; Moore et al.1998). Hypoxia and Fetal Growth This work indicates that hypoxia affects individual biological function and by extension, health and well-being. But does hypoxia also affect reproductive function, fecundity, pregnancy, or fetal and infant growth and development? One of the best demonstrations of the ways in which hypoxia impinges on reproduction is through its effects on fetal growth, which is evident in the birthweight of a newborn. Birthweight is an important determinant of infant survival, as small newborns are particularly vulnerable to morbidity and mortality. Hence birthweight is a good index of adaptive status, as it is linked to ﬁtness outcomes.

28

An Ecology of High-Altitude Infancy

Birthweight (grams)

3,700 3,500 3,300 3,100 2,900 2,700 2,500 0

1000

2000

3000

4000

Altitude (meters)

Figure 2.1. The relationship between birthweight and altitude.

There is a well-established negative correlation between altitude and birthweight, as shown in Figure 2.1. As altitude increases, average birthweight declines, and this effect is evident after controlling for other factors that are known to inﬂuence birthweight (Haas 1980; Jensen and Moore 1997; Yip 1987). There is no evidence of neurological or clinical immaturity or of increased prematurity rates. Instead, it appears that rates of linear growth and fat deposition are slowed during the latter months of gestation, resulting in decreased weight, length, and fat deposition among infants born at high altitude (Haas 1981; Haas et al. 1977; Haas et al. 1980; McClung 1969). The mechanism for this effect is still unclear, but recent work suggests that uterine blood ﬂow may be restricted during pregnancy at high altitude, resulting in substantial reductions in oxygen supply to the fetus (Zamudio et al. 1995), and that hypoxia may alter placental morphology and impair placental function (Zamudio and Moore 2000). Complications of pregnancy such as hypertension and pre-eclampsia are also more common at high altitude and are associated with reductions in birthweight (Moore et al. 1982; Palmer et al. 1999; Zamudio et al. 1995). The link between altitude and birthweight has been conﬁrmed in numerous studies in the New World (Beall 1976; Conlisk 1987; Grahn and Kratchman 1963; Haas 1980; Haas et al. 1977, 1980; Lichty et al. 1957; McClung 1969; Moore et al. 2001b; Unger et al. 1988; Weinstein and Haas 1977; Yip 1987). In carefully controlled studies from Bolivia (Haas 1980; Haas et al. 1980) and Peru (Haas et al. 1977; McClung 1969); depression in the mean birthweight in the high altitude samples has been

Challenges of High-Altitude Living

29

shown to be on the order of 10 percent (200–400 g), attributable solely to the effect of altitude. In La Paz, Bolivia (altitude 3600 m), Haas et al. (1980) found that 10% of the full-term infants born were of low birthweight (LBW; <2500 g) while only 3% of infants born in Santa Cruz (altitude 400 m) were so classiﬁed. Based on an extensive analysis of birthweight data from the National Center for Health Statistics, Yip (1987) found a 10% reduction in mean birthweight in the high-altitude regions of the United States and a threefold increase in the incidence of LBW infants in these areas. In fact, the effect of high altitude was found to exceed all other known health and socioeconomic risk factors in the epidemiology of low birthweight. In Colorado, the state with the greatest altitude differences in residence, the frequency of LBW infants at high altitude (13% of all births) was double that of the low-altitude regions in that state (Unger et al. 1988). Even though the high-altitude Colorado population has substantially better socioeconomic conditions than populations in the Andes, they exhibit much higher rates of LBW than the latter, indicating the importance of evolutionary history to birth outcome. Newborns of recent migrants to high altitude appear to suffer a greater decrease in birthweight than do those born to women native to high altitude. In highland Bolivia, native women delivered signiﬁcantly larger infants in La Paz than did migrant women (Haas 1980). At low altitude in Bolivia, women who were born at high altitude, or who had highaltitude ancestry continued to deliver larger babies than low-altitude natives. In this study, altitude of residence was the most important factor contributing to variation in birthweight, and general genetic and acclimatization factors contributed to a much lesser extent. Comparison of native Tibetan and migrant Han Chinese in Tibet indicates a similar pattern of signiﬁcantly higher birthweights among Tibetans compared to Han migrants (Moore et al. 2001b; Niermeyer et al. 1995). Moore et al. (2001b) have shown that pregnant Tibetan women have greater ventilation and uteroplacental blood ﬂow than Han women, and these characteristics are associated with higher birthweights. As a general rule, males weigh more than females at birth, but at high altitude males appear to suffer a greater reduction in birthweight (roughly 400 grams, or over 10% of the mean birthweight) than do females (150 grams, less than 5% of the mean birthweight) (Haas et al. 1980). Length is also reduced more among males (2.3%) than females (0.4%) at high altitude. This ﬁnding is consistent with the more general observation that males are more vulnerable to stress (e.g., malnutrition) early in life than are females (Stini 1969), which often results in reductions in sexual dimorphism.

30

An Ecology of High-Altitude Infancy

The birthweight situation appears to be somewhat different in the Himalaya. In the ﬁrst study of this phenomenon, Moore (1990) found that the average birthweight of a small sample (n = 15) of Tibetan newborns in Lhasa, Tibet was 3307 grams, which was well above the average birthweights for other populations residing at similar altitudes (see Figure 2.1). Subsequent studies from Lhasa have shown somewhat lower averages of 3140 to 3280 grams, which are more consistent with Andean birthweights (Moore et al. 2001a; Moore et al. 2001b; Zamudio et al. 1993). Zamudio et al. (1993), Moore et al. (2001b), and Smith (1997) all have found that average birthweight declines only slightly among Tibetan or Sherpa populations living at altitudes of up to 4000 meters compared to those at 1200 meters. So the reduction in birthweight associated with high altitude does not appear to be quite so evident in the Himalaya, at least among indigenous Tibetan and Sherpa populations (Sherpas are of Tibetan origin and inhabit northern Nepal). It should be noted that these altitude comparisons are not as marked as those done in the Andes (sea level versus 3600 m), and so these results must be interpreted in that light. At present, however, it does appear that Tibetan newborns have been somewhat buffered from hypoxic stress in utero compared to Andean newborns or those born to Han (migrant) women in Tibet. Hypoxia and Child Health While growth in the prenatal period is notably inﬂuenced by hypoxia, effects of hypoxia on postnatal growth are less clear, in large part because child growth is affected by a more diverse array of stresses that children encounter in their environment. Until recently, it was generally concluded that exposure to hypoxia resulted in slower rates of linear growth (height) among high-altitude populations. Since growth requires energy, and energy metabolism requires oxygen, it makes sense to postulate that growth would be slowed down under hypoxic conditions. Growth might be further compromised by cold, since Andean responses to cold include a higher metabolic rate, which requires additional energy to maintain (Frisancho 1993). In the Andes, several studies have shown that linear growth is slow and prolonged, adult height and weight are reduced, but relatively rapid growth occurs in the cardiovascular system (Frisancho 1993; Frisancho and Baker 1970). However, further studies suggested that hypoxia itself does not have strong consistent or independent effects on child growth, but rather that an individual’s nutritional resources, not altitude, are the most signiﬁcant factors inﬂuencing growth (Leonard 1989; Obert et al. 1994). Nutritional resources are primarily, though not exclusively, related to socioeconomic status at both high and

Challenges of High-Altitude Living

31

low altitudes, and these factors appear to be crucial to child growth. In fact, children of similar socioeconomic status (SES) at high and low altitudes in the Andes have similar patterns of growth (Leonard 1989). Moreover, the height and weight of well-nourished, high SES children in the Andes differ little from that of U.S. children (Greksa et al. 1985; Stinson 1982). Similarly reﬁned studies have yet to be carried out among Himalayan populations, as controlled comparisons between altitudes are more difﬁcult due to genetic and socioeconomic variability (Singh 1989). Socioeconomic status is situated within the ecological conditions of high altitude, and these collectively have a signiﬁcant impact on subsistence possibilities and hence on potential nutritional resources. Highaltitude populations subsist on hardy grains such as wheat and barley, and tubers such as potatoes, all of which provide a carbohydrate base. They rely to varying degrees on animal husbandry for protein. Andean subsistence agriculture is characterized by a low diversity of cultigens, low productivity, seasonal variation in food availability, and interannual ﬂuctuations in productivity; the result is unpredictability of food supplies (Thomas and Winterhalder 1976). Rural households with more resources (land, animals, access to other sources of income) are buffered from these exigencies to a greater extent than poorer households, and thus children in wealthier households have a more consistent source of resources to put toward growth, even in the face of hypoxic stress (Carey 1990; Leatherman 1996). In contrast, poorer children have fewer resources to devote to growth and may also have greater energy expenditures, as their labor is needed for subsistence. Therefore the growth of children in poor households is compromised by the multiple effects of malnutrition, high energy expenditure, hypoxia, and often infectious disease (Berti et al. 1998). Little is known about the distribution and effects of infectious disease in high-altitude environments. An oft-cited work that compared infectious disease patterns at different altitudes in Peru suggested that many common infectious, particularly parasitic, pathogens are much less common at high altitude, probably because ecological conditions are inhospitable to microorganisms that ﬂourish in warm, moist, oxygen-rich environments (Buck 1968). The authors reported that Anopheles mosquitoes, the vector of malaria, were not present at high altitude, while typhus infection was common because people did not bathe often and associated closely with animals on the altiplano. Hookworm infestation rates were low and arboviruses virtually absent, but the common gastrointestinal parasites did not vary noticeably with altitude. In general, respiratory infections seem to be more common causes of serious morbidity at high altitude, whereas diarrheal diseases (usually water- or food-borne) are more common at low altitude. However, poor urban living conditions

32

An Ecology of High-Altitude Infancy

at high altitude can contribute to a high frequency of water-borne gastrointestinal diseases (Berti et al. 1998). Repercussions of respiratory infections are likely to be more serious at high altitude where the respiratory system is already challenged by the stress of an oxygen-deﬁcient environment. Infectious disease patterns and their contribution to health is an important area of future investigation in high-altitude studies, since the combination of malnutrition and infection is usually the proximate cause of poor growth and health among children at any altitude. High Altitude and Infant Mortality The signiﬁcance of reduced child growth in mountain populations depends on the factors that generate the slower development. That is, growth can be impaired by a variety of factors, most notably malnutrition and infectious disease. Under these conditions, decreased growth is an indicator of a problem and is associated with negative health outcomes (Martorell 1980). However, slow or reduced growth that results exclusively from hypoxia may not be associated with those same outcomes. Indeed, the implicit interpretation is that growth deﬁcits from hypoxia are likely to be more benign in their health effects than those caused by malnutrition or infection. The same may be true of fetal growth. When birthweight is compromised by maternal nutritional or health problems, it may be associated with increased risk of morbidity and mortality. Birthweight is well known as a good predictor of mortality in the ﬁrst year of life; up to a point, higher birthweights are associated with decreased risks of mortality, as shown in Figure 2.2 (Ashworth 1998).2 However, when low birthweight derives from prematurity, it is associated with dramatically elevated rates of mortality relative to a similarly low birthweight of a full-term newborn (McCormick 1985). When low birthweight results solely from chronic but moderate hypoxic stress due to pregnancy at high altitude, it does not seem to be associated with the same risk; the whole birthweight distribution is shifted toward lower weights, but the birthweight-speciﬁc mortality curve is also shifted to the left (Beall 1981; Cogswell and Yip 1995; Wilcox 1993; Wilcox 2001). Indeed, Wilcox (2001) has convincingly argued that it is relative birthweight (that is, its deviation from the mean) within a population that is most important in determining mortality risk; he also cautioned that the mechanism by which birthweight (independent of prematurity) affects mortality is not well understood. Beall’s work on birthweight at high and low-altitude in Peru (Beall 1981) illustrates this point. She found that while the frequency distribution of birthweights was shifted downward (indicating a lower average

Challenges of High-Altitude Living

33

100 90

70 60 50 40

500

30

1500

20

2500

10

3500

0 28 30 32 34 36 38 40 42 Gestational Age (weeks)

Bir thw eig ht

Mortality Risk (%)

80

4500

Figure 2.2. Birthweight, gestational age, and neonatal mortality risk.

birthweight) in the high-altitude area, the birthweight associated with the lowest risk of infant mortality was lower at high altitude than at low altitude. That is, the optimal birthweight was lower at high altitude. The ﬁnding of a downward shift in the birthweights associated with lower mortality was replicated by Conlisk (1987) in Bolivia and by Unger et al. (1988) in Colorado; both used retrospective data. The Colorado research indicated that LBW infants had higher rates of mortality than larger newborns at all altitudes. However, LBW infants born at altitudes above 2744 meters had lower mortality risks than their low-altitude counterparts when health resources became more comparable. Thus, despite a greater percentage of LBW births at high altitude, infant mortality statistics were similar throughout the state. Larger newborns have lower rates of mortality regardless of their altitude, but for some reason, smaller newborns at high altitude have lower rates of mortality than they do at low altitude. This most likely reﬂects the different etiology of smallness at low and high altitude. Low birthweight due to hypoxia may be less detrimental than low birthweight due to maternal malnutrition, infection, or premature birth, which may be more often the cause of lower birthweights among low-altitude populations (Cogswell and Yip 1995).3 However, smallness is not adaptive at

34

An Ecology of High-Altitude Infancy

high altitude, because larger babies there have lower rates of mortality. The observation that small newborns have a greater probability of survival at high altitude leads to the conclusion that the adaptive signiﬁcance of birthweight can only be assessed relative to the environment in which it occurs. Lower average birthweight at high altitude is likely the result of some adaptive trade-offs, or compromise, in the sense that reduced oxygen ﬂow to a fetus may be a viable way to maintain a pregnancy at high altitude without dramatically compromising maternal or neonatal health. At the same time, women who are able to produce larger newborns with a reduced risk of dying are likely to have adaptive characteristics relative to women who deliver smaller infants with higher risks of dying. Evidence of higher birthweights among infants born to women native to high altitude indicates adaptation when those infants have greater rates of survival. Broad surveys of infant mortality indicate a link with high altitude. The Pan-American Health Organization (PAHO) noted that Bolivia and Peru, two countries with the largest high-altitude populations in South America, had the highest infant mortality rates of all countries on the continent (1994). Within Bolivia and Peru, infant mortality was highest in the high-altitude regions. In the 1960s, Mazess (1965) noted a clear positive correlation between neonatal mortality (death in the ﬁrst month of life) and altitude in Peruvian census records, and Grahn and Kratchman (1963) reported the same trend in the United States. In the Peruvian study, neonatal mortality rates at high altitude were found to be double those at low altitude (53/1000 vs. 29/1000, respectively), and postneonatal mortality (mortality between 1 and 12 months) was elevated by 20 to 40% (73/1000 and 52/1000 at high and low altitudes, respectively). Neonatal mortality accounted for 42% of infant mortality at high altitude but only 35% at low altitude. Infant mortality was also found to be signiﬁcantly higher among males than females at high altitude, probably because males suffered greater reductions in fetal growth than females (Baker and Dutt 1972). Remember that none of these studies adequately controlled for interaltitude variation in socioeconomic status, access to health care, and other factors that inﬂuence early mortality. The lower average birthweight at high altitude is a likely contributing factor to the magnitude of these rates, especially the neonatal mortality rate, since birthweight exerts its strongest effects on mortality early in the ﬁrst year of life. Though the optimal birthweight may be lower at high altitude, smaller infants are still at a greater risk of mortality at high altitude than are heavier babies, and their prevalence is much greater. Smallness may add to the risk of morbidity, as small infants have fewer biological resources with which to respond to health threats. At high altitude, where hypoxia challenges respiratory function, respiratory diseases are expected

Challenges of High-Altitude Living

35

to be most problematic. In highland Peru, Spector (1971) found that over 87% of infant deaths were attributable to respiratory problems. Likewise in the same area, Beall (1976) found that respiratory diseases were responsible for almost 75% of all neonatal deaths, and the majority (55%) of postneonatal infant deaths. In La Paz, Bolivia, neonatal deaths were more often the result of respiratory diseases than diarrheal diseases, but the opposite effect was found at low altitude in Brazil (Puffer 1973). Respiratory illness is made more life threatening by reduced oxygen saturation and hypoxia-related disease, both common among newborns and young infants at high altitude (Niermeyer et al. 1993; 1995). High Altitude and Fertility The link between high altitude and high early mortality, especially from respiratory disease, has been well established, but the connection between high altitude and fertility remains murky. Although there has long been a concern (since the colonial period when the Spanish tried to occupy the highlands of South America) that hypoxia could somehow reduce reproductive function, there is little evidence that hypoxia has direct effects on fertility (the production of a live birth) or fecundability (the ability to conceive) (Vitzthum 2001; Vitzthum et al. 2000a; Vitzthum and Wiley 2003). The variables that inﬂuence fertility directly and which are in turn affected by larger sociocultural and ecological parameters are referred to as the proximate determinants of fertility. These include exposure variables such as age at marriage, menarche, menopause, and onset of pathological sterility as well as susceptibility variables such as duration of lactational infecundability (the length of time a woman is not able to conceive after pregnancy due to the ovulation-suppressing effects of the hormones of lactation), waiting time to conception, probability of fetal loss and the length of the period of infecundability following fetal loss (Bongaarts and Potter 1983; Bongaarts 1978; Campbell and Wood 1988; Wood 1994). The question is how any of these might be shaped by residence at high altitude, either directly by hypoxia or other ecological variables or indirectly through interaction with sociocultural variables. The most likely candidates for direct inﬂuence by hypoxia are age at menarche, the waiting time to conception, probability of fetal loss, and the duration of lactational infecundability, if it is inﬂuenced by the same factors that inﬂuence the probability of conception under nonlactating conditions. The most basic measure of fertility, the crude birth rate (number of births per population size), of high-altitude populations has not been found to differ signiﬁcantly from the crude birth rate at low altitude

36

An Ecology of High-Altitude Infancy

in the Andes. Estimates of children ever born were well within the normal range for populations with a low prevalence of contraceptive use (6–7 births per woman) in Nunoa, ˜ Peru (Hoff and Abelson 1976). There is some evidence that age at menarche is delayed in high-altitude populations (Frisancho 1993), although the difference is only about one year compared to low-altitude populations, and menarche usually occurs well before marriage. Thus, this delay is likely to have little effect on fertility. Furthermore, whether the delay in menarche is solely hypoxia induced or is a function of child malnutrition or some combination thereof remains unclear. Comparative studies of populations at different altitudes in the Andes have found no signiﬁcant differences in total fertility that could be attributed speciﬁcally to the hypoxic effects of high altitude (Dutt 1980). Abelson et al. (1974) showed that fertility increased with downward migration in Peru due to a shortening of birth intervals. Birth intervals for every parity level were longer among high-altitude natives. This may be due to differences in breastfeeding patterns, which inﬂuence return to fecundity after birth; nutritional factors; or hypoxia factors that inﬂuence the likelihood of ovulation. Without more controlled studies it is impossible to ascertain the cause of this difference in birth interval length. Studies of nonacclimatized mammals exposed to acute hypoxia show disruption in ovarian cycles, although there is at best anecdotal data on this phenomenon in humans (Clegg 1978). As yet there is no evidence suggesting that hypoxia has direct effects on levels of circulating ovarian hormones among long-term residents. Vitzthum and colleagues (2000a) found that circulating progesterone levels of high-altitude native Quechua women in Bolivia were intermediate compared to other industrialized and nonindustrialized populations but did not differ from those of low-altitude Bolivian women. On the other hand, Escudero et al. (1996) reported that progesterone levels were lower among high-altitude than low-altitude women in Peru, although as Vitzthum et al. (2000a) note, it is difﬁcult to control for all the confounding factors that could cause variation in progesterone levels. Menstrual cycle characteristics were within the range of normal sea-level populations (Vitzthum 2001; Vitzthum et al. 2000b). There are conﬂicting data on the incidence of spontaneous abortion or fetal death at high altitude. Although accurate data on these events are often particularly hard to come by, older Andean studies had suggested a decrease in fetal wastage at high altitude (Abelson et al. 1974) while two Himalayan studies suggested an increase (Lang and Lang 1971; Moore et al. 2001a). By monitoring hCG status in pregnant women, Spielvogel and Vitzthum (2003) recently found no difference in fetal wastage at high altitude in Bolivia compared with sea-level standards.

Challenges of High-Altitude Living

37

Exposure to UV light may increase mutations that are deleterious to embryonic or fetal survival. Hypoxia may have more of an observable effect on fetal survival and early infant mortality than fecundability, since there is ample evidence that it negatively affects fetal growth. Based on the limited existing data, it seems unlikely that hypoxia has strong independent effects on the likelihood of conception or on completed fertility. Evidence of lower fertility in a high-altitude population is much more likely to be due to intervening social factors that may be only indirectly related to the mountain ecology (Goldstein et al. 1983; Vitzthum and Wiley 2003). In Ladakh, for example, several studies have shown evidence of low fertility rates among women who were not using any form of technological contraception (Attenborough 1994; Elford 1994; Wiley 1998). There is a low rate of marriage for women and a late average age at marriage, which results in a high prevalence of unmarried reproductive-age women and serves to limit average fertility. This is in part a function of the practice of polyandry, in which one woman may have several husbands; as a result, a large portion of the adult female population is unmarried. However, fertility within marriage is also relatively low (<5 births per married woman). Potential causes of this low fertility include a high frequency of pathological sterility brought on by sexually transmitted diseases (STDs), high rates of fetal loss (possibly related to hypoxia), and possible nutritional constraints on ovarian hormone levels that decrease the probability of conception (Wiley 1998). Other Ecological Stresses of High Altitude The focus here has been on hypoxia as the major stress of high altitude, but other stresses act in conjunction with hypoxia to affect mortality and fertility indirectly. Cold, aridity, seasonality in weather patterns, and UV radiation all inﬂuence subsistence and human biology in ways that affect health and survival. These are to varying degrees more easily buffered by material culture than is hypoxia, but depending on one’s access to the resources necessary to buffer them, they may have important direct impacts on biology. Cold is a daily experience in the Andes and certainly increases energy expenditure, especially during seasons of heavy agricultural work. Cold is more seasonally evident in the temperate-zone Himalaya, but there, particularly in the western regions such as Ladakh, it is extreme, and this necessitates housing and clothing styles that help to reduce heat loss. Aridity is a major problem in the western Himalaya and parts of the Andes as well. Dry air acts as an irritant to the lungs, especially in combination with cold and air-borne particulate matter such as dust, sand, ash, or pollutants that contribute to respiratory problems. Ultraviolet

38

An Ecology of High-Altitude Infancy

radiation is associated with an increased mutation rate, which may inﬂuence adult mortality and also fertility, by contributing to congenital defects. Seasonality in weather patterns, either in temperature or rainfall, contributes to variation in availability of food, energy inputs needed for subsistence, and demands for wage labor, all of which may necessitate migration for some household members. Seasonality concentrates these stresses at particular points in the annual cycle and may generate intraannual changes in biological measures and birth patterns. These are all more or less “natural” stresses, but as should be evident, they are not experienced independently of other stresses that are of social origin. These are the general outlines of high altitude ecology and their likely effects on the reproductive process, inclusive of fertility and early morbidity and mortality. At present the data indicate that infant mortality is more profoundly affected by high-altitude conditions than is fertility. Infant mortality, in turn, derives from the complications of chronic hypoxia in utero that ultimately manifest as lower birthweight. Thus, high altitude acts as a challenge to the well-being of pregnant women and the fetus, and variation in biological and behavioral pathways contributes to individual and population variation in birth outcome and survival. We now need to consider the case of Ladakh speciﬁcally – a mountainous region with a unique history and local social and cultural characteristics that shape current reproductive parameters.

3

Contextualizing Reproductive Health Research in Ladakh

High in a remote mountainous desert of the western Himalaya, Ladakh is geopolitically part of India. It shares unstable, ambiguous, and contested borders with Pakistan and China,1 as shown in Figure 3.1. Popularly referred to as the “Snowland,” “Wonderland,” “Moonland,” “Little Tibet,” or “the last Shangri-la,” its remote location and barren landscape yet colorful and “exotic” culture and the quiet joie de vivre of Ladakhis have intrigued researchers and adventure travelers alike. Descriptions of Ladakh tend to emphasize its contrasts; it is a rare account that does not begin with a dual description of the area as both “otherworldly beautiful” and barren, harsh, and unforgiving. The landscape provides more than a mere backdrop, as the unusual ecology of the trans-Himalayan plateau plays center stage in many analyses of Ladakhi society and culture. While this is appropriate in many ways, we risk concluding that Ladakhi life is entirely determined by the natural environment. Appadurai articulated this problem more generally within anthropology by noting that there is a tendency for places to become showcases for speciﬁc issues over time. . . . [T]he discussion of the theoretical issues tends to surreptitiously take on a restrictive local cast, while on the other hand the study of other issues in the place in question is retarded, and thus the over-all nature of the anthropological interpretation of the particular society runs the risk of serious distortion. . . . A few simple theoretical handles become metonyms and surrogates for the society as a whole. (1986:358)

Knowing the ecology of Ladakh is crucial to understanding Ladakhi culture as well as biology and health, but it should not overdetermine the analysis. One way is to avoid treating the population of Ladakh as a whole, by examining variation among Ladakhis who share the same basic natural environmental context. Further, we can compare Ladakhis to other populations living in high-altitude contexts in other parts of the world as well as to the larger low-altitude South Asian culture area of which Ladakh is a part in order to elucidate how ecological, historical, political, cultural, 39

40

An Ecology of High-Altitude Infancy

LADAKH CHINA

P

A

K

IS

T

A

N

TIBET NEP AL BHUTAN

BANGLADESH

Tropic of Cancer

I N D I A

BURMA

200

0

200 400

KILOMETERS

Figure 3.1. Ladakh in the context of South Asia.

and social-structural factors may be inﬂuencing reproductive health (see Chapter 6). In this chapter, the context that inﬂuences Ladakhi biology and health is described. This includes not only the natural ecology of the region but also the historical and sociocultural background that has shaped these experiences. The account focuses on social factors such as household dynamics and the broader political and economic changes that directly inﬂuence reproduction and infant survival. Further, the process of conducting research on these topics is outlined to provide the reader with a better sense of the local and temporal circumstances that generated the results described in subsequent chapters. Geography Ladakh encompasses about 100,000 square kilometers on the western end of the trans-Himalayan (Tibetan) plateau. Well within the temperate

Reproductive Health Research in Ladakh

41

latitudes (32◦ –36◦ ), its climate is to a limited extent affected by the monsoon cycle of the subtropical areas to its south as well as the weather systems of north-central Asia. Total annual precipitation is between 10 and 15 centimeters, with most occurring in the summer months of July and August when occasional monsoon clouds drift across the Himalaya. Storm systems from the north bring another small burst of precipitation in the winter, but Ladakh is a desert – a mere 0.002% is estimated to be forested (Mann 1986). The average annual temperature in the capital city of Leh is 6◦ C, with highs varying between 25◦ in July and −3◦ in January (Mani 1981). Given the mountainous conditions of Ladakh, there are many microclimates characterized by different temperatures, precipitation, and solar radiation, but the basic conditions of hypoxia, aridity, high levels of UV light, and seasonal extreme cold pervade the region. With a population density of less than one person per square kilometer, Ladakh is the least densely populated district of India. However, cultivation and permanent habitation are sharply limited by access to water and topography, such that only 0.2% of the entire land area is actually used for agricultural production (Mann 1986). Most human settlement is at elevations between 3000 and 4000 meters, and villages and ﬁelds are clustered on the relatively ﬂat plains of the river valleys or along tributary streams (some of which are mere rivulets of melted snow), where ﬁelds are terraced to ﬁt along the contours of the valley.2 The Indus River, best known for the Harappan civilization it spawned at its delta, is the main river in Ladakh and has its origins in the region’s eastern mountains. This glacial river drains several other water systems as it winds westward, in some areas forming a wide ﬂood plain suitable for habitation and cultivation, and in others a deep, narrow gorge (Photograph 1). The population of Ladakh is concentrated along the ﬂood plains of the Indus and its three major tributaries, the Shyok, Drass, and Zangskar, as shown in Figure 3.2, with population density increasing with proximity to the capital, Leh. Leh (Photograph 2) is in a valley of an Indus tributary and is the administrative and commercial capital of the district. Leh has a labyrinthic old town clustered just below the crumbling palace of the former king, but outside the immediate environs of the bazaar much of the town is made up of agricultural households. As a result, in the 1990s Leh still seemed more like a large village than a bustling urban center. This is changing as increasing amounts of land that had been under agricultural production are being turned over to private housing or tourist facilities. An enormous inﬂux of cash from tourism has made this possible, and this transition was most evident to me between visits in the early 1990s and 1996, when houses and hotels with satellite dishes sprouted

Photograph 1. The wide ﬂood plain of the Indus River, where much of the population of Ladakh lives and cultivates crops.

Reproductive Health Research in Ladakh

43

Figure 3.2. Geography of Ladakh.

where barley and potatoes previously were sown. Leh is also the site of the district hospital and the airport, which serves commercial and military ﬂights. The population in this region is largely Buddhist, with clusters of Muslims and a smaller number of Christians living in and around Leh. To the north of Leh is the Nubra Valley, centered along the Shyok River; to the east is the Chang-Tang plateau; and to the west is the intersection of the Zangskar and Indus Rivers. Upstream along the Zangskar and its tributaries are relatively dense settlements, although most of these are accessible only by foot. Farther to the west is the capital of the western district of Ladakh, Kargil, and most of the population in that region is Shiite Muslim. Past Kargil is the Zoji-la, a mountain pass that divides Ladakh from Kashmir. Snow cover on the Zoji-la determines when the 440-kilometer road connecting Ladakh with the capital city of Srinagar opens and closes, generally in mid-June and November, respectively. Because of the ongoing political conﬂict in Kashmir, a road from the south opened in the early 1990s for commercial trafﬁc, although it is poorly maintained. Many villages just off these two roads are connected by link roads and have either state-run or private bus service on a regular (though often infrequent) basis.

Photograph 2. Leh, the district capital of eastern Ladakh.

Reproductive Health Research in Ladakh

45

Early Historical Context The extent to which populations are able to adapt to their environment is predicated in part on the length of their ancestry in that context. Unfortunately, data on Ladakhi origins are scant. Early speculation on the biological origins of Ladakhis in a mid-nineteenth-century report by Cunningham (1854) proposed that they derived from the intermingling of populations from East, Central, and South Asia. Their physical characteristics were attributed to this admixture and adaptations to Ladakh’s ecological rigors. Cunningham’s conclusions have been shared by many subsequent observers, both in their recognition of the pluralistic origins of contemporary Ladakhis and the attribution of Ladakhi characteristics (physical and cultural) to their extreme environment. More recently the Ladakhi scholar Tashi Rabgias (1988) has suggested that human habitation in Ladakh dates back 4000 to 5000 years and derives from populations to the north and east on the central Asian steppes.3 Linguistic data suggest that Dardic-speaking peoples from the northwest may have been the ﬁrst settlers in Ladakh and were later joined by the Mons, who migrated into Ladakh from the south (Mann 1986). Subsequently, between the eighth and tenth centuries political disarray in Tibet spurred the westward migration of numerous noble Tibetan families into Ladakh, where they established a quasi-independent dynasty (Fisher et al. 1963; Rizvi 1983). Although this large inﬂux of Tibetans resulted in greater Tibetan cultural inﬂuence, what came to be known as Tibetan (Lamaist) Buddhism, the religion that predominates in eastern Ladakh, predated this, having been established there sometime between 300 bc and 300 ad and integrated with the local Bon religion (Rabgias 1988). In the middle ages Ladakh became an important way station on crossAsian trade routes (Rizvi 1983). Despite its forbidding geography, the mercantile center in Leh was a fairly cosmopolitan place, its central location providing opportunities for the mixing of goods, cultures and peoples, which further left a mark in the ethnic composition of Ladakh. Some traders settled in Ladakh and took up farming; many of these were Muslims, both Sunnis from Kashmir and Shiites from Baltistan (northwestern Ladakh). Thus, at this point it is reasonable to conclude that the current population of Ladakh is of relatively recent and diverse origins, deriving from groups from numerous areas to the north, east, and south, with most migration and settlement occurring in the past 1200 years. Ladakhi history is characterized by ongoing trends of cultural and biological pluralism.

46

An Ecology of High-Altitude Infancy

Political, Demographic, and Social Changes Ladakh was an independent kingdom with strong cultural, religious, political, and economic ties to Tibet until the mid-nineteenth century, when it was invaded by the powerful Dogras from the southwest in Jammu in their attempt to monopolize the lucrative cashmere trade. By the 1842 Treaty of Leh, the King of Ladakh was stripped of political power, and ties with Tibet were severed as Ladakh became ﬁrmly attached to the emergent state of Jammu and Kashmir under Dogra control. After the British relinquished their rule of the Indian subcontinent in 1947 and partitioned their South Asian empire into India and Pakistan, Ladakh became embroiled in conﬂict between India and Pakistan regarding the alignment of Kashmir, whose leader had opted to join India. Incursions by pro-Pakistan forces into western Ladakh forced the Indian military to respond by hastily constructing an airstrip outside of Leh and a new road from the Kulu valley to the south (Fisher et al. 1963). India eventually recaptured western Ladakh and the Kashmir-Ladakh road, but the agreement signed under the auspices of the United Nations in 1949 allowed Pakistan to annex the northwestern areas of Ladakh (Baltistan). Indian-Pakistani conﬂicts over Kashmir emerged again as military confrontations in 1965, 1971, and 1989, and are ongoing to date. In addition, clashes between India and China over the northern and eastern borders of Ladakh occurred in the 1950s and 1960s as China attempted to assert its sovereignty over the region, citing Ladakh’s historical tributary relations with Tibet. With the 1959 Tibetan revolt and the Dalai Lama’s escape to India, the Chinese occupied the Aksai Chin, the northeast region of Ladakh, to secure a route to Tibet. A short war ensued in 1962, which resulted in the de facto Chinese occupation of the area. The net effect was that Ladakh was further subdivided, the border between Ladakh and Tibet was sealed, and Ladakh became further isolated from both Tibet and Central Asian trade. This displaced a number of Ladakhis formerly employed in trade and left Ladakh bereft of its afﬁliations with Tibet, although at the same time many Tibetan refugees settled there. This period of international conﬂict brought about changes with widespread signiﬁcance to Ladakh, in part by impressing upon the Indian government the region’s strategic importance. The national government sponsored the construction of roads, bridges, transportation services, hydroelectric projects, and educational and health facilities. The route from Srinagar to Leh was improved, thereby dramatically enhancing communication networks and establishing a regular ﬂow of goods into Ladakh

Reproductive Health Research in Ladakh

47

from other parts of India and aiding the massive buildup of the Indian military. In 1974 the Indian government allowed tourists into Ladakh, whose numbers grew from a mere handful in the mid-1970s to over 15,000 annually in the 1980s and 1990s. Foreign and domestic tourists are drawn to the beauty and uniqueness of the Ladakhi landscape and its mountains, and they come to experience one of the only remaining areas of indigenous Tibetan Buddhism and culture. Tourist services became concentrated around Leh; as a result, there has been a centralization of the economy, a massive building boom in Leh and its environs, and rampant inﬂation. These infrastructural, social, and economic changes were associated with striking demographic changes from the 1960s to the 1980s. The population of Ladakh, which had increased very slowly (0.6% annually) in the ﬁrst half of the twentieth century, grew at the unprecedented rate of 2% per year between 1961 and 1981, and it has grown over 3% per year over the past twenty years (Government of India 1961a, 1961b, 1983). Most of this increase occurred in Leh, which in 1961 had an urban population of only 370; currently its population is estimated at over 20,000. This demographic swell was formed both from natural increase and migration of army personnel, merchants from north India, laborers from India and Nepal, and refugees from Tibet into the Leh area. New economic opportunities in the military, tourism, and ancillary occupations undoubtedly lured many of these migrants as well as many rural Ladakhis, and contributed to the urbanization of eastern Ladakh. The weak sanitation infrastructure of Leh has not been able to keep up with the increasing demands placed on it by migrants and tourists. Added to the social turbulence generated by these changes of the past thirty years was an outbreak of religious conﬂict in the early stages of my research in the summer of 1989, when violence broke out in Leh between Buddhists and Muslims. This conﬂict was rooted in the growing perception among the Ladakhi Buddhist population that the Muslimdominated state government was discriminating against them in favor of local Muslims with respect to coveted civil service jobs and economic opportunities. To redress these imbalances the local Buddhist Association was formed as a political organization, prompting the Muslim minority to form a similar coalition. These associations mandated a boycott from 1989 through 1992 that forbade Buddhists and Muslims from engaging in any kind of social interactions, from commerce to meals, under threat of ﬁnes. Among other things, this divisiveness affected people’s diets, as bakers and butchers had largely been Muslims, and Buddhists no longer had access to their products. During the course of my ﬁeldwork I lived with a Buddhist family, and their solution to this problem was to have

Photograph 3. Plowing with the help of dzos in June outside of Leh.

Reproductive Health Research in Ladakh

49

me, as a neutral party, purchase bread for them. As a result of these agitations, Leh District was granted “Hill Council” status in 1995, which provided a measure of autonomy from the state government with respect to internal affairs and more direct relationships with New Delhi. Ladakh is ever more ﬁrmly embedded within the Indian nation state and the global economy at this point, a position solidiﬁed by communication and transportation networks, tourism and the military, and stronger political ties to the federal government in New Delhi. Household Subsistence Although the Leh area has grown disproportionately and the economy has become increasingly diversiﬁed, most Ladakhis, including those in the peri-urban areas, still rely on farming for a large part of their subsistence. Hence household work patterns and the diet they support derive largely from the conditions for agricultural subsistence. Most Ladakhis own their land, and there is little variation in the size of landholdings, which average about ﬁve acres (Osmaston 1994). Barley is the principal crop grown, along with hardy wheats, peas, and potatoes, which were introduced in the late nineteenth century. All agricultural work is done without the aid of machines and with the help of animals only during plowing and threshing (Photograph 3). Field preparation generally begins in April, when ﬁelds are plowed and fertilized with manure. Once planted, these require weeding and regular watering by means of an extensive and complicated network of irrigation channels and propitiation of local water spirits (lhu). Harvesting of crops begins in September and the winter months are spent resting, doing household tasks, or traveling. Sowing and harvesting are the busiest times in the agricultural cycle. Many households also maintain large gardens in which a variety of vegetables such as turnips, radishes, onions, greens of all kinds, carrots, cabbage, and cauliﬂower may be grown, depending on local conditions. Many also have local cows (balang) or dzomos (female yak-cow hybrids) for milk, and dzos (male yak-cow hybrids) for help with plowing and threshing. Some households in the more remote valleys keep large herds of sheep for wool, milk, and meat. Agricultural production may be affected by the vagaries of local climate but in general is not subject to environmental disasters such as ﬂoods and droughts or great interannual variation. Fields are highly productive so that at any given time many can lie fallow (Osmaston 1994; Osmaston, Frazer, and Crook 1994). A household’s ability to maintain a balance between land resources and labor availability is key to its success at labor-intensive subsistence agriculture; those without enough labor

50

An Ecology of High-Altitude Infancy

have been found to produce less per ﬁeld than those with adequate labor (Osmaston 1994). Both men and women work in the ﬁelds at all stages of the agricultural cycle and work patterns are not strongly gender stratiﬁed, although there are activities women are more often involved in and vice versa (i.e., men plow, women weed). Each household is functionally independent and works its land separately, although during busy times in the agricultural cycle, families can share work through phaspun, or mutual-aid networks. Household Dynamics The household frames an individual’s identity, placement in social networks, access to subsistence and other resources, and is the most fundamental unit of Ladakhi society. It organizes the activities of its members, including subsistence work, domestic tasks, child rearing, and social and religious duties. These multiple functions are all related to the health of household members, especially mothers and children, whose lives are profoundly shaped by the overarching processes of household production and reproduction. Households can take a variety of forms but in general are comprised of extended families, often inhabiting different houses, the main one of which is the khang-chen (big house). Subsidiary houses (khang-chung) are usually nearby and inhabited by older parents or siblings (sometimes with their own families). The khang-chen is a multipurpose dwelling, often quite large (2–3 stories), built of sun-dried mud bricks. The ﬁrst ﬂoor houses domestic animals, storage facilities for grains, dairy products, and implements. Also on the ﬁrst ﬂoor is the kitchen, a large room centered around a metal stove, an imposing, often ornately decorated appliance that is at the center of family and social activity. The second ﬂoor is comprised of bedrooms, a composting toilet, and the family chapel, the chodkhang, which may also be in a separate room on the roof. Sleeping arrangements are often communal, and in the winter the entire family may sleep together in the kitchen, keeping warm by the stove, which burns coal and dung. Now simple gas or kerosene stoves are used for the majority of cooking, which often takes place in a small room attached to the kitchen, out of sight of guests. In general the main ﬂoor is dark, with only small windows to retain warmth in the winter and cool temperatures in the summer, while the upstairs rooms are graced with extensive glass windows, which allow for solar heating in the long sunny, but bitterly cold winter months. Marriage is a deﬁning event in the life of a household. It brings in a new member and sets the stage for the reproduction of the patrilineage,

Reproductive Health Research in Ladakh

51

which has its locus in the physical household and whose members are identiﬁed by a common “house name.” Marriages are most often arranged by members of each family in consultation with an astrologer (onpo) and with the consent of the couple. Most occur while the spouses are in their early twenties. Marriages are generally patrilocal, although in the absence of sons daughters can inherit land, and the husband joins the wife’s household. Although gifts are exchanged during the marriage negotiation process, there is neither a formal dowry nor brideprice. Women bring wealth to their marriage in the form of jewelry and textiles, and these continue to be their exclusive property. As is true in other parts of the Himalaya, there is a great deal of ﬂexibility in marital patterns in Ladakh. Historically, monogamous marriages occurred along with those that were polyandrous (usually involving brothers), polygynous, or some combination of those. Polyandry in particular has attracted the attention of numerous researchers, as it is a rare marital pattern in the world, but common to many populations of the Himalaya (cf. Aziz 1978; Crook and Crook 1994; Crook and Shakya 1983; Goldstein 1971, 1976, 1981; Levine 1988a; Majumdar 1960; Peter 1963; Sander 1983; Schuler 1987). Most authors have focused on polyandry’s function as a social institution that limits overall population growth by reducing the number of females in sanctioned reproductive unions while maintaining a large household labor force of brothers, unmarried sisters, and children. These features have been considered adaptations to marginal ecological conditions (Aziz 1978; Carrasco 1959; Crook and Crook 1994; Goldstein 1971, 1976, 1981). Others such as Levine (1988a) and Schuler (1987) have focused on the relationship between polyandry and women’s power, which they see as constrained by this marital practice, since polyandry enhances the patrilineage and marginalizes a woman’s role within the household; some authors, however, have idealized it as contributing to female empowerment (NorbergHodge 1991; Rizvi 1983). Whatever the functions of this marital system, its cultural acceptability probably contributes to a kind of sexual and emotional ﬂexibility, as Ladakhis do not seem to have rigid standards for sexual relationships.4 Marriages are in the service of household wellbeing, and interpersonal relationships among spouses are subsumed to that.5 Internal dynamics of the household are also shaped by the primary institution of Buddhism, the monastery (gonpa). Most villages have at least a small gonpa, which is a place for religious instruction and meditation and a public center for worship ( puja). The gonpa is inhabited by lamas (monks) and presided over by the kushok, the incarnation of a holy man or Bodhisattva (a person who has achieved enlightenment but who

52

An Ecology of High-Altitude Infancy

has forsaken nirvana to return to the world in successive reincarnations to work for the liberation of all souls), who takes the title of Rinpoche. In households with more than one son, the youngest son was traditionally committed to a monastery at a young age to become a lama. An unmarried daughter can become a nun (chomo), but there are relatively few nunneries, so most remain within their natal household. Both nuns and, to a somewhat lesser extent, monks continue to contribute their labor to the household, even when they live separately, and both are expected to remain celibate. Nuns have an ambiguous position in Ladakhi society; those embodying the ascetic existence of someone devoted to religious contemplation are highly respected (then they are referred to ane), but others are marginalized as unmarried and nonreproductive women (Gutschow 1998). This ambivalence does not extend to male monks who are uniformly accorded high social status. New, nonagricultural avenues for employment in the civil service, tourism, and particularly the military have emerged and have changed traditional subsistence strategies. For example, the Ladakh Scouts, a specifically Ladakhi division of the Indian Army, provides jobs for many young men, and joining the military appears to be an increasingly acceptable substitute for entering a monastery. Army employment is lucrative, relatively speaking; families of Ladakh Scouts are eligible for widely coveted reduced-price commodities and foods ﬂown in from the Indian plains; and army employees receive pensions after a relatively short period of duty. At this point, most families have at least one son or a close relative who is or has been a Ladakh Scout or is involved in the military in some way. In addition, many agricultural families continue to pass on their landholdings to the oldest son, but others, especially in the Leh region, divide land among sons and sometimes daughters. In these cases, the khang chung become separate, independent residences of married sons and daughters. Monogamous “love marriages” have become more common as individuals get access to their own land and are able to make a living without having to rely on the extended household. While men are frequently employed in the military, clerical positions in government ofﬁces are often ﬁlled by women, as are staff positions at the hospital. Many young women from villages in the Indus valley come to the hospital for training as auxiliary nurse midwives (ANMs) and medical assistants, so that they can be employed as village health workers. It is not unusual for young women to move to Leh and live alone or with friends or relatives while in training for jobs as teachers or health workers. At the same time, as more men have been drawn into military occupations or the tourist industry, their labor is not available to the household, and agricultural work has become the responsibility of women. This becomes

Reproductive Health Research in Ladakh

53

particularly crucial when men are absent during the most critical points in the agricultural cycle (planting, watering, harvesting). Gender Dynamics The status of women has often been correlated with reproductive health parameters, including infant survival (for discussion of this in India, see Das Gupta 1995; Jeffery et al. 1988). Relative to cultural patterns in the rest of South Asia, women in Ladakh enjoy higher status in society and the household; data from this study (see Chapter 5) and others (Goldstein 1981) do not indicate a strong bias against females that would take the form of female infanticide or higher mortality rates among daughters. While they are considered congenitally weaker than and inferior to men and more subject to pollution through menstruation and birth (Gutschow 1998), women have considerable autonomy and power, especially within the domestic sphere. They travel to the bazaar alone or in groups with other women, freely engage in interactions with unrelated men, and constitute the majority of vendors at the local vegetable market on the sidewalks of the main bazaar in Leh. The produce is from the gardens that they spend much of their time planting, weeding, watering, and harvesting with little or no intervention by (or help from) male household members. Proﬁt from such mercantile activities is theirs, as is the wealth they bring to their marriage. Women usually control the distribution of dietary resources of the household and are also responsible for cooking meals and maintaining a constant supply of gur-gur cha (salt-butter tea) for the family and guests. At the same time, women’s authority and autonomy are constrained by the corporate nature of the household, which emphasizes the obligations of all members to household subsistence and prosperity. As marriage is generally virilocal, a new bride or mother often chafes under the watchful eyes of her husband’s family and her household obligations. A woman’s authority increases as she moves through her lifecycle, especially as her sons marry and bring in their wives, but she is particularly subject to the authority of the patrilineage in the early years of marriage and childbearing. Then she is expected to be subservient, hardworking, respectful, and deferential, which includes feeding herself last and ensuring ﬁrst the wellbeing of other members of the household. As males increasingly seek work outside the household, more labor responsibilities fall to women, and they are quick to articulate the burdens of women’s work, while contrasting it with the relatively carefree lifestyle that many men appear to adopt. Child care is not uniquely a mother’s domain but is shared by all members of the household, and children are treated as members of the

Photograph 4. Grinding barley into tsampa, using a water-driven mill.

Reproductive Health Research in Ladakh

55

household as a whole and not as the speciﬁc responsibility of their parents. Indeed, on a day-to-day basis, children’s household membership is highly ﬂexible as they often spend extended periods of time at the homes of relatives or neighbors. Siblings and grandparents are actively involved in child care as are elder members of the household, who may look after infants while the mother works in the ﬁelds. Children also are given responsibility for various tasks besides child care at a young age, such as washing clothes and dishes, cleaning rice and vegetables, tending animals, and carrying out the daily rituals in the household chodkhang, although in practice these tasks more often fall to girls. Most children, especially in the Leh area, now attend school. Diet Dietary practices are key to health in general, and to maternal and infant health in particular. Not surprisingly, the Ladakhi diet is constrained by what can be grown locally in this relatively isolated region. The staple carbohydrate is barley, roasted and ground into ﬂour called tsampa (Photograph 4). Tsampa is generally eaten at every meal in a variety of forms, the most common one of which is kolak (tsampa mixed with saltbutter tea, rolled into balls and dipped into a sauce of vegetables, potatoes, or dhal ). Other preparations include pawa, a stiff mixture of tsampa, water, and peas; marzan, tsampa mixed with butter and sugar; khambir, ﬂat breads made from a combination of barley and wheat ﬂour; skiu, a pasta-like dish mixed with vegetables and potatoes; and tukpa, a thick soup made of noodles, vegetables, and meat. Although barley products remain the core of most adult diets, over the past thirty years an increasing reliance on imported foodstuffs has developed, especially among communities with easy road access. Rice has become a key imported staple and is widely consumed, and children in particular often profess a preference for rice over tsampa preparations. Substantial seasonal variation in the climate of Ladakh results in variation in the foods that are available. Summer is the season of the greatest variety and abundance as home gardens come to fruition, and the opening of roads in June allows an inﬂux of fruits, vegetables, and processed and staple foods from outside Ladakh. Even before the road ofﬁcially opens, the Kashmiri-run fruit and vegetable market is stocked with bananas, cherries, eggplant, cauliﬂower, eggs, and other items carried by porters over the Zoji-La. These items tend to be prohibitively expensive but highly desirable after a long winter’s bland diet. In the mid- and late summer the Leh bazaar comes to life when merchants from Kashmir and the Punjab open their stores, the Kashmiri fruit and vegetable market sells

56

An Ecology of High-Altitude Infancy

a wide variety of exotic items, and local women sell their garden produce along the sidewalks of the main street. In the fall locally produced apples and apricots become abundant and are widely available. After the roads close, food diversity and availability decline dramatically. The basic winter diet consists of dried vegetables, stored roots, pulses and tsampa. These stockpiled supplies dwindle throughout the winter so that by May, many staples are unavailable (in 1990, tea, sugar, wheat ﬂour, and cooking gas ran out). Families who have or have had a member in the military have access to foods ﬂown into Ladakh throughout the year such as tomatoes, eggplant, eggs, canned ﬁsh, and canned fruits. Dairy products are an important component of the diet in Ladakh and serve as important sources of protein and calories. Most families keep local cows (balang) for milk, though its quality and quantity are often disparaged. In the late 1980s the government introduced a “jerseycow”6 subsidy program to increase milk production, but the project had short-lived success, as few cows remained in 1996. In the more remote areas nomadic groups herd yak and dzomo, which are prized for their high-quality milk. Milk (oma) is widely praised for its strengthening and regenerative qualities; it is made into yogurt, dried cheese, and butter or drunk fresh. Butter is the most frequently consumed dairy product, mostly in the form of gur-gur cha. Adults routinely drink at least a full thermos of this Ladakhi tea each day, and good tea is judged by the quantity of butter in it. Gur-gur cha is made from brick tea, which is boiled into a strong dark syrup and then diluted with boiling water. Together with a generous amount of salt and butter it is put into a churn and blended together. This is the staple drink of Ladakhis, and it accompanies virtually all social interactions. The household I lived with went through 500 grams of butter per day in gur-gur cha. Unlike other dairy products, which circulate among households in an informal exchange network, butter can be bought in the Leh bazaar, but it is very expensive (in 1990, it was 32 rupees/500 g; $1.75). Ration cards provide butter and other commodities to families that have a member who is a civil servant or ex-serviceman, another perk of government employment. For households in and around Leh, meat is most often purchased in the bazaar, and hence its consumption is related to household income. In rural areas, however, households have more routine access to domestically produced animal foods. Goat and sheep (mutton) are the most frequently consumed meats; chickens are occasionally seen in villages and are available in the Kashmiri market, but generally they are not favored for their meat. According to many people, meat consumption used to be more frequent, when more people herded animals and traded with

Reproductive Health Research in Ladakh

57

agricultural households. The social boycott of the early 1990s meant that Buddhists could no longer buy meat from local Muslim butchers. Wealthier Buddhist Ladakhis bought meat that was ﬂown in, but in 1990 it cost 80 rupees (∼4 dollars) per kilo. The wealthiest families in Leh may consume meat almost every day, but average consumption is most likely once every two weeks or once a month, usually in the context of special occasions. Health Care Institutions While diet provides a foundation for health, a diversity of institutions speciﬁcally devoted to health care exists in contemporary Ladakh. Indigenous resources include the amchi and lhaba, while hakim (Muslim healers), Ayurvedic practitioners, and homeopaths are also in residence. Improving access to health care has been an aim of the development policies of the government and private aid agencies, which have added biomedical facilities to the mix. The most widespread of the traditional health resources is Tibetan medicine, a formal tradition whose key text is the Gyud-zhi (Rechung 1973). Tibetan medicine offers a codiﬁed and institutionalized framework of nosology and treatment and derives from larger Buddhist precepts that shape people’s attitudes toward health and disease. Suffering is part and parcel of life from the standpoint of Tibetan Buddhism, which considers attachment to material existence and the notion of an autonomous ego to be the primary causes of suffering. The wheel of life, dramatically emblazoned on monastic walls, symbolizes an endless cyclical return to material existence and suffering, while only the most enlightened and compassionate souls (i.e., Bodhisattvas) are able to attain nirvana, a state in which the ego is abolished and the wheel of life is terminated. This broad template resonates throughout all aspects of life for Ladakhi Buddhists and manifests in a kind of stoicism in the face of pain and death and in the deemphasis on individuality, among others. But at the same time, Tibetan medicine provides mechanisms for alleviating the distress associated with speciﬁc ailments. Tibetan medicine is closely related to Ayurveda, the most ancient textual medical tradition in South Asia, and uses a broad framework of religious, somatic, and psychosomatic etiology for health and illness. Healing techniques include spiritual, psychological, or yogic practices such as meditation, moral development, and prayer. People may travel to gonpas to engage in these practices or engage the services of a lama to perform puja in the family’s chodkhang, a practice more common in wealthier households since it involves a substantial outlay of money and/or food.

58

An Ecology of High-Altitude Infancy

Such pujas are often conducted annually as a kind of preventive measure to protect and ensure the health of the household. Amchi are the primary practitioners of Tibetan medicine and are routinely used by villagers and residents of Leh for a wide range of ailments. Amchi either learn their craft from their fathers or grandfathers or receive formal training. Currently there is a movement to professionalize the tradition with new training and research centers and conferences to discuss problems in diagnosis and treatment, standardizing treatment, and recordkeeping. The bulk of an amchi’s practice is the diagnosis and treatment of somatic disorders, which are conceived as deriving from imbalance among the body’s three humors: air, bile, and phlegm. Humoral imbalance can be induced by diet, climate, life activities, psychological conditions, astrological forces, spiritual conditions, or karma (fate) (Clifford 1984), and is diagnosed through inspection of the pulse, tongue, and urine. Frequently humoral balance can be realigned through dietary modiﬁcation; other times the treatment is pharmacological, based on an extensive pharmacopeia of herbs and minerals derived from the Himalaya (cf. Buth and Navchoo 1988). Other traditional health practitioners are the lhaba (lhamo if female), often referred to as shamans or oracles because they become possessed by spirits and can cure sickness as well as predict the future or interpret past events. Lhaba are outside the formal Tibetan medical tradition, although they utilize many Buddhist symbols in their healing rituals; they are also socially marginalized, as people fear their power. They are not trained in a formal sense; they become aware some time in their development that they are prone to possession and apprentice themselves to established practitioners from whom they learn to control and channel their possession. Lhaba use a variety of ritual objects, along with chanting and rocking to achieve a trance state; when the spirit enters their body patients come forward to state their problem or question. In the case of sickness, the lhaba will often take a long slender straw and apply it to the region where the sickness resides, suck on it, then spit out a dark substance. As with amchi, patients pay for these services in accordance with the complexity of the diagnosis or treatment. Biomedical facilities are increasingly used along with traditional resources. The cornerstone of the biomedical establishment is the 150bed district hospital on the outskirts of Leh, which opened in 1980 and was the locus of much of my research (Photograph 5). The hospital is staffed by eight physicians, four assistant surgeons, and a support staff of ﬁfty nurses, medical assistants and nontechnical personnel, almost all of whom are Ladakhi. A wide array of services are offered, including maternal and child health/family planning, obstetrics and gynecology,

Reproductive Health Research in Ladakh

59

Photograph 5. The Sonam Narboo Memorial Hospital, the only public hospital in Leh District and the site of this study.

pediatrics, optometry, dentistry, internal medicine, radiology, pathology (very limited), and surgery, but the facilities are rudimentary. There is no running water or consistent electricity supply, and toilet facilities are unhygienic. The hospital is supposed to serve mainly as a specialty clinic to which patients with more complicated cases are referred by village health workers, but it defaults to a primary-care walk-in clinic for the population of Leh and environs. Care at the hospital is essentially free; patients pay a nominal fee (2 rupees) to be seen by a practitioner or to be admitted. In addition, it serves as a training center for health workers, both auxiliary nurse midwives (ANMs) and medical assistants. Beyond Leh, Primary Health Centres (PHCs), subcenters, dispensaries, and medical aid centers are spread throughout the district, conforming to the basic governmental health infrastructure common throughout India (Palriwala 1988). Unfortunately, these institutions are poorly coordinated and supervised; long distances between health centers, many of which are accessible only by foot, limit attendance by medical staff and patients alike (Palriwala 1988). Thus, preventive and curative

60

An Ecology of High-Altitude Infancy

services are offered only sporadically at the peripheral stations, depending on the availability of trained staff and medicines. Nongovernmental health institutions also have a presence in Ladakh. The Leh Nutrition Project (LNP), funded since 1979 by the British group Save the Children, focuses on providing mobile health and immunization services in the more rural areas. It also sponsors educational projects and small community development enterprises, distributes food and clothing, and donates medicines and equipment to the hospital. In the course of its work, LNP has gathered some of the few health statistics available for Ladakh. There is also a French-sponsored hospital that mainly serves the needs of the Tibetan refugee community. The Indian army does routine ﬁeld health checks and occasionally offers helicopter transportation of seriously ill patients to the Leh hospital. The Indian military has its own hospital that civilians may use, but they usually prefer the Ladakhi staff at the public hospital. There are also physicians in private practice in the Leh bazaar. Little is known about the major causes of morbidity and mortality in Ladakh, mostly due to the lack of systematic recordkeeping. Army outpost ﬁeld reports show that respiratory and intestinal diseases accounted for about 40 percent of clinical illnesses. Village survey data compiled by LNP showed that respiratory infections accounted for about 20% of morbidity, while anemia, gastroenteritis, and eye problems accounted for another 27% (Palriwala 1988). Of cases admitted to the hospital, the largest percentage was of lower respiratory infections (16%), then injury (13%), obstetric complications (12%), and pulmonary tuberculosis (10%). Causes of death, as enumerated in hospital ward registers, point to pneumonia as the most common (27%), then pulmonary tuberculosis and gastroenteritis (13% each). Caution is advised in the interpretation of these statistics, since the actual numbers of hospital deaths was very small (22 deaths total) and these diseases may not be more broadly distributed in the population. In rural areas, LNP estimated an infant mortality rate (IMR) of 102 to 110/1000, and two other village demographic surveys generated similar estimates between 83/1000 and 190/1000 (Attenborough 1994; Elford 1994). From a nutritional standpoint, the Ladakhi diet is based heavily on fats and carbohydrates for energy, mostly from grains and dairy products, and derives its protein from green vegetables, legumes, and dairy products. Most clinicians and researchers conclude that malnutrition in the form of protein and energy deﬁciency is not widespread. Indeed, a study of agricultural production in Zangskar estimated that 4000 kilocalories per day could be provided, which would be more than adequate for human needs even for energetically expensive agricultural tasks

Reproductive Health Research in Ladakh

61

(Osmaston 1994). Average intake, however, was found to be much lower – about 1500 kilocalories/person per day, which is insufﬁcient for most adults (Attenborough et al. 1994). This discrepancy has not been fully explained. Goiter is rarely seen, although apparently it was more common before iodized salt became widely available about thirty years ago (Stobdan 1990). Salt is consumed in abundance, mostly in gur-gur cha, and a large household can go through three to four kilograms of salt in one month. Hypertension, possibly related to high salt consumption, is reportedly high in Leh (Meyer 1981; Norboo and Yahya 1988). Irondeﬁciency anemia seems to be a problem for both women and children, as may be Vitamin B-12, Vitamin A, and calcium deﬁciencies (Attenborough et al. 1994; Meyer 1981; Stobdan 1990). Reﬂexivity in Human Biology Research The preceding description of Ladakh, especially the area in and around Leh, locates my research on reproductive health in geographical and social space. It is also worth brieﬂy describing the process by which I came to weigh babies in the stark delivery room of a small public hospital in midst of the western Himalayan desert, and the daily life of this particular research experience. Descriptions of the actual “work” of human biological ﬁeld research are few. As a student, the only template I had for conducting anthropological ﬁeldwork derived from ethnographies, and I had virtually no insight into what human biologists did in the ﬁeld; their work described methods of “data collection” without reference to the larger social, cultural, temporal, or personal context in which the research took place. More thorough description of ﬁeld research is essential to human biology, which grounds itself squarely within the life sciences. One of the hallmarks of scientiﬁc research is that it should be unbiased and replicable – anyone else designing the same project and carrying out the same protocol should come up with the same results. And if the project is carefully designed and executed, the interpretation of the results should be consistent as well. However, because of the unique nature of anthropological work, its use of long-term ﬁeldwork during a particular historical moment, and the commitment of time and resources it takes to do this kind of research, it is rare that anyone attempts to replicate a project. Often a single account is the only one dealing with the subject matter in a particular geographical region, and so it stands as the authoritative statement. Thus modern anthropology – especially that which makes claims to scientiﬁc rigor – demands a sincere attempt at reﬂexivity and a greater degree of transparency, so the reader can see how the research might

62

An Ecology of High-Altitude Infancy

be biased in some way by the unique experiences and perspective of the researcher. Background to Fieldwork I came to work in Ladakh through the convergence of two distinct experiences. The ﬁrst was a brief visit to Ladakh in 1985, undertaken as part of a larger trip to South Asia. In the guidebook Ladakh was touted as a place of lunar beauty, impressive mountain peaks, friendly people, and exotic Tibetan Buddhist culture. With the South Asian monsoon, and its oppressive heat and humidity approaching the south side of the Himalaya, this dry, cool Shangri-la seemed to be the ideal place to go. Although the ﬂight into Ladakh was less than an hour from New Delhi, upon landing it felt worlds away from the densely populated, hot and humid north Indian plains. Having always loved mountains, I initially was struck by the beauty of the Himalaya, but subsequently by a more humanistic question: How did people manage in this unlikely Shangri-la, cut off from the rest of Asia for much of the year, especially when the hypoxic stress I experienced left me struggling to catch my breath after only the briefest exertion? That ﬁrst trip lasted only a week, enough for only the most superﬁcial insights, but Ladakh left an indelible impression on my memory as I returned to the United States and pursued PhD work in medical anthropology. In the course of that work, my intellectual interests in diet, reproduction, and human adaptation solidiﬁed and I began to formulate a research plan that would merge with my desire to return to Ladakh for a more extended period. It became clear that there were some important research questions that could be answered by work there. Relatively little was known about high-altitude adaptation in the Himalaya at that time, and anthropological work in this mountain system had been largely conﬁned to small-scale populations in Nepal. Based on some research I had done with Indian census materials, it appeared that population growth and fertility rates were very low, while there were virtually no data on mortality, especially infant mortality. Further, I believed that the unique Ladakhi diet and arid hypoxic environment must have some important implications for nutritional status and infectious disease, two key mediators of child health and survival, which had not yet been elucidated. While these broad issues set up a universe of interesting topics for a potential research project in Ladakh, I settled on the more speciﬁc question of how pregnancy and infancy were affected by the sociocultural and ecological conditions of Ladakh. A focus on newborn characteristics would extend the studies of birth size and altitude to the Himalaya

Reproductive Health Research in Ladakh

63

and show how they related both to maternal characteristics and infant health and survival. I chose to investigate the sources of variation within the Ladakhi population to ascertain how individuals or groups were differentially affected by ecological conditions, as indicated in the speciﬁc questions outlined in Chapter 1. I began the preparations for ﬁeldwork during my ﬁrst return visit to Ladakh in 1989. My ﬁrst and unexpected task was to rebuild my language skills. My two years of Tibetan language instruction generated nothing but blank stares, looks of confusion, or giggles. Although Ladakhi is a Tibetan language, spoken Ladakhi is quite different from spoken Tibetan, which is a tonal language. The multiple consonants used in the written form of Tibetan were replaced, over time, with tones in spoken Tibetan. This did not occur in Ladakhi, which retained a closer afﬁnity to the written form of Tibetan. Tibetan and Ladakhi thus share some vocabulary and grammatical constructions, but pronunciation is quite different, and Ladakhi has incorporated many words from north Indian languages. Therefore I had to spend much of my time learning Ladakhi, which I did with the help of the members of the household where I was staying. I initially imagined a study of home births, but when I contacted Dr. Tsering Ladhol, the obstetrician/gynecologist (Ob/Gyn) at the hospital, she indicated that a large number of births were now occurring at the hospital, making it possible for me to work there. She was interested in my study from a clinical perspective, insofar as an analysis of birth outcome could help her provide better pre- and postnatal care. The resources did not exist for her or her staff to collect systematic and detailed information on newborns or follow up on them, and so my study would provide her with valuable data without interfering with the busy routine of the obstetrics staff. Although working in the hospital had numerous advantages, it did mean that I missed out on the opportunity to observe traditional birthing practices and that the clinical sample would result in some biases such that it probably would not be representative of reproductive health among a broad cross-section of Ladakhi women, an issue to which I return in the next chapter. During the ﬁnal month of this initial stay in Ladakh, things began to look less promising for my project. Buddhist-Muslim hostilities erupted in Leh, neither tourists nor Ladakhis were allowed into the streets, and the bazaar and the shops were closed. There were both police-enforced curfews and Buddhist strikes, both of which disrupted life in the capital and transportation between Leh and other towns. People insisted that such behavior was unlike anything that had ever happened in Ladakh, and they were confused and astonished by the course of events. The future seemed tenuous: international conﬂict between India and Pakistan was

64

An Ecology of High-Altitude Infancy

brewing, and internal tensions threatened to place Ladakh under military rule and close it off to foreigners. I was troubled – both by the negative effects of these hostilities on people I had become close to and, more selﬁshly, by fears of how this would affect my own plans to return the following year to carry out ﬁeldwork. Although the internal conﬂict came to a head in the fall after I left, the situation did stabilize somewhat before my return the following spring. Nonetheless, the fact that my ﬁeldwork took place during this period of social turmoil highlights the ways that the results are historically delimited. What was true about the determinants of reproductive health in 1990–1996 may no longer be true today, or may not have been true ﬁfty years ago. The fact that I was a reproductive-age woman whose work was largely carried out among women and their newborns also biases the conclusions that can be drawn from this work. I had relatively easy access to the world of women and almost all the people in the study were women; males appear as husbands, or sons, healers, but not as central ﬁgures in the research and the story I have gleaned from it. Most statements that generalize about Ladakhis are thus really about women’s experiences, and more narrowly, their reproductive experiences, although certainly these have populationwide ramiﬁcations. Routines of Fieldwork Filled with optimism, I returned to Leh in mid-April 1990 to begin work in earnest. My enthusiasm turned to depression when I discovered the Ladakhi spring to be cold and dreary. The ﬁelds were barren and the few trees were still leaﬂess. The only foods available were barley, potatoes, turnips, dried beans, and rice, and I did not have a fresh vegetable until I had been there for over a month. Existing supplies were running low, and most of the shops in the bazaar were closed until the road reopened, which was not likely until mid-June or so. None of the tourist restaurants were open. Even access to the ubiquitous Central Asian ﬂatbread was problematic when the wheat ﬂour for the commercially produced bread ran out. I came down with a nasty respiratory infection and huddled in my sleeping bag in the sunny patches of my room despairing of my hunger, ill health, and the prospect of a long lonely stretch of ﬁeldwork. But as spring wore on, I reestablished the relationships formed the previous summer and began working at the hospital; I slid into a routine that, for the most part, had a comfortable rhythm to it. For me, the best way of describing the ﬁeldwork experience is that it was not unlike having a job. I would awaken in the morning in the room I rented in a house in Leh and eat breakfast with the adults and children

Reproductive Health Research in Ladakh

65

before the latter went off to school. Like many others, the household I lived with had converted part of their house to a guesthouse for use by tourists during the summer season. I had ﬁrst stayed there in 1989 and found it to be a more or less ideal living situation. The family was welcoming to me and willing to help me with my spoken Ladakhi in exchange for my helping the children with their school lessons. I ate with them most mornings and evenings, sometimes helping with the evening cooking. I occasionally helped the women with weeding the garden, planting potatoes, or watering the barley ﬁelds. I was called “Ache Angmo” – older sister Angmo, the unmarried sister who could be counted on for help around the household and in the ﬁelds and to knit sweaters and sew school uniforms for the household children. I became just another member of the household who went off to work each morning and returned in the afternoon, and contributed to domestic tasks as needed. I went to the hospital six days a week, arriving just before 10:00, when other staff arrived and the hospital clinics ofﬁcially opened. Consequently I began to feel, and was treated, like a regular employee. I was allowed through the guarded doors into the wards and established a base in the labor room nurses’ station, a dark room with benches and a kerosene stove for making tea. The nurses rotated through the wards, so every month or so there was a new set of nurses in the labor room. There was also a group of ANMs in-training who cycled through and who occasionally helped me with my work. The instability of the stafﬁng created minor problems for my research, as each new group had to be instructed on the protocol, and some of the staff were more enthusiastic and competent than others. The actual measuring and interviewing is described in some detail in the next chapter. Sufﬁce it to say here that my “job” consisted of checking the birth registry for births that had occurred in the past twenty-four hours, and visiting the obstetrics ward with the nurses or students, who would then escort willing mothers and newborns to the labor room for assessment. I conducted the interviews in Ladakhi, sometimes with help from my assistants. In the case of the few women who were not native Ladakhis or Tibetans, the nurses translated my questions from Ladakhi into Hindi. Some days there were no births and I went home early or did something else at the hospital; other days there would be four or ﬁve births that would be part of the study and I would be quite busy until late afternoon. The labor room might be occupied by births in progress or other procedures, and I always made sure that my project was secondary to the clinical work of the hospital staff, which often meant waiting several hours for the labor room to be free or missing a set of babies and mothers. Often it provided an opportunity to learn something new about a different aspect of reproductive health.

66

An Ecology of High-Altitude Infancy

One of the drawbacks to being accepted as a routine member of the labor room staff was that the staff and patients expected that I had clinical skills. It was assumed, for example, that I knew how to give an injection, check on the progress of labor through cervical dilation, or even deliver a baby or cut or sew up an episiotomy. I had none of these skills and no formal medical training in midwifery or nursing. Most of the staff assumed, with good reason, that anyone who worked at the hospital must have some medical expertise. My status as a foreigner who was afﬁliated with the obstetrician and had the approval of the hospital superintendent and my facility with an array of equipment implied that I had a certain amount of authority, a conclusion that my childish spoken Ladakhi occasionally belied. My attempts to identify myself as an anthropologist were unsuccessful, as this label had little meaning. I tried my best to explain that I was interested in how newborns made the transition to extra-uterine life in the context of a mountain environment with little exogenous oxygen, and how women’s diet and behavior during pregnancy affected newborn health, but that I was not a medical practitioner.7 After a few months of my presence in the labor room, most of the nurses understood my role there and no longer requested my assistance in clinical procedures. At the same time, as I witnessed numerous labors and deliveries, I began to feel more comfortable in the delivery room context and with the general procedures of childbirth. I liked to think of my status in the labor room during a birth as a kind of doula – a labor assistant – and after the birth as a researcher doing assessment of newborns and interviews of mothers. One outcome of this experience for me was a complete demystiﬁcation of the birth process. After witnessing numerous births, I can only describe it as a very mechanical process, characterized by stages of cervical effacement and dilation brought about by contractions, and ﬁnally the pushing out of the baby. Birth required patience as these events took place in a temporal context that could not be radically altered. Certain complications were not uncommon. In other words, I began to take on the nurses’ attitudes toward birth and to see it as a kind of routine, largely predictable process. Prior to this project, I had an overly romanticized notion of birth in non-Western cultures, where women would be surrounded by supportive and experienced older women and gently coached through this important rite of passage. That may more closely approximate the traditional form of birth in Ladakh, but birth in the hospital labor room was a stark contrast to this ideal. Women most often were alone during labor; some were accompanied by husbands or more often, mothers-in-law or other female relatives. They were conﬁned to metal beds with thin mattresses and covered with heavy quilts. Nurses would occasionally check to see

Reproductive Health Research in Ladakh

67

how their labor was progressing by assessing their cervical dilation and fetal heartbeat, and then they would return to the nurses’ room to rest, chat with other nurses, and wait for the labor to progress to the point where delivery was imminent. The nurses who were currently doing a rotation in the labor/delivery room conducted most births; the obstetrician was involved only in the case of complications. There was a tremendous amount of variation in the skills of the nursing staff; some had had very limited experience with birth, having been trained in other specialties. I cannot say that they seemed particularly sympathetic to laboring women; there was tacit agreement that labor was difﬁcult, no doubt about it, but that women had a responsibility for maintaining a stoic demeanor during the process. Women who moaned or cried out “excessively” were chided to endure, or chastised. This kind of emotional reserve is expected behavior in Ladakh; strong expressions of emotions are discouraged. At the same time, it was clearly evident that women, especially those who were young and primiparous, were terriﬁed and in a great deal of distress. For the actual birth women were moved, alone, to the delivery room where they endured the rest of the process on a ﬂat enamel bed with their legs up in stirrups and no pain relief. Once the baby was born, its nose was suctioned or given oxygen brieﬂy, if necessary; it was hastily wiped off, and quickly wrapped in a ﬂannel sheet provided by the mother. Bundled up in this way, the baby was handed directly to waiting relatives (usually the mother-in-law) while the placenta was delivered and repairs made to the perineum. The mother often looked rather bewildered, as the baby disappeared before she got a chance to get a look at it. Sometimes, the nurses failed to note or convey to the mother the sex of the baby before handing it off. All in all, I would have to say that hospital birth in Leh was an alienating experience. However, I did not ask Ladakhi women directly about their birth experiences in the hospital, and I have limited insight into their expectations. I am, perhaps, more horriﬁed by this scenario now than I was then. As I said, I became familiar with a kind of ennui toward birth while I worked there, but while writing and revising this book, I was pregnant twice and gave birth to my two sons, and the terror on the faces of laboring women whom I could only help with weak words of encouragement is more vivid to me now in light of my own labors. Beyond the trials of labor, there were other occasions for this work to be unsettling at best, and horrifying at worst. Part of the risk of giving birth in Ladakh is that it is not infrequently linked to death. Stillbirths were not uncommon: of the 563 births that took place in 1990, 25 (4.5%) were stillborn. While I worked there, three women died immediately postpartum. Then of course, there were infants who died after birth, some within

68

An Ecology of High-Altitude Infancy

hours or days, others weeks or months later. Since a central concern of my research was the survival probabilities of newborns, I expected there to be deaths, but the actual experience of them was something I found myself unprepared for, both in the physical reality of a dead fetus or infant and the emotional consequences. I was constantly afraid that a particularly vulnerable-looking newborn would die in my hands as I measured him or her (and they all looked frail compared with the robust newborns I had worked with in the United States). I made the decision early on that newborns or mothers who were obviously ill would not be included in the study. I did not want to place them at whatever marginally increased risk they might have incurred by being taken out of the ward. The hospital itself was not a pleasant place. Constructed and maintained with a minimum of governmental resources yet having to serve multiple functions for a large population without access to good local primary care, it lacked many of the amenities that we take for granted in the most rudimentary hospitals of the West. During the time I worked there, there was neither running water nor regular electricity, although the hospital had its own backup generator for emergency operations. There were no adequate toilet facilities; the nurses had access to locked rooms that were nothing but ﬁlthy earthen ﬂoors with holes in them, and which, therefore, few people used. If you needed to use the bathroom you were directed outside. The few outhouses were less appealing than the grounds of the hospital, where goats and cows roamed in search of clumps of grass stubble and which were a mineﬁeld of human and animal feces. In the hospital, needles and other instruments were boiled for sterilization, and basic drugs (antibiotics, pitocin) were usually available. The gynecology ward housed both gynecological and obstetrics patients in one large communal room. In uncomplicated births, women spent about twenty-four hours there with their newborns nestled beside them and with family members coming for extended visits. The Ob/Gyn clinic was always swamped with patients and waits of several hours were routine. Prenatal visits took up two days each week; other days were taken up with patients coming for infertility, family planning, and other gynecological concerns. The success of this clinic, evidenced by the rising numbers of women who seek prenatal care, deliver in the hospital, or use family planning services, can be largely attributed to Dr. Ladhol (cf. Wiley 2002). The only Ob/Gyn in the district, she is from one of the most respected families (the Kahlon) in the traditional social hierarchy of Ladakh and she has dedicated her life to her work, eschewing the normative behavior of women in Ladakh – that is, marriage and motherhood. Despite her social position, Dr. Ladhol is an extremely humble individual who derives a great deal of personal satisfaction from

Reproductive Health Research in Ladakh

69

her work, which she sees as more a form of public service than a career or job. Working with her was one of the most pleasurable aspects of my research, and I continue to value her kindness, collegiality, and integrity. There were also more informal aspects of ﬁeldwork. I interviewed women I met through the family I lived with about their reproductive histories. I closely observed mothers and infants wherever I went in Leh, in the villages around Leh, and in villages off the road that I visited on treks in the mountains. During visits to households with children in the study I observed interactions between caretakers and children and gathered information on child health issues, household composition, and activities. The data derived from these observations is less systematic that that from the hospital study because these visits were opportunistic and not formally scripted. Nonetheless they provide important contextual and ethnographic data for child health. More informal interviews and observation occurred with amchi, lhamo, physicians, and other health care workers. During a return trip to Ladakh in the summer of 1991 and ﬁve years later in 1996 I spent much of my time trying to locate children who had been born in the study during 1990. I also spent numerous hours going over pediatric records looking for patterns of disease among children and for references to children who were born in the study in 1990. I collected data from the birth and prenatal records from 1990 to 1995 to generate a more complete picture of reproduction in Ladakh beyond the sample. I was also able to compile information on several aspects of family planning. Beyond the hospital routine and other aspects of the research, I experienced frustrations and periods of depression, especially when I had spent a long fruitless day at the hospital or when one of the babies in the study died. Anxiety, loneliness, and hunger were all part of my life in Ladakh. I keenly missed opportunities to discuss my research and get feedback on it from anthropological colleagues, but at the time, electricity was erratic; there was no e-mail, fax, or phone; and the mail moved at a snail’s pace. My anxieties were pervasive about data collection (Was I being thorough enough? Would I be able to measure and follow up enough babies?), the constant surveillance by armed Indian soldiers posted throughout Leh who it seemed scrutinized my daily commute, and access to food. I ate breakfast at home, but I regularly missed lunch while at the hospital, and the demands of agricultural tasks often meant that dinner was not served until late evening, by which time I was starving and mildly hysterical. My solution to this was to help prepare dinner; this did not make it happen any earlier, but it did reassure me that I would eventually eat! It also provided me with an opportunity to learn about diet and food

70

An Ecology of High-Altitude Infancy

consumption by a household in Ladakh that was fairly typical of those in the area. Such were my daily experiences of ﬁeldwork, each shaped by anxieties but also small triumphs and satisfactions. The morning’s wonder of what I would encounter in the labor room (Would it be a “baby boom” or “bust” day?), the fun of interviewing mothers and holding their newborns, the relaxed pleasure of the company of women and children in the sunny courtyard of the household, the harvesting of the year’s ﬁrst produce from the garden, the anticipation of mail from my friends and family, and a visit from some friends and my sister were all parts of this process.

4

Big Mountains, Small Babies

Babies are born exhibiting the full range of human variation. They differ with respect to weight, length, head size, fatness, trunk and limb circumferences, skin color, hair (color and absence/presence), personality, and in innumerable other ways. They also vary with respect to age. Although the way we assign ages to individuals is based on everyone starting at age 0 at birth, in reality newborns are already of different ages, some having been in utero for over forty-two weeks, some as little as half of that! In the United States, certain characteristics of newborns are particularly closely attended to by family members and friends. The ﬁrst questions asked about newborns are “Boy or girl?” “How much did he/she weigh?” and, less often, “How long was he/she?” These statistics appear on birth announcements, which is the only time in the lifecycle when the speciﬁcs of one’s size are touted so publicly. To medical personnel, other measures are equally important. The APGAR score, done at one and ﬁve minutes after birth, assesses color, pulse, reﬂex irritability, muscle tone, and respiratory effort. Blood samples may be taken to ascertain glucose and bilirubin status. All of these measures are culturally and medically signiﬁcant in their own ways and generally help ascertain the health status of the newborn. The newborn measurement most often used as a summary index of a newborn’s condition is weight, but other measures provide different or more speciﬁc information. Skinfolds measure subcutaneous fat and are more precise measures of energy reserves, but they are more difﬁcult to do and vary from site to site on the body. Head circumference can provide a rough estimate of gestational age and brain development. Although newborns at the low and high end of the size spectrum have the highest risks of various maladies, it is important to remember that there is a wide range of what can be considered “normal” and “healthy” anthropometric measures. Much of the variation among newborn size cannot be accounted for in a systematic way because it is due to the unique combination of genetic information that every individual inherits from his or her parents 71

72

An Ecology of High-Altitude Infancy

and the contingencies of the prenatal environment. But some of the differences can be explained by the progress of the pregnancy, which is profoundly affected by maternal characteristics and behaviors. These, in turn, are complexly determined by social, cultural, and natural environmental aspects of maternal ecology. Through various pathways these ultimately inﬂuence the growth and development of a fetus and thereby set the stage for the health and survival prospects of an infant. But prenatal conditions also have effects well beyond the infancy period. In a series of papers, Barker and colleagues (e.g., Barker 1995) have demonstrated that biological markers of prenatal life such as birthweight may continue to exert their inﬂuence throughout life and affect the risk of cardiovascular disease, respiratory disease, and diabetes, among others. Research such as this makes the investigation of the determinants of birth outcome ever more urgent. Pregnancy and Birthweight Birthweight is the most widely used descriptor of birth outcome in the public health and clinical literature, as it is a summary measure of the resources acquired by the fetus in utero, and, in turn, it reﬂects the biological resources that a newborn has to meet the challenges of life outside the womb. There is a wealth of evidence that birthweights at the low and very high end of the weight spectrum incur increased risks of mortality (McCormick 1985). Lower birthweights are a more signiﬁcant global health problem, as they are much more frequent than very high birthweights. The World Health Organization deﬁnes birthweights less than 2500 grams as low birthweight (LBW) and LBW as a primary risk factor for infant mortality (World Health Organization 1977). Approximately 16% (20.5 million) of all infants born each year weigh less than 2500 grams; over 10% of all newborns in developing countries are LBW, and the rate is highest in South Asia (de Onis et al. 1998). Where such high percentages of LBW newborns are evident, the World Health Organization urges public health action to prevent fetal growth retardation and thereby reduce infant mortality. Lower birthweights among full-term newborns stem from impaired resource ﬂow to the fetus and fetal growth retardation, but the means by which birthweight is related to maternal resources are somewhat complicated. The fetus is neither a “perfect parasite,” meaning that the fetus can extract whatever resources it needs from the mother regardless of the cost to her health, nor is its health perfectly correlated with that of the mother. The most useful determinants of birthweight are maternal age, parity (the number of previous births a woman has had), nutritional status

Big Mountains, Small Babies

73

(both current dietary behavior and weight, height, or fatness), infectious disease history, whether she smokes, the altitude where the mother lives, and sex and gestational age of the newborn (Cogswell and Yip 1995).1 Many of these factors are synergistic and their independent effects are difﬁcult to assess. From a comparative perspective, the birthweight distribution that characterizes any given population is likely to differ from other populations as a function of a distinct array of these variables. Some of these variables reﬂect conditions that exist independently of a given pregnancy, but others act during the process of gestation and are affected in important ways by cultural norms or economic constraints that shape a woman’s behavior during pregnancy. Recognition of pregnancy as a stage in a woman’s lifecycle that requires some degree of behavioral change is common to all human societies. Dietary changes are common aspects of the management of pregnancy, although the list of “do’s” and “don’ts” varies cross-culturally. Many foods are prescribed or proscribed based on the effects they are thought to have on the growing fetus, and hence dietary changes of pregnancy provide insight into ethnomedical and ethnophysiological understandings. Laderman (1983) has suggested that these dietary expectations, which may include cravings for unusual foods, also serve as a kind of “cultural map” of the process of gestation, to allay anxiety and fears and protect the health of mother and infant. Such cultural mandates for dietary change during pregnancy occur within a larger context of subsistence practices, work, and the daily and seasonal rhythms of life. There may be, for example, signiﬁcant variability in food availability as a function of the agricultural cycle or other economic factors that vary seasonally (such as employment). Likewise, food intake must always be evaluated relative to energy expenditure, and the latter often varies in a seasonal pattern, sometimes quite markedly. Thus, while there are often ideologies or “cultural maps” for pregnant women that recommend behavioral change in the service of successful birth outcome, it is important to investigate what women are actually doing during pregnancy. What people say pregnant women “should/should not” do is always contingent on the material constraints on what they “must” or “can” do. Collectively these conditions are likely to impact birth outcome in either positive or negative ways. Measuring Mothers and Newborns To understand the determinants of birthweight in Ladakh, I took neonatal and maternal measurements in the Leh hospital labor room and interviewed women about their pregnancies within twenty-four hours after they had delivered. This occurred over a six-month period in 1990 (April

74

An Ecology of High-Altitude Infancy

through October). During this time, there were 240 deliveries in the hospital, and of these, 168 mothers and infants became part of the study. These mothers and their newborns, who were recruited to participate in the study by nurse assistants, were opportunistically sampled. That is, I did not try to dictate the composition of the sample by selecting only mothers and newborns with certain characteristics (e.g., sex, ethnicity, age, parity). My only two criteria for inclusion in the study were that (1) newborns had to be less than twenty-four hours old to minimize the effects of normal postpartum weight loss, since newborns may lose up to 10% of their birthweight over the ﬁrst days after birth; and (2) neither the mother nor the newborn could be seriously ill, as diagnosed by the obstetrician or nurses. When the mothers and newborns were escorted to the labor room by the nurse assistants, my ﬁrst task was to explain in detail what I would be doing. Most mothers were eager to have their newborns weighed and measured, and they were willing to answer questions about themselves and their pregnancies. Some women consented to the assessment of their newborns but were themselves too tired to participate so soon after the birth. I tried to maintain an informal atmosphere during the assessment, but I also tried to be as efﬁcient as possible in doing the neonatal measurements so as to return the baby to the mother as soon as possible. The ﬁrst task was the neonatal assessment, which involved taking the following measurements: circumferences of the head, chest, arm, calf; skinfolds, which are measures of subcutaneous fat, at four sites (subscapular, tricep, suprailiac, and calf); weight; and length.2 Neonatal anthropometry is a challenge, as infants, including newborns, generally do not like to lie still for any length of time when they are awake. Swaddling them in a ﬂannel sheet helped during the weighing procedure and reduced their wriggling about on the tray, which helped me get an exact weight. Circumferences were not hard to do, although the head circumference measure was complicated by the cranial distortion that often occurs during the birth process. Some of the births had required vacuum extraction, which resulted in a temporary bulbous protrusion on the top of the head. Skinfolds, while not painful, do involve slight discomfort, since they lightly pinch the skin to measure subcutaneous fat. They also had to be done three times at each of the four sites to assure greater reliability of the measurement. Moreover, the Harpenden skinfold calipers I used to measure skinfolds look like large metal clamps, which often caused the mothers to look on with some alarm. If the mothers seemed concerned when I got out the calipers, I demonstrated their use on them ﬁrst, to show that they were not painful or damaging. I encouraged the mothers to assist in all of the newborn assessment by holding or soothing their

Big Mountains, Small Babies

75

babies. An accurate length measure was probably the most difﬁcult to procure. Newborns instinctively resist having their legs straightened out, probably because they have been curled up in the fetal position in utero. Trying to get the legs straight, straightening the head so that the crown was against the headboard, and taking a measurement often required several pairs of hands! Throughout the entire process, I tried to keep the babies covered up as much as possible. I had quickly learned that Ladakhi mothers are very concerned about their babies being cold, and the labor room, though heated with solar panels, was not very warm. Indeed, when the babies were handed to me, they were often swaddled in several layers of cotton ﬂannel, wool, and acrylic blankets, and they all wore knitted hats. Newborns are not washed after birth in the hospital, and many were still covered in vernix, a white, cheesy substance that protects fetal skin. There was often meconium, the thick black substance that makes up the newborn’s ﬁrst stools, on their skin and wrapping. Since warm water was not usually readily available, I would wipe away as much of this as possible with a dry cloth. Once weighed, measured, cleaned, and assessed, the newborns were quickly wrapped back up. The obstetrician had described to me how newborns were traditionally wrapped in dough made from barley ﬂour and water, and wiped clean with butter, which seemed preferable to the materials available in the hospital. After the newborns were resettled and soothed, the mothers were measured. Like their infants, women were often dressed in several layers of clothing, starting with a cotton top and pants, over which they wore one or two sweaters. On top of these, they wore the ubiquitous Ladakhi gos, an overcoat with a gathered skirt, usually made of maroon wool or a darkly colored velvet material. Over the gos women often wore a wool or ﬁberﬁlled vest. Women were asked to remove their outer coats and sweaters, and were weighed on a standard adult spring scale wearing their cotton shirts and pants. A correction factor was subtracted from their weights to accommodate the weight of their clothing. Height was assessed to the nearest millimeter using an adult height rod, which I had attached to the wall of the labor room. Upper arm circumference and triceps skinfold measurements were taken on their left arm, as with infants. After this relatively brief measuring session, women put their outer clothes back on and sat down with me for the interview. Some interviews lasted only ten minutes or so; others went on for a half-hour or more. Sometimes if a female relative had accompanied them, both would respond to the more general questions and the conversation would often expand. If the nurse assisting me was from the same village as the woman, she would often join in the discussion. The interviews ranged across a

76

An Ecology of High-Altitude Infancy

variety of topics related to pregnancy and reproductive health. They took place in Ladakhi, with occasional translating help from my assistants. I recorded my notes in a mixture of Ladakhi and English. In the course of the interview, the basic information from the mother’s hospital chart and prenatal records was reviewed to insure its accuracy. These records contained the following data: the infant’s sex and estimated gestational age (EGA: estimated by fundal height at the time of admission for labor), type of delivery, maternal age, parity (number of prior live births), gravida (number of prior pregnancies), residence, number of prenatal checkups, blood pressure from each checkup, hemoglobin levels from different times during pregnancy, date of last menstrual period (LMP), vitamin and mineral supplements taken while pregnant, tetanus innoculations, and special health conditions or prescriptions. For multiparous women, much of the time was spent going over the reproductive histories summarized in the prenatal records, checking the age, sex, and survival status of previous children. If any of these children had died, I asked women to elaborate on the events leading up to the death, a subject that most women talked about in fairly vague terms. During the interviews women also provided me with the following information: length of residence in Ladakh, household size and composition, education and occupation (including that of their husbands), size of landholdings, religion, and ethnicity. The household questions proved to be the most difﬁcult. Given the complexity and ﬂuidity of Ladakhi households and extended families, women were bafﬂed by my questions about household membership, and I was often thrown into utter confusion by their answers regarding the relationships among household members at any given time. After these questions I went on to ask about diet and activity during pregnancy, generally looking for changes from prepregnancy patterns. Dietary questions were straightforward (cf. Quandt 1987 for a description of these methods for determining dietary intake). First I asked about general consumption patterns – that is, “Did you eat more/same/less during your pregnancy than when you were not pregnant?” I then homed in on the frequency of consumption of nutrientrich foods such as meat, eggs, and milk (i.e., how often did you eat meat while you were pregnant – never, rarely, occasionally, or every day?). I then would ask the woman to recount what she had eaten over the past twenty-four hours. This information is not really helpful in understanding eating patterns during pregnancy, but it does describe postpartum dietary consumption. Questions about activities followed, using the same format: Did you work more/the same/or less while you were pregnant? For most women who identiﬁed themselves as housewives or farmers (almost 80% of the sample), the term “work” meant agricultural work. During the

Big Mountains, Small Babies

77

seasons when the study was carried out (spring through fall) this entailed planting, watering, weeding, and harvesting crops from the ﬁelds and house gardens and tending to domestic animals. For about 7 percent of the women it meant engaging in agricultural labor or hard physical work (such as carrying heavy loads) for a wage. For all women it included domestic chores such as cooking, washing clothes and dishes, and cleaning the house. As it turned out, activity was one of the most important variables affecting birth outcome, and I wish I had collected more precise data on what individual women were doing while they were pregnant (such as time budgets, or actual measures of calories burned during various tasks). Last, I inquired about health problems experienced during the pregnancy and the practitioners that had been sought out in response to these or other problems. Sample Characteristics It is crucial to know how this sample of 168 mothers and newborns is related to the overall population of mothers and newborns in Ladakh. That is, does this sample reﬂect the variability and trends in the entire population of mothers and newborns, and thus can it be used to describe a general “Ladakhi reproductive health experience”? Or does the sample speak only for itself or some small segment of the population? Answering these questions is complicated by the lack of information on infant mortality, obstetric complications, and reproductive histories from the general population, and therefore it is difﬁcult to determine whether the health characteristics of the sample varied from those of the entire population of mothers and infants. Table 4.1 describes some of the socioeconomic characteristics of women in the sample relative to the data that can be gleaned from the 1981 Indian national census of Ladakh.3 Only census data from Leh district were included, as this is the area from which most of the sample derived and it is also where most people in Ladakh reside. This area encompasses eastern Ladakh and does not include the area around Kargil at the western end of Ladakh, or Zangskar, the mountainous region in the south. Table 4.1 also compares the sample with the entire hospital birth cohort of 1990 (n = 563) to see whether this is a good representation of hospital births and to see if the entire hospital birth cohort of 1990, which will be used more extensively in Chapter 5, is representative of the population of eastern Ladakh. First, the distributions of residence, ethnicity, and religion are comparable in the samples and the census. The notable exception is the distribution of villages from which women come. Only thirty-seven villages are

78

An Ecology of High-Altitude Infancy

Table 4.1. Sample characteristics relative to census data for Leh District

Residence # of villages % from Leh

Leh district census data

Study cohort (n = 168)

1990 birth cohort (n = 563)

112 23

37 23

43 24

Religion % Buddhist % Muslim % Hindu

86 12 1.4

80 15 5

80 14 5

Ethnicity % Ladakhi % Tibetan % nonnative

86.5 1.5 12

86.3 6.6 7.1

87 5 7

Occupation % farmers

58

49

NA

Landholdings (hectares) Education Mothers % 0 years % 1–5 years % 6–10 years % 11+ (Matric) NA Fathers % 0 years % 1–5 years % 6–10 years % 11+ (Matric) NA

1.5

2.5

NA

88 2 3 4 3

43 7 15 9 26

NA NA NA NA

64 11 13 12

24 10 26 8 32

NA NA NA NA

Source: Directorate of Economics & Statistics Planning and Development Department (1987).

represented in the sample and only slightly more are in the whole cohort. This is largely because most women using the hospital come from the main Indus valley region, from which Leh is relatively easily accessible by a network of roads and bus or taxi service. Many villages in the district are connected to this network only by footpaths and are often several days’ journey from the road. However, relative to the population as a whole, the number of people living in these villages represents only a small proportion, since the majority of the district’s population lives in the main Indus valley or in tributary valleys connected by link roads. Women residing in these distant villages are more likely to deliver their infants at

Big Mountains, Small Babies

79

home, although they sometimes stay with relatives in or around Leh as the birth approaches. Important differences between villages in the Indus valley region and those distant from the road network are most likely to be evident in the availability of imported goods, especially foodstuffs, and services, such as health care, education, and electricity. The same percentage of women was from Leh in the samples and the census, so the samples are not biased toward urbanites in the district. Furthermore, in 1990 Leh had only about 15,000 people; it does not really qualify as an “urban” area, and most of the women in the sample who reported Leh as their residence were landowning farmers. The variable ethnicity distinguishes between women native to the region and those who are migrants and unlikely either to share ancestry with the Ladakhi population or to have a long history of high-altitude habitation. Only 7% of both samples were migrants from the plains of India or Kashmir (who live temporarily or seasonally in Ladakh as spouses of men employed in the military or commerce), or from Nepal (working as laborers on construction projects). Tibetans also form a small minority inhabiting the region around Leh and comprised about 7% of the sample. This ethnic distribution is at odds with the census only insofar as the distribution of migrants is different. In part this discrepancy is due to the enumeration of Tibetans, who were identiﬁed in the census only if they had been born in China, whereas in the sample many women identiﬁed as Tibetan had been born in Ladakh. Thus the percentage classiﬁed as Ladakhi in the census is slightly inﬂated by those who identify themselves as ethnically Tibetan but who were born in Ladakh, and the percentage of Tibetans is underestimated. The “nonnative” category is larger in the census than the sample for a number of reasons, the most signiﬁcant of which are likely to be that pregnant spouses of men working in Ladakh temporarily are likely to return to their homes for the duration of their pregnancy and birth and that the number of migrants had actually diminished as a likely result of the hostilities of Buddhist Ladakhis toward Kashmiris in 1989. This stimulated some Kashmiris to leave Ladakh permanently or to choose not to return for the summer season in 1990. Occupation is difﬁcult to compare between the sample and the census and to ascertain precisely because of the multiple domains in which any individual might be simultaneously employed. This is especially true of men’s employment. For example, the modal occupation for husbands of the women in the samples was reported to be the Ladakh Scouts, a local military unit. Following that were the occupations of driver, farmer, then other military occupations. Aside from those who gave farmer as their only employment, each of the other modes can be, and usually is, combined with agricultural activities. This is possible in extended households

80

An Ecology of High-Altitude Infancy

where wives and other relatives contribute to agricultural work. Workers in the formal sector, including professionals, often take leave in seasons of heavy agricultural work, especially during the harvest, to help on the family farm. Women also may have multiple occupations. Among the nurses I worked with were women who worked a full shift at the hospital, managed a guest house in the summer, and carried out important agricultural and animal husbandry work on the family’s landholdings on a daily basis. Thus while only 49% of women reported “farmer” as their occupation, many more were in fact likely to have been involved in agricultural activities on a steady basis if they came from landowning households, as most of them did. The average level of education achieved by women in the sample is much higher than that recorded in the census in 1981, as is that of their husbands, although the differences are much smaller for men. The census ﬁgures were calculated from the population of males and females aged twenty to thirty-four. These ﬁgures are undoubtedly at variance in part because the census was taken over ten years before the study, and education at more advanced levels has become more accessible and normative. It also seems likely that those in the sample were better educated than the general population, since they come from the region with more diverse and accessible educational facilities. Furthermore, they can realize the beneﬁts of education in terms of increased employment options and are more likely to be from families with members engaged in other modes of employment. They may also be more likely to use the hospital for birth. Of note is the difference in landholdings between the census and sample populations, of about one hectare. Perhaps women using the hospital are from wealthier families (with respect to land at least). However, landholding estimates from the census are notoriously inaccurate; often they are given by village headmen and not by individual households, or individuals may report having less land than they do in order to avoid taxation. Moreover, when I asked women about their landholdings, I inquired as to how many zhing (ﬁelds) their household owned. Zhing range between one-quarter and one-half hectare, and hence are not standardized units. In sum, it is important to keep in mind that the sample does not represent the whole population of Ladakh, but more likely those who live in eastern Ladakh, in closer proximity to Leh. For simplicity’s sake, unless a speciﬁc subgroup is being referred to, I refer to the sample as the “Ladakh sample” (especially in comparisons with studies from other geographical areas), as in fact most of the population does live in this district. Hence what characterizes the population in this area increasingly

Big Mountains, Small Babies

81

represents contemporary Ladakhi life. Women in the samples do have access to a greater diversity of foods, education and occupation options, and health care in ways that vary somewhat from the access of women in outlying villages. At the same time, they share a larger Ladakhi “culture” and behaviors with people from other parts of the district. Furthermore, based on an estimated birth rate of 30/1000, the 563 live births that took place in the hospital in 1990 represent over one-third of all births in the district.4 Potential Sample Biases in Health Parameters Valid questions can be raised about the representativeness of the health characteristics of the mothers and newborns at the hospital. Possibly, women using the hospital facilities for giving birth were more likely to have experienced previous obstetric problems or the death of an offspring. On the other hand, almost half of the women who gave birth in the hospital in 1990 or who were part of the prospective study were primiparous (this was their ﬁrst birth), and 43% and 39% were primigravid (this was their ﬁrst pregnancy), respectively (see Table 4.2). One reason for this is that Dr. Ladhol recommends that women of higher parities who have previously delivered without complications do so at home. Therefore statistics on reproductive health problems such as stillbirths, spontaneous abortions, and neonatal mortality may be overestimated in the sample(s). However, the study cohort deliberately excluded newborns or mothers who were acutely ill in the day after birth, so the argument could be made that the health statistics of the sample are better than those that describe the larger population. In reality, there were only two or three mother-infant pairs who were excluded on this basis. The study in which anthropometric measurements were made took place during the spring, summer, and fall and does not include newborns born in the winter. Seasonal differences in Ladakh are extreme and affect the behavior of pregnant women and the health of newborns. One further issue is the sex ratio of the samples. There was a disproportionate number of females in the anthropometric sample (86 females and 82 males; sex ratio = 95/100). Since females generally are smaller at birth than are males this may downwardly bias the average neonatal measurements, but separate analyses were run for males and females, and the effects of sex were controlled for as needed in other statistical analyses. The sex ratio for the entire 1990 birth cohort was 115/100, with more males than females (the average for most large-scale populations is around 105/100).

82

An Ecology of High-Altitude Infancy

Table 4.2. Maternal and neonatal characteristics of the sample

Mean Mothers Parity Primiparous Primigravid Age (years) Weight (kg) Stature (cm) Body mass index (kg/m2 ) Arm circumference (cm) Triceps skinfold (mm) Hemoglobin in pregnancy (g/dl) Anemic (Hb <12 g/dl) Education (years) Husband’s education (years)

0.8 48% 39% 24.7 48.9 152.8 20.7 22.2 9.4 10.4 78% 3.8 5.7

Standard deviation

1.1

4.3 6.2 5.4 2.2 1.7 3.0 1.7 4.9 5.0

Male Female Standard standard standard Mean deviation Males deviation Females deviation Newborns Birthweight (g) Birth length (cm) Head circumference (cm) Chest circumference (cm) Arm circumference (cm) Sum of all skinfolds Ponderal Index (g/cm3 ) a Male-female

2764 48.0 34.2 30.9 10.2 15.4 2.5

435 2.4 1.4 1.9 1.1 2.9 0.2

2853a 48.6a 34.5a 31.1 10.2 15.2 2.5

441 2.4 1.5 1.9 1.1 3.1 0.2

2678 47.8 34.0 30.8 10.2 15.6 2.5

412 2.3 1.2 1.8 1.0 2.6 0.2

differences in these measures are signiﬁcant (p < .01)

Descriptive Anthropometry Table 4.2 summarizes the demographic and anthropometric characteristics of the mothers and newborns in the study. The average age of women who were measured and interviewed in this study was 24.7 years, and ranged from 18 to 42. Almost half of the mothers were primiparous, but less than 40% were primigravid. Over 8% of the women had experienced fetal loss before successfully delivering their ﬁrst live birth. Only two women had more than four prior births, and they had only ﬁve each, so there were no very high parity women in the sample. Even though they had just given birth and retained the extra ﬂuid and tissue necessary to support pregnancy, the mothers in the sample weighed

Big Mountains, Small Babies

83

relatively little (∼49 kg, or ∼108 lbs). Their low weight is reﬂected in their relatively little subcutaneous fat as measured by the triceps skinfold. Average height was 153 centimeter (just under 5 ft). They also had quite low hemoglobin levels when these were measured during pregnancy (10.4 g/dl). Hemoglobin levels generally decline during pregnancy due to the increase in blood ﬂuid volume, but we expect that the baseline should be elevated due to the high altitude. Hemoglobin concentrations less than 12 grams/deciliter are considered evidence of anemia in sealevel populations, although India uses a lower standard of <11 grams/ deciliter. A fundamental aspect of neonatal assessment is the estimation of gestational age, as the achieved size of a newborn is ﬁrst and foremost a function of time. The most accurate measures of gestational age use the date of the last menstrual period (LMP), with thirty-seven to forty-two weeks LMP weeks considered to be full-term pregnancies. In many cases women do not keep track of LMP, so other measures may be used such as fundal height, the distance between the pubic bone and the top of the uterus. Or, a series of maturational measures of the newborn can be taken at birth (i.e., the Dubowitz score). In a pinch, head circumference can be used as a crude measure (Miller and Hassanein 1971). In 32% of the sample, dates of the last menstrual period were recorded, but in many cases they yielded wildly impossible pregnancy durations. Estimated gestational ages (EGAs) reported in the birth records were from fundal height measurements done at the time of admission for labor and were available for 79% of births, 7% of which were classiﬁed as premature (EGA <37 weeks). The mean was thirty-eight weeks and ranged from thirty-two to forty weeks. Based on head circumference measurements, only 1.2% could be classiﬁed as premature (head circumference <31.2 cm). Previous studies had indicated that head circumference is not affected by altitude (Haas 1980; McClung 1969), so it is possible to use this sea-level standard. Since there was no evidence of high rates of prematurity based on these two estimation procedures, all analyses were done using the full sample unless otherwise indicated, and when differentiated by maturity status, EGA based on fundal height was used. Neonatal anthropometric characteristics for males, females, and the entire sample are also described in Table 4.2. The mean birthweight was 2764 g (6.08 lbs), with a range between 1665 and 3905 grams, as shown in the birthweight distribution in Figure 4.1. The median birthweight was also the mean (2764 g), indicating that the sample was normally distributed. Twenty-seven percent of all neonates weighed less than 2500 grams.

84

An Ecology of High-Altitude Infancy 0.25 average bwt=2764 g males=2853 g females=2678 g

Percentage

0.2

0.15

0.1

0.05

0 1500 1750 2000 2250 2500 2750 3000 3250 3500 3750 4000 Birthweight (grams) sample

males

females

Figure 4.1. Birthweight distribution.

At sea level, male newborns are generally heavier than females at a given gestational age (Cogswell and Yip 1995). However, they appear to be more vulnerable to stresses in the uterine environment, and under stressful conditions their birthweights are closer to those of females (Stini 1969). In the Ladakhi sample, males were signiﬁcantly heavier (175 grams) than females. Only 17% of males weighed less than 2500 grams, while 37% of all female newborns were in this category. As there were more females than males in the sample, this is biasing the average birthweight downward. Males in the sample were also signiﬁcantly longer and had larger heads than females. Other anthropometric characteristics were not signiﬁcantly different between the sexes, but they did illustrate the standard pattern of males having larger circumferences but less fat than females. Without a standard of reference, it is difﬁcult to judge the signiﬁcance of these anthropometric measures. First, note that the above descriptors apply to the entire sample, not just to ethnic Ladakhis. However, there are no noticeable differences in the newborn and maternal means between a sample restricted to ethnic Ladakhis and the entire sample, largely because ethnic Ladakhis make up 86% of the entire sample.5 In Table 4.3 a comparison is made between Ladakhi and other broad population groupings. The women in the Ladakhi sample are slightly lighter and shorter with much smaller triceps skinfolds than other populations. As

Big Mountains, Small Babies

85

Table 4.3. Ladakhi women and infants compared with other general populations Australian Aborigines

Ladakhis

Europeans

Africans

Asians

Male infants Weight (g) Length (cm) Head circumference (cm) Chest circumference (cm)

2853 48.6 34.5 31.1

3500 50.2 34.9 33.1

3400 51.6

3200 50.2

3100 50.5 34.0

Female infants Weight (g) Length (cm) Head circumference (cm) Chest circumference (cm)

2678 47.8 34.0 30.8

3320 49.1 34.0 32.9

3200 52.2

3100 49.6

3200 49.8 33.5

Women Weight (kg) Height (cm) Triceps skinfold (mm) Arm circumference (cm)

48.9 152.8 9.4 22.2

54.5 161.3 16.1 24.0

50.4 154.0 11.6 22.0

46.5 156.8 21.3 22.7

50.0 153.4 8.4 24.3

Source: Eveleth and Tanner, 1990

for the newborns, both males and females are much lighter, shorter, and have smaller chest sizes, but their head circumferences do not vary from those in European populations. Male-female differences in birthweight vary between 100 and 200 grams across all populations. The smaller chest sizes among Ladakhi newborns are somewhat surprising, given that larger chest sizes tend to characterize adults at high altitude, including Ladakh, and this appears to be under some degree of genetic control (Frisancho et al. 1997; Malik and Singh 1984). The comparison of the sample with other high-altitude populations is analyzed in more detail in Chapter 6. The Ponderal Index (PI: weight (g)∗100/length (cm)3 ) was calculated to differentiate between chronic and acute growth retardation (stunting vs. wasting), with the cutoff of PI <2.25 indicating disproportionate, or acute growth retardation (low weight for length) (Haas et al. 1987). Growth of the fetal skeleton occurs very rapidly in the ﬁrst and second trimesters of pregnancy and slows in the third. Weight continues to increase throughout the last trimester, when it is primarily deposited as fat. Newborns who are small for gestational age (SGA; usually deﬁned as <2500 g or <10th percentile in weight) usually have experienced acute growth disruption in the last trimester. Infants that are light and short at full term are likely to have suffered chronic growth retardation throughout

86

An Ecology of High-Altitude Infancy

the second and third trimesters (proportionate growth retardation; stunting). Among the Ladakhi newborns, average PI was 2.49 g/cm3 and 13% of Ladakhi neonates had low PI values. Males and females did not differ in their Ponderal Indices. Among full-term newborns, the sample was divided into those who were SGA (<2500 g), and all others. Among the SGA babies, 71% had PI scores over 2.25, indicating chronic growth retardation, while 29% had PIs less than 2.25, indicative of appropriate skeletal growth but diminished weight accumulation (acute growth disruption).6 Intrapopulation Variation in Birth Outcome There is clearly variation in birth outcome within the sample of babies born in Ladakh. Some babies are heavier, longer, or fatter than others. There are two main ways to understand some of the sources of that variation. First, regression analysis allows us to test the statistical signiﬁcance (deﬁned as p < .05) of correlations between maternal or family characteristics and birthweight, and to ascertain which variables best predict birth outcome and hence are sources of variation within the sample. Second, we can compare subgroups within the sample for evidence of statistically signiﬁcant differences in birth outcome as described by a variety of measures. Because this is a study of only one population, the members of which all inhabit the same altitude, it is not possible to test the speciﬁc effects of altitude on birth outcome. However, it is possible to test for differences between subgroups to see which have better or worse birth outcomes, given that they were all exposed to the same hypoxic conditions. Newborns of subgroups that vary with respect to high-altitude habitation history may reﬂect differences in adaptation to hypoxia. The above types of analyses provide a jumping-off point for exploring the context of these statistical associations. We can work our way outward from the newborn’s body to the mother’s body, to the household and family, to the healing resources available during pregnancy, and to the larger social and cultural context that gives these form and meaning. These are all framed by the high altitude, arid conditions, and seasonally variable environment of Ladakh. Here I focus on the domains of maternal nutritional status; diet during pregnancy; activity and workloads during pregnancy; ethnicity and religion as meaningful mediators of birth outcome; and the healing traditions used by pregnant women. Maternal Nutritional Status and Birth Outcome Of the biological variables that are likely to affect birthweight, in this study maternal postpartum weight and newborn gestational age accounted

Big Mountains, Small Babies

87

Table 4.4. Regression equations for predicting birthweight Coefﬁcient

Partial F

p value

Gestational age Maternal weight Intercept R2 = .16

101.3 21.26 −2094

5.33 8.99

p < .02 p < .003

Gestational age Maternal activity Intercept R2 = .30

171.9 −227.2 −3170

5.47 15.41

p < .02 p < .0003

for about 16% of the observed variation in birthweight, as shown in Table 4.4. The positive coefﬁcients on gestational age and maternal weight indicate that these variables increased birthweight; for a given maternal weight, every one-week increase in gestational age resulted in a 101 gram increase in birthweight. The widely used standard for prenatal growth of Lubchenco et al. (1963) has a ∼200 gram increase for each week of gestation from thirty-ﬁve to thirty-eight weeks, with growth slowing after thirty-eight weeks. The Ladakhi sample clearly deviates from that standard, indicating slower growth among Ladakhi fetuses during late gestation. Maternal weight and fatness are well known to be positively correlated with birthweight (Silbert et al. 1978; Vobecky 1986). Since Ladakhi mothers are relatively light, this contributes to the smallness of their newborns. A 1 kilogram increase in maternal weight added over 20 grams to infant birthweight, compared to a similar study in Gambia in west Africa where 14 grams was added to birthweight for a 1 kilogram increase in maternal weight, although that study did not control for the effect of gestational age (Prentice et al. 1987). For comparative purposes, if a woman weighed 65 kilograms (143 lbs) as is more typical in the United States, the regression equation predicts that she would have a newborn weighing 3031 grams (6.88 lbs). This is lower than the average birthweight in the United States and provides support for the idea that additional factors contribute to the low birthweights evident in Ladakh. Maternal height, as an indicator of lifelong and intergenerational nutritional status, is usually positively correlated with birthweight (Lechtig et al. 1977), but it was not related to birthweight in this study. Maternal age and parity are also known to have positive effects on birthweight, although this effect reverses at older ages and higher parities. Firstborns are generally about 100 grams lighter than are subsequent infants (Cogswell

88

An Ecology of High-Altitude Infancy

and Yip 1995). In this sample maternal age and parity each had signiﬁcant and positive effects on birthweight. This effect was evident after the second birth – ﬁrst- and second-born babies had similar average weights, but birthweight went up by 100 grams among third-born babies. However, the effects of maternal age and parity disappeared when they were added to the regression equation with maternal weight and EGA. That is, a mother’s weight was much more important than her age or parity in predicting birthweight (and neither age nor parity was correlated with weight). With regard to sex as determinant of birthweight, when maternal weight and EGA were controlled for, the newborn’s sex ceased to predict its weight. Diet during Pregnancy Maternal weight is a good summary of the resources that a woman can use to support a pregnancy and channel to fetal growth; it is a product of a woman’s prepregnancy reserves and her diet during pregnancy. Nutrient needs increase during pregnancy to support the growth of fetal and maternal tissues. Most important are calories, as pregnancy is very energetically expensive, but additional protein and micronutrients such as calcium, folic acid, and iron are also needed (Adair 1987). Studies have shown that supplementing the diets of pregnant women with calories can result in heavier babies – additions of up to 30% of normal caloric intake have produced a maximum increase in birthweight of 200 grams (Lechtig et al. 1975, 1978). The impact on birthweight was most signiﬁcant among newborns of undernourished mothers and among male infants (Adair and Pollitt 1985; Lechtig 1982, 1975; Mora 1981). Protein supplements had no additional effect, and when nutritional status was already adequate, dietary supplementation did not increase birthweight (Habicht et al. 1974; Lechtig 1982). Note, however, that while diet played a signiﬁcant role in determining pregnancy outcome, it explained only a small percentage of the variance in birthweight; height, weight, SES, age, parity, gestational age, diet and supplementation, and arm circumference collectively accounted for less than 18% of the variation in birthweight (Habicht et al. 1974). In this study, less than half of pregnant women reported that they were able to augment their diet, as shown in Table 4.5. In fact, similar percentages of women said they ate more while pregnant as said they ate less (44% and 40%, respectively), while 15% said they ate the same amount as before they became pregnant. Overall food consumption (eating more than usual) was associated with a signiﬁcant increase in neonatal weight. However, as the supplementation studies concluded, more frequent

Big Mountains, Small Babies

89

Table 4.5. Dietary practices and activity during pregnancy

Meat Eggs Milk

Dietary intake Activity

Never eaten

Rarely eaten

Occasionally eaten

Eaten everyday

18% 31% 30%

40% 25% 11%

26% 26% 19%

17% 18% 40%

Less than usual

Same as usual

More than usual

40% 28%

15% 72%

44% 0%

consumption of any of the protein-rich foods (milk, eggs, and meat) was not associated with signiﬁcant increases in newborn size in the regression analysis. The majority of women ate eggs or meat never or only rarely during their pregnancy. Within Leh and the surrounding area, meat has to be purchased in the market; hence, access to it is constrained by a household’s available cash. Eggs are mostly available through the Kashmiri market, which is open only during the summer months – and which was avoided by Buddhists in 1990. Most women drank milk at least occasionally and 40% said that they drank it every day, but 30% said they never drank it. Households (even those in the center of Leh) generally keep some dairy animals, and milk is traded among households and relatives, so most women had access to milk; yet, almost one-third of the sample said that they never drank milk. The reasons behind this variation are obscure – lactose intolerance is reportedly very rare in Ladakh, and milk consumption was actually negatively related to land ownership, indicating that residence in Leh did not limit access to milk. Landholdings are a good index of a household’s wealth although not necessarily their cash supply. When Ladakhi women talked about wealth, their comments inevitably centered on how many zhing, or ﬁelds, their household had. The number of zhing had no effect on maternal weight, height, fatness, overall diet, or consumption of meat, eggs, or milk. Although food taboos during pregnancy are common in many cultural contexts, they were not widely articulated by Ladakhi women. No dietary proscriptions are speciﬁcally mentioned in the Tibetan medical texts on pregnancy and embryology, and women consistently told me there were no speciﬁc dietary restrictions. Women are encouraged by family, friends, and Dr. Ladhol to eat well, to increase their consumption of strengthening and nourishing foods such as dairy products, meat, and eggs, and to

90

An Ecology of High-Altitude Infancy

decrease their involvement in strenuous physical work. These messages are supported by widely held cultural ideologies, deriving in part from Tibetan medical doctrine (see below) indicating that women should eat more strengthening foods and work less during pregnancy. Informants frequently commented that the only time women are given comfort and attention is during and immediately following pregnancy and birth. Primigravid women often return to their natal homes in the later months of pregnancy for extended periods of time, although it is considered inauspicious to give birth there. At home they are freed from their tasks in their husband’s household and are fed well with lots of meat tukpa. Informants told me that the ideal is for a sheep to be slaughtered and given to the expectant mother to eat, but observed that during the year of the study sheep were hard to come by, especially for Buddhist families, due to the boycotting of Muslim vendors during this time of communal hostility. Based on the data in Table 4.5, it appears that while the ideal is for women to eat more in general, and more “strengthening” foods while pregnant, in reality most women experience little signiﬁcant change in diet during pregnancy. In my observations, it is only during the immediate postpartum period that women consume ample quantities of rich foods brought to the hospital by relatives to help them recoup their health, which is considered to be compromised by the birth event. Marzan (butter and barley ﬂour), copious quantities of gur-gur cha, and meat tukpa are commonly consumed by postpartum women, who are supposed to be fed in this way for forty days after birth. Depending on the wealth of the household and agricultural cycle, this may or may not occur. Anemia, resulting from deﬁciency of iron, folic acid, or Vitamin B-12, is recognized as a very common nutritional problem for women during pregnancy by the head of the maternal and child health department in the hospital and Dr. Ladhol. Indeed, the average hemoglobin reading for pregnant women was 10.4 grams/deciliter. This may be due to shortages of meat or dark green leafy vegetables that supply folic acid (especially during the winters), although meat consumption during pregnancy was not correlated with hemoglobin in the study. Surprisingly, neither hemoglobin status nor classiﬁcation as anemic (Hb<12 g/dl) had a signiﬁcant effect on birthweight in regression analysis, although the coefﬁcient was negative. Similarly, Haas (1980) and Moore (2001b) found that hemoglobin was inversely related to birthweight in La Paz, Bolivia, and Tibet, respectively. In the Ladakh study, a low hemoglobin reading would lead to the prescription of folic acid or iron tablets, which may have reversed the anemia, although no interaction effect between supplementation and hemoglobin was observed. Moore (2001b) argued that a fall in

Big Mountains, Small Babies

91

hemoglobin is normal in pregnancy due to the expansion of plasma volume and that other mechanisms such as increasing uterine artery blood ﬂow appear to be more important for fetal growth under hypoxic conditions. So, the signiﬁcance of low hemoglobin levels in pregnancy is not clear in high-altitude contexts. Another speciﬁc nutrient of interest during pregnancy is salt. One feature of the typical Ladakhi diet is the daily consumption of large quantities of salty gur-gur cha. This is potentially worrisome because of the very high salt intake that results from its routine consumption, as salt has been implicated as a risk factor for hypertension. During pregnancy, hypertension can escalate into edema, proteinuria, and eclampsia, which can be fatal. Research in Colorado suggests that pregnancy-induced hypertension (PIH; deﬁned as blood pressures above 140/90) may also be related to hypoxia. PIH was three times more frequent at high altitude (12%) than at low altitude (4%) (Moore et al. 1982); and at high altitude, blood pressure did not fall during the second trimester (Palmer et al. 1999). Furthermore, PIH has been associated with lower birthweights at high altitude, possibly due to insufﬁcient expansion of uteroplacental blood ﬂow, and is more common among nonnative women (Zamudio et al. 1995). As in the Colorado study, in Ladakh, 12% of women in the study were hypertensive during the second trimester of pregnancy. They were prescribed a low salt diet and bed rest. The frequency of hypertension in the ﬁrst trimester was 15%, and 8% in the third trimester. Blood pressure in each trimester of pregnancy had no effect on birthweight in regression analysis. However, women with blood pressures >140/90 during the third trimester did have newborns that were substantially smaller than women with lower blood pressures (2418 g vs. 2765 g). It is not clear how blood pressure is related to diet in this study, as there were no detectable dietary differences between women with high or low blood pressures. Salt probably plays less of a role in the etiology of PIH than it does in hypertension in the general population, as other hormonal factors appear to mediate PIH, and hypoxia and a woman’s adaptations to hypoxia appear to inﬂuence PIH risk. Workloads during Pregnancy Diet during pregnancy, especially the consumption of energy, needs to be evaluated relative to energy expenditure. A particular combination of work and food intake affects the supply of energy that the fetus can draw upon and is likely to vary seasonally in subsistence agricultural societies. The study in Ladakh occurred during the seasons of peak agricultural work. It started during the planting season, when stored foods may have

92

An Ecology of High-Altitude Infancy

been running low and work was particularly demanding. When pregnancy is added to the mix there is the potential for seasonal differences in birthweight as fetal growth is constrained when calories are short and work is energetically demanding. Profound seasonal variation was documented in a study in Gambia, in which pregnant women expended more energy than they took in during the wet season, and this resulted in average birthweights that were lower by 200 to 300 grams than in the dry season (Prentice et al. 1981). Activity patterns during pregnancy turned out to be very important predictors of birth outcome in Ladakh. As is evident in Table 4.5, most women (72%) reported no change in their activity patterns during pregnancy. Only 28% reported working less or not at all. A regression equation using EGA and maternal activity during pregnancy (as a categorical variable: working less, the same, or more during pregnancy) resulted in a higher proportion of the variation in birthweight being accounted for (r2 = .30) than in the earlier equation using maternal weight and EGA (see Table 4.4). The negative coefﬁcient on maternal activity indicates that this variable had a negative impact on birthweight. Controlling for gestational age, women who continued to work during pregnancy had substantially smaller infants (by >200 g) than those who worked less. This effect of activity was signiﬁcant even when maternal weight was added to the prediction equation; in fact, it nulliﬁed the effect of maternal weight. In other words, the effects of a woman’s activity patterns during pregnancy on birth outcome are not mediated by her weight, but work has strong independent negative effects on birthweight. However, when we compare the extremes of the birthweight distribution in this study, we see that mothers of newborns weighing <2200 grams (n = 19) were both signiﬁcantly smaller (5 kg) and reported working signiﬁcantly more than those who gave birth to newborns weighing >3200 grams (n = 23). A similar negative effect of activity during pregnancy on birthweight has been observed in other studies, including one among Sherpas at high altitude in Nepal (Smith 1997). An Ethiopian study comparing birthweights of women who worked hard and those who engaged only in light work found the same effect: a 200 gram reduction in birthweight regardless of the woman’s prepregnancy weight (Tafari et al. 1980). Apparently women in that study were purposely engaging in hard work during pregnancy speciﬁcally to avoid having a large newborn and difﬁcult birth, and this appears to have been a successful strategy! In another African study (Manshande et al. 1987), resting for a month in a special “rest village” before delivery compared to resting minimally increased birthweight on the order of 200 to 400 grams, even when maternal weight and diet were similar.

Big Mountains, Small Babies

93

Given that most of the women in the sample were involved in agricultural work, reduction in activity should result in a substantial reduction in calories expended, which are then available to the growing fetus. The evidence does suggest that the energetic economy of work directly affects the fetus, independent of existing maternal energy stores, and that Ladakhi women are aware of this relationship. Similar to the Ethiopian study previously cited (Tafari et al. 1980), Kuhn (1988) reported that Ladakhi women said that they would work hard after the seventh month of pregnancy in order to decrease the likelihood of having a large baby and thus a difﬁcult birth. However, the PI distribution of Ladakhi newborns suggests that they are more often chronically rather than acutely stressed in utero, so energy expenditure late in pregnancy is not the sole factor. And while women expressed concern over having a large baby, no one in my sample said that she worked especially hard with the intention of having an easier birth. This may be because women were already working hard during these months of agricultural activity. The effect of women’s work should not be underestimated. Ladakhi women work hard at an array of domestic and agricultural tasks when they are not pregnant, and they continue to do so during pregnancy. Household Dynamics and Women’s Workloads. The demands of work are likely to be determined by the availability of labor a household can draw upon, and in the absence of sufﬁcient labor from other household members or the ability to afford paid laborers, a pregnant woman may have no choice but to continue to work throughout her pregnancy. The ﬂuidity of household arrangements made it difﬁcult to analyze further the relationship between household size and composition to women’s activity during pregnancy. Household complexes may contain several houses, and members of these separate houses may or may not be available for agricultural labor, depending on their relationship to the main house. There was wide acknowledgment of the demands of work and their detrimental effects on women’s health, and while women complained about the burdens of work, they also were resigned to it; after all, who else was going to do it? A woman’s position in the household is also relevant, as it impacts her workload. Women who had already had one child worked signiﬁcantly less than those who were primiparous, and there was a steady decline in reported work with greater parity. Moreover, women with at least one child gave birth to signiﬁcantly larger neonates than those with no previous children. It appears then that women who are more established in their marital households have a greater ability to modify their work schedules while pregnant, and this has signiﬁcance for birth outcome.

94

An Ecology of High-Altitude Infancy

On the other hand, the amount of food a woman reported consuming while pregnant did not change with parity. So, early in a woman’s reproductive career, not only are energetic demands great, but a young woman does not yet have sufﬁcient status in her husband’s household to claim additional resources or exemptions. The role of a primiparous woman in an established household and the ﬂuidity of household membership are illustrated by the case study of Nyilza, a young woman who had temporarily joined the household in which I resided in 1991. Nyilza was the wife of the deceased grandmother’s younger brother, who was a member of the Ladakh Scouts and rarely at home. As the youngest adult female and married to a peripheral member, her own status in the household was ambiguous, and her duties not well deﬁned. She answered my questions about her pregnancy with a shyness and reticence that is rare among adult women, but it certainly reﬂected her uncertain position and the fact that it was her ﬁrst pregnancy. She reported consuming her regular foods, albeit in smaller quantities due to her stomach feeling unsettled, and she was able to procure two glasses of milk per day. Nyilza participated in the daily work of the household, mostly in and around the household, milking the cows, feeding the animals, weeding the garden, and preparing meals. She joined the other household adults in the planting and watering of the ﬁelds, but returned home before the others to attend to domestic tasks. Since this was a household with access to several working adults, her labor in the ﬁelds was not crucial, but her help around the house was appreciated by the other women, who recognized her “condition” but clearly expected her to participate in the household’s work. The Seasonal Ecology of Work. Researchers in the Andes (Kashiwazaki et al. 1995; Leonard et al. 1995) have found that agricultural work (planting, watering, weeding, carrying, threshing) at high altitude is particularly arduous and calorically expensive, and requires adequate energy and responses in oxygen transport.7 As Ladakhi women display high rates of anemia during pregnancy, perhaps the amount of oxygen they can transport to the fetus is constrained, although there may be other mechanisms that compensate for this. Because this study was undertaken from April to October, the period of most intensive agricultural activity, the effect of activity on birthweight may be magniﬁed and the effect of maternal weight minimized due to the increase in food availability. There was a gradual seasonal increase in average birthweight, from ∼2700 grams in April to May, to ∼2950 grams in September to October, an increase of almost 10%. Month of birth was a signiﬁcant variable with a positive coefﬁcient when added to the regression equation with

Big Mountains, Small Babies

95

maternal weight. Seasonal effects on fetal growth thus act independently of maternal weight, and there was no perceptible trend in average maternal weight, triceps skinfold, or food consumption over the six-month period. Hence the seasonal impact on birthweight is largely mediated by changes in maternal activity. Having participated in agricultural tasks throughout the time I lived in Ladakh, it is my experience that spring work is the most arduous, since it consists of preparing ﬁelds (including hauling night soil from the household composting toilet) and planting, which is particularly difﬁcult for potatoes, because women must dig hundreds of holes. Summer work involves watering and weeding, which, though tedious, are not energetically extremely costly and provide more opportunities for rest. Threshing had begun when the study ended, but it is not as difﬁcult as spring planting. In the spring, arduous work during the last month of pregnancy can result in acute growth disruption for the fetus, especially when food consumption is limited because stores are dwindling from the winter and the road has not yet opened to allow the import of new supplies. Different results might have been achieved if the study had taken place over the relatively restful winter months. Intrapopulation Variation in Maternal Workloads. The sources of variation in maternal diet and work patterns among pregnant Ladakhi women are hard to pin down. The most obvious place to look would be socioeconomic status, which should correlate with dietary intake and the need to do physically demanding labor during pregnancy. However, women from households with more land worked slightly more than those from households with less land, which points to the crucial importance of household composition and available labor relative to landholdings. Similarly, landholdings had no effect on any neonatal anthropometric characteristics. Perhaps the strong effect of activity on birth outcome derives from the recent socioeconomic changes in Ladakh that so many authors have lamented, particularly the increasing participation of Ladakhi men in military occupations or ancillary services or tourism-related activities, which are concentrated in the summer months. Most husbands had some occupation related to these. Must women whose husbands are employed in these ways work more and hence have smaller babies? The answer to this question appears to be no, as no clear effect of male absenteeism on birth outcome emerged from the analysis, even among households with lots of land. However, the categories for male occupation may not capture male absenteeism well, as many men who are Ladakh Scouts or drivers may in fact be at home during periods of heavy agricultural work.

96

An Ecology of High-Altitude Infancy

Furthermore, households with cash derived from male employment can pay workers to help in the ﬁelds. In any event, work is likely to have been an ongoing reality for women in Ladakh even before these new economic opportunities came about. Education level of either the mother or father could have an effect on birthweight, by inﬂuencing household wealth, food intake, or work patterns. The link between education and birthweight itself was not signiﬁcant, but that relationship was mediated by increased food consumption among mothers who were more educated or married to men who were more educated, and also by decreased activity among educated women, who are more likely to hold clerical, professional, or other types of nonagricultural jobs. The ﬁnding that there are not clear socioeconomic differences that create variation in birth outcome is consistent with observations of other workers in Ladakh that socioeconomic differences are not really visible or meaningful except at their extremes, and these are not well represented in the study.8 Social stratiﬁcation does not appear to have meaningful effects on birth outcome in Ladakh. Women whose households owned no land or who had no education did not have signiﬁcantly different birth outcomes from those who had more of these resources. In sum, we see that birth outcome is constrained by two major factors: maternal weight, as a measure of the mother’s biological resources, and maternal activity level during pregnancy. As expected, higher maternal weights and increased food consumption were associated with larger newborn size, whereas higher levels of activity reduced resource ﬂow to the fetus, resulting in lower birthweights. These are the important “proximate” factors inﬂuencing birth outcome, but their effects derive from the larger household and ecological context of life in Ladakh. Moreover, while there are often ideologies or “cultural maps” for pregnant Ladakhi women that stress protection for them and the fetus (e.g., rest, increased food consumption), the behavior of most women was constrained in ways that did not allow them to conform to these mandates, with negative consequences for fetal growth. Ethnicity and Birth Outcome Within the sample, there are small but distinct groupings. One is denoted by ethnicity, which is used here in part as a marker for high-altitude ancestry, although it has other local political and social meanings. Ethnic Ladakhi mothers and newborns are compared with Tibetans and migrants from low altitude in Table 4.6. Tibetan women had babies that were heavier (145 gms) and longer (1.6 cm), on average, than those of

Big Mountains, Small Babies

97

Table 4.6. Comparison of maternal and neonatal parameters among Ladakhis, Tibetans, and migrants from low altitudes in Ladakh Ladakhi

Tibetan

Migrants

Mothers N Parity Age (yrs) Weight (kg) Stature (cm) Arm circumference (cm) Triceps skinfold (mm) Hemoglobin (g/dl)

145 0.9 24.4 48.8 153.3 22.2 9.3 10.5

11 1.4 27.6 50.2 153.5 22.8 11.2 9.2

12 1.2 24.9 49.0 156.8 21.9 9.3 10.6

Newborns Weight (g) Length (cm) Head circumference (cm) Chest circumference (cm) Calf circumference (cm) Calf skinfold (mm)

2745 47.9 34.2 30.8 10.3 4.0

2990 49.5 34.3 31.6 10.5 4.3

2782 48.3 34.1 31.5 10.2 4.7

Ladakhi women. Newborn sizes were not statistically signiﬁcant probably because of the dramatically unbalanced sample sizes (there were only 11 Tibetans sampled relative to 145 Ladakhis), but the magnitude of the difference (5% of average birthweight) is striking. Chest circumferences (as an indirect measure of lung size) were not different, but Tibetan newborns were signiﬁcantly fatter than Ladakhi newborns. On average, Ladakhi mothers were smaller than Tibetan women in weight (1.4 kg) and triceps skinfold (1.9 mm), while Tibetan mothers were older by three years and of higher parity. None of the maternal differences were statistically signiﬁcant. When Tibetan identity was added to the regression equation as a dummy variable (1 = Tibetan; 0 = Ladakhi) with maternal weight and EGA, it added no signiﬁcant predictive power but substantially reduced the difference in birthweight to 72 grams. This suggests that differences in weight between Ladakhi and Tibetan newborns are largely a result of Tibetan mothers weighing more. Differences in activity between Ladakhis and Tibetans were quite evident. Tibetan women usually do not work at agricultural tasks, since they do not generally own land. Most Tibetans live in Ladakh as refugees (or now, offspring of refugees), having ﬂed Tibet in the late 1950s and early 1960s, and hence they are landless. In some ways, however, they are better off than many Ladakhis. They are recipients of numerous sources of

98

An Ecology of High-Altitude Infancy

foreign aid, and women tend to work in tourism and handicrafts, among other nonagricultural tasks. They may also derive from a population with substantially longer antiquity at high altitude. These differences between Ladakhi and Tibetan newborns are consistent with data from Lhasa, Tibet, indicating much larger mothers and newborns (see Chapter 6). There is a small Indian population in Ladakh, especially during the summer months of intensive commercial activity. There were no significant differences in birthweight or length between newborns of native Ladakhis and those born to nonnative (non-Tibetan) women, but the observed differences (Table 4.6) were in an unexpected direction – that is, the migrants had marginally larger (37 g) babies. Newborns of migrant women were distinctly fatter than Ladakhi newborns, but they did not differ in chest circumference or other dimensions. Ladakhi women were signiﬁcantly shorter than nonnative women by 3.5 centimeters, although they did not differ signiﬁcantly in weight. When migrant status was added to the regression equation predicting birthweight (Table 4.4) as a dummy variable (1 = migrant; 0 = Ladakhi), the coefﬁcient on this variable became negative and reduced birthweight relative to Ladakhis by 36 grams. Thus, when maternal weight was taken into account, being a recent migrant from low altitude was associated with a very small reduction in infant birthweight. Still, compared to other studies that have shown substantial birthweight differences between newborns of native women and newborns of recent migrants (cf. Haas 1980, Haas et al. 1980), the difference between these groups in Ladakh was negligible. As with the comparison to Tibetans, activity differences during pregnancy cannot be assessed fully because nonnative women are mostly wives of military ofﬁcers or merchants and they do not generally engage in agricultural work. The exception are the Nepalis, who are usually in Ladakh as migrant laborers. They do very hard physical work and physicians note that they have worse health than Ladakhis, including poor birth outcomes. Buddhist-Muslim Differences in Birth Outcome Religious afﬁliation is also an important aspect of sociocultural variation in Ladakh. Indeed, it became an increasingly important part of people’s identity during the conﬂict of the early 1990s, although it had not traditionally been so. Signiﬁcant differences were found between birthweights of newborns of Ladakhi Buddhists (n = 118) and Muslims (n = 26); newborns of Buddhist women were on average 193 grams heavier than those born to Muslim women. There were no signiﬁcant differences in maternal anthropometric characteristics between these two groups

Big Mountains, Small Babies

99

of women, although Muslim women averaged more previous pregnancies. When maternal weight and EGA were controlled for, the difference between Muslim and Buddhist birthweights remained substantial, with Muslim newborns weighing 160 grams less. Dr. Ladhol had suggested that Muslim women were more vulnerable to urinary tract infections, which are associated with reduced birthweight (Braun et al. 1971; Sever et al. 1977), but there was no evidence of this from their records, and Muslim women did not report more problems during pregnancy. Although women in Ladakh generally do not smoke, smoking is on the rise among men, and Muslim women were more likely to report that their husbands smoked. Smoking is a well-known risk factor for fetal growth retardation and has independent effects beyond other factors such as height, weight, parity, prenatal care, and previous obstetric history (Meyer 1977; Secker-Walker et al. 1997). The effect of secondhand exposure to carbon monoxide and other air-borne noxious substances can mimic the effects of smoking, insofar as carbon monoxide competes with oxygen for uptake by hemoglobin, which has a much higher afﬁnity for carbon monoxide. For women who are passively exposed to domestic pollution, this could have a similarly negative effect on fetal oxygenation and hence birthweight. At present the birthweight literature does not support a strong effect of domestic pollution, at least in the form of passive exposure to tobacco smoke (Windham et al. 1999), although increased exposure (time and amount) during pregnancy seems to result in greater reductions in birthweight. In regression analysis, the effect of the husband’s smoking was not signiﬁcant in addition to maternal weight among Muslims – or Buddhists for that matter – and there were no differences in birthweight between newborns whose fathers smoked and those who did not. Surprisingly, considering that meat production was in the hands of Muslims and eggs were available only through the Kashmiri (Muslim) market during the boycott in 1990, Muslim women did not report consuming more meat, eggs, or milk during pregnancy than Buddhist women. There were, however, substantial differences in activity during pregnancy between the two groups. Muslim women reported working signiﬁcantly more during pregnancy and came from families that owned somewhat more land. When differences in activity and EGA were taken into account, Muslims had larger newborns than Buddhists, and although this difference was statistically insigniﬁcant, its magnitude was impressive – 111 grams! This again points to the signiﬁcance of work patterns during pregnancy as important determinants of birth outcome. Overall, then, religious identity served as a poor predictor of differences in birth outcome, and the analysis underscores the commonality of experience

100

An Ecology of High-Altitude Infancy

during pregnancy among Ladakhi women in the sample. However, since Muslim women did report working more during pregnancy, it does suggest that this is a meaningful axis of variation that needs further exploration.9 The Institutional Management of Pregnancy and Birth Ladakhi women triangulate among the health resources available to them in the management of pregnancy and childbearing. These include the biomedical resources of the hospital, private clinics in Leh and village health posts, Tibetan medical practitioners (amchi ), and shamans (lhaba/lhamo), among others. These practitioners are used for different reasons by pregnant women, and each addresses their concerns to varying degrees. Biomedical institutions in Ladakh attempt to confront reproductive problems and are increasingly utilized by pregnant women, but they are faced with severe ﬁnancial constraints and are based on training that is not always relevant to local reproductive health problems. On the other hand, Tibetan medicine, a textual tradition with a long history in Ladakh, suggests a long-standing, local, highly institutionalized and organized framework for pregnancy management. Shamans are often the ﬁrst ones sought out for routine health problems and they work within a more localized framework. All of these practitioners diagnose and prescribe changes in diet and behavior in the management of pregnancy, and the more formal traditions (biomedicine and Tibetan medicine) utilize an array of pharmacological substances. In general, Ladakhis consider pregnancy to be a healthy phenomenon, a stage in the normal process of childbearing and rearing that should begin within the ﬁrst year after marriage, while a woman is in her twenties. Women in the sample conformed to these norms for the most part; the average age at marriage for Ladakhi women in the sample was twenty, and the average age at ﬁrst birth for women in the sample whose current birth was their ﬁrst was twenty-three, although this did range from eighteen to thirty-seven. Over half (59%) of women surveyed noted that they had not experienced any health problems during their pregnancy. The most common complaints reported were pains in the back, heart, or stomach (34%, 31%, and 19%, respectively; the latter two are often combined by reference to the nyingka, or “heart-stomach”). In response to these discomforts, women sought relief from biomedical or traditional sources. Virtually all women in the sample had come for some prenatal care at the hospital, regardless of whether they perceived a speciﬁc problem, and 80% said they sought only the care of the obstetrician. Ten percent said they had been to see the lhaba/lhamo as well, most often for pains

Big Mountains, Small Babies

101

in the nyingka, the latter being associated with the nausea and vomiting of early pregnancy and heartburn-like symptoms. Another 6% said they had been to a lama or amchi as well as the Ob/Gyn for prenatal care. It thus appears that women are using prenatal care as a form of routine preventive care, while they make use of other nonbiomedical practitioners almost exclusively for curative care of speciﬁc problems that arise during pregnancy (Wiley 2002).

Biomedical Care For women who live in the main Indus valley and are thus connected to Leh via the main road and link roads, prenatal care by the Ob/Gyn at the district hospital is becoming increasingly popular, and women often travel distances of up to 200 kilometers to reach the hospital. When the hospital opened in 1981, only thirty babies were delivered there that year; in 1990, over 500 deliveries took place in the hospital. In 1995, there were 840 hospital births. Of women who seek prenatal care at the hospital, almost all go on to deliver there; likewise, of those who deliver there, the vast majority came for prenatal care. Through education, radio campaigns, and word of mouth, the Ob/Gyn has advocated widespread prenatal care as a way of combating maternal, perinatal, and infant mortality, and women have responded eagerly. Dr. Ladhol’s high rank in the traditional Ladakhi hierarchy and her empathy undoubtedly facilitate women’s decisions to have a hospital birth, despite breaking with the tradition of giving birth in the husband’s household and risking the evil eye (mik-ya) in this public context. Because of Dr. Ladhol’s authority, mothers-in-law are also less likely to object to hospital births and women are likely to feel more conﬁdent in the biomedical hospital context. The majority of women in the study had come for ﬁve or more prenatal visits. Monthly visits are advised through the eighth month, after which weekly visits are suggested. However, the number of prenatal visits was not signiﬁcantly related to birthweight. These visits may have acted to prevent complications from manifesting, or more likely, prenatal care has little impact in a context where the forces affecting birthweight are well beyond biomedical control (i.e., a woman’s diet, weight, and activity).

The Role of Tibetan Medicine in Pregnancy Management Tibetan medicine has substantial treatises on pregnancy and birth that describe in detail the course of embryology (Dhonden 1980b; Rechung 1973), specify a pharmacoepia of herbs and pills for the treatment of

102

An Ecology of High-Altitude Infancy

speciﬁc ailments (Buth and Navchoo 1988), and offer general advice on diet, activity, and morality. For example, During pregnancy, the mother should avoid eating food given by those who have not kept their moral obligations, nor should she be a guest of such people. She should not wear second-hand clothes. Her food should be neither very hot nor very cold, and strong liquors, large amounts of beer and hot, spicy and sour food should be avoided. She should eat nourishing food, taking care not to overeat. Heavy work and frequent sexual relations should also be avoided. If these restrictions are not observed, there is no need to mention the great pain caused to the unborn child. . . . While being concerned about her health, the expectant mother should not remain too inactive. She should do light housework, circumambulate temples, take walks, etc. Walking exercises the child, making its body supple, paving the way for an easy birth. . . . When the time of birth is near, the mother should take strong nourishing food such as milk, meat soup or broth with rice to suppress the airs in her body. (Sangay 1984:5)

As this passage indicates, Tibetan medical treatises are especially concerned with how the course of fetal development is affected by maternal behavior. There is a very close relationship between the fetus and the mother, so the activities, thoughts, diet, and the physical, mental, or spiritual health of the mother impact immediately upon the fetus, which in turn is said to use the mother as a vehicle for experiencing the external environment. Doctrine suggests that in the third week after conception the mother can alter the karmic destiny of the fetus by the administration of special medicines, by observing religious rites, and by wearing amulets. Diagnosis of the “pregnancy pulse” can reveal the sex of the fetus, as can the position of the fetus in the womb. A son is indicated when the right side of the abdomen is higher.10 Tibetan medicine is based on a humoral framework for interpreting states of the body, and pregnancy is considered a “hot” state. During this time “cold” foods should be consumed less frequently or avoided altogether, so as to avoid accumulation of water and wind in the body. In addition, women should avoid becoming cold by wearing woolen garments and avoiding windy places, even in the heat of summer. On the other hand, excess heat generated by the pregnancy may result in vomiting, which may be relieved by the consumption of “cold,” sour foods such as apricots, which are grown in Ladakh. Cravings for such foods are considered to be expressions of the fetus’s wants. In some ways formal Tibetan medicine has provided Ladakhis with ideas about the progress of pregnancy and appropriate behavior necessary to produce a healthy child, but it does not play an active role in the formal management of the reproductive process in contemporary Ladakh. Amchi have not been closely involved in providing prenatal care or delivering

Big Mountains, Small Babies

103

babies. Relative to other practitioners, amchi are rarely sought out for pregnancy complaints, and they currently refer obstetrics patients to the obstetrician, admitting that they do not have the ability to care for such clients. Moreover, most women have only the vaguest ideas about embryology and the physiological progress of pregnancy as outlined in Tibetan medicine. As one woman noted, “There’s some way of knowing if it’s a boy or a girl – something about where you feel pain, but I don’t know what this is.” Although some authors, impressed by the image of Tibetan medicine promoted by the Tibetan population in exile in India (centered in Dharamsala, in the foothills of the Himalaya), have stressed the “naturalness” and “effectiveness” of Tibetan medicine’s treatment of pregnancy and childbirth (cf. Dhonden 1980a), this is not at all evident in Ladakh. Furthermore, note that based on the prescriptions and proscriptions outlined earlier, much of Tibetan medicine’s basic approach to pregnancy is very similar to current biomedical practice, although the explanatory frameworks for them are quite different (Wiley 2002). Other Healers Shamans (lhamo/lhaba) are sought out for complaints during pregnancy, especially for heart or stomach problems, but no more so than they are consulted by the general populace. One common problem associated with pregnancy from the women’s perspective is the mik-ya, which may be cast against the fetus by others envious of the baby or its mother, and the shaman may be able to uncover the culprit. Women will often hide or conceal their pregnancy to avoid the evil eye, and they are often reluctant to discuss their condition while pregnant for fear of attracting it. Indeed, to identify a woman as pregnant is often difﬁcult because traditional Ladakhi dress is so bulky that it can easily conceal her swelling abdomen. Unlike many other cultural contexts, in Ladakh there was no tradition of midwifery (often referred to as traditional birth attendants, TBAs). Women often sought the help of an older woman such as their mother or mother-in-law to help them with problems encountered during pregnancy or birth. In the villages there are auxiliary nurse-midwives (ANMs) and they may assist at a home birth, but many women continue to deliver with only the assistance of female relatives. Thus the management of pregnancy and birth has been very localized and informal. I think it is likely that the absence of a tradition of midwifery or the involvement of amchi in birth is what has made it possible for biomedical institutions, particularly the hospital, to expand its purview to include pregnancy and birth and to become the central locus for the management of these processes

104

An Ecology of High-Altitude Infancy

(Wiley 2002). The eagerness with which Ladakhi women appear to have embraced biomedical control over reproduction indicates that there may have been a vacuum in this area that was not being ﬁlled by traditional means. It may also suggest that women felt the need for greater institutional management of a process that is clearly fraught with difﬁculties in this particular context. Summary Among this large sample of Ladakhi newborns, birthweight was low, and a large proportion of newborns were in the low-birthweight categories. Many had suffered both chronic and acute growth disruption, indicating increasing stress as pregnancy progressed. Variables that accounted for some of the variation in birthweight included sex and EGA, but also maternal nutritional status and maternal workloads. These in turn are products of a household ecology that requires women to work at agricultural tasks during their pregnancy, particularly women having their ﬁrst child, as they have lower status and power in the patrilocal household. There were other axes of variation; Buddhists had larger babies than Muslims, in part because Muslim women appeared to work harder during pregnancy; Tibetan women had larger babies than Ladakhis, perhaps reﬂecting their longer ancestry at high altitude but also overall better maternal nutritional status and reduced workloads; and indigenous women and migrant women delivered similarly sized babies, when maternal weight was controlled for. Socioeconomic status does not seem to be a major axis of variation in birth outcome in Ladakh; maternal diet and work, whose relationship to SES is not well understood, appear to be more important.

5

An Ecology of Infancy in Ladakh

The previous chapter documented the small average size of newborns in Ladakh and substantial variation within the population with respect to birth outcome that could be attributed, in part, to differences in maternal background and behavior during pregnancy. Size at birth is an important descriptor of birth outcome, but it is important to keep in mind that from an evolutionary perspective, birth outcome is signiﬁcant insofar as it is related to fertility and mortality. Stillborn offspring and those who die before maturity potentially reduce a woman’s reproductive success (ﬁtness) because a women has at most about thirty years during which she can reproduce, and each birth takes up at least nine months and usually much more, given a period of postpartum infertility due to breastfeeding. Each premature death therefore reduces the overall number of surviving offspring that a woman can have. Therefore, if there is an association between measurable birth outcomes and early mortality risk, then those birth outcomes that are associated with reduced mortality can be considered adaptive because they increase the ﬁtness of a woman and her offspring. With regard to the speciﬁc case of birth outcome and infant mortality at high altitude, Haas (1980:261) has noted: It is likely that variation in fetal growth at high altitude in the Andes is due in part to maternal phenotypic variation, which is a consequence of each woman’s individual adaptation to hypoxic and nutritional stress during her period of growth. . . . Selection could have operated through the differential reproductive success of women who were better able to respond to high-altitude stress during their period of growth by developing an adult phenotype capable of producing relatively more viable offspring. The birthweight of the neonate, [and] the perinatal and infant morbidity and mortality of the offspring . . . should provide good indications of the relative reproductive capacity of the female at high altitude. Furthermore, these measures of reproductive capacity should vary with the level of maternal adaptation to the environment.

This variation is observable both within high-altitude populations (independent of ancestry) and between populations inhabiting different 105

106

An Ecology of High-Altitude Infancy

altitudes. Thus far, most of the studies of birthweight and altitude have concentrated on interpopulation variation to assess the role of hypoxia. But as we saw in the previous chapter, hypoxic stress does not occur independent of the other stresses that women face before or while they are pregnant. Thus there is substantial intrapopulation variation in birth outcome and presumably in overall reproductive success. In this chapter I investigate the determinants of early mortality, particularly mortality in the ﬁrst month of life. These determinants were selected from the measures of birth outcome described in Chapter 4. If they are found to be signiﬁcant determinants of survival, then we can trace the pathways of inﬂuence on infant survival from maternal context, through pregnancy and fetal growth and development, to the contexts of infancy and childhood. The analysis of the neonatal anthropometric data will be complemented with data from a large set of reproductive histories of women who delivered in the hospital, as well as pediatric admissions data that provide insight into speciﬁc causes of infant morbidity and mortality. Determinants of Infant Mortality A useful framework for understanding the paths of inﬂuence on infant survival is Chen’s proximate determinants model (Chen 1983), diagrammed in Figure 5.1. Chen outlined four sets of proximate determinants of infant mortality: parental (maternal), nutritional, infection, and child care. The ﬁrst three act in both the prenatal and postnatal periods, and, importantly, birthweight is the mediating factor between pre- and postnatal conditions. This framework emphasizes individual characteristics of infants, their mothers, and their caretakers that lead to the survival or death of an infant; these then need to be placed within a larger social, cultural, and ecological context. For example, the resources available to a mother, family, household, community, or region inﬂuence an infant’s health, as do natural and socioecological conditions such as infectious pathogens and their vectors, contaminated water, crowded and inadequate housing, or abiotic features such as hypoxia or pollution. Millard’s more recent model (Millard 1994; see Figure 1.2) organized these social and ecological factors hierarchically. Infectious disease and malnutrition, which often act synergistically to produce a “downward spiral of health,” are the most common proximate causes of early mortality and are outcomes of an “intermediate” sphere of inﬂuence at the household level. The household, in turn, is nested within a natural environment, a cultural context, community, nation state, and the global economy (the “ultimate determinants”). So, we can start with characteristics of birth outcome and use them to predict mortality risk, but then examine the

An Ecology of Infancy in Ladakh

107

PARENTAL FACTORS Maternal age and parity Birth intervals Maternal nutritional status

NUTRITIONAL FACTORS Fetal Factors Maternal diet and nutrition during pregnancy

INFECTION FACTORS

BIRTHWEIGHT

Child Factors Diarrhea Respiratory infections

Child Factors Breastfeeding Supplementation Food intake

CHILD-CARE FACTORS

Fetal Factors Infections during pregnancy

GROWTH AND DEVELOPMENT; ILLNESS

Quality of child care Availability and use of health services MORTALITY SURVIVAL

Figure 5.1. Chen’s model of the proximate determinants of infant mortality (Redrawn from Chen 1983, with permission of the author.).

larger contexts that both determine their relationship and mediate their effects. The factors outlined here affect the risk of mortality differently as an infant ages. In the ﬁrst month of life, neonatal biological characteristics that derive from the fetal and maternal environment are most crucial to survival. As the infant grows and develops over the course of the ﬁrst year, the determinants of its survival change. Neonatal characteristics, derived from maternal factors, have less salience while other environmental conditions play a more important role as infants are increasingly exposed to a wider array of environmental risks. Exposure depends on maternal, familial, and community resources, which are in turn inﬂuenced by macro-historical, political, and economic factors, as well as ecological, demographic, and sociocultural conditions that ultimately inﬂuence

108

An Ecology of High-Altitude Infancy

an infant’s likelihood of survival or death (Hrdy 1999; Millard 1994; Scheper-Hughes and Sargent 1998). The determinants of infant mortality/survival in Ladakh are laid out in this chapter. In particular, the connection between newborn and maternal health, mediated by hypoxia and maternal diet and activity patterns, is analyzed. As the infant ages, this close association gradually dissolves as the infant becomes more autonomous and hence more directly affected by the household and community environment. Infant-caretaker and household relations and the social value attached to a child’s characteristics manifest in infant health parameters, and these in turn derive from larger suprahousehold forces that impinge on household decision making, economic viability, and ecological vulnerability. Although the focus is on neonates and infants, some comparative data on children (ages 1–4) are presented to provide a sense of the uniqueness of the infant (especially neonatal) period with respect to health and survival. Follow-up of Infants in This Study At the postpartum interview, mothers were asked to return to the hospital one month after the birth for a follow-up interview. New parents often observe a period of postpartum seclusion, but it does not generally go beyond one month. They were speciﬁcally told that they did not need to bring their infant because I anticipated that some of the infants would die in that ﬁrst month and I did not want to bias the sample in favor of surviving infants. At the follow-up, the mothers were again weighed; arm circumference and triceps skinfold measurements were taken; and they were questioned about their own health as well as dietary and activity patterns. Any infant who was brought in was weighed, and its length and head circumference were measured. Women were asked about the health of their infant, whether it had been taken to any healers and for what problems, if it had been named, what its diet and breastfeeding patterns were as well as the infant’s caretaking and sleeping arrangements. The follow-up procedure proved problematic because many women did not return after one month. There are probably several reasons for this. First, many women who delivered in the hospital were from surrounding villages, and returning to the hospital would have required motorized transport by bus, taxi, or truck. Coming for the postpartum visit would have entailed a day’s travel and exposed the mother and her very young and vulnerable charge to various hazards including the mik-ya, noxious fumes, crowds, and other dangers. Second, since I had intentionally left the explicit purpose of the visit vague, emphasizing that the visit was to see how the mothers were getting along, not the infants, it is possible that

An Ecology of Infancy in Ladakh

109

during this busy agricultural season, women did not see the value in the follow-up visit. Third, I suspect that women whose infants had died saw no point in returning, assuming that in fact I was most interested in their infants or because they wanted to dissociate themselves from the previous birth. Also, because the labor room was located in the hospital wards, to which access was restricted except during visiting hours (which in reality were not strictly enforced), it is possible that some women came but were not admitted, despite instructions to the guards on duty that they should be. For women who did not return, I employed other means to obtain information on the status of their infants. Some women saw Dr. Ladhol for postpartum health care or for contraception in the months after the birth and I was able to obtain information on them and their infants from her or her records. If the infant had been sick and brought to the pediatrician, I could obtain a record of that visit. In other cases, I inquired among acquaintances who lived in the same village or neighborhood or I went to the village and asked myself. If I could locate the women from the study, I conducted the interviews, but not the anthropometric measurements. When I returned to Ladakh to do more extensive follow-up in 1991 and 1996, I located many of the children who had been born in 1990, either by village, or by ﬁnding reference to them in obstetric or pediatric records at the hospital. Inevitably, follow-up information was not available for all women and infants, as some could not be located.1 This presents some problems for the analysis because the reasons follow-up information was available could have been related to the health or survival status of their infant. For example, some of the data are from hospital pediatric admissions records and hence disproportionately reﬂect children who were sufﬁciently ill to be admitted. On the other hand, for the women and infants for whom I have no follow-up information, it is possible that a high percentage of these infants died. Infants born to women from more remote villages are disproportionately missing from the follow-up. Thus the follow-up data are not complete, nor do they constitute a random subset of the study sample. However, it is not clear in what direction the bias may lie because infants who had died were not seen, or those who were ill, and thus at a higher risk of mortality, were more likely to have been seen at the hospital. The data from the follow-up take the form of the age at follow-up and survival status. These became the dependent variables in hazard analysis (also known as Cox Regression), which generated mortality estimates based on characteristics of the infants at birth (and other independent variables, as relevant). Hazard analysis allows those infants who were not followed for the entire period of interest to be used in the estimation

110

An Ecology of High-Altitude Infancy

procedure with their follow-up age, which is an advantage over logistic regression analyses of mortality.2 Speciﬁcations for the hazards model can be found in SAS Institute Inc. (1991). Neonatal Characteristics and Infant Mortality Of the 168 infants in the prospective study, there were seven conﬁrmed deaths. Their ages at death were scattered from three days after birth to seven months. There were no known deaths to older infants or children (based on follow-up in 1991 and 1996). Contrary to my initial fears, no babies died in my hands or shortly after being assessed. The infants who died are described below: 1. A three-day-old male, the ﬁrst child of a Buddhist couple from Alchi, a village about four hours from Leh. He weighed 1998 grams at birth, and was reported as “premature” – though this was based on his small size rather than being an accurate assessment of his maturity. The mother had been to numerous prenatal visits and received the usual supplements of iron and folic acid along with a tetanus shot. He was reportedly not feeding properly, and the pediatrician simply noted that he died of “cardiorespiratory arrest.” 2. A ﬁve-day-old female, the second child of a Buddhist couple from the village of Saboo, near Leh. At birth she weighed 2495 grams, and was full term. The ﬁrst child of this couple, a son, had also died shortly after birth. The mother had come for numerous prenatal visits, and there was no sign of any problems: her blood pressure was normal, and she was given iron and folic acid supplements and a tetanus shot. The baby was jaundiced and dyspnoic (having difﬁculty breathing) just before she died. 3. A twenty-three-day-old male, the second child (ﬁrst son) of a couple who were not Ladakhi but from Himachal Pradesh. The husband was in the Indian military. The baby weighed 1722 grams at birth. The mother had come for several prenatal visits and received the usual supplements and tetanus shot. She reported no problems during the pregnancy. The child died of a respiratory infection. 4. A ninety-day-old male, the fourth child of a Buddhist couple from Spitok (a village close to Leh). He weighed 2608 grams at birth. The parents had an older son and daughter, and another son who had died at eighteen months. The baby was reportedly not sucking properly, then died of respiratory distress. 5. A 173-day-old female, the second daughter of a Muslim couple from Thiksey (a large village, easily accessible from Leh). She weighed 2523 at birth and was full term. The couple had a previous daughter who had

An Ecology of Infancy in Ladakh

111

died at eighteen months. The mother had come for several prenatal visits and gotten the routine supplements and tetanus shot. Cause of death was bronchial pneumonia/respiratory distress. 6. A six-month-old boy, the ﬁrstborn of a Muslim couple from Leh. He was full term and weighed 2636 grams at birth, having been born by cesarean section. His mother had numerous prenatal visits that included normal blood pressure, iron and folic acid supplements, and a tetanus shot. Cause of death was not reported. 7. A 202-day-old male, the ﬁrst child born to a Buddhist couple from Chemre (just outside of Leh). He weighed 2721 grams and was full term at birth. The mother had come for prenatal care and received the usual blood pressure monitoring, supplements, and tetanus shot. Cause of death was respiratory distress. Although there were only seven conﬁrmed deaths, their descriptions do not point to any clear similarities among the dead infants or their parents. These infants are a heterogeneous group, of different sexes, religions, villages, parities, and so on. Note, however, that they all weighed below the mean birthweight, most died of respiratory disease, and they all died within the ﬁrst seven months of the ﬁrst year, which provides the ﬁrst clues as to the importance of birthweight and respiratory problems as predictors of infant mortality and a pattern of high early infant mortality. A negative relationship between birthweight and early mortality has been widely demonstrated, with the relationship being strongest in the neonatal period, although smaller newborns retain much higher mortality risks in the postneonatal period as well (Ashworth 1998). Babies with fewer biological resources are much more likely to suffer from a variety of health problems and are more vulnerable to mortality from infectious disease in the ﬁrst year of life than infants of higher weights because of compromised immune function (Ashworth 1998; Victora et al. 1988). This vulnerability is modiﬁed by other factors. Sex, gestational age, growth disruption at different stages of pregnancy, and altitude, for example, have all been associated with different risks for a given birthweight or for the LBW infant, indicating that a given birthweight does not predict mortality similarly across populations (Wilcox 1993, 2001) The results of hazard analysis linking measures of birth outcome to mortality in this study are summarized in Table 5.1. In general, increases in size at birth reduced the probability of death, but some measures were more predictive of mortality than others. Birthweight, head circumference, and fatness (all of which are highly correlated) all predicted mortality in the ﬁrst month. The Ponderal Index signiﬁcantly predicted mortality within the ﬁrst six months. Chest circumference and length were never signiﬁcantly associated with mortality, nor were gestational

112

An Ecology of High-Altitude Infancy

Table 5.1. Coefﬁcients and signiﬁcance of anthropometric indices of newborns for predicting early mortality

Birthweight Birth length Head circum Chest circum Ponderal Index Sum of skinfolds

0–1 month

0–6 months

0–12 months

1–6 months

6–12 months

−.004; p < .03 NS

−0.002; p < .005 NS

NS

NS

NS

NS

−0.891; p < .05 NS −6.445; p < .06

−0.003; p < .01 −0.288; p < .06 −.7300; p < .02 NS −4.783; p < .05

NS

NS

NS

NS −4.001; p < .04

NS NS

NS NS

−0.792; p < .02

−5.017; p < .01

−0.311; p < .05

NS

NS

Notes: NS = not statistically signiﬁcant at p < .05.

age or sex. While the anthropometric measures predicted mortality up to one year, none had signiﬁcant effects between one and six months or six to twelve months, which indicates the high baseline level of neonatal mortality and its strong inﬂuence on mortality probabilities in the ﬁrst year. Birth size thus contributes most importantly to the risk of death in the ﬁrst month but is less important in determining later mortality risk. The average birthweight for those infants who died in the ﬁrst month was 2072 grams (−690 g from average); the average for all infants who died was 2390 grams (−375 g less than average); and the average birthweight for those who survived was 2770 grams. Many of the smaller infants were thus “weeded out” in the ﬁrst month, but at the same time, the average weight of those who survived is very close to the mean birthweight, and so many small newborns survived. Because weight is a convenient summary of the circumferences and fatness of a newborn, I focus here on the birthweight/mortality relationship. Figure 5.2 illustrates the mortality probabilities generated from hazard analysis for the range of birthweights. Increasing birthweight was clearly associated with lower probabilities of mortality, but this effect was strongest within the ﬁrst month and for the lower weights. At six months and one year, the mortality curves had the same shape and differed only slightly, even though the time periods represented by the curves were disproportionately larger (i.e., 0–6 months; 0–12 months). Interestingly, there was no evidence for an increase in mortality among larger newborns, as most studies show, but that is most likely because there were

An Ecology of Infancy in Ladakh

113

0.5 0.45 Mortality Probability

0.4 1 month 6 months 1 year

0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 1750

2000

2250

2500

2750

3000

3250

3500

3750

Birthweight (g)

Figure 5.2. Birthweight and mortality in the ﬁrst year of life

no large (>4000 g) newborns in the sample, which is the weight at which mortality generally begins to rise (see Figure 2.2). If we consider the birthweight–mortality relationship in combination with the birthweight distribution from Chapter 4, it is immediately clear that there is likely to be a very high level of neonatal mortality in Ladakh, as many newborns fall into the high-risk groups. That is, 27 percent weighed below 2500 grams, which remains a convenient cut-off, more or less, for when mortality risk begins to increase rapidly; infants weighing below 2000 to 2500 grams had more than four times the mortality of those weighing at least 2500 grams, which is similar to what is observed in other populations with similar frequencies of LBW (Ashworth 1998). And 50 percent of all newborns weighed less than 2750 grams and hence encountered probabilities of dying of greater than 1 percent to 2 percent in the ﬁrst month. Below 2000 grams, mortality increases sharply (>10% risk of death); indeed, Ladakhi mothers and the nursing staff, as well as Dr. Ladhol identiﬁed 2000 grams as the weight at which newborns are perceived as sickly and vulnerable. Their perceptions are well founded in reality, as these infants do have much higher chances of dying. From a Ladakhi perspective, 2000 grams makes a reasonable standard for assessing newborns with a higher relative risk of death. However, 2500 grams, the standard in wide usage, is still associated with a substantial risk of mortality. But when that is the norm, it is understandable that only infants in dramatically higher risk categories are singled out as particularly vulnerable.

114

An Ecology of High-Altitude Infancy

Low PI measurements (below 2.25) and small skinfolds were associated with increased risks of mortality within the ﬁrst 180 days, again suggesting that decreased fat deposition relative to length may pose a mortality risk. Overall, neonates who weighed less than 2500 grams and were of low PI made up almost 13 percent of the sample and thus were at risk not only because of their low weight but also because of the body composition that contributed to that weight. While the literature contains conﬂicting results vis-`a-vis the contribution of PI to mortality (cf. Ashworth 1998; Haas 1990; Neuﬁeld et al. 1999), this study conﬁrms the ﬁndings of Haas, Balcazar, and Caulﬁeld (1987) that wasting, as evidenced by low PI, had more negative effects on early survival. Further support for this conclusion is the ﬁnding that length was never a signiﬁcant predictor of mortality, which suggests again that reductions in weight – not length – are more signiﬁcant in determining mortality risk. Because chronic fetal growth retardation (stunting) is the norm in Ladakh, it is the background against which acute growth disruption may occur in a small subset of infants. Those who face additional acute shortages of nutrients in the last trimester are at greater risk after birth. Interestingly, chest circumference was never associated with mortality during the ﬁrst six months of life. If newborn chest size reﬂects lung size, this indicates that larger lungs offer no advantage to neonates, which is surprising in this hypoxic environment. Estimated gestational age had no signiﬁcant effect on survival, either as a single independent variable or in combination with birthweight. Previous studies have consistently indicated that for a given birthweight, there is a decline in the mortality rate with increasing gestational age up to forty-six weeks, while at a given gestational age, there is a decrease in neonatal mortality with increasing birthweight (Lubchenco et al. 1972; Figure 2.2). The deviation from this well-established pattern can be explained in part by the limited variation in EGA among the Ladakhi sample (77% of the gestational age estimates were recorded to be 38 weeks). Furthermore, fundal height is a very indirect measure of gestational age and produces a crude categorization of fetal age that does not allow us to state deﬁnitively its role in determining infant mortality risk in Ladakh. However, as this and other studies have shown, lower gestational age is not the driving force behind reduced birthweight at high altitude, and it is thus not a primary causal factor for infant mortality in this context. Although, on average, females weighed 175 grams less than males at birth and were disproportionately represented among infants of low birthweight, they were not more likely to die than males. Other studies have shown this same effect. That is, the mortality risk of a male infant at a particular birthweight tends to equal the mortality risk of a female

An Ecology of Infancy in Ladakh

115

100 to 150 grams lighter (Lubchenco et al. 1972; Wilcox 2001). Therefore, lower birthweight does not carry the same risk for females as it does for males. In this study, this resistance was not evident in their average birthweights but only in the reduced mortality risks that these weights incurred. In sum, birthweight is a good overall measure of birth outcome as a predictor of mortality risk. It thus stands to reason that the determinants of birthweight are important indirect inﬂuences on mortality, and so maternal weight, diet, and activity patterns during pregnancy all ultimately affect the fate of the neonate. The effects of maternal biology and behavior during pregnancy are most apparent in the early part of infancy and diminish as the infant ages and develops more independent relationships with his or her environment. As time passes, neonatal characteristics become less inﬂuential. To begin the analysis of other household and maternal variables that might be affecting neonatal mortality in conjunction with birthweight, the following characteristics were each added as independent variables in hazard analyses: parity (birth order of newborn, i.e., ﬁrstborn, secondborn, etc.), month of birth, number of previous stillbirths or abortions, religion, ethnicity (Tibetan or Ladakhi, and migrant or native), mother’s and father’s occupation, mother’s and father’s education, and landholdings. Interestingly, none of these factors added any predictive power to the basic equation. That is, with birthweight already taken into account, these other variables did not seem to matter. What this suggests is that they act through birthweight, which is likely considering that some of these factors were found to inﬂuence birthweight (ethnicity, religion, parity, month, occupation). Others had no effect on birthweight or neonatal mortality, suggesting that they are not important sources of variation in reproductive health in Ladakh. Hospital Birth and Pediatric Admissions Records The prospective data analyzed earlier provide insight into the neonatal and hence maternal determinants of early mortality. The analysis of birth records and reproductive histories, gathered from women in the sample and from a larger sample of all women who delivered in the hospital in 1990 (n = 576) and a smaller sample of those who delivered in 1995 (n = 244), provides further information on the level, pattern, and other determinants of infant mortality among a larger cross-section of Ladakhi infants. I was able to record their entire prenatal ﬁle, and the sex of their newborn, estimated gestational age, type of delivery, and any problems noted while the mother and infant remained in the hospital, and the

116

An Ecology of High-Altitude Infancy

infant’s age at discharge (usually 24 to 48 hours for those without any notable disease). For women with previous births, reproductive histories described in the records contained data on the ages and sex of their living children, and age at death and sex of those who had died. These data from 1990 and 1995 were used to develop cohort-speciﬁc statistics on neonatal, infant, and child mortality, calculated as the number of children who died before a given age divided by all of the children born for whom there was information on age and, if available, sex. These statistics are not exactly analogous to the standard infant mortality rate, which is calculated as number of deaths before age one in a given year/number of births in that year.3 In addition, I compiled data from the records of births, stillbirths, and neonatal deaths (by month and sex) that occurred in the obstetrics ward for the years 1990 through 1995. These were useful in uncovering seasonal variation in births and very early deaths, as well as sex differences in the death rates. With the cooperation of the pediatrician at the Leh hospital, I was also able to obtain pediatric admissions records for 1991–1995. Information available from these records included age at admission, diagnosis, sex, village of residence, date of admission and discharge, and survival status. These were helpful in identifying children who had been in the study who had subsequently been admitted to the hospital, and also in understanding age, gender, and seasonal patterns of sickness among infants and children. They can be used in conjunction with the prospective data and reproductive histories to elaborate patterns and causes of infant morbidity and mortality. As an important caveat, these data say nothing about the overall prevalence of disease or death among children in the population, but only the relative frequency of different causes of morbidity or mortality among pediatric hospital admissions. Pattern and Level of Infant Mortality Table 5.2 reveals an infant mortality rate of 117/1000 among previously born children of known sex and age for parities one through ﬁve (n = 274), calculated from the reproductive histories of women who delivered in 1990. The vast majority (84%) of this mortality occurred in the neonatal period. If only ﬁrst births were considered (n = 164), higher mortality rates were found: IMR = 152/1000, NMR = 128/1000, with the latter again accounting for 84% of all infant mortality. Compared with the mortality probabilities of children of all parities, children of low parity had higher mortality rates, and ﬁrstborn children had higher rates of mortality than secondborn children. However, in the whole sample there

An Ecology of Infancy in Ladakh

117

Table 5.2. Neonatal and infant mortality rates from reproductive histories, 1990 and 1995 Perinatal mortality 1990 All parities (n = 274) Males Females First births (n = 164) Males Females First & second births (n = 239) Males Females All parities – including those with no sex recorded (n = 310) First births – including those with no sex recorded (n = 185) 1995 All parities (n = 192) First births (n = 94) First & second births (n = 130)

∼210/1000

160/1000 125/1000 147/1000

Neonatal mortality

Post-neonatal mortality

Infant mortality

NMR/ IMR

98/1000 103/1000 94/1000 128/1000 129/1000 127/1000 105/1000

19/1000 29/1000 7/1000 24/1000 36/1000 12/1000 16/1000

117/1000 132/1000 101/1000 152/1000 165/1000 139/1000 121/1000

84% 78% 93% 84% 78% 91% 87%

101/1000 108/1000 148/1000

25/1000 9/1000 35/1000

126/1000 117/1000 184/1000

80% 92% 81%

173/1000

54/1000

227/1000

76%

104/1000 74/1000 85/1000

47/1000 75/1000 61/1000

151/1000 149/1000 146/1000

69% 50% 58%

were very few children of parities greater than two (only 13% of sample). Males and females had similar neonatal mortality rates, but postneonatal mortality rates were two to four times higher among males. Because many children were reported in the reproductive histories without their sex noted, but simply as alive or dead at a given age, analyses were performed including these infants, which increased the sample sizes. Not surprisingly, most of the infants reported without sex had in fact died, which dramatically increased the mortality rates (IMR = 184/1000; NMR = 148/1000), especially for ﬁrstborns (IMR = 227/1000; NMR = 172/1000). This disparity may be explained by a bias in the sample in which women whose ﬁrst offspring had died very young were predisposed to deliver their next child in the hospital. At the same time, it is worth noting that parity was a strong positive predictor of birthweight in the prospective study; ﬁrstborns weighed the least, and those of parity three and higher weighed over 100 grams more on average than

118

An Ecology of High-Altitude Infancy

ﬁrst- and secondborn offspring. This may contribute to the higher mortality of ﬁrstborn children. The same analysis was done using the reproductive histories of women who delivered in 1995. Because the sample was much smaller, all previous births were included regardless of whether the child’s sex was noted. The IMR was 151/1000, and the NMR = 104/1000; NMR was 69% of IMR. For ﬁrst births, the IMR was 149/1000; NMR was 74/1000 (50% of IMR). Rates were similar for parities one and two. Thus, in 1995, lower parity offspring had much lower neonatal mortality but higher postneonatal mortality than those of higher birth orders. Hospital births had been increasing over this time period, and it would be nice to be able to attribute the lower mortality rate to the success of Dr. Ladhol’s policy of encouraging women who are pregnant with their ﬁrst child to deliver in the hospital (i.e., ﬁrstborns reported by women who delivered in 1995 were more likely to have been born in the hospital than those in 1990), although this is a conjecture. Overall, the rate of infant mortality seems to have gone down slightly over the previous ﬁve years. The contribution of neonatal to infant mortality has declined, while there appears to have been a modest increase in postneonatal mortality. In both 1990 and 1995, child deaths were negligible – only ﬁve were reported in 1990 and three in 1995 out of over 500 births – but this is also a reﬂection that most of the previous children born were still under ﬁve years of age. Overall, these results indicate a consistent pattern of very high mortality very early in life, with a dramatic decline in mortality after the ﬁrst month. Related to this is the high rate of perinatal mortality (late fetal deaths and those occurring <1 week after birth) in this population, which was 210/1000 in 1990 and 160/1000 in 1995. Over 25% of the women who delivered in 1990 and over 9% of the women who delivered in 1995 had experienced at least one stillbirth (this lower number reﬂecting, perhaps, broader usage of the hospital in 1995). Stillbirth rates of hospital deliveries from 1990 to 1995 averaged 4%; neonatal deaths occurred in the hospital at a rate of 2% (as a proportion of hospital births). Thus almost 6% of hospital deliveries resulted in death before the mother was discharged. Late pregnancy and the early neonatal period are treacherous times for the fetus and newborn. Hospital Deaths Between 1991 and 1995, 118 infant and 23 child deaths occurred in the hospital. These deaths include neonates who died in the obstetrics ward (derived from birth records) and infants and children admitted to the pediatrics ward. As Figure 5.3 indicates, the majority of pediatric deaths

An Ecology of Infancy in Ladakh

119

0.6 All deaths Females Males

% of all deaths

0.5 0.4 0.3 0.2 0.1 0 <1 month 2-6 months

7-12 months

13-23 months

24-35 months

36-60 months

Age

Figure 5.3. Patterns of infant and child hospital deaths.

occurred to neonates (57%); among infant deaths, 67% were neonates. There were sex differences in the age pattern of infant deaths – more males died during the neonatal period, but slightly more females died during later infancy (7–12 months), a pattern somewhat at odds with that derived from reproductive histories. As for causes of death, more than 60% of all pediatric hospital deaths (not including those in the obstetrics ward for whom I do not have cause of death information) were attributed to some form of respiratory disease. The relative frequency of diagnoses varied by age as shown in Figure 5.4. The most common cause of death for neonates was septicemia, classiﬁed here in the “other” category, followed by respiratory disease. Proteinenergy malnutrition (PEM) and gastrointestinal diseases were very rare causes of neonatal death. For infants age 1–11 months, respiratory ailments were major threats, with the remaining causes relatively evenly split between gastrointestinal diseases, PEM, and “other.” After the ﬁrst year, respiratory disease remained the most common cause of mortality but decreased relative to other categories, whereas deaths due to PEM increased. There were no gender differences in the frequency with which respiratory illness, gastrointestinal disease, or malnutrition resulted in death. Most of the pediatrics hospital deaths occurred in the winter months, with a small rise in summer. Two-thirds of the deaths of neonates who had been brought to the hospital (that is, they were not in the obstetrics

120

An Ecology of High-Altitude Infancy

0.8 Resp. Disease Gastro. Disease PEM Other

% of deaths in each age group

0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 <1 month (n=80)

1-11 months (n=38)

13-23 months (n=12)

24-35 months (n=6)

36-60 months (n=5)

Age group

Figure 5.4. Causes of pediatric hospital deaths.

ward) occurred in the months between November and February. With regard to the monthly variations in neonatal deaths in the obstetrics ward (which were not predicated on a neonate’s being brought into the hospital, the likelihood of which varies by season), stillbirths, and births illustrated in Figure 5.5, none were statistically signiﬁcant based on analysis of variance. Neonatal deaths were spread evenly across most months, but their rate doubled (albeit from .001 to .002) in the late spring–early summer. The neonatal death rate and the stillbirth rate were not correlated, suggesting different etiologies and reafﬁrming the conclusion that the postnatal environment poses different challenges such as respiratory infection or hypothermia. Note that there was a seasonal trend in birthweight: the lowest weights occurred in the spring, with a gradual rise through the summer and fall. Neonatal deaths in the hospital were highest during the months when birthweights were lowest (at least as far as we can tell from the six-month prospective study). Further, there were more births in the fall, indicating more conceptions during the relatively restful winter months. Although neonatal deaths in the obstetrics ward were not especially high in the winter, neonates brought to the pediatrics ward had the highest death rates in the winter, suggesting that a conﬂuence of high birth rates in the fall and high death

An Ecology of Infancy in Ladakh

121 0.1

0.005 0.005

0.095

0.004

0.09

0.003 0.003

0.085

0.002

% of births

Neonatal and stillbirth rate

0.004

0.08

0.002 0.001

0.075

0.001

Stillbirth Rate

Neonatal Death Rate

December

November

October

September

August

July

June

May

April

March

Feb

0.07 Jan

0.000

Births

Figure 5.5. Monthly variation in hospital stillbirths, neonatal deaths, and births.

rates in the winter could help explain the overall high rate of early infant mortality. Patterns of Morbidity There is a high rate of neonatal mortality in Ladakh, and it is related to the low average birthweight. But birthweight itself is not a cause of mortality; instead, it acts as an indicator of an infant’s vulnerability to serious disease.4 Respiratory disease is clearly an all-too-routine cause of death, as is evident in the diagnoses of deaths among infants in the pediatrics ward and of those whose deaths were reported in the prospective study. Hospital records of diagnoses for infants and children admitted to the pediatrics ward between 1991 and 1995 (n = 150 cases) are useful for more detailed evaluation of morbidity. Diagnoses fall into the following categories: respiratory, gastrointestinal, protein-energy malnutrition (PEM), and “other,” which included measles, hepatitis, congenital diseases, accidents, and so on. Between 1991 and 1995 there was no evident change in the numbers of admissions, and the diagnostic categories also remained relatively constant, with no clear trend of increasing or decreasing relative prevalence of respiratory, gastrointestinal, or malnutrition disease. However, there was variation in morbidity by age, gender, and season,

122

An Ecology of High-Altitude Infancy 0.8 females males total

0.7

% of admissions

0.6 0.5 0.4 0.3 0.2 0.1 0 0-1 mo

1-11 mo

12-23 mo

24-35 mo

36-60 mo

Age

Figure 5.6. Pediatric hospital admissions by age and sex.

which provide clues to the unusual pattern of infant mortality in Ladakh and elaborate issues of sickness and death among Ladakhi children. The overall age and sex distribution of hospital admissions is presented in Figure 5.6. Infants made up about one-third of all pediatric admissions. Unlike the deaths, neonates did not predominate among infant admissions, which are spread fairly evenly over the child’s ﬁrst year, with slight peaks at three and nine months. As Table 5.3 indicates, the most common admission diagnosis for neonates was “other,” the majority of which were septicemia cases. Otherwise, 26% were diagnosed with respiratory disease, and less than 5% with gastrointestinal disease. Almost 25% of all neonatal admissions ended in death in the hospital, so these were quite serious conditions for neonates. In later infant and child diagnoses, the “other” category is eliminated because it included a trivial number of diagnoses, aside from a measles epidemic in 1991. There was remarkable stability in the diagnoses over the course of childhood: respiratory diseases consistently made up 60 to 70% of diagnoses and gastrointestinal diseases accounted for 22 to 26%. PEM became a more common diagnosis with age. As for gender, males were admitted more often than were females. The gender disparity was greatest among the youngest infants and decreased with age, as Figure 5.6 illustrates. There were also sex differences in diagnosis that changed over the age groups. As Table 5.3 shows, both sexes were admitted most often with respiratory disease. However, females were more likely to be admitted with a respiratory disease than were males after

An Ecology of Infancy in Ladakh

123

Table 5.3. Age and sex patterns in hospital admission diagnoses Age

N

Respiratory

G-I

PEM

Other

<1 month 1–11 months Females Males 12–23 months Females Males 24–35 months Females Males 36–60 months Females Males

43 652 266 380 537 243 302 281 127 141 437 208 173

26% 70% 73% 75% 63% 70% 62% 64% 76% 61% 70% 74% 71%

5% 24% 20% 20% 27% 20% 27% 25% 17% 26% 22% 15% 22%

0% 6% 7% 6% 10% 9% 11% 11% 7% 13% 8% 11% 8%

70% –

–

–

–

Note: N = Number of admissions; G-I = Gastrointestinal Infection; PEM = Protein-Energy Malnutrition.

infancy, but less often with gastrointestinal disease. PEM accounted for similar percentages of admissions among males and females, with small differences by age group. Thus there appear to be some differences in the disease ecology of males and females and these vary slightly by age. More pediatrics admissions occurred during the winter months between November and March. As Figure 5.7 indicates, respiratory diagnoses reached their peak in January and declined steadily over the course of the year, although they remained the most common diagnosis in all months but September. This decline is mirrored by a gradual increase in gastrointestinal diseases. This general seasonal pattern was fairly consistent across all postneonatal age groups. Seasonal differences in diagnosis or admissions were not notably different among males and females of all postneonatal ages, although there was a slightly greater increase in gastrointestinal disease in the summer for males than for females. Collectively, the analysis of these different sources of data reveal a somewhat peculiar pattern of early mortality, in that risk is strongly focused in the neonatal period. The transition from prenatal to postnatal life appears to be particularly crucial for Ladakhi newborns, much more so than the vagaries of the external environment to which the infant is increasingly exposed as it matures. A more dispersed pattern of infant mortality is common in contexts where infant mortality is similarly high, as infants have increasing contacts with environments that are often contaminated or dangerous and as they become more mobile (Wiley and Pike

124

An Ecology of High-Altitude Infancy 0.92

RI

GI

PEM

All Admissions

0.82 % of diagnoses

0.72 0.62 0.52 0.42 0.32 0.22 0.12 0.02 1

2

3

4

5

6 7 Month

8

9

10

11

12

Figure 5.7. Seasonal variation in admissions and diagnosis. (RI = Respiratory Infection; GI = Gastrointestinal Infection; PEM = ProteinEnergy Malnutrition)

1998). Usually when neonatal mortality makes up such a high percentage of infant mortality, as it does in countries such as the United States, it reﬂects the risks associated with prematurity (Ashworth 1998). This is not the case in Ladakh, where the relatively high rate of neonatal mortality is likely related to the effects of hypoxia both during pregnancy (when it constrains fetal growth) and in the neonatal period when it increases vulnerability to respiratory disease and indirectly (via reduced birthweight) increases the risk of other infectious diseases such as septicemia. It is an unavoidable aspect of high-altitude life, although some newborns are less affected by it than others (i.e., they are larger at birth), indicating that it is not an immutable risk factor but can be shaped by maternal diet and behavior. The pattern of high early mortality risk suggests that the sources of very early death are likely to be closely associated with maternal health and behavior during pregnancy, and hence reﬂect an elaboration of stresses that were already apparent in utero. After the neonatal period, the impact of hypoxia on infant mortality is lessened, and morbidity falls into a pattern that remains more or less stable over the ﬁrst ﬁve years of the child’s life, with seasonal variations in the frequency and severity of gastrointestinal and respiratory diseases. The Ecology of Infant Health and Death All newborns in Ladakh encounter an environment that is hypoxic, arid, and seasonally very cold, yet many must make the transition to postnatal

An Ecology of Infancy in Ladakh

125

life with relatively few biological resources to draw on. There is variation in the ways that newborns experience this transition that derives from interactions of their prenatal legacy with their local household environment, situated within the larger natural and socioecology of Ladakh in the 1990s. The analysis begins with the household, as the immediate “developmental niche” of a newborn, and the locus for the production of health throughout an individual’s life. Dynamics within the household are inﬂuenced by household resources, cultural norms, and social institutions as well as the broader mountain ecology. Salient factors inﬂuencing infant health via the household that are amenable to investigation here are those that inﬂuenced the infant’s birthweight, and, in the postnatal environment, hypoxia, respiratory disease, domestic pollution, diet, child care, and gender dynamics. What we need to assess is how these factors contribute to the very high rate of neonatal morbidity and mortality, particularly among small infants, as well as the predominance of respiratory disease, particularly during the winter months, the lack of strong gender differences in morbidity and mortality, and the relative health enjoyed by older infants and children. Neonatal Ecology When a mother and newborn are discharged from the hospital after birth, the baby receives BCG and polio vaccines and then is bundled up in sweaters, hats, and blankets for its journey home. Fear that the infant will become cold is pervasive among parents, even during the warm summer months, and for this reason, infants are rarely bathed or removed from their swaddling. Buddhist infants receive a smudge of soot on the forehead shortly after birth to ward off evil inﬂuences, and their sex may not be announced for several days or weeks in order to discourage the evil eye, especially in the case of a male child. Additionally, they (especially males) often receive a cord from a lama to be worn around the neck for protection. For a month after birth the mother and infant are supposed to stay inside the house and observe a period of postpartum seclusion; hence, the domestic sphere is a critical aspect of neonatal ecology. After this time, a “rdun” is held for the ﬁrst child. Rdun means seven in Ladakhi, but it also refers to a large celebration held one to three months after the birth on an auspicious date set by an onpo. Relatives and guests are invited to a large party of food, chang (barley beer), music, and dancing to celebrate the child and the mother. At some point the infant is given a name, but there appears to be no hurry to do so. Parents may name the child, or, preferably, they wait and have a respected lama or Rinpoche bestow an auspicious name on the baby.

126

An Ecology of High-Altitude Infancy

Respiratory Disease and Seasonality. The immediate postnatal period is characterized by a profound switch from fetal liquid breathing to neonatal air breathing. In Lhasa Tibet, Niermeyer et al. (1995) found that the establishment of mature cardiopulmonary patterns was delayed in many neonates, who exhibited signs of clinical hypoxemia (a critical lack of oxygen delivery to cells, manifesting as low levels of oxygen saturation) over the ﬁrst four months of life. This effect was particularly evident among non-Tibetan (Han) newborns, who were also small at birth (average birthweight was 2773 g). Oxygen saturation was not measured in this study in Ladakh, but it seems highly likely that many of the neonatal deaths in the hospital resulted, at least in part, from hypoxemia. Although some newborns receive oxygen right after birth, none are monitored for oxygenation status or provided with continuous oxygen. Recall that these very early deaths had no particular seasonal pattern, and thus are likely to result from a stress that is ubiquitous throughout the year. After a newborn goes home, its exposure to hypoxia is mediated by a household environment that may exacerbate this stress. Ladakhi houses are most often heated with dung, coal, kerosene, or wood. They are purposely poorly ventilated to conserve heat; many do not have chimneys. Especially during the cold winter months, families cluster in the kitchen, close to the central stove. Household air contains high densities of particulate matter and high levels of carbon monoxide (CO) when stoves are lit. Ladakhi physicians have found that adult respiratory function is compromised in the winter compared to the summer and is correlated with high levels of domestic pollution (Norboo et al. 1991). Although this effect has never been speciﬁcally measured in children, women aged twenty to twenty-nine had signiﬁcantly reduced lung capacity and were more likely to experience chronic bronchitis than women of other ages or men. This is probably attributable to their spending more time indoors in the kitchen than men. In Bolivia, Albalak (1997) found high levels of carbon monoxide in kitchens and in the exhaled air of women cooking under these conditions, which correlated with the amount of time a woman spent and the level of CO in the kitchen. Indoor cooking was associated with higher rates of acute respiratory infection, especially for young girls. Domestic pollution of carbon monoxide is particularly insidious in hypoxic contexts in that it further reduces the availability of oxygen, which is most critical for newborns. Increases in neonatal mortality due to hypothermia and bronchiopulmonary infections in the winter have been documented in other temperate regions (Hare et al. 1981; Sakamoto-Momiyama 1977), and vulnerability of young infants during months with low temperatures is magniﬁed in socioeconomic contexts where climatic factors cannot be effectively

An Ecology of Infancy in Ladakh

127

modulated and where domestic pollution is common. In Ladakh, domestic pollution is exacerbated during the cold months, but at least household heating should serve to reduce cold stress. While outdoor temperatures may drop as low as –40◦ C, the temperature in households in the Ladakhi study rarely exceeded 12◦ C in the winter, which is still relatively cool (Norboo et al. 1991). Newborns rarely leave the house in the ﬁrst month, particularly in the winter when there is no demand for agricultural labor. They are kept particularly close to heat sources, both during the day and at night when they sleep with their mothers close to stoves. So, although they are kept warm, doing so may contribute to their respiratory problems. Small newborns are those most at risk of mortality in the neonatal period. They have relatively little fat for insulation, and have more difﬁculty maintaining a constant body temperature. In Peru, Frappell et al. (1998) found that small newborns at high altitude were less able to maintain body temperature, and they attributed this to reduced thermogenic capacity related to hypoxic stress. Small newborns are also more vulnerable to lower respiratory tract infections (Victora et al. 1988) and diarrheal disease, although very few neonates in Ladakh suffered from the latter. Furthermore, Niermeyer et al. (1995) found that smaller newborns in Tibet also had lower levels of oxygen saturation. So, in the face of hypoxia exacerbated by domestic air pollutants and seasonal cold, small infants may suffer from extreme oxygen deprivation, which is likely to be worse in newborns who live in particularly small or crowded houses in which air-borne particulate matter and respiratory infections can circulate more densely. Smallness is also likely to be related to the very common diagnosis of septicemia (severe blood-borne infection) among neonatal deaths in the hospital. Without bacterial cultures it is impossible to know which species are likely culprits, but the most common form is Staphyloccus aureus (group B strep) and other bacteria of the birth tract that may infect a newborn during delivery. A Nigerian study (Airede 1992) found that septicemia was most common among small newborns and was associated with a very high rate of mortality. At this point the data say little about this source of disease among Ladakhi newborns but suggest that the maternal disease environment and the treatment of maternal infection during the prenatal period and birth are key to understanding early mortality patterns. Interestingly, neonatal tetanus is not a common problem, in part due to the widespread vaccination of pregnant women with tetanus toxoid.5 While I was working on this book, my infant son was hospitalized three times with bronchiolitis, characterized by respiratory distress and

128

An Ecology of High-Altitude Infancy

hypoxemia. Twice the condition was due to infection with respiratory syncitial virus (RSV). RSV is a common (though little known outside of the medical community!) respiratory virus that has a distinctive seasonal pattern to it (October–April), and while is it relatively benign in older children and adults, it can be life threatening in young infants, especially those who are premature. RSV causes inﬂammation of the bronchioles, which are very small in young infants, so they struggle to obtain sufﬁcient oxygen. It now strikes me that many infants admitted to the Leh hospital with the diagnosis of respiratory distress/bronchial pneumonia are likely to have had RSV, although there was no testing for it. Studies from other north Indian cities have found RSV to be a common etiology for respiratory disease among infants, and it has a seasonal distribution in the winter months (Chattopadhya et al. 1992; Patwari et al. 1996). RSV, as with other lower respiratory tract infections, is much more dangerous in an ecological context that is already hypoxic, since it can dramatically compromise oxygen uptake in young or small infants. Diet. Although maternal diet was an important inﬂuence on birthweight, diet is of more direct and crucial importance to an infant’s health after birth. Cross-culturally, infant diets vary tremendously, with some societies making widespread use of specialized infant formulas, others using the milk of other mammals (especially cows, goats, or sheep), others practicing unrestricted breastfeeding for many months or years, and many combining all of the above. In the ﬁrst months of life, infants grow at a very rapid rate and require very concentrated sources of nutrients in a more or less liquid form. By about six months most infants require supplementary food, and then gradually increase their consumption of adult foods. Much research has pointed to the superiority of breastmilk as opposed to formula or cow’s milk as infant food (cf. Stuart-Macadam and Dettwyler 1995). Breastmilk contains appropriate nutrients, growth factors, enzymes, hormones, and immunological factors to facilitate growth and development and protect infants from common infections (Cunningham 1995). Formula-fed infants, especially in places where the formula is prepared with contaminated water, are at increased risk of diarrheal and other infectious diseases. However, formula feeding may be more compatible with maternal wage labor and has come to take on speciﬁc positive associations with modernity, among other things, in different cultural contexts (Scheper-Hughes 1992; Van Esterik 1989). One qualitative difference between infants who are breastfed from birth and those who are not is that the latter experience an abrupt shift from maternal resources to those of the external environment. While this transition

An Ecology of Infancy in Ladakh

129

is true of birth in general, breastfed infants continue to be buffered from the external environment by deriving nutrients and health-promoting factors from their mothers. This means that they continue to be affected to some degree by maternal biology and behavior, and experience the local dietary environment in a more gradual way as they age. Virtually all Ladakhi newborns are breastfed, and it is common for this practice to continue for two to three years (Cvejic et al. 1997; NorbergHodge and Russell 1994). Women whose infants were followed up in the prospective study reported nursing on demand “whenever the baby cries,” although the frequency of nursing varied from three times/day to “very frequently.” Newborns sleep with their mothers at night, and women reported nursing one to six times per night. Initially babies are given only breast milk, but gradually small quantities of tsampa and butter are added to their diet. Supplementation occurs relatively early, with some women reporting that they fed their neonates this butter-tsampa pap. Some women supplemented with balang milk or formula; indeed, in many houses, formula tins are ostentatiously displayed along with the household’s ornaments in the public kitchen. However, formula use is still primarily a form of supplementation for newborns, and Dr. Ladhol strongly advocates breastfeeding. Women’s work demands shape breastfeeding patterns. Although the ideal is to observe a postpartum seclusion period, women may return to work very quickly if they give birth during periods of heavy agricultural demands. If a suitable caretaker such as an aunt or grandmother is available, a young infant may be left at home; if not, mothers wrap the infant in a shawl and tie it tightly to their backs, or bundle up the baby in a basket carried on their back to the ﬁelds (Photograph 6). When left at home, newborns are likely to be given supplemental feedings earlier. If the mothers keep their babies with them while they are working, infants are generally attached to their mothers’ backs, or left sleeping in baskets, so they nurse only when a mother rests from work. If a child is born during the winter, when agricultural demands are minimal, it is more likely to be kept at home and tended by the mother; as a consequence, the neonate should be breastfed more frequently and supplementation should be delayed. Child Care and Mother-Infant Attachment. Infant feeding practices are just one part of a larger package of child-care behaviors. Child care can be deﬁned as practices that are, on a basic level, designed to protect the health and safety of children, and they also shape the landscape of the child’s early physical, psychological, and social development. In most subsistence societies, child care occurs within an extended family

130

An Ecology of High-Altitude Infancy

Photograph 6. Mothers carry their bundled-up infants and toddlers in baskets on their backs.

household, and caretakers are usually (though not always) female relatives or neighbors (Weisner 1987). In all contexts, child-care strategies derive from the contingencies of daily life and work, and child care can only mitigate or exacerbate risks that ultimately derive from sociocultural forces, including, for example, the values placed on reproduction and children; household composition; pollution; violence; and access to health care. Infants and young children have few skills to buffer themselves from risks and must rely on their caretakers to do so. The complexities of child-caretaker relationships have long interested psychologists, who focus on the ways these affect later emotional and social well-being. Bowlby’s famous work suggested that the development of an attachment to a primary caregiver (usually the mother) during the

An Ecology of Infancy in Ladakh

131

ﬁrst months and years of life was a key part of a child’s psychobiological development (Bowlby 1982). A caretaker who responds consistently and reassuringly to infant needs creates a secure attachment; those who respond inconsistently, capriciously, or with hostility may contribute to a sense of insecurity in an infant. Much debate currently surrounds whether there is only one type of healthy attachment (i.e., secure) or whether other forms of attachment are either culturally prescribed or are adaptations to local social or natural conditions and hence not necessarily “unhealthy” (Ainsworth 1965; Belsky 1997; Chisholm 1993; Chisholm 1996; Hinde 1982; LeVine 1998; Main 1990; Sroufe and Rutter 1984; Wiley and Carlin 1999). There is also continuing controversy surrounding the presence of a “critical period” after birth during which foundational “bonding” occurs (Lamb et al. 1985). In Ladakh, the arrival of a newborn is a joyful occasion but not an overly sentimental one, as it is an integral and, for most, inevitable part of the household cycle. Ladakhi mothers, though attentive and nurturing, also adopt a practical “wait and see” attitude toward their babies. It would be easy to posit a kind of emotional reserve or lack of interest on the part of women toward their newborns, but this must be understood in context. It is a function of the high probability that the child will die very early, an event perceived as unavoidable if it occurs; the routinized nature of birth in the household and agricultural cycle; and a Buddhist detachment from the material world (including other individuals) and an accompanying restrained emotional expression. The death of an infant is met with sadness but also with a sense of resignation and acceptance. There is no ceremonial funeral for infants, and they are buried, not cremated like adults. To illustrate, one day when I was in the hospital, a blood-curdling cry rang out through the wards. A three-week-old infant (one of the infants in the neonatal assessment study), the ﬁrst son of a north Indian woman who was in Ladakh as the wife of a man in the Indian military, had just died of a severe respiratory infection and the mother was screaming in grief. The nurses and young women around me shook their heads somewhat disapprovingly, shrugged with a sense of the inevitability of infant death, and stated simply, “A child has died,” and went on with their conversation. While this could be seen as a common response among those who work in a hospital context where death is not uncommon, the emotional display of the mother’s grief was viewed as excessive (as are expressions of pain during childbirth). On a related note, when women are asked how many children they have had, it is not unusual to get very vague responses such as “three or four” or “eight or ten.” This reﬂects both ambiguity about the number of living

132

An Ecology of High-Altitude Infancy

and dead children and a sense that the real number doesn’t matter when it is more than “three or four.” However, it also points to a downplaying of the importance of children as unique individuals and an emphasis on the family unit over its individual members.6 When I began working in the labor room, I was taken aback by the earliest moment of the mother-infant relationship, which was a far cry from my experience in the United States. As described earlier, very quickly after delivery the newborn is wrapped up and handed not to the mother but to waiting relatives. The mother has scarcely any time to see, let alone hold or spend a few moments with, her new baby. She may not even know whether it is a boy or a girl until the birth is completed by the delivery of the placenta and the conclusion of any surgical repairs and she is reintegrated with her newborn and family in the obstetric ward. After the birth, she is the center of attention, not the newborn. The mother is plied with foods and visitors, urged to rest, eat, and regain her strength. Her health is paramount – to care for the baby and to get back to the routine household and agricultural tasks upon which the success of the household depends – while household members simply hope for the best with regard to the newborn. What this implies to me is the primary signiﬁcance of the birth as a family or lineage event, the predominant concern with the function of the household, and the secondary nature of the mother-infant relationship or attachment. This connection is not imbued with the meaning that it has in cultural contexts such as the United States, where it is assumed to be the primary and most signiﬁcant relationship, beginning immediately after birth. The ﬁrst moments after birth, once thought to be critical to mother-infant attachment (Lamb et al. 1985), are clearly not considered so in Ladakh and may be used instead to emphasize the newborn’s membership in the household and lineage. Mothers are primarily responsible for feeding and protecting their newborns, but even these tasks may be taken on by other caretakers. In short, a newborn is a member of a lineage and household, and members share responsibility for its care and upbringing; the maternal-infant relationship is not privileged. Behavior toward a newborn may not appear to be overly solicitous, particularly given its vulnerability, but it should not be viewed in any way as neglectful. Outright or purposeful neglect is rare and considered aberrant, as the following case exempliﬁes. A young woman had come to Dr. Ladhol to request an abortion. The doctor refused, since the woman said she was married and there was no medical indication for the procedure. The woman went on to deliver a baby, but later the newborn was discovered where it had been left in the street and half eaten by a dog. The mother had been unmarried, and while illegitimacy is not terribly

An Ecology of Infancy in Ladakh

133

rare, to have a child out of wedlock is stigmatized, especially if the father is married or denies paternity. Illegitimate births have a much higher rate of death as a consequence. If the father does admit paternity and supports the child and mother, the relationship is considered an informal marriage, although the child retains no claims to the father’s property. Dr. Ladhol noted at least three similar cases. In another case, again highly aberrant, a young woman became pregnant as a result of rape. She was extremely unwell during the pregnancy, with both hypertension and severe anemia, but the baby died of “unknown causes” during the ﬁrst month. Thus neglect so severe that it results in infant death appears to be largely conﬁned to cases where childbearing has occurred under circumstances considered socially deviant. The “wait-and-see” attitude that families adopt toward newborns does not preclude taking them to health practitioners, although many families apparently did not perceive the need to do so. At follow-up, most women reported that their newborns had not been to any healers. One had been to the hospital pediatrician for jaundice, which is not uncommon among Ladakhi and other high-altitude newborns (cf. Leibson et al. 1989). Two neonates had been to a lama for coughing or crying. As for illnesses serious enough to warrant hospital admission, there was no evidence of special privileging of newborns for hospital care; neonates do not predominate among infant admissions (see Figure 5.7). On the other hand, of infants taken to the hospital, neonates have by far the highest mortality rates there, indicating the seriousness of neonatal disease and the rapidity with which their health may deteriorate, but also that they may not be taken until they are perceived to be very ill. Gender. Child-care practices and attachment can be understood in part as a function of individual or lineage reproductive strategies to obtain a particular family size and age/sex composition through the manipulation of child mortality (Miller 1987; Scrimshaw 1978, 1983, 1984). Behavior ranging from infanticide to “benign” neglect to active health promotion may be part of conscious family planning choices. Scrimshaw has described circumstances under which infant/child mortality is manipulated, either overtly or unconsciously. “Depending on the culture, these children might be of the wrong sex in a family that already has children of that sex, closely spaced, of high birth order, or from a very large family, or they might be ‘difﬁcult’ or unattractive” (1978:397). Such behavior is well documented in South Asia, where females are disproportionately represented among ill, untreated, and dead children; male children are more often brought in for treatment; and amniocentesis often results in the selective abortion of female fetuses (Chen 1981;

134

An Ecology of High-Altitude Infancy

Das Gupta 1987; Miller 1981, 1984, 1987). Among Tibetan communities in an isolated and marginal region of Nepal (a useful population for comparison with Ladakh), Levine (1987) concluded that there was a preference for sons. Because inheritance was patrilineal, sons remained in the parental home after marriage and were therefore a source of old age support and, more than daughters, provided economic contributions to the household through their labor and remunerative occupations. In contrast, daughters were generally considered a net drain because they eventually left the household and their productive assets moved elsewhere, and because wedding costs could be quite high (although dowries were minimal). Mothers might want daughters only for sentimental purposes. Neglect of girls (and other categories of less-desired children) generally took the form of maternal absence and early return to agricultural work (interrupting breastfeeding) as well as denial of high-quality foods, clothing, and health care. In contrast to Levine’s conclusions, Goldstein (1976) and Ross (1984) have argued against deliberate sex-speciﬁc neglect among other Tibetan populations in Nepal and have offered alternative explanations for high mortality. Although the reproductive history data do not show a strong gender bias among neonates, there is a gender bias among neonatal hospital admissions (Figure 5.6) that diminishes in the older age groups. Over twice as many male neonates were brought to the hospital, and the percentage of male neonates who died was 21% (6/29) whereas the percentage of females who died was 29% (4/14). Thus, females are brought in less often and perhaps at a more advanced stage of disease so that they are at somewhat greater risk of death. This bias is not evident in the neonatal deaths in the obstetrics ward (these infants were born in the hospital and died before discharge), where 7% more males died than females. When there is no decision to be made about bringing a young infant to the hospital, the standard pattern of slightly higher neonatal mortality among males prevails. Ecology of Infant Health As the hazard analyses demonstrated, after the ﬁrst month of life, neonatal anthropometric parameters cease to be as crucial to infant survival. Furthermore, reproductive histories indicated that mortality in the ﬁrst month was quite high but then diminished to much lower levels. Postneonatal mortality rates ranged from 38 to 61/1000, which is still relatively high but less than half the neonatal rates. This suggests that whatever is generating neonatal mortality (e.g., low birthweight) is not impacting older infants so strongly. At the same time, infants aged one to eleven

An Ecology of Infancy in Ladakh

135

months accounted for over one-third of all pediatric admissions and over one-quarter of pediatric hospital deaths, so these babies are clearly vulnerable to other threats to their health and survival. Child mortality rates (mortality between ages 1–4) were negligible, based on both hospital deaths and reproductive histories. Infectious Disease. As described in Table 5.3, there is a remarkably consistent pattern of infectious disease across later infancy and into childhood. Gastrointestinal disease was extremely rare among neonates, but in older infants gastrointestinal diseases caused about 25% of their hospital admissions. Respiratory disease remained the most common diagnosis at any age and accounted for 60 to 70% of all morbidity as well as mortality. The only deviation from this pattern was seasonal. In the summer and fall, gastrointestinal diseases increased in frequency for all age groups after the ﬁrst month of life (Figure 5.7). Respiratory infection and the respiratory distress that is often associated with it were common twin diagnoses and derive from interrelated sources. Infectious diseases of the respiratory tract are generally directly transmitted from person to person, so as infants become more mobile and interactive they increase their opportunities for acquiring pathogens. This developmental pattern is exacerbated in the Ladakh case by hypoxia and dry, cold air, which irritates delicate lung bronchia. This ecological effect becomes more pronounced in winter, with its extreme cold. During the winter infants and children are most often indoors, where respiratory pathogens circulate more densely. At the same time, they are also exposed to domestic pollution, which as we saw before, was associated with reduction in lung function and chronic cough. Infants who are not yet mobile may be exposed to more stove fumes; as they become able to move about on their own, they are less likely to be in one place inhaling lots of smoke, but they are more likely to be in contact with other sources of infection. So maturation reduces some household risks but increases others. Outdoor air pollution also exacerbates respiratory disease in the summer, as the area around Leh is crowded with diesel buses and trucks that belch plumes of noxious fumes as they ply the roads during the few months that ground transport into Ladakh is possible. This trafﬁc became particularly congested in the 1990s with the opening of a second road into Ladakh for commercial vehicles. Diesel fumes are likely to be more of a problem for the infants examined in this study, as they were generally from around Leh and villages along the main road. Infants are somewhat protected as they ride bundled up on their mothers’ backs, with their heads well-covered; their mouths and noses do not have direct contact with diesel smoke, and this may reduce their inhalation of

136

An Ecology of High-Altitude Infancy

carbon monoxide and particulate matter. However, others have found that while the practice of swaddling infants among the Quechua of the Andes increases the humidity and temperature of their immediate environs, it also increases the child’s re-uptake of carbon dioxide at the expense of oxygen (Tronick et al. 1994). A high level of respiratory disease sets the background for the emergence of gastrointestinal disease in the summer. In general, infectious pathogens proliferate in warmer months as many require higher temperatures and humidity. When they ﬂock into the region in the summer, tourists, seasonal workers, and truck drivers bring with them pathogens acquired in other parts of South Asia. These make their way into the local water supply, are consumed on imported food, or are directly transmitted through person-to-person contact. As with air pollution, such problems are likely to be especially acute in the area in and around Leh, where water supplies are contaminated by waste and where visitors, seasonal migrants, and imported foodstuffs are concentrated. Though the proportion of morbidity from gastrointestinal disease increases during the summer months, it is not associated with mortality for infants or children. Out of almost 2000 pediatric admissions from 1991 to 1995, only nine deaths in the hospital were from gastrointestinal disease, while there were thirty-nine deaths from respiratory disease. Thus gastrointestinal disease is seasonally acute as a source of morbidity but, fortunately, not of mortality. Drinking water in Leh derives from snowmelt accumulated in a reservoir and piped into various spouts throughout the town. There are also streams that are used extensively for washing, bathing, irrigation, watering animals, playing, and occasionally cooking. In villages, streams are used as the sole source of water for all purposes. Although Ladakhis are acutely aware of the importance of water to their livelihood and have a long tradition of venerating water spirits (lhu), at this point the conﬂuence of a number of factors has resulted in widespread contamination of water supplies. There are no facilities for the collection or disposal of excessive waste created by seasonal and internal urban migration, tourism, and the inﬂux of unrecyclable commodities. Many lodgings have Western-style ﬂush toilets whose waste makes its way into streams or irrigation ditches. Among the items routinely observed in the streams in and around Leh were batteries exploding from their casings, plastic bags, fruit and vegetable leavings, fecal material from humans and animals, cans, bottles, clothing, shoes, and bags of household garbage. Ladakhi children, especially boys, play in the streams in the summer, while their sisters and adults wash clothes, dishes, and themselves in these same streams. In many houses (especially those that double as guesthouses in the summer),

An Ecology of Infancy in Ladakh

137

water from local streams is pumped to the roof, to be delivered to sinks and showers. Adults and children alike use this water for bathing and other activities. Infants are likely to have direct contact with contaminated water while being bathed (which occurs rarely), but they have more, indirect contacts from food, dishes washed in streams, and older children. While the summer months bring a surge in gastrointestinal infections, they do not also bring the spectrum of infectious diseases known at low altitude in South Asia. Malaria, the most common protozoan disease associated with infant and child mortality in the tropics, is virtually unheard of except among seasonal migrants, and other parasitic diseases are also conspicuously absent, again reﬂecting the inhospitable conditions of the high altitude and cold, dry environment. There is evidence, however, that other parasitic diseases are not uncommon among children. A multivillage survey of 477 children from six months to eight years of age indicated that 9% were infected with Enterobius vermicularis (threadworm), 44% with Entamoeba coli, and 47% with Ascaris lumbricoides (round worm) (Wilson et al. 1990). These parasites were not more common among underweight children, but they were more common among those living in villages in the valley region (elevation 3000–3700 m) than in the highland villages (elevation 3800–4100 m). Epidemic diseases also cycle through Ladakh. In the spring and summer of 1991, a measles epidemic contributed to widespread sickness and at least four deaths among Ladakhi infants and children, which attests to the inadequacy of vaccination coverage for this preventable disease. Diet. Older infants continue to be breastfed, both during the day and as they sleep with their mothers at night. Supplementation occurs at an average age of seven months, but one survey indicated that 27% of all infants had been given additional foods before three months (Cvejic et al. 1997). Bread, rice, fruits, and vegetables are gradually introduced, but barley and butter, milk, and yogurt remain staples for infants and children. Small children rarely drink the butter tea that is a staple of the adult diet, consuming it only when it is used to make a porridge with tsampa. While breastfeeding is protective of infant and child health, and it continues into the toddler period in Ladakh, infants do not nurse frequently, in an unrestricted way, or on a strict schedule. This is conﬁrmed by contraception records, which indicate that, on average, women returned to menses 10.4 months (median = 9 months) after a birth (Wiley 1998), which is relatively quickly compared with other subsistence populations (Wood et al. 1985).7 Infrequent nursing is associated with a more rapid return to menses among lactating mothers (Ellison 1995). As infants age,

138

An Ecology of High-Altitude Infancy

mothers are increasingly likely to leave them at home with another caretaker while they work in the ﬁelds. When mothers do carry their infants with them, the babies are wrapped carefully and strapped to their backs so that infants do not have nursing access. Moreover, the practice of dressing both infants and mothers in numerous layers of clothing further restricts nursing. So, while breastfeeding may continue for two years, it appears to be relatively infrequent on a daily basis. Diseases of uniquely nutritional origin are rare. However, beyond pediatric admissions records, there are no good data on these problems among children. The Leh Nutrition Project does some surveillance of childhood diseases, but as yet there is no systematic study of childhood nutrition in Ladakh. I was told by physicians and health workers that malnutrition was virtually unknown, but 6% of post-neonatal admissions included a diagnosis of PEM, and this increased to 8 to 11% after the ﬁrst year. There were thirteen pediatric deaths that included a diagnosis of PEM. PEM did not correlate with gastrointestinal infection (although it did with respiratory infection), indicating that diarrheal disease was not the underlying cause of undernutrition, which is consistent with the village survey conducted by Wilson et al. (1990). Anemia was diagnosed among less than 1% of all infant admissions, but in 4 to 6% of children aged one to two years. There was some overlap between anemia and PEM in that slightly less than half of all diagnoses of anemia occurred in association with PEM. It is not entirely clear what the source of the anemia is; deﬁciencies in iron, folic acid, and vitamin B-12 all manifest as anemia, and all are likely in Ladakhi diets (Attenborough et al. 1994; Meyer 1981; Stobdan 1990). Unlike infectious diseases, PEM and anemia did not show any seasonal pattern in their frequency. This is somewhat surprising given the strong seasonal variation in diet, but seasonality may not have a great impact on the diets of infants. Young infants are probably protected from PEM and other nutrient shortages by breastfeeding, but as they make the transition to other dairy products and grains, they risk becoming nutrient deﬁcient. While it is safe to say that dietary diseases are probably not the most critical source of morbidity and mortality for Ladakhi infants and children, remember that pediatric admissions represent just the tip of the iceberg of infant and child morbidity, and that subclinical undernutrition is likely to have a much broader distribution. A 1983 survey of child growth showed that village children are at risk of growth retardation (Wilson et al. 1990). Overall, 16% of boys and 26% of girls aged less than one year to eight years were of low weight-for-height (indicating acute undernutrition). These numbers are strikingly similar to the percentages of males and females who weighed <2500 grams at birth and raise the possibility

An Ecology of Infancy in Ladakh

139

that this pattern of growth retardation continues during early childhood. Underweight was more common among infants and toddlers than among older children and was more common among girls of every age group. No infants and relatively few children showed severe growth retardation, which is consistent with the hospital admissions data indicating that PEM was a rare diagnosis, especially among infants. A more recent survey found that 53% of infants and over 80% of children were stunted in height, an index of chronic undernutrition (Cvejic et al. 1997). Malnutrition often interacts with infection to produce a “downward spiral of health” characterized by growth reduction, morbidity, and, in extreme cases, mortality (Millard 1994). Often the infection is gastrointestinal, generating diarrhea and exacerbating nutrient loss and deﬁciencies. This pattern often becomes particularly apparent around the time of weaning, when children move from the protective effects of breastmilk to dependence in their own immune systems (Molbak et al. 1994). In unhygienic environments, this transition is particularly problematic, and it is not uncommon to see higher rates of morbidity and mortality among weanlings. Ladakhi infants and children deviate from this pattern, however. First, PEM diagnoses do not overlap with gastrointestinal diseases; rather, they overlap with respiratory diseases, which are the most common diagnosis of infection for infants and children of all ages. Second, there is no rise in morbidity or mortality around the time of weaning perhaps because weaning is not an abrupt transition, but follows from early supplementation and infrequent nursing of infants and toddlers. On a related note, a study in Bangladesh found that diarrheal disease in infants increased when they began to crawl (Zeitlin et al. 1995). Because Ladakhi infants are not encouraged to crawl, they may be protected against gastrointestinal diseases by being less likely to contact pathogens in ground debris. Child Care. Child-care patterns shape the experience that infants and young children have with infectious disease, undernutrition, and other threats to their health. Care encompasses numerous activities – from simple watchfulness, to feeding, clothing, and washing; protecting from automobiles, stoves, or streams; seeking health care, and so on. Many different household members care for young children, although no one is necessarily specially designated as a child minder for older infants or toddlers. Indeed, only slightly older siblings, relatives, or neighbors are often their caretakers, and these mixed-child groups move freely in and out of the household environs. So, the category of “child-care” is ﬂuid and informal, as is common in extended family households. Indeed, it is not unusual for mothers, fathers, or both to leave their young children

140

An Ecology of High-Altitude Infancy

for extended periods of time (days, weeks, or even months) with trusted neighbors or relatives while they attend to nonhousehold obligations or careers. This laissez-faire form of child care does not seem to have negative ramiﬁcations as reﬂected in infant or child mortality, although it does not preclude an association with morbidity especially in the summer. Kerosene poisoning is apparently on the rise, and automobile accidents are increasing in Leh and villages along the main road. Child-care groups that include infants tend to stick to their household compounds or those of neighbors. In general, it appears that Ladakhi newborns and older infants are protected from a variety of serious ailments, but when their symptoms are perceived as sufﬁciently serious (prolonged coughing, lassitude and lack of appetite, crying, abcesses, etc.), they are taken to a healer, most often by the mother. The long lines at both the clinic of the hospital pediatrician and the general practitioner in private practice in Leh attest to the widespread usage of biomedical facilities for children. For families who live in the Leh area, biomedical facilities are often the ﬁrst source of care for sick children; for those who live beyond the Leh environs, families may wait until serious symptoms develop and until local resources have been exhausted before resorting to a long trip to Leh, where biomedical resources are concentrated. There is a deﬁnite seasonal pattern to hospital clinic usage, with declining usage during the summer. This may reﬂect an actual reduction in serious respiratory disease in the warm months, but summer is also when parents and other household members are busy with agricultural tasks and when hospital staff, including doctors, may be absent while attending to their own household responsibilities. The decision to go to a practitioner or the hospital is inﬂuenced by more than the ill-health of an infant. Long distances to the hospital and long waits in the hospital discourage its use; in addition, crowded buses and crowded hallways in the hospital may increase contact with infectious diseases. One strategy I observed was that families would wait until more than one person was sick before seeking help at the hospital so as to maximize the efﬁciency of their time there. Currently there is no coordinated public health immunization or wellbaby program for the monitoring of infant health. The Leh Nutrition Project has initiated a widespread education campaign for child health. One aspect of the program is the distribution of posters in both Tibetan and Urdu script throughout the Leh area that illustrate aspects of proper diet and hygiene for the child. But this private program offers no institutionalized program nor is it coordinated with governmental efforts. To ﬁll this gap, Dr. Ladhol has been attempting to establish a follow-up system

An Ecology of Infancy in Ladakh

141

for infants. As Ladakhi women have embraced her services for preventive care during pregnancy, this might be an opportunity to translate that trust into the monitoring of their infants as well. The satellite health centers in villages beyond the environs of Leh are often poorly staffed and have little to offer in terms of medicines. There the most common and accessible method of intervention is the lhamo or lhaba. Amchi are also consulted and can provide medicines and amulets to cure and protect infants and children from disease and harmful inﬂuences. Lamas also provide protective amulets. While the effectiveness of such interventions cannot be assessed here, they are the ﬁrst level of healers to whom village mothers are likely to bring their ailing infants and children. Tibetan medicine, as practiced by amchi, has a formal textual tradition that outlines the causes and treatment of infant and childhood ailments. Medical treatises discuss twenty-four illnesses that affect children and their cures. In general, infant illnesses are considered to result from improper care of the mother during pregnancy and breastfeeding, or of the infant. If cures are prescribed for any ailment, they are given to the mother if the child is solely breastfed, to both the mother and infant if it is receiving some solid foods, and to the infant if it is completely weaned. A close link is thus articulated between maternal and infant health in Tibetan medicine, and there is recognition that this link becomes less signiﬁcant as the infant ages. Diagnosis of infantile illnesses is based on examination of the veins, voice, and mother’s milk. Veins in the earlobes reveal conditions of the organs, and illnesses are categorized by the color of the veins. The voice can reveal illnesses due to bile, cold, lung ailments, diarrhea, or possession by spirits. Mother’s milk is diagnosed by color and its ability to suspend in water; its examination reveals which of the humors is out of balance. Speciﬁc illness symptoms recognized by the scriptures include the following: high temperatures, coughing, pulsating veins and arteries, scratching the chest or even the mother, sweating and vomiting, nose-bleeding, loud stomach rumbles, diarrhoea of different colours, strong thirst, rubbing the nose and mouth with the hands, a swollen navel, a desire to eat but not actually eating, etc. Symptoms of dangerous illnesses include a drying and hardening of the inside of the nostrils, shriveled ears, thin limbs, sunken eyes, a dry tongue, a hard stomach, short breath, watery diarrhoea, strong rolling movements of the head, puffed up skin etc. If the liver drops to the level of the navel this is particularly dangerous and there is little chance of the child surviving. Otherwise if the eyes, ears, nose, mouth and complexion are looking healthy, if the child breathes well, if its veins and arteries are not pulsating, if the palms and its fouroles have a ruddy colour and he eats well, the illness is curable, and a wise doctor should be immediately consulted. (Sangay 1984:14)

142

An Ecology of High-Altitude Infancy

Illnesses may be treated with mantras in combination with medicines that are suited to the humor out of balance. So, the amchi tradition provides a comprehensive nosology as well as therapy for infant and child diseases, and in many ways their diagnoses and therapies are similar in spirit, if not in content, to biomedical practice. The use of a complex pharmacoepia, attention to diet and physical condition, the recognized close link between maternal and infant health, and the general concern with physiology are allied with the practice of biomedicine, and it may be this correlation that has facilitated broad usage of biomedical resources for infant and childhood diseases when available. Gender. Usage of health care facilities is an important component of child care, and the decision to make use of a particular institution may be contingent on such factors as a child’s sex or birth order. Pediatric hospital admissions data indicate that 16% more males were admitted from 1991 to 1995 than females. The bias was strongest among neonates, and apparent among older infants, but it disappeared after the ﬁrst year (see Figure 5.6). Overall, there are no meaningful differences in the death rates of males and females once admitted (4.1% for females vs. 3.7% for males), although this does vary somewhat by age. Although reproductive histories indicated no major differences in overall male and female infant mortality rates, there are substantial differences in postneonatal mortality rates, with males having two to four times the rate of death from ages one to eleven months. I suspect that these rates, derived from reproductive histories of women who delivered in the hospital, bias mortality for males in an upward direction because women were more likely to seek biomedical help for pregnancy and delivery if a previous child, and more often if a previous male child (especially a ﬁrstborn), had died. The rates listed above also support this interpretation insofar as ﬁrst- and secondborn children (with or without their sex reported) had higher mortality rates than infants with several living siblings, although there were very few of these among the reproductive histories. This pattern was not evident in the 1995 sample, perhaps indicating the broader usage of the hospital for birth. However, if this is the source of gender bias, one wonders why it is evident in the postneonatal period but not especially so in the neonatal period. Since males already are likely to have higher neonatal mortality (as we see in the death rates in the obstetrics ward), it seems odd that there would be such disparity in the NMR and PNMR sex differences. All the Ladakhi women I spoke with said they wanted a son, but interestingly, they generally wanted only one, and they professed a wish for a son and a daughter. Traditionally, sons took over the family household,

An Ecology of Infancy in Ladakh

143

worked the land collectively, and were necessary for the continuation and economic viability of the patrilineage. However, with increasing observance of the Indian law favoring equigeniture, families may prefer to have only one or two sons at most so that their land is not subdivided. From the women’s perspective, inheritance is matrilineal as daughters inherit wealth from their mother in the form of clothing and jewelry, sometimes a sizable amount. This is one reason women want at least one daughter. Mothers also spend a lot of time with their daughters – grooming, working, and joking. Daughters help more with household and garden tasks, and relieve the adult women of some of their workload. Observation of how children are treated suggests that sons are more indulged than are daughters, getting priority for special foods and often fed ﬁrst, but there is no blatant systematic discrimination against girls. The survey by Wilson et al. (1990) indicated that girls were more likely to be undernourished than boys, but this was not borne out in the hospital data: among serious cases necessitating hospital admission, there was no strong gender bias, except that males were more commonly admitted with PEM between two and three years of age. Boys and girls do exhibit slightly different disease patterns. Although respiratory diseases predominate for both sexes, females routinely experience relatively more respiratory diseases than do males, but fewer gastrointestinal diseases after the age of one year. After this age, males are admitted to the hospital with up to twice as many gastrointestinal disease diagnoses as females (see Table 5.3). As girls age, they are more often within the domestic sphere where respiratory diseases are spread more easily. They spend more time around cooking stoves and in the kitchen, and this may both exacerbate respiratory infection and decrease lung function by exposure to domestic pollutants. Boys, on the other hand, are more likely to play in streams during the summer and wander farther from home, increasing their contact with gastrointestinal pathogens from beyond the household compound. In sum then, there is no evidence for a strong bias resulting in differential infant and child mortality by gender among Ladakhi families. And males and females do not seem to experience signiﬁcantly different disease ecologies. Levine (1987) has argued that in Tibetan communities in Nepal, however, certain children are at risk of neglect: females, secondor third-order sons, children of unstable marriages, and illegitimate children. She contended that this neglect explained the very high rate of infant mortality she observed (192/1000 to 226/1000). It seems unlikely that neglect of children of higher birth orders is a major determinant of infant mortality in Ladakh, as the highest mortality is seen among ﬁrstand secondborn offspring, and mortality decreases with increasing birth

144

An Ecology of High-Altitude Infancy

order, at least among women who delivered those higher parity infants at the hospital (although there were relatively few high parity women delivering there, and one could make the argument that women who already had several children and who were pregnant with an unwanted child would be less likely to deliver in the hospital). Moreover, women consistently told me that while they all wanted one son, females were valued as companions, inheritors of their prized possessions, and most important, as a source of labor. Given the low level of technology used in agricultural work in Ladakh, labor inputs are extremely important, and children play major roles. In the context of increasing education for children and the rising expenses of educating them, women show an eagerness for family planning, which is readily available, easily accessible, and culturally acceptable (Wiley 1999). This reduces the burden of “unwanted” children who might otherwise be subject to neglect. Summary Analysis of data from the assessment and follow-up of neonates, birth records from 1990, and reproductive histories consistently points to a high level of infant mortality, and neonatal mortality in particular, in the population of Ladakh. Reproductive histories showed that neonatal mortality determined to a large extent the magnitude of infant mortality. Neonatal characteristics such as low birthweight play important roles in determining this pattern, and predict mortality risk well in the ﬁrst month of life. These characteristics result from forces acting on the fetus in utero, which produce newborns at high risk of mortality precisely because of their small size. Small size appears to place neonates at increased risk of succumbing to septicemia and to respiratory infections that are more prevalent or virulent in the winter. In Ladakh small neonates face an oxygen-depleted environment, and after birth must meet their respiratory functions independently. Since the respiratory system is vulnerable to challenges after birth and small infants have few energy reserves to meet their increased needs, the newborn in Ladakh is compromised by the combined effects of the prenatal and postnatal environments. After the neonatal period, however, infants and toddlers seem to be less acutely stressed by hypoxia and are somewhat protected by their mountain ecology from a wide array of tropical pathogens and diarrheal diseases. Respiratory diseases remain a somewhat ubiquitous threat to health and survival, especially in the winter months. The incidence of gastrointestinal diseases does rise in the summer, but these are rarely so serious as to cause acute growth disruption or death. As toddlers become more autonomous and move farther aﬁeld, they encounter new threats

An Ecology of Infancy in Ladakh

145

from the water, their companions, and food; and while their morbidity may increase, they are less vulnerable to mortality. There do not seem to be major gender differences in disease ecology, but minor variations emerge as females remain more often within the household and males range farther away; the result is more respiratory disease among females, but more gastrointestinal disease among males.

6

Comparative Perspectives on Reproductive Health in Ladakh

The previous two chapters described a number of striking features of reproductive health in Ladakh. Some result from the ecology of this mountainous region, while others derive from the social and cultural forms extant in Ladakh in the 1990s. Aspects of these are unique to Ladakh, but many are common to the Himalayan culture area. In addition, some derive from similar social and historical processes of the Indian subcontinent. Hence, the signiﬁcance of the Ladakhi material is highlighted by three comparisons: (1) mountain populations from the New World that share a similar mountain ecology, for which there are extensive data; (2) other Himalayan populations that share ecology, social organization, and cultural traditions, albeit with different historical trajectories; and (3) other populations in India, with which Ladakh shares some infrastructural similarities and social trends. Each of these three comparisons provides different ways of understanding the possible roots of the reproductive health situation observed in Ladakh. Comparisons focus on the signiﬁcant aspects of maternal and child health in Ladakh, and include birthweight and Ponderal Index; maternal biology; diet and work patterns during pregnancy; neonatal and infant mortality rates and their relationship to patterns of morbidity; child care and gender differentials in health and survival; and household and family dynamics. Ladakh in Relation to New World High-Altitude Populations Neonatal Characteristics We start with Figure 6.1, which we saw in Chapter 2 but which is redrawn here with particular studies labeled. While the negative relationship between altitude and birthweight is obvious, there is tremendous variation in birthweight at the high altitudes (up to 700 grams). The average birthweight of the Ladakh sample is much lower than predicted 146

Perspectives on Reproductive Health in Ladakh

147

3700

3500 Tibetans, Lhasa Tibetans, Lhasa

Birthweight (g)

3300

Bolivia

3100

Khumbu

2900

Ladakh

2700

Colorado

Han, Lhasa Han, Tibet

2500 0

500

1000

1500

2000

2500 3000 Altitude (m)

3500

4000

4500

5000

Figure 6.1. Birthweight and altitude: A summary of the evidence.

for its altitude, and is clustered with three nonindigenous populations below the regression line. Surprisingly, the highest average birthweight within these altitudes is from a sample from Lhasa, Tibet (Moore 1990). Other Tibetan averages are somewhat lower than this, and the Sherpa sample from Khumbu reported by Smith (1997) is slightly lower still. The South American estimates are intermediate. This variation in the experience of pregnancy and birth at high altitude needs to be explained. More detailed data from published studies of New World high-altitude neonatal anthropometry are summarized in Table 6.1. The comparative data are drawn from samples that are made up of native or mestizo populations at altitudes similar to Ladakh; the Ladakh sample used here includes only native Ladakhi mothers and infants from my study. Studies done from birth records are not included in this table; only those in which a trained anthropometrist did the assessments and only those for which information on maternal characteristics was available were included. The Andean sample sizes are small relative to the Ladakh sample. Ladakhi newborns were 350 to 500 grams lighter than Andean newborns, and many more of them were in the category of low birthweight (27% vs. 9–10%). These are substantial differences and most likely reﬂect variation in risk factors. Haas (1980) compared native Andeans with those of low-altitude ancestry living at high altitude and found that newborns of the latter weighed >100 grams less, which is still higher than

1 includes

47.6 0.9 ± 1.1 24.4 ± 4.1 48.8 ± 6.4 153.3 ± 5.5 22.2 ± 1.6 9.3 ± 2.9 10.5 ± 1.6

34.2 ± 1.3 30.8 ± 1.9 3.8 ± 0.8 37.8 ± 1

3534 145 0.84 2745 ± 418 2818 ± 413 2684 ± 415 27.0 47.9 ± 2.3 2.50

149.7 ± 5.2

25.8 ± 5.0 61.4 ± 7.22 149.8 ± 5.1 16.2 ± 4.9 13.7 ± 1.6

41.1

39.1 ± 0.83

2.72

3251 ± 417 3149 ± 438 10.0

3600 51 0.65

Bolivia (Haas et al., 1980)

36.6

4.4 ± 1.1 39.0 ± 0.7

34.2 ± 1.3

48.9 ± 1.8 2.71

3600 41 0.46 3165 ± 428

Bolivia (Haas, 1980)

only native Ladakhis; 2 weight measured in third trimester

Maternal Characteristics Parity (% primipara) Average parity Age (years) Weight (kg) Height (cm) Arm circumference (cm) Triceps skinfold (mm) Hb in pregnancy (g/dl)

Infant Characteristics Altitude (m) Sample size Sex ratio Birthweight (g) Males Females % <2500 grams Birth length (cm) Ponderal Index (g/cm3 ) Circumferences (cm) Head Chest Subscapular skinfold (mm) EGA (weeks)

Ladakh1

16.9 ± 4.3

33.3 3.3 ± 2.6 25.2 ± 6.3 54.6 ± 6.8 148.7 ± 4.9

34.2 ± 1.2 32.6 ± 1.4 4.4 ± 1.3

3416 73 1.09 3091 ± 458 3130 ± 458 3053 ± 425 8.6 49.6 ± 1.6 2.53

Peru (McClung, 1969)

Table 6.1. Comparison of neonatal and maternal characteristics in Ladakh and the Andes

22.0 3.4 ± 2.4 27.0 ± 7.6

3.5 ± 0.9

350 ± 0.9

49.2 ± 1.6 2.74

3870 27 1.70 3258 ± 369

Peru (Haas et al., 1977)

Perspectives on Reproductive Health in Ladakh

149

the Ladakhi average. Ladakhi newborns were, on average, younger by at least one week than Andean newborns, but this level of variation is within the range of error found even with relatively accurate estimates of gestational age (Caulﬁeld 1991). Head circumference, as a crude index of gestational age, was consistent across populations. The lower mean weight of the Ladakhi newborns is not a function of the disproportionate number of females in the sample, since the sex ratio varied in the other studies from 0.5 to 1.7, and comparison of same-sex birthweights across populations indicated that their differences were similar in magnitude to overall differences. Ladakhi newborns were thinner than most of the Andean samples, although similar to those in Haas et al.’s (1977) Peruvian sample. Ladakhi newborns were also at least 1 centimeter shorter, but with respect to PI, the mean for the Ladakh sample was between 0.03 and 0.24 g/cm3 lower than in the other studies. A substantially greater percentage of the Ladakh sample was of low PI than another Bolivian high-altitude sample (Conlisk 1987), which, in combination with the overall lower average PI in Ladakh, indicates that Ladakhi neonates are lighter for both length and age than Andean neonates. Thus, while Andean fetuses appear to endure mild chronic stress during gestation, Ladakhi fetuses appear to experience increasing levels of stress as gestation advances. Sex differences in neonatal parameters in the Ladakh sample are at odds with the results of McClung (1969) and Haas et al. (1980). In those studies, sexual dimorphism in birthweight was reduced. The sex difference in birthweight among Ladakhi newborns was 134 grams (which was reduced 120 g when maternal weight and EGA were taken into account). The maintenance of this degree of sexual dimorphism (similar to that observed in well-nourished low-altitude populations) in both weight and body composition suggests that males are not more stressed than females in utero in Ladakh. Indeed, stillbirth data suggest that more fetal deaths were females in Ladakh. Both of these patterns warrant further investigation. Differences in maternal characteristics provide some clues to the divergence of the Ladakh sample from other high-altitude samples. Overall, Ladakhi mothers were slightly younger and of lower parity, and a greater percentage were primiparous than Andean mothers, and these factors are known to be associated with lower birthweights. Ladakhi women also weighed 6 kilograms less than those in McClung’s sample, the only one in which postpartum weight was reported. They had much less fat (triceps skinfolds = 9.3 vs. 16.2 mm), but were 3.5 to 4.6 centimeters taller than the Andean women. Given the importance of maternal weight as a predictor of infant birthweight, it is likely that the lower average weight

150

An Ecology of High-Altitude Infancy

of Ladakhi women is an important factor contributing to the lower birthweight in Ladakh. If average Ladakhi maternal characteristics (gestational age, triceps skinfold, and hemoglobin) are used in Haas’s (1980) prediction equation for birthweight among native women in La Paz, the predicted birthweight is 2903 grams.1 However, if the model for nonindigenous women is used, the predicted birthweight is 2748 grams, which was almost exactly the average birthweight in the Ladakh sample! Unfortunately we cannot standardize the Bolivian data using the Ladakhi prediction equation, since Haas measured maternal weight prenatally and I measured it postnatally. However, the fact Haas’s equation, so accurately perdicted the Ladakhi average birthweight suggests that the factors contributing to birthweight in Bolivia are also important in Ladakh, and that differences in maternal weight, gestational age, and high-altitude ancestry produce the lower birthweights there. The small size of Ladakhi women, which contributes to the small size of their newborns, results in part from constraints on their own growth. Haas et al. (1982) found that lower weights among older infants at high altitude in Bolivia were due to a persistent effect of their small size at birth. The small size of Ladakhi infants may persist throughout growth, as two surveys have suggested (Cvejic et al. 1997; Wilson et al. 1990), and result in smaller adult size. Furthermore, it is possible that the low average PI of newborns persists in the adult phenotype, since Ladakhi women are substantially thinner but taller than Andean women. Diet and activity patterns during pregnancy had dramatic effects on birthweight in Ladakh, but no data are available on these factors among Andean women from the sources in Table 6.1. Haas’s Bolivian sample comprised newborns delivered in large city hospitals to middle-class, fairly well-educated women, and Haas et al.’s Peruvian sample derived from a similar population, albeit with slightly less education and a higher percentage of women described as “lower class.” The work or dietary behavior of these mothers was not reported, although McClung noted that none of the women in her sample smoked or chewed coca. I suspect that the Ladakhi sample, though also drawn from an “urban” area, was more likely to include women who worked harder at agricultural tasks than those from the Andes, and this may further reduce birthweight relative to the Andes samples. Differences in population history contribute to variation between these two sites. As we saw in Chapter 3, the best extant information on population habitation in Ladakh dates it back only 4000 to 5000 years, and the area has experienced waves of immigration since then. This history is subject to change as more data become available, but no one has thus far suggested deeper antiquity for permanent settlement there. In contrast, there

Perspectives on Reproductive Health in Ladakh

151

are abundant data indicating that the Andes have been settled for at least 10,000 years (MacNeish and Berger 1970). Greater length of exposure to high-altitude conditions has been found to be associated with higher birthweights, presumably because genes generating higher weights and lower mortality have had longer to spread through the population (Haas 1980; Moore et al. 1998). The fact that Ladakhi birthweight was more accurately predicted by Haas’s equation for nonnative women indicates that lack of high-altitude ancestry contributes to variation between these two populations. However, the nonnative newborns in Haas’s sample had birthweights that were only 110 grams less than those of native women, while differences between Ladakhi and native Andean birthweights were much greater than that. Thus, differences in maternal nutritional status and work patterns (as described above) also play signiﬁcant roles. Neonatal and Infant Mortality Several reports show that infant mortality is higher at high altitude compared to low altitude in the Andes, and this effect has remained relatively stable over the course of these investigations (late 1950s–1990s) and when other factors are controlled for (access to medical care, education, etc.) (Edmonston and Andes 1983; Mazess 1965; PAHO 1994). There are two types of data on neonatal, infant, and child mortality from the highaltitude Andes summarized in Table 6.2: censuses and other large-scale regional surveys, and smaller-scale, anthropological community surveys. The larger surveys span a thirty-year period, but indicate infant mortality rates of 70 to 120/1000, with most results clustering around 100/1000. The community study of early mortality by de Meer, Bergman, and Kusner (1993) among the Aymara and Quechua of southern Peru yielded a higher estimate of 156/1000 from reproductive history data. Only these authors and Mazess broke the IMR down into neonatal and postneonatal rates, and both found that neonatal mortality represented 42 percent to 54 percent of total infant mortality. Comparing de Meer, Bergman, and Kusner’s community survey estimates with those from Ladakh, it is evident that infant mortality was higher in Ladakh (184 vs. 156/1000), and that neonatal mortality made up a much greater percentage of death in the ﬁrst year (81% vs. 54%). Perinatal and neonatal mortality rates in Ladakh were about twice the magnitude of those in Peru, while postneonatal mortality was only about half of the Andean rates (35 vs. 72/1000). The work of Beall (1981) and Conlisk (1987) on birthweight and neonatal/infant mortality demonstrates that variation in the risk associated with a given birthweight exists between Ladakhi and Andean populations. Beall (1981) calculated birthweight-speciﬁc mortality rates for

Peru (Mazess, 1965) Bolivia (PAHO, 1994) La Paz Cochabamba Potosi Peru (McClung, 1969) Peru (Beall, 1976) DeMeer et al. (1993) Southern Peru Ladakh ∼110/1000

71/1000 150/1000

14/1000 90/1000

Perinatal mortality

61/1000

Stillbirths

84/1000 148/1000

52/1000 52/1000

Neonatal mortality

156/1000 184/1000

122/1000 120/1000 70/1000 78/1000 118/1000

Infant mortality

72/1000 35/1000

70/1000 68/1000

Postneonatal mortality

54% 81%

38%

42% 43%

NMR/IMR

Table 6.2. Neonatal, infant, and child mortality in the highland Andes and Ladakh

60/1000 –

45/1000 57/1000 50/1000

Child mortality

Perspectives on Reproductive Health in Ladakh

153

0.4 0.35

Mortality rate/probability

0.3

Conlisk -7 days Ladakh -7 days Beall - 1 year Ladakh - 1 year

0.25 0.2 0.15 0.1 0.05 0 1750 2000 2250 2500 2750 3000 3250 3500 3750 Birthweight (g)

Figure 6.2. Birthweight and mortality in Ladakh and the Andes.

one year, and Conlisk (1987) calculated them for one week. These rates are presented in Figure 6.2 along with the predicted mortality rates from the Ladakh sample.2 At one week, birthweight-speciﬁc mortality is similar in Ladakh and Bolivia, except that it is higher in Bolivia for very small newborns, perhaps because these were live births in this major urban hospital and would have been stillbirths in Ladakh. Major differences emerge between Ladakhi and Peruvian birthweight-speciﬁc mortality up to one year. This gap is most profound at the lower weights, as there is convergence at 3000 grams, but for a given birthweight below 2750 grams, Ladakhi infants have much higher rates of mortality. Because most Ladakhi birthweights were lower than 3500 grams, and over 25% were less than 2500 grams, the birthweight distribution, in combination with the higher birthweight-speciﬁc mortality in Ladakh, contributes to the overall higher rates of infant mortality there. It seems likely that the much lower average birthweight found in Ladakh is inﬂating neonatal and infant mortality (given the preponderance of neonatal deaths making up infant mortality), but Ladakhi infants have an advantage in the postnatal period over their New World counterparts. Since New World populations do not show the very high rates of neonatal

154

An Ecology of High-Altitude Infancy

mortality that are seen in Ladakh, even for a given birthweight, different factors are likely to be contributing to the risk that a certain birthweight entails, such as less acute growth disruption during Andean pregnancies, and neonatal conditions (e.g., indoor pollution, breastfeeding, child-care practices, infections) that enhance the survival of Andean neonates. In contrast, Ladakhi infants are subsequently advantaged and so regional differences in the infant period need to be explored. De Meer and colleagues (1988; De Meer et al. 1993) argued that infant mortality in the Peruvian Andes could be attributed largely to routine causes such as poor sanitation and inadequate nutrition, perhaps exacerbated by the adverse ecological conditions of the altiplano, and, most important, infanticide. In the communities in which they worked the normative pattern for birth and early child care was to have the mother and neonate stay at home in bed, with the baby heavily swaddled to protect it from cold and nursed immediately when it cried. Leaving infants alone, not responding to their cries or adequately protecting them from cold, would rapidly compromise a young infant’s health and resulted in death from exposure through passive neglect. De Meer et al. noted that informants did not describe incidences of fever (indicative of neonatal infections) or make use of primary health care facilities for their neonates who had died. Multiple perinatal deaths were clustered among families with several living children, but they also occurred among some smaller families. De Meer proposed that infanticide was used to reduce family size when resource crises occurred or where landownership had declined but outmigration was not common. Gender selection was not evident, as deaths were slightly biased toward males, and ethnographic investigation did not suggest the existence of strong gender hierarchies. If De Meer and colleagues are correct in their interpretation, infanticide could certainly be one source of variation in the early mortality proﬁles between Ladakh and Peru, as there is no evidence that it played an important role in shaping infant mortality in Ladakh, but it does little to resolve the question of why neonatal mortality should be so much higher in Ladakh than the Andes. Breastfeeding is virtually ubiquitous among Andean mothers and continues well into the second year of life although there is substantial variation in this practice by social class (Vitzthum 1989; 1992). As in Ladakh, nursing generally begins immediately after birth, most often within twenty-four hours. However, it seems likely that Peruvian infants nurse more frequently, as they are more often in the company of the mother than are Ladakhi infants, who may be cared for by other

Perspectives on Reproductive Health in Ladakh

155

individuals from a younger age. Andean women wear their infants in a sling on their back and nurse while engaged in various tasks. Both Vitzthum and I noted the constraint that cold places on ad libitum nursing, especially at night, since both babies and mothers are heavily clothed. There is substantial variation in the age at which supplemental foods are given – in Ladakh, they are given within the ﬁrst few months, whereas in Peru infants are often over one year old before their mothers report that they are eating solids. However, Vitzthum (1992) noted that infants do consume small amounts of transitional foods long before this, including potatoes and bread, as well as tea or broth. Ladakhis appear to have greater access to dairy products, which are frequently used as supplemental foods early on, along with barley ﬂour. The distribution of diseases and their contribution to infant mortality in Ladakh conform to what is found in the Andes, where respiratory diseases were consistently the most frequent cause of mortality for neonates and infants (Beall 1976; Puffer 1973; Spector 1971). In Beall’s study, over half of all infant deaths were due to respiratory disease; the percentage was highest in the ﬁrst month (74%) and lower in the postneonatal period (55%). Likewise, in Ladakh, over two-thirds of the infants who died in the hospital in the 1990s succumbed to respiratory disease. Respiratory disease was also a more common source of morbidity at high altitude (∼35%) than was gastrointestinal disease (∼25%). Somewhat surprisingly, Beall also found that morbidity risk did not vary by birthweight. Seasonality in disease patterns has not been examined in the Andes and is likely to differ from Ladakh because the Andes are a tropical mountain system with more diurnal than seasonal variation in temperature. Respiratory and gastrointestinal diseases may thus have a more uniform monthly distribution than in Ladakh. Summary: Ladakh and the Andes Compared What can be ascertained from this comparison of birth outcome and early mortality between Ladakh and the Andes? Ladakhi newborns are smaller, possibly because of fewer energy reserves among mothers, greater maternal workloads, and a shallower ancestry at high altitude. As for mortality, peri- and neonatal mortality is much higher and postneonatal mortality much lower in Ladakh. The lower average birthweight in Ladakh is, at present, the best explanation of the much higher early mortality rates there. There does appear to be variation in neonaticide, with the practice being more routine in the Andes than in Ladakh. Patterns of infectious disease are similar, although Beall found risk of infection did not vary by

156

An Ecology of High-Altitude Infancy

birthweight as I have hypothesized in Ladakh. Breastfeeding is common in both areas, though it may be more frequent in the Andes, where supplemental foods are delayed. More ﬁnely tuned observations of breastfeeding behaviors, similar to those conducted by Vitzthum in Peru and Bolivia, are needed in Ladakh. Indoor pollution has not been examined in relation to maternal-infant health in the Andes, but similar fuels are used and cold is a chronic stress in both locations (Albalak 1997). There is likely to be more seasonality in morbidity and mortality in Ladakh as a function of its temperate location. In both locations, infant care is shared by siblings and other household members while mothers engage in agricultural work. It is not clear why the postneonatal period is characterized by lower rates of mortality in Ladakh than in the Andes. Some researchers in the Andes have argued that widespread poverty and economic marginalization resulting from a colonial legacy and neocolonialism have contributed to deﬁcits in infant and child growth (Carey 1990, Leatherman 1996, 1998). Growth retardation among children may originate in the lower birthweights of high-altitude children, as Haas (1982) and Beall (1976) found that older infants at high altitude had relatively low weight for age, although socioeconomic differences were not analyzed in these reports. Such marginalization and social stratiﬁcation has not been a historical trend in Ladakh, but surveys of infant and child nutritional status do indicate continued slow growth during childhood (Cvejic et al. 1997; Wilson et al. 1990). Similarly, admissions of infants to the hospital are spread across the ﬁrst year of life and are thus not biased toward neonates. So there is morbidity among older infants in Ladakh, but, unlike in the Andes, it does not result so often in death. It is also possible that a selection bias may be operating in Ladakh: where so many infants succumb in the ﬁrst month, those remaining may be more robust than infants in the Andes, who start off with more resources. On the other hand, it may be that living conditions in most Ladakhi towns and villages are better than in the Andes so that when infants become more independent, they encounter fewer virulent pathogens. This is an empirical issue in need of further comparative study. Ladakh in the Himalayan Context Ladakhi birth outcomes are also at signiﬁcant variance from those described in studies from high-altitude Tibet and Nepal among native Tibetans or Sherpa populations. Ladakhis share many ecological and cultural characteristics with them, but differences in birth outcomes suggest ways in which their histories have been quite distinct.

Perspectives on Reproductive Health in Ladakh

157

Birthweight Table 6.3 reveals the variation in birthweight between Tibetan newborns in Tibet, Sherpas, and Ladakhis, and includes only prospective studies. Overall, Ladakhis are the smallest of the Himalayan newborns, and the largest are those of Tibetan women in Lhasa (Zamudio et al. 1993). Other studies have shown birthweights between 3067 and 3280 grams among indigenous Tibetans (Moore et al. 2001a, 2001b; Niermeyer et al. 1995). In Nepal, a study of Sherpa birthweights indicated an average of 3099 grams (Smith 1997). Tibetans in Ladakh and Sherpas were on the low end of the spectrum of Himalayan birthweights (2990 to 3099 g), and Ladakhis had the smallest newborns of all (2745 g, 300 to 500 g less than other Himalayan populations). Ladakhi birthweights were much more similar to those of the Han, recent migrants to high-altitude areas of Tibet (2745 vs. 2610–2773, cf. Moore et al. 2001a; Niermeyer et al. 1995). Differences among Himalayan populations were substantial, surprisingly greater than the differences between them and Andean populations. Sources of variation in Tibetan and Ladakhi birth outcomes could derive from several sources. Ladakhi women were lighter and thinner, and their hemoglobin levels were 20 to 40% lower than those of other Himalayan mothers, though they were within the range of other populations for age and parity. This suggests that conditions of growth and development, current diet, and adaptation to hypoxia may be substantially different in the samples. None of the women in any of the samples smoked during pregnancy. Zamudio et al. (1993) indicated that the women in their Lhasa study were most often employed in the home, worked in clerical jobs, or were involved in trade. They had been for an average of 5.4 prenatal visits (similar to 4.7 in Ladakh) and were of relatively high socioeconomic status. On the other hand, in the larger sample described in Moore et al. (2001a), 83% of Tibetan women were involved in occupations requiring “heavy labor,” although apparently this did not contribute to birthweight variation and it was not clear if this included work during pregnancy. In the Ladakhi sample, male and female birthweights differed by 134 grams. The studies by Smith (1997) and Moore et al. (2001b) showed male-female differences on the order of 200 to 300 grams while Zamudio et al. (1993) found no signiﬁcant sex differences. The ﬁnding that most Himalayan populations maintain sexual dimorphism among newborns while those in the New World do not is worth further exploration, particularly as it relates to early mortality. Tibetans in Ladakh had larger infants than Ladakhis, but their average birthweight of 2990 grams was lower (100 to 300 g) than the

43 1.4 ± 1.6 27.4 ± 6.3 53.8 ± 6.7 153.2 ± 4.6 24.8 ± 2.2 21.9 ± 5.3 12.3 ± 1.1

Maternal Characteristics Parity (% primipara) 47.6 58.0 Average parity 0.9 ± 1.1 0.4 ± 0.1 Age (years) 24.4 ± 4.1 25.0 ± 0.5 Weight (kg) 48.8 ± 6.4 58.0 ± 1.02 Height (cm) 153.3 ± 5.5 158.0 ± 1.0 Arm circumference (cm) 22.2 ± 1.6 Triceps skinfold (mm) 9.3 ± 2.9 Hb in pregnancy (g/dl) 10.5 ± 1.6

1 includes only native Ladakhis; 2 weight measured in the third trimester; non-pregnant in the case of weight

3

39.6 ± 1.05

39.7 ± 0.3

3.0

21 1.33 3099 ± 438 3183 ± 364 2987 ± 389 14

145 0.84 2745 ± 418 2818 ± 413 2684 ± 415 27.0 47.9 ± 2.3 37.8 ± 1

2.6 27

87% full-term

165 1.63 3140

0.4 ± 0.2 24.0 ± 1.0 50.0 ± 2.0 157.0 ± 1.0 14.4 ± 0.4

12.6 ± 0.2

47.0 ± 1.0 39.4

13 1.25 2645 ± 96 2558 ± 194 2379 ± 71

0.5 ± 0.1 25.0 ± 1.0 49.0 ± 1.0 157.0 ± 1.0

49.0 ± 1.0 40.6

18 0.86 3280 ± 78 3425 ± 105 3100 ± 85

Tibetans 3–4000 m Lhasa – Tibetan3 Lhasa – Han3 (Moore et al., 2001a) (Moore et al., 2001b) (Moore et al., 2001b)

maternal characteristics drawn from larger sample of pregnant women, or

Lhasa (Zamudio Sherpa et al., 1993) (Smith 1997)

34 0.89 3222 ± 90

Infant Characteristics Sample size Sex ratio Birthweight (g) Males Females % <2500 grams Birth length (cm) EGA (weeks)

Ladakh1

Table 6.3. Comparison of neonatal and maternal characteristics among high-altitude Himalayan populations

Perspectives on Reproductive Health in Ladakh

159

differences of Tibetans in Tibet. Birthweight differences between these two related groups can be attributed to socioeconomic differences, as Tibetans in Ladakh are political refugees. Tibetan women in Ladakh were four centimeters shorter than those in Tibet, which suggests chronic undernutrition during childhood among the migrants. This group also had substantially lower hemoglobin readings during pregnancy, suggesting current dietary differences between the two Tibetan populations. It is interesting that the average birthweight in another Sherpa sample (Wiley 1992b) is closer to the mean birthweight of Tibetans in Ladakh (2957 and 2990 g, respectively) than to the Tibetan samples. Thus, there may be important differences among populations at the core and peripheries of the Tibetan culture area, where there is likely to be more genetic and lifestyle heterogeneity. Nevertheless, Ladakhi newborns are substantially smaller than Sherpa newborns (by ∼300 to 400 g). Similarity in lifestyle suggests that further comparison between these groups would be useful to help explicate the source of this difference. Smith’s (1997) comprehensive report of characteristics among Sherpa mothers in the Khumbu region provides such a comparison. Ladakhi and Sherpa women are both involved in agricultural production and experience substantial energy costs due to that involvement during pregnancy. Smith found that Sherpa women did not increase their food intake or reduce their workloads during pregnancy, and that energy expenditure and calorie consumption during pregnancy were signiﬁcant predictors of birthweight among high-altitude Sherpa women. She noted that “although many women would have liked to reduce activities in the third trimester, actual reduction was based on the economic status of the family, the presence of servants, and season. If the third trimester fell during planting or harvesting, women often worked in the ﬁelds” (1997: 760). Both Sherpa and Ladakhi males are engaged in tourism-related activities that take them away from their households during the summer months. The Sherpa women were slightly older, of higher parity, and had much higher prepregnancy weights (5 kg) than the Ladakhi women’s postpartum weights (which are likely to be higher than their prepregnancy weights). Sherpa women were not any taller than Ladakhi women (unlike Tibetan women), but they were much fatter – triceps skinfolds were over twice as large as Ladakhi women’s (21.9 mm vs. 9.3 mm). Hemoglobin measures were higher (12.3 vs. 10.5), but Sherpa women reportedly did not increase their iron consumption during pregnancy. They did not acknowledge any food proscriptions during pregnancy, and believed that pregnant women should increase their consumption of fruits

160

An Ecology of High-Altitude Infancy

and vegetables since these were considered “strong” foods. Like Ladakhi women, in the postpartum period they increased their consumption of meat and dairy products and decreased their activities, contingent on household economy and season of the birth. It seems likely, then, that the differences observed between these two populations stem from variation in women’s nutritional status prior to pregnancy. Sherpa women are much heavier for their height and have over twice the energy stores of Ladakhi women. The nutritional background for Sherpa women appears to be more robust than that of Ladakhi women, although it is not immediately apparent why this should be so. Although the area in which they live is not accessible by road, frequent porter trafﬁc and trade result in a variety of foods being available in the area. They may also be wealthier than Ladakhi families, as a consequence of their long-term involvement in lucrative tourism and climbing expeditions in the Mount Everest area. Other studies in Nepal conﬁrm the ﬁnding that Himalayan women continue to work hard at subsistence activities during pregnancy. PanterBrick’s work with the Tamang, a Tibetan-derived group inhabiting mid-range altitudes (∼1870 m), pointed out that pregnant and lactating women were generally not able to reduce their workloads (1989, 1993). During the seasons of most intensive agricultural work, these women worked the same number of hours at agricultural tasks as nonpregnant or lactating women. Panter-Brick noted, “Mothers perform the most demanding tasks, such as hoeing or transplanting . . . sustaining their agricultural activity from dawn until early evening” (1989:214). The selfsufﬁciency of nuclear households places substantial demands on female labor, which must be divided among multiple subsistence tasks, many of them arduous (agricultural work, travel, gathering wood). Pregnant women were able to work less only during the seasons with fewer intensive agricultural activities. Furthermore, pregnancy did not result in special food provisioning. Anemia was found to be widespread, with average hemoglobin levels of 10.7 among these women, including those who received iron supplementation. Thus, there are striking parallels among pregnant women of the Tamang and Sherpa, and those in Ladakh, resulting in competing demands by the fetus and mother for maternal energy and oxygen due to agricultural participation in this mountainous context. Differences between Ladakhi newborns and those in Lhasa and Nepal may also stem from differences in population history. The current population of Tibet may have ancestral roots there as early as the Upper Paleolithic (Moore et al. 1998), although the earliest conﬁrmed date of settlement is 7000 bp (Chang 1986). Sherpas have their origin in Tibet, having moved into Nepal in the 1500s (Ortner 1989). Consequently,

Perspectives on Reproductive Health in Ladakh

161

Ladakhis may represent a more genetically heterogeneous population, with many groups of relatively recent antiquity at high altitude and without sufﬁcient time in that environment to have evolved uniform efﬁcient adaptive responses that would result in larger newborns. It is interesting that the sample that most closely matches the Ladakhi birthweight data is the Han Chinese, just as Ladakhi birthweights were most similar to recent migrants in the Andes. Moore et al. (2001b) were able to demonstrate that pregnant indigenous Tibetan women had higher uterine artery blood ﬂow than Han women, and this diversion of oxygen ﬂow to the fetus was protective of birthweight among Tibetans. It would thus be useful to make a similar comparison with pregnant Ladakhi women to see if this physiological mechanism is more akin to what was observed among Han mothers. On the other hand, variation in maternal anthropometric status suggests that Ladakhis were poorly nourished relative to their Tibetan and New World high-altitude counterparts. Ladakhi women are lighter and thinner and have smaller triceps skinfolds. While dietary data gathered in Ladakh do not indicate critical deﬁciencies in protein and energy-rich foods, agricultural demands on maternal energy constrain maternal resources, including those necessary for fetal growth, more so than they do among urban populations in Tibet or the Andes. Neonatal and Infant Mortality There are relatively few published data on neonatal and infant mortality from high-altitude Himalayan populations, and they are summarized in Table 6.4. Most estimates of infant mortality are over 150/1000; neonatal mortality rates are 50 to 60/1000; and stillbirth rates are variable – 34 to 90/1000. Geetha et al. (1995) linked high perinatal mortality rates to prematurity, followed by asphyxia and pre- and postpartum infection. Hypothermia, especially among LBW infants was a likely contributing factor as well. Among the Tamang, at a more intermediate altitude, Koppert (1998, cited in Panter-Brick 1989) attributed a high infant mortality rate to routine respiratory and gastrointestinal causes. Gupta et al. (1989) reported a relatively low IMR of 72/1000 in upper Khumbu, Nepal, and noted that this was not signiﬁcantly different from what was found among low-altitude Sherpa populations. However, among Tibetans, Moore (2001a) found that both pre- and postnatal mortality tripled at altitudes of 3000 and 4000 meters (101, 123/1000, respectively), compared with those <3000 meters. These rates were twice as high among Han than among Tibetans at the higher altitudes (210, 207/1000, respectively).

Ladakh (this study) Jumla, Nepal (Geetha et al., 1995) Tamang (Panter-Brick, 1989) Tibetans in Nepal (Levine, 1987) Sherpas (Lang & Lang, 1971) Tibet (Chin, 1992) 96/1000

34/1000

47/1000

150/1000

Perinatal mortality

90/1000

Stillbirths

51/1000

62/1000

144/1000

Neonatal mortality

107/1000

38/1000

Postneonatal mortality

Table 6.4. Rates of infant mortality from the Himalaya

136/1000

158/1000

130/1000

100–210/1000

192–226/1000

–

Child mortality

88/1000

32%

79%

NMR/IMR

175/1000

182/1000

Infant mortality

Perspectives on Reproductive Health in Ladakh

163

Unfortunately, there have been no other Himalayan studies connecting birthweight to infant mortality, and our understanding of the determinants of early mortality remains very sketchy. Chin (1992) conducted a survey of growth among a small sample (n = 48) of children in three Tibetan villages and found that among ages zero to four years, males and females had attained less than the third percentile of the World Health Organization (WHO) standards for weight and height. The youngest infants in the survey were ten months of age, but this does suggest that if larger birthweights are also found in rural Tibet, they are not necessarily translating into better growth or survival rates there. Indeed, with respect to postnatal growth, these Tibetan children were smaller than Andean infants and children. Chin’s data on infectious disease were consistent with Ladakhi data; respiratory and gastrointestinal diseases accounted for 39% and 17% of under-ﬁve mortality, respectively. Child-care factors have been emphasized as key determinants of infant mortality in the Himalaya, often to the exclusion of other proximate determinants. Levine’s analysis of infant mortality in western Nepal (1987, 1988b) provides the most detailed insights into how child-care patterns might contribute to early death. In this region, many men outmigrate to work, leaving agricultural work almost entirely to women, who have to balance food production and child rearing. Breastfeeding is ubiquitous and prolonged (>2 years), but supplemental foods are introduced early and facilitate a mother’s return to work in the ﬁelds. In a village noted for extremely heavy workloads for women, over 75% of all one-month olds had been supplemented. Likewise, in rural Tibet, Chin (1992) found that most women returned to work within one month after birth, and consequently over 90% of all Tibetan infants had been supplemented with barley gruel within the ﬁrst month. In Nepal, supplements were grains mashed with water or butter, often contaminated by ﬂies and bacteria, especially in the summer (when more supplements were given), thus contributing to a seasonal rise in infant diarrheal disease and deaths (Levine 1988b). Although the ideal is for mothers to remain at home for at least a month with their newborns, in fact, mothers often must return to work and leave their infants at home, which is considered safer than lugging them to distant ﬁelds. Infants are particularly vulnerable to demons and ghosts that haunt public roads, spirits in the ﬁelds, rain and sun, and they are difﬁcult to transport over the challenging terrain. As Levine notes: “What a woman does depends on whether her household can afford the loss of her labor, whether there are other workers to replace her in the ﬁelds, whether there are responsible adults or children to leave the infant with who are not themselves needed in agricultural work, and also depends

164

An Ecology of High-Altitude Infancy

on seasonal labor demands” (Levine 1988b:237). Wealthier households can afford to hire additional help and reduce infant supplementation. In the winter when there are reduced demands on a woman’s agricultural labor, supplementation and separation are delayed until the infant is older. In contrast, among the Tamang, Panter-Brick found that nursing was blended into agricultural and domestic work schedules, and infants were suckled every 100 minutes or so during rests (1989). Child-care responsibilities were shared among groups of women in the ﬁelds, but toddlers were often left behind with older siblings (themselves only 3 years old or more) when mothers needed to attend to tasks distant from the household. Busy periods of work correlated with weaning and resulted in “maternal neglect of the older children left behind and also to a peak in the incidence of diarrhea” (Panter-Brick 1989:220). So while the health of younger infants is protected by frequent nursing and contact with the mother while she works, older infants and children may suffer from a lack of care or supervision. Levine observed evidence of child neglect in western Nepal, including naked toddlers sitting in the snow in winter, which surely contributes to hypothermia and vulnerability to disease. Such a situation is virtually unimaginable in Ladakh, at least for infants, who are considered very susceptible to cold and kept swaddled. On the other hand, it is not unusual to see toddlers dressed only in a shirt and well-worn, dirty cotton pants under the care of older siblings. (However, since I’ve never been in Ladakh during the coldest months of winter, I have not had the opportunity to observe how toddlers are dressed then.) This is but one example of what Levine describes as a pattern of widespread neglect of numerous categories of children, most importantly girls, higher birth-order sons, illegitimate offspring, or those in unstable families. Such children may be discriminated against if they are considered a net drain on household productivity and prosperity through a process of shrewd and unsentimental household calculus. Likewise, in Tibet, Chin (1992) found that while socioeconomic status did not seem to correlate with variation in child health parameters, household conﬁgurations did. Children had better health and survival indices when they lived in households made up of monogamously married parents, or those of their mother’s family. Chin speculated that women had higher status and more autonomy in their natal households than in those of their husbands, where they were under the surveillance and authority of their in-laws. Similarly, in polyandrous unions, a woman is dominated by her husbands and their families and may not have the authority to act on behalf of her children or to protect her own health. In contrast to

Perspectives on Reproductive Health in Ladakh

165

Levine’s work, Chin found no gender bias in child health parameters, and that daughters were brought to practitioners for health problems twice as frequently as sons. Although Levine makes a compelling case that neglect occurs in western Nepal, its contribution to the high rate of infant mortality there is difﬁcult to judge, and other authors have come to different conclusions (Goldstein 1976; Ross 1984). Since Levine did not break down infant mortality into neonatal and postneonatal rates or provide data on the causes of infant deaths, it is difﬁcult to ascertain what other factors might be playing a role. Similar ecological, cultural, and social-structural conditions are found in Ladakh, but they do not generate widespread neglect of infants. Certainly one son is desired by all families, but as equigeniture becomes more common, more than one son is problematic, especially if landholdings are small. In polyandrous families, which are becoming rarer, this is not a problem; in fact, it is an advantage in terms of increasing the pool of committed labor. Daughters are valued precisely for their labor in a variety of domains: the kitchen and household, daily religious observances, child care, and agricultural work, both in the household garden and in the ﬁelds. The loss of their labor occurs when daughters marry, which is usually in their early twenties, but this may be offset by a son’s marriage which brings a daughter-in-law into the household. As in Tibetan communities in Nepal, substantial dowries are not required for a daughter’s marriage. That less desired offspring might be discriminated against in some way comes as no surprise but is not sufﬁcient reason to explain high infant mortality in Ladakh. Ladakhis are extremely pragmatic about the limited resources available to them (stated most often in terms of availability of arable land) and the costs and beneﬁts of relationships in general (see also Ortner 1978 for a discussion of this quality among Sherpas). The fact that Ladakhis now eagerly seek means to control their fertility implies that they have a clear sense of the costs as well as beneﬁts of children (Wiley 1999) and that historically they may have been confronted with having to make hard decisions concerning higher parity children. In that case, we would expect infant mortality to have been much higher in the past. Remember that the data documenting high infant mortality from Ladakh derive from reproductive histories of women who chose to deliver in the hospital, most of whom were of relatively low parity. It is reasonable to conclude that they sought prenatal care and delivered there because they wanted a successful birth outcome and that their previous children had been desired as well. So, the Ladakh data are biased against an interpretation of high infant mortality deriving from intentional neglect.

166

An Ecology of High-Altitude Infancy

Summary: Ladakh in the Himalayan Context These studies direct our attention to the crucial importance of household composition and subsistence demands, since these shape the resources available for child care and maternal health in this cultural and ecological context. We have seen these operating in Ladakh, along with a kind of “laissez-faire” child-care system. Stillbirth, perinatal, neonatal, and infant mortality rates are high throughout the Himalaya, probably largely as a consequence of similar kinds of maternal conditions: hypoxia, workloads (often exacerbated by male absence), maternal nutrition, the need to balance infant feeding and care with work, and the availability and early use of supplemental foods, etc. But birthweights are lowest and neonatal mortality is highest in Ladakh, perhaps reﬂecting the additional problem of inadequate adaptive mechanisms for hypoxia during pregnancy due to the relative recency of biological ancestry in this mountain environment. Ladakh in the Context of South Asia Ladakh is increasingly linked politically, economically, and socioculturally to the lowlands of north India. Thus a comparison with populations in this area provides some insight into the relative importance of high altitude to Ladakhi birth outcomes, given similarities in health infrastructure, lineage systems, and extended family households with subsistence patterns that often require women to work hard at agricultural and other household tasks, among others. At the same time there are not only striking ecological differences but also cultural differences in gender dynamics, diet, and economy. Discussion of reproductive health in India inevitably centers around the broad themes of poverty and gender discrimination, which act to restrict the amount of resources put toward female health in general and reproductive health in particular. The household is the locus of such behavior; its form is a product of a strongly patriarchal social organization that emphasizes male power often at the expense of female well-being (Das Gupta 1995). At the heart of this differential treatment is the relative cultural and economic value accorded males and females. Sons inherit land (which they often work jointly) and wealth, support parents in their old age, light their parents’ funeral pyres, and bring in wives, who come with substantial dowries. Hence, the economic value of males greatly exceeds that of females for landed households. Das Gupta (1995) argues that female status is low in early childhood, rises somewhat during adolescence, drops sharply at marriage, and remains low until the later reproductive years when a woman becomes

Perspectives on Reproductive Health in Ladakh

167

a mother-in-law and grandmother, but declines again with widowhood and old age. Devaluation by her natal family can take the form of reduced investment in a daughter’s well-being, but surviving females are probably better treated in their natal homes than their husband’s home. At marriage, North Indian women move into their husband’s extendedfamily household, where they are often quite distant from their natal families. There she has very low status and is subservient to her husband, his brothers and their wives, and her mother-in-law. Women are likely to be nutritionally most vulnerable during the early years of marriage and their active reproductive years, before they have produced surviving sons. As women age and their sons grow up, their status increases; they acquire important decision-making roles and can exercise authority over younger women (e.g., daughter-in-laws). A similar pattern of household dynamics that impacts negatively on reproductive health is found in Ladakh, but gender discrimination is ameliorated by dowries being relatively minimal. Females are subordinate to males, but that inequality is not as extreme as the lowlands of North India and does not manifest in dramatic differences in expected behaviors. Nor are the extreme poverty and exploitative economic relationships common to rural India characteristic of Ladakh. The themes of poverty and gender discrimination are discussed as they mediate birth outcome through more proximate maternal conditions such as food intake and nutritional history (weight and height), physical activity, anemia, age, parity and interbirth interval, household income, and utilization of health care services. Birthweight Relative to other parts of the world, South Asia has the highest rates of low birthweight newborns and small size for gestational age (de Onis et al. 1998). Between 25% and 50% of newborns have been found to be LBW in studies from Nepal and India (Sethi et al. 1991; Singh and Paul 1988). With 27% of all births classiﬁable as LBW, the data from Ladakh are consistent with broader regional trends. Among poor and middle-class rural Indians, birthweights tended to fall within the range of 2600 to 2900 grams compared to 2745 grams in Ladakh and 3400 grams among a high-income sample in New Delhi (Srikantia 1989). So, even without hypoxia as a factor, the statistics for rural South Asia suggest that other factors, such as maternal diet, activity during pregnancy, and female health status, contribute to this birthweight distribution. That is, what we observe in Ladakh may not necessarily be attributable to altitude but may result from similar maternal factors, themselves related

168

An Ecology of High-Altitude Infancy

to common social, cultural, and economic factors in South Asia. On the other hand, perhaps the birthweight distributions are similar, but are the outcome of qualitatively different causes. The nutritional conditions of pregnancy derive from the more general dietary background of women in India. Although they are responsible for preparing and serving food, women generally eat last, after elders, men, and their children. Household food resources are contingent on economic status, but even in wealthy households, reproductive-age women may not get enough to eat. Pregnancy status does not change that. Women generally either maintain their prepregnant eating patterns or decrease their consumption. Ramachandran (1989) reported that rural pregnant women consumed between 1200 and 1600 kilocalories/day during pregnancy, and gained relatively little weight (<6 kg), and that caloric intake varied by seasonality in household income and agricultural production. Socioeconomic status affects maternal weight and weight gain, and subsequently the infant’s birthweight, with differences between wealthy and poor women on the order of 10 to 15 kilograms in maternal weight, 6 kilograms in weight gain, and 400 grams in infant birthweight. The average weight of rural Indian women in two studies was 43 kilograms and height was 151 centimeters (Ramachandran 1989; Sethi et al. 1991); similar statistics are widely reported from throughout rural India. Triceps skinfolds during late pregnancy and arm circumference were 9.2 millimeters and 22.0 centimeters, respectively (Ramachandran 1989). Maternal weight is a signiﬁcant predictor of birthweight, and maternal weights less than 45 kilograms are routinely associated with dramatic increases in LBW and perinatal mortality (Mavalankar et al. 1994). Relative to these parameters, Ladakhi women are heavier (49 kg; though their weight was measured postpartum, when they retained some birth tissues and ﬂuid) and taller (153 cm), but they have similar arm circumferences and skinfolds. Given that birthweight averages and ranges are similar, yet Ladakhi women are larger overall, other determinants must be driving low birthweights in Ladakh. As in Ladakh, North Indian women say that they should consume strengthening foods while pregnant, but they often face constraints that prevent them from doing so. As one woman noted, “The pregnant women should have ghi (clariﬁed butter), milk, fruit, coconut, anything she desires,” while at the same time, “pregnant women should get good food, but can you tell me where they do so? Who eats well in a village? Anyway, we can only afford fruit or milk occasionally” ( Jeffery et al. 1988:77–78). Pregnant women, especially when newly married, have little control over their diet, being subject to both household economic constraints and their

Perspectives on Reproductive Health in Ladakh

169

husbands and mothers-in-law. To request special or additional food is considered shameful (Jeffery et al. 1988). Anemia (deﬁned in India as Hb <11.0 g/dl) is extremely widespread among pregnant women; studies routinely ﬁnd frequencies of 40 to 50%, in some areas reaching virtual 100% prevalence, especially where hookworm is endemic (Agarwal and Agarwal 1987; Sethi et al. 1991). Hemoglobin levels during pregnancy have been found to average 9.5 g/dl, with little variance between studies (Ramachandran 1989). Anemia was associated with reduced birthweight by 60 to 320 grams, depending on its severity, and as much as a tenfold increase in perinatal mortality (Ramachandran 1989). Anemia usually correlates with other indices of maternal malnutrition, and iron intake increases with increased food consumption in rural India, where meat and other iron-rich foods are rarely consumed. In comparison, the mean hemoglobin level in the Ladakh sample was 10.4 grams/deciliter, and there were very few cases of severe anemia (Hb <5.0 g/dl). Hemoglobin status did not signiﬁcantly predict birthweight in Ladakh, but other studies (e.g., Haas 1980; Moore et al. 2001b) suggest that its relationship to birthweight is not straightforward at high altitude. Most rural Indian women experience little change in their workloads when they are pregnant. Studies in Uttar Pradesh (Northern India) found that pregnant women continued to work twelve to eighteen hours a day at tasks requiring physical labor, as wage laborers, in the household, or in subsistence agriculture (Jeffery et al. 1988; Sethi et al. 1991). Hauling water and ﬁrewood, tending animals, collecting dung, or working in the ﬁelds are part of a woman’s daily work and are not usually alleviated by her need for rest during pregnancy. This is especially true for primiparous women, who are usually young daughters-in-law. Even though pregnant women know they need additional rest, economic and domestic demands take precedence, and few women can afford to reduce their physical labor during pregnancy (Jeffery et al. 1988). Other factors interrelate with nutrition and work. They are the young age at which many rural Indian women begin their reproductive careers and repeated, closely spaced pregnancies that can lead to maternal depletion. Household surveys in northern India indicate that between 20 and 40% of girls are married before the age of ﬁfteen (average age at marriage ∼16–17 years), and 10 to 15% of all births occur to teenage mothers (Jejeebhoy and Rao 1995). Indian researchers have routinely identiﬁed teenage motherhood as a risk factor for both low birthweight and premature birth as well as high perinatal and neonatal mortality (Agarwal and Agarwal 1987; Hirve and Ganatra 1994). As reproduction is a new bride’s primary role in her husband’s household, women

170

An Ecology of High-Altitude Infancy

often become pregnant shortly after marriage, and ﬁrstborn offspring often suffer much higher mortality rates than later children. However, when many births result from pregnancies spaced so close together that women have scant time to recover their energy resources before these are again put toward fetal growth and breastfeeding, these pregnanies are also associated with low birthweights and high rates of early death (Hirve and Ganatra 1994; Singh and Paul 1988). This reproductive scenario is at substantial variance from Ladakh, where on average, women marry at age twenty and begin their ﬁrst pregnancy about three years later, so that there is no overlap between their own growth and that of their fetus and infant. Moreover, average interbirth intervals in Ladakh are about three years, partly because of increasing use of contraceptives to space births (Wiley 1998, 1999). A salient difference that emerges between Ladakh and other regions of India is the extent to which women make use of prenatal services. Numerous reports indicate that most Indian women are unaware that such services exist. Those who are aware still do not use them, and those who do average only one visit, usually late in pregnancy (Jejeebhoy and Rao 1995). A broad network of Auxiliary Nurse-Midwives (ANMs) is supposed to serve rural India, and many villages have health centers that employ an ANM who is supposed to provide regular medical checkups, distribute iron and folic acid tablets, monitor fetal development, give tetanus toxoid injections, and refer high-risk women to specialized clinics or hospitals. In reality ANMs are underutilized and their services are not as complete as envisioned. As a consequence, national surveys show that less than 25 percent of women received iron-folic acid supplements or tetanus toxoid inoculation; very few (3–22%) had ever gone to a prenatal checkup or were visited by a health worker before or after the birth (Mathai 1989). Two reasons are most often articulated for the underutilization of these services. The ﬁrst is a cultural explanation: pregnancy and birth are simply not considered processes of medical importance and hence medical services are perceived as unnecessary. Traditional birth attendants (dai) are widely used in rural areas, but they play a very circumscribed role and women do not use them for prenatal care. Young women may not be able to claim special dispensation for prenatal medical care, and public acknowledgment of pregnancy (associated with special trips for antenatal care) is considered shameful. But further, women who did seek prenatal care widely reported dissatisfaction with the services they received. ANMs frequently were not at their post or had no supplies with which to provide care; they were not known among the local community, and many were young women who were perceived to have little knowledge

Perspectives on Reproductive Health in Ladakh

171

of reproductive health. Economic constraints, in the form of not being able to take off time from work to seek care or not affording transportation also reduced the usage of these free government-sponsored health services. Ladakh shares this Indian health infrastructure. Few village women make use of local clinics, which are both poorly staffed and insufﬁciently stocked with supplies, so a visit is not considered worthwhile. But instead of not seeking prenatal care at all, increasing numbers of women make the trip to Leh to attend Dr. Ladhol’s prenatal clinic at the hospital, despite having to take off a whole day of work to make the trip and to wait to see her. The success of the prenatal clinic is to be credited to Dr. Ladhol’s status as a well-known, highly respected, and compassionate Ladakhi woman with a great deal of cultural and biomedical authority who is understanding of and responsive to local women’s reproductive health issues (Wiley 2002). Neonatal and Infant Mortality Table 6.5 summarizes published data on perinatal, neonatal, infant, and child mortality in India. While all of these rates are high, perinatal and neonatal rates are particularly high. Neonatal mortality relative to infant mortality is fairly constant across studies (∼60%) and has remained fairly constant over time, despite some reduction in overall infant mortality. This pattern is likely to derive from the high frequency of LBW and low average birthweights and their antecedents in India, a connection that many researchers have suggested and/or statistically demonstrated. Important contributors to high rates of early mortality included, most signiﬁcantly, low birthweight, prematurity, neonatal tetanus, and birthinduced trauma. Hospital studies of birthweight and neonatal mortality have shown the standard pattern of dramatically increased risk below 2500 grams. Low birthweight associated with prematurity accounted for up to 50% of all neonatal deaths, 70% of perinatal deaths, 90% of neonatal deaths, and 50% of infant deaths were to LBW infants (Singh and Paul 1988). Government surveys in 1987 found that prematurity and neonatal respiratory infection were responsible for 40% and 15% of infant mortality, respectively, while other respiratory disorders contributed 15% and diarrheal disease up to 14% of mortality in the ﬁrst year (Jejeebhoy and Rao 1995). Gastrointestinal infections accounted for 6 to 21% of neonatal and 19 to 31% of postneonatal deaths; respiratory diseases caused 19 to 22% of neonatal and 25 to 31% of postneonatal deaths. Hence infectious diseases are major determinants of infant death, and undoubtedly interact

Ladakh Hospital Reproductive histories

Agarwal & Agarwal (1987) Bihar (household survey) Uttar Pradesh (household survey) Singh & Paul (1988) (All India rural survey data) Kielman et al. (1983) Narangwal study, Punjab, 1970 Jejeebhoy & Rao (1995) Registrar General, 1987 Rural India, 1976–1987 Jeffery, Jeffery, & Lyon (1988) 2 village survey of births 1966–1985 51/1000

>30/1000 55/1000 160/1000

50/1000 52–77/1000

13/1000 9–19/1000

36/1000 90/1000

46–80/1000 87/1000

Perinatal mortality

57/1000

Stillbirth rate

Table 6.5. Infant mortality statistics from North India

144/1000

52/1000

58/1000 64–88/1000

73/1000 66/1000

86/1000 78/1000

Neonatal mortality

38/1000

33/1000

38/1000

56/1000 39–48/1000

92/1000 64/1000

Postneonatal mortality

182/1000

86/1000

95/1000 114–140/1000

129/1000 105/1000

178/1000 141/1000

Infant mortality

75–80%

60%

61% ∼62%

∼60% ∼62%

48% 55%

NMR/IMR

Perspectives on Reproductive Health in Ladakh

173

with birthweight and household patterns of birth, infant feeding, and child care. Birth trauma (asphyxia) accounts for a substantial percentage of neonatal deaths (20–27%) (Singh 1986) and is generally attributed to the conditions under which most Indian infants are born. Less than 15% of rural deliveries occur in health institutions (Jejeebhoy and Rao 1995). Women are extremely reluctant to go to a hospital and rarely will the mother-inlaw or husband permit a hospital birth. Aside from a cultural mandate for birth to occur in the father’s home, hospital deliveries incur substantial monetary costs, separate a laboring woman from her kin and surround her with strangers, and breach her modesty. Thus most births occur at home with no professional assistance, attended by female household members and the dai, whose principal role is to deal with the pollution attendant to childbirth. The cutting of the umbilical cord is among the practices cited for its negative effects on early mortality, for it is usually done with unsterilized instruments such as scissors, razor blades, or sharpened pieces of reed from a thatched roof. This undoubtedly contributes to the high frequency of neonatal tetanus, when such instruments are contaminated by Clostridium tetani. Since only a small percentage of women are immunized against tetanus during pregnancy (which can prevent tetanus in newborns), this disease is a very common cause of neonatal death under these birthing conditions. Up to 62% of all neonatal deaths and 10 to 29% of postneonatal deaths have been attributed to tetanus in north India (Agarwal and Agarwal 1987); the rate is strikingly lower in hospital settings (<10%). After birth, mothers and newborns remain in seclusion for up to forty days, in part because they are still retain the pollution of childbirth, but also because they are considered to be vulnerable to cold and the evil eye (Jeffery et al. 1988). Unlike pregnancy, after the birth women do observe dietary changes, although these may entail eating no solid foods for several days. Newborns are fed only a syrup of boiled water mixed with sweet ingredients for the ﬁrst three to four days after birth. The mother’s ﬁrst milk, the immunoglobulin and fat-rich colostrum, is considered far too “heating” for a newborn, and is commonly discarded; only after a regular milk supply is established is the baby nursed. Breastfeeding continues throughout the ﬁrst year, and solids are introduced after six months, an age at which growth faltering begins to manifest (Ramachandran 1989). Breastfeeding is arranged around a woman’s work schedule, which quickly returns to its previous routine. The infant is generally left at home while a woman works and is brought to her to be fed if she is working outside the home. Older siblings, sisters, and mothers-in-law or local young girls are called on for help, but this means little other than

174

An Ecology of High-Altitude Infancy

keeping an eye on the infant and notifying the mother if it needs feeding or cleaning. If an infant gets sick, the costs of treatment must be balanced by other household needs; consequently, treatment is often delayed until the symptoms become serious. Virtually all studies in North India have found that females have higher rates of infant and child mortality. Males generally have higher neonatal mortality rates, as is common across populations (Das Gupta 1987; Srinivasan and Kanitkar 1989). Higher postneonatal female mortality has been widely attributed to gender discrimination in the household. First, females are more likely to be malnourished: 25% of females are malnourished compared with 15% of males; girls receive only 86% of the calories as males and are disproportionately represented among infants and children dead from malnutrition (DeSweemer et al. 1983). They are also taken to a clinic less often and with more serious symptoms. In North India, Miller (1981) has shown that the ratio of male to female hospital admissions was greater than 2:1, while Das Gupta (1987) showed that household expenditures on medical care were over two times higher for sons than daughters. Similarly, Kielmann et al. (1983) found that 48% of female children who died had received care in the ﬁrst twenty-four hours of the illness, compared with 64% of males. Summary: Ladakh in the Indian Context With this as an admittedly broad description of reproductive health in North India, how does it compare with reproductive health indices in Ladakh? First, the high rate of LBW infants and high levels of neonatal mortality are strikingly similar, and it is clear that low birthweight contributes to the high neonatal mortality level in both contexts. However, the birthweight distributions derive from somewhat different determinants. Maternal weight and activity drive birthweight in both cases, but Ladakhi women are larger, have their ﬁrst birth at a later age, and do not generally have closely spaced births. Anemia is a signiﬁcant problem in both contexts, but there is not such a high frequency of severe anemia in Ladakh. Ladakhis make wider use of prenatal care, although currently there are not enough data on pregnancy and birth among women who do not deliver at the hospital. So, from the perspective of the antecedents of birthweight, it would seem that Ladakhi women should deliver larger newborns; the fact that they do not suggests that hypoxia is adding to fetal growth retardation there. Perinatal, neonatal, and infant mortality statistics from the Ladakh study show a slightly different pattern from that observed in other parts of India. Perinatal mortality in the hospital was within the same range as

Perspectives on Reproductive Health in Ladakh

175

found in other studies; but neonatal mortality was much higher, while postneonatal mortality was lower, resulting in a higher percentage of infant deaths occurring in the neonatal period. The best current explanation may be hypoxia, for there is reason to believe that neonatal mortality should in fact be higher in other parts of India than in Ladakh, given the divergent conditions under which birth occurs and the different risks that newborns and infants encounter. Thus, any suggestion that hypoxia is relatively benign in its effects on birthweight and early mortality (e.g., Cogswell and Yip 1995) is not supported by the Ladakhi data. Tetanus is not a problem in Ladakh; there were no diagnoses of tetanus in the pediatric or obstetrician’s records, and women in the study had been immunized against tetanus as part of their prenatal care. Home deliveries may be associated with higher rates than hospital deliveries, but Ladakhi physicians have commented on the lack of neonatal tetanus in Ladakh, despite the fact that adults generally have tetanus antibodies, indicating exposure (Norboo and Yahya 1988). It is hard to say how delivery practices compare, as all of my work was done in a hospital context. Traditionally (and still in the majority of births), delivery occurred at home, with no trained attendants other than more experienced female kin, and this undoubtedly was associated with a higher frequency of birth trauma resulting in perinatal death. Severe bloodborne infections (septicemia) are important sources of neonatal mortality in Ladakh, and so other sources of perinatal infection exist, and small newborns are most vulnerable to them. In both Ladakh and North India, the ideal is for birth to occur at the father’s home, and women and their newborns observe a period of postpartum seclusion as they recover from the pollution of childbirth. However, increasing numbers of births in Ladakh are taking place at the hospital, while no such trend is apparent in other parts of India. The Leh hospital is not such an alien environment; the staff and doctors are Ladakhi, although not all of the staff are well trained in obstetrical practice. Birth conditions and outcomes are not ideal in the hospital and result in substantial morbidity and mortality, but they are likely to be better than home births, especially in cases involving complications or neonatal disease. In other parts of India, hospital birth and antenatal care are both associated with improvements in birth outcome. So this should have reduced perinatal and neonatal mortality in Ladakh, but rates there are still higher than those from other Indian studies. Respiratory diseases in the neonatal period are more prevalent sources of mortality in Ladakh, whereas gastrointestinal diseases are rare among these neonates yet quite frequent among North Indian neonates.

176

An Ecology of High-Altitude Infancy

Respiratory diseases are likely to represent a greater mortality risk in Ladakh due to the infants’ interaction with exogenous hypoxia. The greater prevalence and diversity of infectious diseases in the lowlands have a greater impact on infant mortality than in Ladakh, especially in the postneonatal period. They also affect maternal health, as hookworm infection is a major source of anemia. Infant feeding practices are similar in some regards, with breastfeeding the predominant pattern. Ladakhis do nurse right from birth, not discarding the immunoglobulin and fat rich colostrum produced right after birth, as Indian women do. However, Ladakhis make earlier use of supplemental foods than do Indian women, who wait until well after the child’s sixth month. In both cases, nursing occurs within the context of maternal work demands, with infants often left at home in the care of other female household members and attended to more often when the mother is doing domestic tasks. The pattern of gender differences is similar but much more muted in Ladakh compared with North India. Women are subject to the authority of their husbands’ households, which restricts their claims to food and rest during pregnancy and compromises birth outcome. Sons are brought to the hospital more often than daughters for treatment of an illness, and they have marginally lower death rates there. Boys are also less likely to be underweight relative to their height (Wilson et al. 1990). These differences diminish as children age, unlike the case in North India, where they diverge with age. There is a common bias toward female respiratory illness throughout India that emerges over the female lifecycle (though is evident after infancy in Ladakh) as a consequence of females spending more time in smoky, unventilated kitchens cooking (Caldwell and Caldwell, 1990, cited in Kishor 1995), and these are likely to become more serious illnesses in the hypoxic context of Ladakh. Although Ladakh has been part of the nation state since India’s inception in 1947, communication in the form of air and road transport, telecommunications, tourism, and direct political relations with Delhi have accelerated over the past decades. From India, Ladakh derives media images, migrants, health and educational infrastructures, and national political agendas, all of which inﬂuence social and cultural institutions and ideologies. The process by which marginalized groups (or scheduled castes/tribes, as they are ofﬁcially labeled), such as Ladakhis, are drawn into the larger Indian sociocultural sphere has been described as Sanskritization, which is characterized by the emergence of similar patterns of gender relations, dietary habits, marriage and household formation patterns, to name a few. Although Ladakhis have not thus far been absorbed by pan-Hinduism, it will be interesting to observe how they

Perspectives on Reproductive Health in Ladakh

177

begin to take on a national as opposed to regional identity and how this impacts reproductive health. Increases in stratiﬁcation, gender discrimination and constraints on gender roles, or economic marginalization could erode an already precarious reproductive health situation, especially within the unique ecological context that adds the exacerbating effects of hypoxia. Conclusions These three comparisons point to some general conclusions about the determinants of reproductive health in Ladakh, especially infants’ small size at birth and high neonatal mortality. Compared with those in the Andes and Tibet, Ladakhi newborns are smaller and more vulnerable after birth. Variation in maternal nutritional status, work patterns, and high-altitude ancestry are likely to account for some of this variation, as other factors – such as hypoxia exposure, breastfeeding, and respiratory disease – are similar. Ladakhi newborns are similar in size to those born in the lowlands of north India. Because Ladakhi mothers seem to be somewhat better nourished, their pregnancies are monitored by an obstetrician, birth occurs in the hospital, and they nurse their offspring from birth, this similarity in birth outcomes is particularly striking and implies that hypoxia may be contributing independently to low birthweight and high rates of early mortality among Ladakhi infants. Ladakhi infants are protected in the postneonatal period relative to other populations, perhaps due to better nutrition, fewer infectious diseases that result in mortality, and child-care patterns that do not actively discriminate against them, especially females. This is not to say that postnatal health is ideal but rather that Ladakhi infants suffer disproportionately in the neonatal period and do not experience some of the routine threats to infant health that are evident in South Asia and the Andes. Hence, ecological factors appear to play important roles in shaping the risks inherent to the reproductive process in Ladakh.

7

Toward Relevant Research: Adaptation and Policy Perspectives on Maternal-Infant Health in Ladakh

There are two broad issues that this work must now address. The ﬁrst is theoretical and concerns the concept of adaptation. What do the ﬁndings from this investigation of maternal-infant health tell us about the adaptive status of Ladakhis vis-`a-vis their socio- and natural ecology, and about evolutionary processes acting in this population? The second is more applied. How do these results inform health policy or help in the design of effective health interventions targeted at reducing the signiﬁcant burden of early mortality in Ladakh? Furthermore, do the answers to these two questions have anything to do with one another? Evolutionary perspectives have rarely been linked to policy, and in some cases they have been interpreted as antagonistic, especially to social justice–based health policies (Singer 1989). Since the term evolutionary implies long-term biological processes, it does at ﬁrst glance seem incompatible with policy making, which is concerned with issues of more immediate importance. But what an evolutionary perspective, especially one concerned with both biology and culture, tells us is that there are rarely effective “quick ﬁxes” to endemic health problems that are tied to a complex of local behavior, ecology, and resources. In addition, it forces us to consider the long- and short-term costs and beneﬁts of any sort of intervention in relation to the whole of a population’s adaptations. After such considerations, even with no particular theoretical agenda, some possible domains of intervention can be suggested to address the very high rate of neonatal mortality revealed in this study. To address the issues of adaptation and health policy, the relevant determinants of maternal-infant health and infant mortality need to be laid out. This allows us to see the likely consequences of particular interventions or the pathways through which hypoxia impinges on infant survival. Earlier I have described two proximate determinants models of infant and child mortality (see Figures 1.2; 5.1), and I intend to blend these models and specify the determinants in the Ladakhi situation. In doing so, I hope to take into account how ecology and that complex of behavior, 178

Perspectives on Maternal-Infant Health in Ladakh

179

tradition, history, and ideology called culture ultimately inﬂuence infant survival. The Proximate Determinants of Infant Mortality in Ladakh Figure 7.1 is a schematic representation of the forces that inﬂuence infant survival in Ladakh. Like Chen’s model (Figure 5.1), it features birthweight at the center – since birthweight is a key determinant of infant survival – and its link to maternal characteristics. This results in a twostage model in which ﬁrst the determinants of birthweight are outlined, and then birthweight becomes a proximate determinant for infant mortality that interacts with other biological, behavioral, and structural factors. Unlike Millard’s model (Figure 1.2), which is arranged in a hierarchical, linear fashion, this one nests the determinants within concentric domains of inﬂuence, reﬂecting the pathways through which more distant forces impact the proximate determinants of infant death. Pervading the whole model is history; that is, this set of determinants derives from a complex process of population settlement and organization, subject to more recent regional (i.e., South Asian) and global (e.g., tourism) inﬂuences. History also plays out in individual life histories, themselves adaptive responses to shorter term processes and contingencies that mold development. It is the addition of the historical component that allows us to assess the net effects of both evolutionary and social processes in shaping birthweight and infant survival. Considering both the general domains of inﬂuence and their speciﬁc features, we work outward from birthweight at the center of Figure 7.1. The most immediate (proximate) factors inﬂuencing birthweight are the biological conditions of mothers (weight and parity) and infants (sex, gestational age). Affecting biology is individual behavior; for both mothers and infants, behavior is primarily reﬂected in diet (during pregnancy for the mother; nursing, consumption of supplemental foods for the infant) and activity (work for the mother, exploratory behavior for the infant). Behavior is locally inﬂuenced by household dynamics which include household membership (especially in relation to landholdings), economic status, physical environment (such as house size), domestic pollution, hygiene, child caretaking, and health care usage patterns. These in turn are inﬂuenced by larger social and cultural dynamics including involvement in tourism or the military; agricultural subsistence; social distinctions/ conﬂict around religion and privilege; seasonal migration; and an infrastructure of roads, health care institutions, and cultural ideologies about

180

An Ecology of High-Altitude Infancy

SEASONAL COLD

ARIDITY

SOCIO-CULTURAL CONTEXT ideology, gender, religion, involvement in military, tourism, agricultural practices

Biological history

ECOLOGICAL CONTEXT HYPOXIA

HOUSEHOLD Size, composition, subsistence, economic status, landholdings

Maternal weight infection

BEHAVIOR

BIOLOGY Fetal age, sex

HISTORICAL CONTEXT

Maternal diet, work

Domestic pollution & hygiene

Birthweight

Neonatal Mortality

Caretaking Undernutrition

Infant mobility

Social history

Infection

Infant diet Breastfeeding Supplementation

Post-neonatal Mortality

Figure 7.1. Proximate determinants of infant mortality in Ladakh.

reproduction, lineage, and gender. All of these exist within the mountainous ecology, including the abiotic and immutable features of hypoxia and seasonality in temperature. Birthweight The most important proximate determinants of birthweight were maternal weight and gestational age, which both positively affected birthweight. Larger infants were those who were older and more mature, and those born to mothers who were heavier. Males were signiﬁcantly larger than females, but sex was not a signiﬁcant predictor of birthweight once maternal weight and EGA were factored in. Affecting maternal and infant biology were maternal dietary and work practices during pregnancy. Eating more was associated with a small increase in birthweight and working less resulted in a substantial increase in the biological resources of a newborn. Household dynamics were the crucial inﬂuences on maternal behavior during pregnancy, especially workloads. The household

Perspectives on Maternal-Infant Health in Ladakh

181

labor-land relationship, in conjunction with household economic status, which inﬂuences the likelihood of hiring additional workers among other things, and a woman’s status in the household shape the amount of work she must do, regardless of reproductive state. These may also inﬂuence her ability to seek prenatal care, as this requires time off from household tasks and is more problematic for women who live far from Leh. Economic status inﬂuences the amount and diversity of foods available to a pregnant woman and interacts with religion and ethnicity, as pregnant Muslim women appear to work harder at agricultural tasks than Buddhist women in Ladakh, while Tibetan women have access to modes of employment that do not require them to do physically demanding work. Household dynamics are in turn a product of local agricultural subsistence strategies that require large labor inputs to be productive. Economic opportunities in the military and tourism result in shortages of labor, since tourism is concentrated during the months of most intensive agricultural tasks. At the same time, both these occupations bring in income and beneﬁts such as food and other commodities. During the early 1990s, religion restricted the resources available to a household; a social boycott limited Buddhists’ access to meat, eggs, and imported produce in the Kashmiri bazaar. Curfews limited mobility and access to health care resources, especially the hospital, although the number of prenatal visits and supplements derived from those visits did not have strong effects on birth outcome in this study. A poor infrastructure of biomedical health care in the villages, road access, and transportation further constrain access to quality health care. Along with these structures are ideologies that on the surface seem protective of reproductive health (by promoting rest and consumption of high-quality foods for pregnant women; by granting women liberty to seek health care on their own), but in reality they fail to reduce the double burdens of reproduction and economic production that women bear. Further, women recognize that smaller babies are easier to birth, and as Kuhn (1988) found, women may explicitly choose to work hard during pregnancy to avoid laboring hard during birth. The mountainous ecology certainly challenges attempts to make infrastructural improvements in Ladakh, since the terrain makes access to many villages difﬁcult. It further sets the boundaries to agricultural production in a geographic but also a temporal sense by forcing production into the short summer months. Women who are in the latter stages of pregnancy during these months are likely to be especially burdened by work demands. On the other hand, produce from their gardens becomes available during the summer and may offset some of the energetic drain caused by increased work. Over the course of the spring, summer, and fall months, birthweight of infants went up, perhaps as the result of additional

182

An Ecology of High-Altitude Infancy

nutrients devoted to fetal growth since there were no seasonal changes evident in women’s postpartum weight or skinfolds. Finally, high altitude is associated with hypoxia, which acts as an independent determinant of birthweight, constrains fetal growth throughout pregnancy, and may increase the risk of PIH, which negatively impacts birth outcome (Moore et al. 1982; Palmer et al. 1999). When work is undertaken under hypoxic conditions, or when mothers’ lung function is compromised by chronic respiratory infection or domestic pollution, oxygen scarcity may become more acute for the fetus, especially in the last trimester. Collectively, these multiple factors contribute to the high prevalence of low birthweight among newborns in Ladakh (27%), and the low average birthweight (2764 g). Fetal growth is relatively slow and fetuses accumulate few energy stores to facilitate their transition to postnatal life. As a result, many infants are born with few resources to draw on, and this contributes to their vulnerability to infection as well as any other stress that demands energy in the postnatal environment. Neonatal Mortality In Ladakh, the majority of infant deaths take place in the ﬁrst month, although morbidity is more diffusely spread. Because the period of highest risk is clustered so close to the fetal period, the factors affecting fetal growth act indirectly as determinants of early death. The resources that an infant starts out with in the postnatal world are signiﬁcant predictors of its survival prospects, as evidenced by the strong relationship between birthweight and neonatal survival in this study. Ponderal Index, a measure of acute relative to chronic growth retardation, was also associated with mortality in that newborns whose growth was compromised by increasing stress as gestation progressed had higher probabilities of death. Age affects the relative functional maturity of a newborn’s physiological systems, but EGA was poorly described in this study and did not predict mortality independent of birthweight. Sex also played an independent role; that is, males weighed more than females at birth but had higher neonatal mortality rates, the standard pattern reported in the literature (Chase 1969; Cogswell and Yip 1995). Both reproductive histories and hospital neonatal mortality rates showed a bias favoring female survival in the ﬁrst month. Beyond these biological factors, severe blood-borne and respiratory infections are important causes of death for neonates, and these stem from broader environmental factors. Septicemia can result from maternal reproductive tract infections, and probably accounts for many of the obstetrics ward neonatal deaths and those cases documented among the

Perspectives on Maternal-Infant Health in Ladakh

183

pediatric admissions. These are particularly serious for small newborns. Respiratory disease can result from a complex of problems including hypoxia, infection, and indoor pollution, all of which inhibit adequate oxygen uptake by a neonate. The latter two derive from a household environment characterized by poor ventilation and indoor cooking and heating with fuels that release noxious gases and particulate matter. Smaller or more densely inhabited houses may exacerbate these effects. Child-care practices that place neonates close to heat reduce the risk of hypothermia but further compromise their respiratory function. Virtually all Ladakhi newborns are breastfed and this practice likely protects their health. Gastrointestinal disease that results in serious morbidity or mortality is rare among neonates. Although in describing the cause of death for neonates women often said that they refused to eat, this was probably indicative of another underlying problem as anorexia is often a symptom of infection. Mothers nurse at night and with varying frequencies during the day during this ﬁrst month, when they remain at home during the period of postpartum seclusion. This time may be interrupted by the need to work, and thus depends on the season and household labor resources. If a woman returns to work within the ﬁrst month, breastfeeding may be less frequent and supplemental foods introduced earlier. Such early supplementation may explain the few cases of gastrointestinal disease noted among neonates. These household dynamics interact with ecological processes, but the social and cultural processes that intervene between these two domains play a small role in health in the ﬁrst month of life. Access to medical care is probably an important factor here, but neonates are taken to the hospital less frequently than older infants. Respiratory disease is most problematic in winter months when windows are closed and additional fuels are used to keep warm. Given the extreme winter cold in Ladakh, hypothermia is a serious threat and Ladakhis are acutely concerned that young infants will become cold. So the seasonality of the Ladakhi environment contributes to variation in neonatal and infant mortality. Hypoxia interacts with other sources of respiratory disease to add yet another stress to small neonates who have few resources to meet such challenges. Postneonatal Mortality Mortality in the postneonatal period is dramatically reduced relative to neonatal mortality; it becomes increasingly dissociated from maternal factors and neonatal characteristics and increasingly associated with infant behavior (dietary, mobility) and the larger ecology of the household compound and neighboring environs. As their survival is more assured

184

An Ecology of High-Altitude Infancy

(although 2% to 5% of infants do die in these months), infants are named and take on full household membership status, which remains their primary “niche.” They are less often cared for exclusively by their mothers and more often watched over by other adults and especially older children. As infants begin eating more household foods, their exposure to gastrointestinal pathogens increases, although these are rarely fatal. Their mobility likewise brings them into contact with other threats to their wellbeing (toxins, pathogens, accidents); but while these may be sources of morbidity, they rarely convert to mortality. Although male and female infants do not have strongly different disease ecologies, boys, as they age, have more gastrointestinal diseases than girls, perhaps as a function of their wider ranging behavior in the summer, and girls have more respiratory disease, a function of their primary locus in the household (a pattern that persists across the life span). Males are taken to the hospital more often, but there are not strong differences in hospital mortality among those admitted. So, gender ideologies that are modestly biased in favor of males during infancy result in slight differences in morbidity and mortality patterns. Morbidity and mortality take on a more seasonal pattern among older infants and toddlers, with respiratory disease and death peaking in the winter months, and gastrointestinal disease (and much more rarely, mortality) rising in the summer months. This is part of a natural seasonal cycle, but it is exacerbated by household heating strategies in the winter and water supplies that are contaminated in the summer, when infants have more exposure to them. This contamination is related to the inﬂux of summer residents and tourists and the poor public health infrastructure of Leh and its environs. The proximate determinants model points to the importance of both natural ecology, its variation by season, and the ways it is modulated by the household environment and associated patterns of behavior, to reproductive outcome (birthweight and infant survival). It pinpoints the demands on maternal resources that compromise maternal well-being and infant survival. We now turn to issues of history and adaptive processes and how these interact with household dynamics and the larger mountain environment. Adaptive Status: Ladakhi Mothers and Infants Historically, research in Ladakh has emphasized the role of adaptive processes, especially with regard to subsistence in what has been described as an extremely marginal environment. What has resulted from that body of work has been a very encouraging picture of high levels of agricultural productivity, maintained with few negative impacts on the environment,

Perspectives on Maternal-Infant Health in Ladakh

185

and in balance with a household demographic system (primogeniture and polyandry) that reduced land ﬁssioning and kept overall population growth to sustainable levels (Crook and Crook 1994; Norberg-Hodge 1991; Osmaston and Crook 1994; Osmaston et al. 1994). Ladakhis begin to sound like Sahlins’s depiction of hunter-gatherers as the “original afﬂuent society” (Sahlins 1968), managing their subsistence and social activities so as to produce harmony within society and with the environment. In other words, cultural adaptations have been seen as approaching some kind of optimum. But there are other places to look for evidence of adaptation. Mazess (1978), and Little (1989) provide a list: diet and nutrition, growth, adult size and body composition; activity and physical ﬁtness; reproduction; disease. Their links to survival mean that they can be considered domains of adaptation, but because they are all so crucial, one cannot be evaluated fully in isolation from another. The domains of reproduction, disease, diet and nutrition, and maternal anthropometry and workloads have been explored in this research, and provide a different perspective on adaptive processes in Ladakh by unpacking that hallmark of “traditional” Ladakhi lifeways: the ﬁnely tuned household organization of reproduction and production. Household dynamics, particularly those impinging on maternal and infant health, are further analyzed in light of the reality of contemporary life in Ladakh, characterized by political and social turbulence, global economic penetration, and increasing integration within an Indian governmental infrastructure. The biocultural nature of this research in addition forces consideration of biological history (ancestry) relative to environmental stresses, the household as adaptive locus and women’s roles in the household, and the changing cultural landscape. These factors operate as interactive, dynamic processes that have resulted in particular reproductive outcomes at a particular historical moment in Ladakh. Chapter 6 demonstrated that birth outcome in Ladakh is at substantial variance from that in other high-altitude populations. What differs, especially between Ladakh and other Tibetan populations (in which behavior is very similar), is history. Ladakhis have limited ancestral depth at high altitude and are of diverse genetic origins. They have lower birthweights than both Andeans and Tibetans, including Tibetans in Ladakh, even when differences in maternal weight are controlled for. On the other hand, and consistent with other studies, Ladakhi women have slightly larger babies than do migrant women when maternal weight is taken into account. Ancestry is generally correlated with efﬁciency of response to an environmental stress; it may be that there has not been sufﬁcient time for selection of genes for increased birthweight to have evolved in the Ladakhi population. In this scenario, intrapopulation differences are in

186

An Ecology of High-Altitude Infancy

part attributable to variable genetic inheritance. Or, it is also possible, given that other genetic adaptations have been known to evolve over relatively short periods of time, that these genes exist and manifest in some individuals in Ladakh, but have met with constraints on their ability to increase fetal growth in others. In this case, intrapopulation variation is the outcome of social distinctions among women. Most likely, some combination of these scenarios explains the variation in reproductive outcome. Given the signiﬁcance of birthweight to infant mortality, it would appear that Ladakhis are imperfectly adapted to this environmental context. To put it more technically, the adaptive process has not resulted in optimal phenotypes; natural selection, as a process that plays out in part through differential mortality, is still very much at work, as is evident in the fact that the average birthweight is still associated with greater mortality than higher birthweights. Adaptations may also be running up against constraints that continue to prevent optimal outcomes. The Household: Compromise and Adaptation Although older cultural ecology models emphasized limits on population growth as an important part of adaptation, natural selection clearly would not favor population-reducing mechanisms for their own sake. On the other hand, as in the Ladakhi case, generating a sustainable householdto-landholdings ratio over generations could be adaptive relative to other options, given the constraints on arable land and water in this context (Crook and Crook 1994; Durham 1991). This ratio might be achieved by limiting household fertility through polyandry, but also through early mortality (Goldstein 1981). High infant mortality has had a substantial dampening effect on overall population growth while simultaneously forcing an increase in fertility to replace or insure against dying infants. Infant mortality rates of up to 20% increase by at least that amount the total energetic and nutrient demands that reproduction places on a woman over the course of her lifetime. This is an expensive means for curtailing population growth, especially for women. It is perhaps ironic that polyandry, combined with high neonatal mortality, dramatically reduced the reproductive costs to the relatively large pool of unmarried women while increasing reproductive costs to married women. It would be interesting to investigate how or if married women might be able to offset some of the energetic costs of production by relying on unmarried women in their households. Women play central roles in agricultural production and organization of the domestic sphere in Ladakh. During their early childbearing years, women have little authority in their husbands’ households, and the

Perspectives on Maternal-Infant Health in Ladakh

187

demands of agricultural production weigh heavily on them at the same time that they take on the burdens of reproduction. Women are caught in a compromising situation that brings into relief the very nature of adaptation. Adaptation in one domain (say, to hypoxia) does not manifest independent of other domains (subsistence demands). Work and scarce maternal resources (fat stores) constrain fetal growth, even in the presence of genes that increase the ﬂow of oxygen to the fetus. The effect is the same as the secular trend in height in industrialized societies over the past century. Height has increased, not because genes for tallness have been selected for but because nutrition and infectious disease conditions improved so as to allow the full expression of existing genes (Bogin 1998). To return to the original point, the subsistence system requires input by women to remain productive; it also requires reproductive input by women to be sustained. As a result, reproductive health is compromised while subsistence beneﬁts from energetic inputs from other household members. Organized around production and reproduction, the household provides the parameters within which Ladakhis live their lives and organize those two primary activities. It functions admirably to optimize productivity by channeling labor through a variety of mechanisms to agricultural activities. But the same cannot be said about reproduction, despite the ostensible links between the two domains. Perhaps production is perceived as more immediately and tangibly important, while reproduction is remarkably robust over a wide range of suboptimal conditions (note that even in Ladakh, where infant mortality is quite high, over 80% of children do survive). Thus households prioritize their needs, and the energy of adult females is put toward subsistence rather than reproduction. Acknowledgment of the double burden that women face manifests in widely articulated ideologies that emphasize the need for pregnant women to rest and consume high-quality foods. Indeed, such ideas have a broad distribution among societies, including the United States, where they are part of standard obstetric advice. As always, such ideals can be met only when resources are sufﬁcient. And in subsistence societies especially, resources rarely permit, so that women can neither access nutrient-dense foods nor exempt themselves from physical work. In Ladakh, the ideal is for women to rest at their natal home and consume foods specially acquired and prepared for them. This is considered especially important for primigravid women, who are very anxious about pregnancy, birth, and the transition to motherhood. However, neither increased food consumption nor reduced work was the norm for primigravid women in this study, who instead worked more than older women with established reproductive records. Thus what people say they should do is not always

188

An Ecology of High-Altitude Infancy

Photograph 7. Ache Tsewang with the produce from her garden for sale along the gutters in the main Leh bazaar.

what they must do, and articulated prescriptions for behavior are not always evidence for adaptation (or maladaptation for that matter). The fact that Ladakhis talk about valuing reproductive health does not mean that they act in ways that are consistent with expressed sentiments. It is always necessary to observe what people actually are doing. Along similar lines is the consideration of women’s status in Ladakh visa` -vis their health. Recently, demographic policy makers have promoted women’s autonomy as the crucial factor inﬂuencing maternal-child health (Singh 1998), so women’s status ought to be correlated with positive health indices. Women in Ladakh have considerable autonomy in the sense that their mobility is not strictly limited; they engage in economic enterprises, selling the produce from their home gardens in the bazaar (Photograph 7), and they participate in household decision making. In the South Asian context, they have relatively high status. But Ladakhi women are positioned to demonstrate their worth to their marital households by working hard (one of the criteria for a “good” Ladakhi wife), putting the needs of other household members before theirs, and bearing children. In the service of the household, women have considerable

Perspectives on Maternal-Infant Health in Ladakh

189

latitude, but this does not extend to personal freedom to act in what they may perceive as their own best interest. As the household is a corporate unit in Ladakh, there is inevitably tension over individual and group interests, and given the historical importance of household membership to individual survival, household goals tend to preempt individual expressions of independence. The care of infants is loosely organized within the household. As I posited earlier, the child belongs to the household, not uniquely to the mother or parents, and that status is solidiﬁed as the neonate becomes an older infant and child. Since the pioneering work of John Bowlby (1982), psychologists have emphasized the mother-infant bond as primary, an evolved mechanism to enhance an infant’s survival. The question has remained, however, whether other types of attachment, particularly multiple attachments, are adaptive (Chisholm 1996). In Ladakh, where household identity is a key part of an individual’s identity, it seems that multiple attachments are the norm, although this has not been investigated using standard psychological protocols. They facilitate the attachment of the infant to the household while downplaying the relationship with the mother, who, after all, is not part of the patrilineage (Wiley and Carlin 1999). Insofar as care is responsive and predictable, there is no reason to think that infants cannot thrive under these conditions. Moreover, this type of care and its resulting attachment form begins the socialization process by which an individual comes to understand herself as part of a household, and the rights and responsibilities deriving from that membership. Beyond these aspects of household behavior, there are other issues over which the household has little control or that it cannot effectively modulate. These include, of course, the pervasive problems of hypoxia and seasonal extreme cold. Attempts to confront the problem of seasonal extreme cold by using multiple fuel sources, reducing ventilation, and clustering together around heat sources in the winter does reduce cold stress (though the indoor temperature of Ladakhi houses in the winter is a far cry from what would be called “room” temperature in the United States). However, household heating strategies exacerbate acute respiratory problems for neonates and generate chronic respiratory compromise, especially for females, both made worse by ambient hypoxia. So, from the perspective of adaptation, we can see how response to one stress not only affects but exacerbates another stress. Impact of Social and Economic Change on the Household Women who contributed to this study were generally born in the 1960s and their life histories have occurred under notably turbulent conditions.

190

An Ecology of High-Altitude Infancy

An increasing military presence; Indian infrastructural penetration and governmental oversight; tourism; internal communal tensions over autonomy, power, and corruption; and the ongoing conﬂict between India and Pakistan over the status of Kashmir have all characterized the past thirty to forty years. Of these, the increase in nonagricultural occupations has been one of the most signiﬁcant. It has resulted in a decoupling of household membership and individual well-being through the breakup of landholdings and the proliferation of nuclear family households. Some of these are constructed on land that was previously under agricultural production. This process is most evident in Leh and its environs, but increasing numbers of families have members who do not contribute to agricultural subsistence but who earn wages at other types of jobs. As a result of this process, there have been profound alterations in household structure, while agricultural production has not changed for those households still involved in subsistence activities. Nuclear families have little additional labor to draw on from within the household, although as a concomitant effect, landholdings per household have decreased. With adult men involved in ancillary occupations in the military, government, or tourism, however, women’s workloads may have increased substantially. I was unable to ﬁnd a strong effect of male absenteeism on maternal work in the 1990 data, and household composition during a woman’s pregnancy was difﬁcult to assess; but it seems likely that some women’s workloads have expanded dramatically while others have declined as a function of both husbands and wives being employed in nonagricultural occupations. The pace of this transition has increased over the course of the 1990s, and as a result, future studies may ﬁnd a different pattern and quite possibly a greater degree of variation with regard to work during pregnancy and infant birthweight. Other sources of differentiation in health indices are emerging as a function of these social and ecological changes. Dietary changes must follow from an increase in nonsubsistence employment. Rice is already ubiquitous in the diets of people living in and around Leh or the main road and is rapidly replacing tsampa as a staple food, especially for children. Urban dwellers become increasingly dependent on imported and relatively expensive foods available in the bazaar. Dr. Ladhol lamented that women who work in Leh often subsist on a very poor diet of Indian tea (sweet milk tea), biscuits, and rice, all imported and relatively cheap; they scarcely have time to prepare or consume “proper” meals of traditional Ladakhi foods. Other imported foods – including fruits and vegetables as well as processed foods that are common throughout India – are widely available, especially in the summer. The nutritional impact on

Perspectives on Maternal-Infant Health in Ladakh

191

women and children has yet to be assessed. Probably effects are likely to be mixed; while imported foods increase dietary diversity, that effect may be tempered by decreased ﬁber intake, increased salt and sugar, and changes in other nutrients. Cavities among children and hypertension among adults are becoming more common, especially in the Leh area (Norboo and Yahya 1988). New patterns of infectious and respiratory disease have become established, and this is especially evident in Leh. More people have been attracted to the capital in search of civil service employment, which has resulted in rapid urbanization, crowding, and unsanitary conditions in Leh. In the summer the bazaar is overrun with diesel trucks that transport items from Kashmir, the Punjab, and Himachal Pradesh as the demands for imported goods have skyrocketed, especially to serve the tourist industry. Air quality in Leh is extremely poor, especially around the roads. Pollution from diesel smoke irritates infant lungs, and in an environment that is hypoxic to begin with, carbon monoxide and airborne particulate matter inhibit adequate oxygenation of the lungs. Child deaths from vehicle accidents and kerosene poisoning have also increased. The easy availability of drugs increases the risk of toxicity among children treated with medicines purchased in the bazaar, a problem repeatedly referred to by doctors. There are few public sanitation facilities in Leh, and the smell of urine and feces permeates some regions of the bazaar. The streams around Leh, where children play and women wash their clothes and vegetables, are polluted with garbage of all kinds. Drinking water is available only at a limited number of public spouts for a limited amount of hours, and long queues are common. These conditions are especially acute in the summer, when the population of Leh swells with the inﬂux of tourists and vendors, and when most of the year’s commercial and agricultural activities take place. Tourists generate waste – fecal and otherwise – that pollutes the land and water and increases the risk of exposure to infectious agents. Crowded conditions and more contact with individuals from many places increases the number and variety of pathogens, and exposure to them, whether on the bus or in the bazaar. Thus, as Leh becomes increasingly urbanized and the center of mercantile, tourist, and governmental activities, living conditions there have deteriorated in many ways, although many of the trappings of Westernization are also available to those with resources to afford them (phones, satellite dishes, etc). Pollution, dietary and infectious diseases, and social conﬂict and differentiation by economic status, occupation, religion, and living conditions are new aspects of life there. Leh has become more, for lack of a better word, Indianized, as it begins to take on aspects of any

192

An Ecology of High-Altitude Infancy

town its size in India. In other words, although Ladakh has a distinct natural ecology, its increasing participation in national and international economic and social trends has caused it to lose part of its local uniqueness and manifest conditions common to other “developing” areas. Adaptation and Reproductive Health: Concluding Remarks Returning to the theme of adaptation, two points remain. First, the adaptive process is most evident in a population when there is observable variation in reproductive outcome. We saw that there was variation that could be attributed to two main issues: differential work and energy resources of pregnant women (weight). There was also variation by ethnicity and to a limited extent by religion. In countries like India, workloads and energy resources are generally outcomes of socioeconomic stratiﬁcation, whereby poorer women have to work harder and have access to less food. In this study, landholdings did not explain why some women had to work harder than others or why some women had higher weights than others. Wealth, measured by landownership, did not predict work or weight; in fact, on average, women from households with more land worked harder than women from households with fewer ﬁelds. Nor did a woman’s education contribute signiﬁcantly to the variance in her work or weight. As the earlier discussion suggests, however, this dynamic is likely to change, as some families proﬁt from participation in new economic opportunities, while others become increasingly marginalized by the rampant inﬂation that the global economy has introduced to Ladakh. But in the early 1990s, poverty and stratiﬁcation did not, fortunately, seem to be the avenue by which women experienced differential reproductive health, or by which infants came to have different probabilities of survival. Second, the profound and accelerating changes in the social and economic landscape in Ladakh over the course of the past forty years create a very ﬂuid and, to some extent, unpredictable environmental context. While some of the reproductive problems identiﬁed here are likely to be long-standing, others result from these recent phenomena. Adaptation in the context of rapid and unpredictable change is not likely to be achieved. Instead, what we can see is the process by which people respond to novel conditions and try to meet their needs for sustenance and reproduction. It is impossible to establish a deﬁnitive set of identiﬁable, quantiﬁable adaptations; such a list would be static and would not reﬂect the current dynamic context of Ladakh. Instead, reproductive health indices reveal a number of compromises that represent attempts to balance conﬂicting challenges and evidence of the ongoing and underlying problem of

Perspectives on Maternal-Infant Health in Ladakh

193

hypoxia. As the situation in Ladakh unfolds further and the social and natural environment undergoes further modiﬁcation, new adaptive scenarios for reproductive health are likely to manifest. Intervention Now we come to the most important part of this work. What is the relevance of this project to the design of policies to enhance reproductive health and survival probabilities for neonates? Before addressing that question, though, it is important to note two contrasting perspectives on the high rate of neonatal mortality in Ladakh that would result in radically divergent health policies. The ﬁrst is that the death of 10 to 15% of infants in the ﬁrst month of life is obviously a problem in need of a solution. It places enormous energetic and emotional burdens on mothers, not to mention the suffering that precedes it among young infants as they struggle with the transition into the postnatal environment. The second perspective is that the high rate of neonatal mortality is actually a blessing in disguise: if infants die early, their suffering is not prolonged, and with age they do not face increasing threats to their survival, as occurs in other resource-poor societies. From this perspective, the high NMR is not to be lamented but perhaps accepted as the “least worst” outcome for the Ladakhi situation, especially as it may have played an important role in population regulation. Putting scarce resources into saving frail young infants may simply delay their inevitable death or contribute to unsustainable population growth. The problem with this second interpretation as a basis for designing health policy is that it assumes high infant mortality to be normative or inevitable, part and parcel of the cultural and ecological context of Ladakh (cf. Farmer 1999). Furthermore, whether investing in the survival of small neonates (almost 50% of newborns) would increase postneonatal mortality is an empirical issue, not a valid assumption. In addition, contraception is easily available in Ladakh and its use is spreading, so controlling reproduction by accepting high rates of neonatal death is entirely unjustiﬁable. Hence, I adopt the perspective that while the ultimate roots of high neonatal mortality are social, cultural, and ecological, the work of medical anthropologists should help to clarify avenues of appropriate action to improve the well-being of the populations among whom we are privileged to work. The proximate determinants analysis outlines the pathways that led to neonatal survival or death, and thereby suggests opportunities for intervention. Two qualitatively different types of intervention can be imagined. One targets speciﬁc reproductive health problems that are preventable or

194

An Ecology of High-Altitude Infancy

curable without radical social and economic change in Ladakh. For example, if septicemia is a major cause of neonatal mortality, then it can be addressed by testing women for reproductive tract pathogens (such as group B streptococcus), providing antibiotics during labor for women who test positive, and setting stricter standards for hygiene in the obstetrics ward and delivery room. All of these would require some ampliﬁcation of hospital resources. I outline some narrowly focused possibilities later; however, anthropologists are trained to think holistically, so they have tended to be critical of such targeted policies because they fail to address the larger cultural and ecological context that gives rise to any given health problem. Indeed, medical anthropology got its start in part by offering explanations for why speciﬁc health interventions failed (Paul 1955). Often the latter are culturally inappropriate, at odds with belief systems, inconsiderate of other more pressing problems or crucial behaviors, or inappropriate for ecological reasons. And infant and child health interventions sponsored by donor countries or international health organizations are subject to political and economic agendas and more general fads in health policy (Farmer 1999; Justice 1986; Scheper-Hughes and Sargent 1998). So, any sort of intervention has to be very carefully thought out, justiﬁed, and designed in consultation with those who are likely to be affected by it. The alternative approach takes a more systemic view in locating health problems. According to this view, by emphasizing the context that gives rise to disease, intervention must occur at a more systemic level and its effects will ﬁlter down to the locus of the immediate problem. The emphasis is on the broader mechanisms that ultimately act to prevent ill-health, and are similar in spirit to McKeown’s famous analysis of the decline in mortality in Europe that began in the eighteenth century (McKeown 1976). He traced this decline to dramatic increases in the overall food supply; public health interventions were secondary, and much to everyone’s surprise, clinical medicine (the use of vaccination or antibiotics) played a decidedly minor role. Many anthropologists who take this position would argue that most disease can be understood as a function of political and economic forces that conspire to bring about ill-health by, for example, limiting people’s access to food and clean water or providing them with unsafe working conditions. In her seminal book, Death without Weeping, Scheper-Hughes (1992) argued that high infant mortality in northeast Brazil could be understood proximately by both the terribly unhygienic conditions in which infants live and by child-care patterns. These factors could be understood ultimately, however, only as a result of the economic marginalization of women brought about by governmental economic development policies designed to increase production of exports

Perspectives on Maternal-Infant Health in Ladakh

195

at the expense of subsistence agriculture, which in turn forced relocation of people into peri-urban shantytowns. In this case, then, the ideal “intervention” would be sweeping social changes that include a more egalitarian social organization and infrastructural reform (e.g., public health interventions such as clean water) that act to prevent rather than cure disease. Even more local, small-scale projects (e.g., child-care cr`eches) still require political will and economic resources to be sustained. Speciﬁc biomedical interventions are only adjuncts to broader structural reform, but as the physician-anthropologist Paul Farmer (1999) notes, when these are made widely and easily accessible, they can be extremely effective. Proximate determinants analysis is useful for both speciﬁc and systemic approaches. It can point to speciﬁc determinants that are signiﬁcant in a given population, which can thereby be targeted for intervention. But it also places those determinants within a broader context and brings into relief other factors that affect each determinant and that are likely to facilitate or impede a given intervention. Proximate determinants analysis also suggests that without consideration of the more distant forces inﬂuencing a proximate determinant, a given intervention is likely to have only transient effects because it does not address the fundamental conditions that give rise to a health problem in the ﬁrst place. Targeting Neonatal Mortality in Ladakh Infant Mortality Statistics in Global Perspective The ﬁrst issue to consider in the Ladakhi situation is the unusual pattern of infant mortality. In general, in areas where infant mortality rates are high, neonatal mortality makes up less than half of infant mortality, whereas in contexts of low IMR, neonates account for the vast majority of infants who die. Table 7.1 illustrates these broad trends but also indicates that there is variation within and overlap between countries with respect to the ratio of neonatal to infant mortality. Infants in poor countries where the IMR is relatively high tend to face environments characterized by malnutrition and infectious disease, and these threats become more tangible and problematic as the infant ages and has more contacts with its environment. As a result, infants’ rates of mortality increase throughout the ﬁrst year of life. In wealthy countries where the IMR is low, infectious pathogens and malnutrition play a minor role in shaping infant mortality; this results in a pattern of low infant mortality, with the majority of deaths occurring in the ﬁrst month of life. A relatively large number of premature and critically ill newborns contributes to the high proportion of neonatal deaths. The statistics for India and Pakistan are more in line

196

An Ecology of High-Altitude Infancy

Table 7.1. International infant mortality rates (IMRs)1 and per capita wealth2

Country

GDP per capita ($US)

United States Japan Germany Netherlands France United Kingdom Italy Australia Canada average

31,746 29,956 26,183 24,956 24,739 23,934 20,659 20,125 19,642 24,660

7.2 3.6 4.9 5.2 4.8 5.9 6.6 5.3 5.5 5.4

4.8 2.0 2.9 3.8 3.1 3.9 4.9 3.6 3.9 3.7

2.5 1.7 2.0 1.3 1.7 2.0 1.7 1.7 1.9 1.8

0.67 0.56 0.59 0.73 0.65 0.66 0.74 0.68 0.71 0.66

8257 4921 2793 1941 1302 1211 894 458 422 219 2242

19.1 10.9 14.2 15.0 17.8 29.7 19.0 100.4 129.0 42 39.8

11.7 6.3 9.1 5.4 3.7 10.1 9.9 62.0 73.0 10.9 20.2

7.5 4.6 5.1 9.6 14.0 19.7 9.0 38.5 56.0 31.4 19.5

0.61 0.58 0.64 0.36 0.21 0.34 0.52 0.62 0.57 0.26 0.47

184.0

148.0

35.0

0.80

Argentina Chile Costa Rica El Salvador Morocco Egypt Philippines Pakistan India3 Tajikistan average Ladakh 1

422 (India)

IMR (0–1 year)

NMR (0–27 days)

PNMR (28–364 days)

NMR/ IMR

United Nations (2001b); 2 United Nations (2001a); 3 Jejeebhoy & Rao (1995)

with Ladakh, as all have a relatively high percentage of infant mortality occurring in the neonatal period, reﬂecting the high frequency of low birthweight in South Asian countries (de Onis et al. 1998). Despite the magnitude of infant mortality in Ladakh, which is high even relative to those countries with high IMRs,1 the percentage accounted for by neonatal mortality is more akin to countries with low IMRs. This suggests that the high rate of infant mortality in Ladakh clouds the true picture of infants at risk, and that as a statistic, the IMR for Ladakh has a different meaning than what one might expect in a poor country. Appreciation of the variation in the causes and patterns of infant mortality in different populations in different environments is essential to avoid imposing deterministic models of infant mortality onto populations manifesting statistical or apparent similarity. Only with sensitivity to the particular

Perspectives on Maternal-Infant Health in Ladakh

197

forces acting in any population and the local responses to these forces can we begin to address the idea of solutions for the multitude of infants who struggle and ultimately die in the ﬁrst year of life. As Nichter and Kendall (1991:196) noted, “In an international health setting in which ‘body counts’ and extrapolated statistics carry immense weight in health and development planning, anthropologists have not been vocal enough in questioning the meaning of health statistics and helping to construct more appropriate indicators of ill health.” International health and development policy makers create interventions based on infant mortality statistics, but it is clear that in Ladakh standard approaches designed to deal with high rates of postneonatal mortality are not likely to be relevant. At the same time, however, it might be tempting to focus interventions there very narrowly in the peri- or neonatal period, given what appears to be an acute problem during this delimited time. In considering health policy or intervention to reduce neonatal mortality in Ladakh, two complementary pathways are possible. Because of the strong link between birthweight and early mortality, the determinants of low birthweight should be considered as possible domains for intervention. In addition, since low birthweights make neonates more vulnerable to other challenges, those threats can be addressed. Obviously, both these causal factors need to be addressed, but they are interrelated; factors that may increase birthweight may also reduce neonatal morbidity. Interventions outlined in the remaining sections consider biomedical, household, and social avenues of reform, following the model depicted in Figure 7.1. Targeting Birthweight As this work has established, a primary factor contributing to the unusual pattern of infant mortality in Ladakh is likely to be hypoxia. It constrains fetal growth, producing newborns with few energy resources and a high risk of early death. Because hypoxia is inherent to this environment and constitutes a signiﬁcant “ultimate” determinant of neonatal mortality, it is difﬁcult to imagine what kind of intervention might ameliorate its negative effects. However, some neonates evidently experience more problems than others, and this study has identiﬁed them: smaller newborns are those at greatest risk. Speciﬁc interventions such as maternal and neonatal oxygen supplementation might be possible, but would require identiﬁcation of high-risk individuals, for whom this treatment would be tailored. Identiﬁcation and care of high-risk women is probably more difﬁcult than recognizing infants at risk. Routine testing for oxygen saturation among neonates, especially those of LBW, is noninvasive and

198

An Ecology of High-Altitude Infancy

easy to perform, although it requires a pulse oximeter and mechanisms for providing supplemental oxygen if necessary. It would require longer hospitalization for mothers during the last trimester of pregnancy, as well as longer postpartum stays for neonates. Currently the Leh hospital does not have such facilities, but this is a potential biomedical avenue for intervention. Note that in the state of Colorado, where low birthweights and pregnancy complications (such as PIH) are more common at high altitude, infant mortality, especially that related to low birthweight, has declined dramatically in the high-altitude regions. This is likely a function of specialized tertiary treatment centers and the availability of pre- and postnatal air transport to such centers (Unger et al. 1988). While such a suggestion might seem laughably unaffordable for Ladakh, it does indicate that biomedical institutions can have a profound effect on reducing risks to infant survival at high altitude. This study does point to certain determinants that place some neonates at greater risk of death than others because of their smaller size. Maternal workloads during pregnancy and maternal weight both predicted birthweight, which in turn predicted early mortality. Maternal work resumes quickly after pregnancy during busy months and is associated with earlier infant supplementation and reduced nursing. Though a number of studies in developing countries have shown that food supplementation for pregnant women can result in moderate increases in infant birthweight (Adair and Pollitt 1985; Prentice et al. 1987), this is not a sustainable or practical solution to a very widespread pattern. At present we do not know the extent to which maternal size is related to dietary intake or to constraints on growth and development in this environment (remember that Ladakhi women are lighter but taller than Andeans, and both lighter and shorter than Tibetans). Smaller adult size due to constraints during growth and development is an intergenerational issue that short-term supplementation would not affect. The “rest villages,” which generated quite positive effects on birthweight in Africa, are also untenable, as women’s labor is crucial to subsistence (Manshande et al. 1987). However, one might imagine that instead of training ANMs to work in the poorly maintained and supplied health outposts, pregnant women might be better served by having them as “agricultural assistants.” If pregnant women were relieved of their heavy work in the ﬁeld, they could redirect some of the energy taken up by agricultural tasks toward fetal development. Maternal work and diet are embedded in Ladakhi subsistence practices that in turn are rooted in ecological and social conditions. As such, they do not seem amenable to public health mandates to reduce work and consume more food in the service of reducing infant mortality, especially when the mediating factor is birthweight, and women are not likely to

Perspectives on Maternal-Infant Health in Ladakh

199

perceive larger newborns as inherently preferable. Moreover, these mandates already exist in traditional ideologies, yet because maternal work and food consumption is most often the result of what must or can be done, not what should be done, I am not optimistic about this kind of intervention. This conclusion leads to the question of the authority of public health ofﬁcials and messages in Ladakh. The government- and nongovernmentsponsored public health infrastructure is relatively poorly developed. Moreover, traditional institutions that would have dealt with reproductive health were notably absent in Ladakh (e.g., traditional midwives), and amchi played virtually no role in birth, leaving a vacuum that was easily ﬁlled by biomedical services. Within those services, Dr. Ladhol has become the primary source of information and the locally recognized authority on such matters; people give her advice strong consideration, as indicated in their attendance at her prenatal clinic and their hospital deliveries. It remains to be seen whether other biomedical initiatives for reproductive health could be successful in her absence. She recognizes, more than anyone else, the need for pregnant women to rest and eat well, and during prenatal visits makes use of traditional ideologies in urging women to do so; at the same time she is cognizant of the forces that preclude such changes. Dr. Ladhol is not reluctant to point the ﬁnger at other household members, especially husbands, who could alleviate some of the burdens that fall to women during pregnancy; however, they have not as yet been brought into the public health discussion aimed at reducing infant mortality. An important implication for how this has unfolded in Ladakh is that reproductive health care is going to require more resources and should, in any case, become more decentralized. Dr. Ladhol can no longer handle so many patients and take on the infant monitoring that she would like to implement. To be more effective, reform of ANM training and the funding and maintenance of villagelevel health resources is desperately needed, in consultation with village women. I suspect that a further reason for the dramatic increase in the use of prenatal services and hospital over the past twenty years (and one that may be independent of Dr. Ladhol) has been because Ladakhi women themselves perceive a close connection between maternal and infant health. The local focusing of risk within pregnancy (evident in the high rate of fetal death) and immediately following birth (neonatal death) makes prenatal services seem more compelling to women. That women take time off from other pressing demands and are willing to endure long waits to see Dr. Ladhol suggests to me that they ﬁnd these visits to be valuable, either in terms of assuaging their fears (especially among new mothers),

200

An Ecology of High-Altitude Infancy

providing a justiﬁcation for taking a day off from work, or helping ensure a positive outcome to their pregnancy. It further implies that this care is effective in dealing with forces that women previously felt were beyond their control. While a salutary effect of prenatal care on birthweight has not been demonstrated conclusively in this or other studies (Alexander and Korenbrot 1995; Fiscella 1995), at this point investments in prenatal care appear to be an appropriate use of resources, given the close connection between maternal health and neonatal survival. The use of prenatal services has likely reduced perinatal mortality, which may be why very early mortality reported in reproductive histories was lower in the 1995 sample than in those from 1990. One primary mechanism by which prenatal care may have this effect is through the identiﬁcation of individual women at high risk: those with pregnancy-induced hypertension, extreme anemia, or a history of poor birth outcomes. Prenatal care might be made more effective by assessing weight gain (which would require only a scale) to identify women at risk of delivering small newborns. Reproductive tract infections could be identiﬁed so as to reduce the risk of neonatal septicemia. The use of prenatal care as a screening device is warranted, but thus far there is no information on what happens to at-risk women or their households when they are admitted (or if they agree to be admitted) to the hospital for extended stays, particularly during busy agricultural seasons. Given its success as a preventive health service, infant health monitoring might be integrated into this clinic, a change that Dr. Ladhol has been trying to institute. Infectious Disease Another major domain of intervention could be aimed at reducing the problems that accompany respiratory disease in Ladakh. While hypoxia is the backdrop that exacerbates respiratory ailments, other conditions conspire to increase morbidity and mortality associated with them. The focus here should be on reducing levels of domestic pollution, particularly airborne particulate matter and noxious gases such as carbon monoxide. Efforts in this direction have already been initiated by the Ladakh Ecological Development Group (LEDeG). This group has focused on the development of ecologically sound and sustainable technologies appropriate for use in Ladakh. For heating they have emphasized solar sources, which are ideal given the high level of UV radiation and little cloud cover in Ladakh. Solar panels and stoves have been designed and put into use in households in Leh and numerous villages. Solar panels (trombe walls) should increase ambient household air temperature, while obviating the

Perspectives on Maternal-Infant Health in Ladakh

201

problems associated with the lack of ventilation. Other types of solar panels have been installed primarily as a source of energy for electricity or to heat water. Unfortunately, solar stoves are not widely used because cooking time is much longer than with conventional methods, and clean alternatives to kerosene are needed. The effects of these modiﬁcations on respiratory disease, particularly among women and children, and household air quality have yet to be measured. At this point gastrointestinal diseases are not major sources of mortality for neonates or infants, and data from 1990–1995 do not show any evidence of a rise in these illnesses during this period. Public health investments should be made to ensure a safe water supply. LEDeG has also promoted the traditional composting toilet as a way of reducing fecal contamination of water and recycling organic human waste. However, new hotels and houses use Western-style ﬂush toilets, and the need for compost has diminished as fewer households are involved in agricultural production. The tradition of venerating lhu (water spirits), prevented the pollution of water sources (by defecating, urinating, or littering in them) crucial to subsistence. As more Ladakhis leave farming to take up other nonsubsistence tasks, however, the perceived force of these spirits may lessen, which may help explain the rampant pollution of streams in and around Leh. Of course that is also a function of the lack of an infrastructure to dispose of garbage safely and hygienically, the importation of commodities that do not decompose quickly into inert materials, and the use of water sources by individuals without similar traditions. The Future of Reproductive Health in Ladakh As in much of the Himalayan culture area, extreme poverty and the social stratiﬁcation that generates it are not major “ultimate” causes of infant mortality. Much of the literature on infant health has emphasized the role of poverty and social and economic marginalization in creating conditions inimical to infant health and well-being (cf. Scheper-Hughes and Sargent 1998). Indeed, these are the conditions that most infants are born into in poor countries such as India. But the Ladakh data indicate that differences in wealth and social status are not signiﬁcant sources of variation in birth outcome and infant survival, at least as far as these issues can be analyzed at present. In the pessimist’s view of this scenario, even under relatively benign social conditions, health outcomes are still astonishingly poor. And given the increasing absorption of Ladakh into mainstream Indian society, increasing stratiﬁcation and impoverishment could possibly exacerbate what are already serious problems. The optimist’s view is that the situation is not so bad. True, an alarming number of infants die,

202

An Ecology of High-Altitude Infancy

but they die quickly; and for the 85 to 90% of infants who make it past one month, health is relatively good, and they do not face the myriad infectious and nutritional diseases common to infants and toddlers in other places. Regardless, for now Ladakhi mothers and neonates face a unique set of ecological and sociocultural circumstances; some of these protect their health while others place them at increased risk relative to other mothers and infants around the world. As we have seen, some of these are potential targets for intervention; others are more intractable. Social differentiation has become part of recent history, and tourism has played a major role. Tourism not only adds to the burdens placed on the local infrastructure and work demands during the peak summer season, which is also the time of intense work and activity among agricultural Ladakhis, but it also has served to differentiate households more distinctly into those that have access to an extra source of cash income and those who do not. It also exacerbates inﬂation, and prices have skyrocketed in recent years. Along with military and civil service employment, tourism represents an avenue to acquiring commercial goods to which the rest of the population has restricted access. This means that involvement in the tourism economy has come to determine to a certain extent the access people have to dietary and health resources, and it may ultimately manifest in marked stratiﬁcation in reproductive health indices. Currently Ladakh is experiencing immense political instability because of the ongoing escalation of conﬂict between India and Pakistan over the status of Kashmir, the state to which Ladakh belongs. Uncertainty about the future intensiﬁes and subsides in parallel with military skirmishes in Ladakh. The effects are manifold. As India seeks to strengthen its dominion over Ladakh as a way of safeguarding its northern border, it pulls Ladakhis into occupations deﬁned by national interest (civil service, military). Ladakhis are taking advantage of these opportunities and adjusting their behavior to ﬁt Indian norms. At the same time, resources may be diverted to serve that cause instead of being devoted to the welfare of the local populace. While I have painted a somewhat bleak picture of the likely effects on infant survival, there are positive aspects as well. For example, fertility has been declining in Ladakh, largely as an effect of increasing contraceptive usage. Ladakhi women are eager to use birth control, and Dr. Ladhol is a champion of family planning. Fertility decline is associated with a growing sense that children are expensive and can no longer be relied on as a source of labor for a household. Meanwhile agricultural landholdings are declining, so children are not needed for labor. This is a complex process that I detail elsewhere (Wiley 1999), but it certainly has relevance for reproductive health insofar as it provides women with increasing control

Perspectives on Maternal-Infant Health in Ladakh

203

over their reproductive lives. However, with no improvement in infant survival, family planning goals will be achieved for many women only after substantial reproductive wastage. So, the reproductive health situation in Ladakh is very much in ﬂux at the moment, reﬂecting the major changes that Ladakhi society is undergoing at present. The ways in which household production affects reproductive health are undergoing profound changes, and we can expect that birth outcomes will change as well. It remains to be seen whether these changes will be for the better or worse. The burden of hypoxia will remain, but its effects on birth outcome may be ameliorated for some. Seasonality may decline in importance as fewer households are organized around the seasonal agricultural cycle and as household environments buffer changes in temperature without the concommitant rise in indoor pollution. Women’s diet and activity patterns are clearly changing as a result of new opportunities outside agriculture. New public health initiatives will be proposed and possibly carried out. The hope is that these will be locally relevant and cognizant of the context that generates reproductive health challenges in Ladakh, rather than driven by mandates from New Delhi or beyond. For the moment, however, reproductive health remains compromised by the natural environment and the prevailing social and cultural conditions that have privileged household productive efforts over reproductive health.

Notes

CHAPTER 1 1. Much of the following material outlining deﬁnitions used in the biocultural approach is derived from a previous paper in Medical Anthropology Quarterly, “Adaptation and the biocultural paradigm in medical anthropology: a critical review” (Wiley 1992a). 2. Life expectancy is deﬁned as the average number of years a person in a given population can expect to live. CHAPTER 2 1. There is a large literature on the biological effects of acute exposure to extreme hypoxia that is mainly of relevance to mountain climbers; it will not be reviewed here. 2. Very large newborns also suffer from higher rates of mortality, usually because they are more likely to suffer trauma during birth or because their large size results from gestational diabetes. 3. Conlisk (1987) compared premature and full-term newborns at high and low altitudes in Bolivia and found that premature infants at high altitude retained the same perinatal survival advantage at high altitude as their full-term counterparts. CHAPTER 3 1. The borders of Ladakh look very different depending on the nation doing the cartography. India continues to claim a large part of northwestern Ladakh that is effectively held by Pakistan and has been since 1948. India also lost a sizable portion of northeastern and eastern Ladakh to the Chinese during a war in 1962, although this area is still labeled as “India” on Indian maps. 2. Many authors suggest that Ladakh lacks the water resources necessary to support a large population. However, it appears that the constraining factor is not lack of water per se, but access to it, which in turn is a function of the amount of labor that people are willing and able to commit to expanding or intensifying their water usage through irrigation. More land could be brought under cultivation but only with substantial communal labor inputs, and thus it seems that the Ladakhis have struck a shifting balance between their population size and their resources, mediated by the amount of labor they can allocate to building 205

206

3.

4.

5.

6.

7.

Notes to Pages 41–81

more extensive irrigation networks, some of which already extend for miles, and all of which must be built by hand. Himalayan sites in Pakistan and Tibet have yielded surface ﬁnds of stone tools that may be up to 25,000 years old (the Upper Paleolithic) (Dennell 1988; Zhiman 1982). Such sites are too distant from Ladakh to be of use in determining the antiquity of Ladakhi habitation. The only chronometric date for a settlement site on the Tibetan plateau is less than 7,000 years bp (Chang 1986). This is not to say that extramarital relationships are culturally sanctioned, but based on the gossip I was often party to, it seems to me that they occurred with some frequency and with tacit knowledge. There seem to be numerous ways in which individual autonomy is subsumed by household membership in Ladakh. This is evident in personal names; individuals sharing a lineage are identiﬁed by a house name, and individual names are a combination of two of the limited supply of Tibetan-derived names. With a few exceptions, these names are non-sex speciﬁc, but are religion speciﬁc, so that Muslim, Buddhist, and Christian Ladakhis can be immediately differentiated by their personal names. Though Ladakhis referred to these cows as “Jersey cows,” they were not the fawn-colored cow known in the United States and Europe. These cows are a variant of Indian brahma cows. The fact that I was, at the time, unmarried and childless did not seem to constrain my inquiries, in part because women of that status are not uncommon in Ladakh. Indeed, the obstetrician herself was unmarried and childless. My association with her certainly enhanced the perception that I was to be trusted when examining women and their newborns and probing the details of their reproductive histories despite my lack of personal experience.

CHAPTER 4 1. Although paternal biological factors inﬂuence the size of the fetus, their significance is slight in comparison with maternal contributions, since the mother contributes not only genes, but the entire fetal environment. Haig (1993) has suggested that paternal genes may act to increase resource ﬂow to the fetus, while maternal genes counter such efforts, as they are likely to come at high cost to the mother’s health. This has been termed “maternal-fetal conﬂict.” 2. The procedures and speciﬁc analysis of each of these measures is described more completely in Wiley (1994). 3. Because of the political instability that continued in Ladakh during 1991, the year when the decennial Indian census was conducted, there are no census data available for that year. 4. Although births are supposed to be registered by village authorities, the most reliable birth records come from the hospital. The annual number of “ofﬁcial” births enumerated for the early 1980s ranged between 370 and 540 (Directorate of Economics & Statistics Planning and Development Department, 1987), yielding a birth rate of 7/1000, which is far too low.

Notes to Pages 84–127

207

5. For a full description of the ethnic Ladakhis only, see Tables 4.6 and 6.1. 6. When this was done using birthweights <10th percentile, very similar results were found: 22% were wasted, while 78% were stunted. 7. In highland Peru, Thomas (1976) found that households employed child labor and engaged in sedentary activities as much as possible to reduce adult energy expenditure in a hypoxic context. 8. Chin (1992) found a similar situation in a survey of three Tibetan villages, where variation in child health parameters could not be traced to socioeconomic differences between households. 9. At present, most of the research in Ladakh has been done on Buddhists and Buddhist institutions. Muslims are not well represented in the published literature on Ladakh. 10. The obstetrician explained to me that since 75% of all fetuses are carried high on the left side, there is no way that this could be true, a statement she shared with the amchi at a recent conference on health in Ladakh.

CHAPTER 5 1. During the summer, habitation tends to be ﬂuid, especially among women, who often head up into the mountains to graze animals or relocate to other households in need of their labor. 2. I also conducted this analysis using logistic regression, which can be used only on the portion of the sample whose survival characteristics at each age (i.e., 1 month, 6 months, 12 months) are known. The two procedures yielded very similar estimates, and they were highly correlated (Pearson’s r = .90 − .96). Logistic regression tended to overestimate mortality in this sample because only those infants who were followed for the entire time under consideration could be used, and they disproportionately represent infants who died. 3. The infant mortality rate (IMR) is traditionally calculated as the number of deaths in a time period (usually 1 calendar year) per 1000 live births. This is a peculiarity of the IMR, as the deaths are not necessarily drawn from the same population of births. That is, babies born at the end of the time period (say, in December) do not experience a full year of risk; or, deaths that occur in January may be of infants who were born in the previous year. In both cases, the deaths and the births can constitute different populations. In the data presented here, this is not the case, as the deaths are drawn from a cohort of births; therefore, the numerator and the demoninator are the same population. This generates a cohort death rate, or n qx , where n = number of days and x = starting age. These rates are expressed in deaths/1000 by convention. 4. How exactly birthweight constitutes a risk factor for morbidity and mortality is not known with any certainty (Wilcox 2001). 5. Clostridium tetanus spores have been isolated by microbiological examination in local dung. Expression of the disease is rare, and Ladakhis seem to have a high degree of resistance to tetanus and have high levels of circulating antibodies (Norboo and Yahya 1988).

208

Notes to Pages 132–196

6. I suspect that this is likely to change dramatically as women in Ladakh increasingly use birth control to limit their family size and as nuclear family households proliferate with nonagricultural opportunities for employment. 7. This is similar to Vitzthum’s observations in the Andes, where she reported that lactating women returned to menses at 10.2 months (Vitzthum et al. 2000a), although in an earlier report (Vitzthum 1989) she noted that this varied from 8.8 months among wealthy women to 21.6 months among poorer women. CHAPTER 6 1. Haas’s (1980) equation for birthweight at high altitude in Bolivia is this: BWT = (36.25∗EGA(days)) + 154.56∗ethnicity (0 = nonnative; 1 = native) + (107.43 ∗(0 = primiparous; 1 = multiparous)) + (11.33∗Maternal triceps) – (44.85∗Maternal hemoglobin) – 6536.38. 2. The Andean data are birthweight-speciﬁc mortality rates derived from birth records for one week and one year, while those from Ladakh are birthweightspeciﬁc mortality probabilities for one week and one year derived from hazard analysis (remembering that mortality up to one year largely reﬂected high mortality in the ﬁrst month). CHAPTER 7 1. Of course Ladakh is not a country, but a small region. In any given country there are likely to be regions or populations with mortality that is much higher or lower than the mean. So the comparison between the IMR statistics for Ladakh and national level statistics is not entirely straightforward.

References

Abelson, Andrew E., Baker, Thelma S., and Baker, Paul T. 1974. Altitude, migration, and fertility in the Andes. Social Biology 21: 12–27. Adair, Linda S. 1987. Nutrition in the reproductive years. In Johnston, Francis E. (ed.), Nutritional Anthropology. New York: Alan R. Liss, Inc., pp. 119– 154. Adair, Linda S., and Pollitt, Ernesto 1985. Outcome of maternal nutrition supplementation: A comprehensive review of the Bacon Chow study. American Journal of Clinical Nutrition 41: 948–978. Agarwal, D. K., and Agarwal, K. N. 1987. Early childhood mortality in Bihar and Uttar Pradesh. Indian Pediatrics 24: 627–632. Ainsworth, Mary D. Salter. 1965. Further research into the adverse effects of maternal deprivation. In Bowlby, John (ed.), Child Care and the Growth of Love. Baltimore: Penguin Books, pp. 205–211. Airede, A. I. 1992. Neonatal septicaemia in an African city of high altitude. Journal of Tropical Pediatrics 38: 189–191. Albalak, Rachel. 1997. Cultural Practices and Exposure to Particulate Pollution from Indoor Biomass Cooking: Effects on Respiratory Health and Nutritional Status among the Aymara of the Bolivian Highlands. PhD dissertation, University of Michigan. Alexander, G. R., and Korenbrot, C. C. 1995. The role of prenatal care in preventing low birth weight. The Future of Children 5: 103–120. Appadurai, Arjun. 1986. Theory in anthropology: Center and periphery. Comparative Studies in Society and History 28: 356–361. Armelagos, George J., Goodman, Alan, and Jacobs, Kenneth H. 1978. The ecological perspective in disease. In Logan, Michael H., and Hunt, Edward E. (eds.), Health and the Human Condition: Perspectives on Medical Anthropology. North Scituate, MA: Duxbury Press, pp. 71–84. Ashworth, A. 1998. Effects of intrauterine growth retardation on mortality and morbidity in infants and young children. European Journal of Clinical Nutrition 52: S34–S42. Attenborough, Robert. 1994. The population of Stongde, Zangskar. In Crook, John, and Osmaston, Henry (eds.), Himalayan Buddhist Villages: Environment, Resources, Society, and Religious Life in Zangskar. Bristol, UK: University of Bristol, pp. 295–330. Attenborough, Robert, Attenborough, M., and Leeds, A. R. 1994. Nutrition in Stongde. In Crook, John, and Osmaston, Henry (eds.), Himalayan Buddhist 209

210

References

Villages: Environment, Resources, Society, and Religious Life in Zangskar. Bristol, UK: University of Bristol, pp. 383–404. Aziz, Barbara Nimri. 1978. Tibetan Frontier Families: Reﬂections on Three Generations from D’ing-ri. Durham, NC: Carolina Academic Press. Baer, Hans A., Singer, Merrill, and Johnsen, John H. 1986. Introduction: Toward a critical medical anthropology. Social Science & Medicine 23: 95–98. Baker, Paul T., ed. 1978. The Biology of High-altitude Peoples. New York: Cambridge University Press. Baker, Paul T., and Dutt, J. S. 1972. Demographic variables as measures of biological adaptation: A case study of high altitude populations. In Harrison, G. A., and Boyce, A. J. (eds.), The Structure of Human Populations. Oxford, UK: Clarendon Press, pp. 352–378. Baker, Paul T., and Little, Michael A., eds. 1976. Man in the Andes: A Multidisciplinary Study of High-Altitude Quechua. Stroudsburg, PA: Dowden, Hutchinson & Ross. Barker, David J. P. 1995. Fetal origins of coronary heart disease. British Medical Journal 311: 171–174. Beall, Cynthia M. 1976. The effects of high altitude on growth, morbidity, and mortality of Peruvian infants. PhD Dissertation, Pennsylvania State University. Beall, Cynthia M. 1981. Optimal birth weights in Peruvian populations at high and low altitudes. American Journal of Physical Anthropology 56: 209–216. Beall, Cynthia M. 2000. Tibetan and Andean patterns of adaptation to highaltitude hypoxia. Human Biology 72: 201–228. Beall, Cynthia M., Almasy, Laura A., Blangero, John, Williams-Blangero, Sarah, Brittenham, Gary M., Strohl, Kingman P., Decker, Michael J., Vargas, Enrique, Villena, Mercedes, Soria, Rudy, Alarcon, Ana Maria, and Gonzales, Cristina. 1999. Percent of oxygen saturation of arterial hemoglobin among Bolivian Aymara at 3,900–4,000 m. American Journal of Physical Anthropology 108: 41–51. Beall, Cynthia M., Blangero, John, Williams-Blangero, Sarah, and Goldstein, Melvyn C. 1994. Major gene for percent oxygen saturation of arterial hemoglobin in Tibetan highlanders. American Journal of Physical Anthropology 95: 271–276. Beall, Cynthia M., Strohl, Kingman P., Blangero, John, Williams-Blangero, Sarah, Almasy, Laura A., Decker, Michael J., Worthman, Carol M., Goldstein, Melvyn C., Vargas, Enrique, Villena, Mercedes, Soria, Rudy, Alarcon, Ana Maria, and Gonzales, Cristina. 1997. Ventilation and hypoxic ventilatory response of Tibetan and Aymara high altitude natives. American Journal of Physical Anthropology 104: 427–447. Belsky, Jay. 1997. Attachment, mating, and parenting: An evolutionary interpretation. Human Nature 8: 361–381. Berman, P., Kendall, C., and Bhattacharyya, K. 1994. The household production of health: Integrating social science perspectives on micro-level health determinants. Social Science & Medicine 38: 205–215. Berti, Peter R., Leonard, William R., and Berti, Wilma J. 1998. Stunting in an Andean community: Prevalence and etiology. American Journal of Human Biology 10: 229–240.

References

211

Bogin, Barry. 1998. The tall and the short of it. Discover 19: 40–44. Bongaarts, J., and Potter, R. G. 1983, Fertility, Biology, and Behavior: An Analysis of the Proximate Determinants. New York: Academic Press. Bongaarts, John. 1978. A framework for analyzing the proximate determinants of fertility. Population and Development Review 4: 105–132. Bowlby, John. 1982. Attachment and Loss: Vol. 1. Attachment. New York: Basic Books. Braun, Peter, Lee, Yhu-Hsiung, and O’Klein, Jerome, et al. 1971. Birth weight and genital mycoplasmas in pregnancy. New England Journal of Medicine 284: 167–171. Buck, Alfred A., Sasaki, Tom T., and Anderson, Robert I. 1968. Health and Disease in Four Peruvian Villages: Contrasts in Epidemiology. Baltimore, MD: Johns Hopkins University Press. Buth, G. M., and Navchoo, Irshad A. 1988. Ethnobotany of Ladakh (India): Plants used in health care. Journal of Ethnobiology 8: 185–194. Campbell, Kenneth L., and Wood, James W. 1988. Fertility in traditional societies. In Diggory, Peter, Potts, Malcolm, and Teper, Sue (eds.), Natural Human Fertility: Social and Biological Determinants. London: Macmillan Press, pp. 39–69. Carey, James W. 1990. Social system effects on local level morbidity and adaptation in the rural Peruvian Andes. Medical Anthropology Quarterly 4: 266–295. Carrasco, Pedro. 1959. Land and Polity in Tibet. Seattle: University of Washington Press. Caulﬁeld, L. E. 1991. Birth weight as an indicator of fetal nutritional status. In Himes, J. H. (ed.), Anthropometric Assessment of Nutritional Status. New York: Wiley-Liss, pp. 259–271. Chang, K.-C. 1986. The Archeology of Ancient China. New Haven, CT: Yale University Press. Chase, Helen C. 1969. Infant mortality and weight at birth: 1960 United States birth cohort. American Journal of Public Health 59: 1618–1628. Chattopadhya, D., Chatterjee, R., Anand, V. K., Kumari, S., and Patwari, A. K. 1992. Lower respiratory tract infection in hospitalized children due to respiratory syncytial (RS) virus during a suspected epidemic period of RS virus in Delhi. Journal of Tropical Pediatrics 38: 68–73. Chen, Lincoln C. 1983. Child survival: Levels, trends, and determinants. In Bulatao, Rodolfo, and Lee, Ronald D. (eds.), Determinants of Fertility in Developing Countries. New York: Academic Press, pp. 199–232. Chen, Lincoln C., Huq, Emdadul, and D’Souza, Stan. 1981. Sex bias in the family allocation of food and health care in rural Bangladesh. Population and Development Review 7: 55–70. Chin, Nancy Perini. 1992. Child growth in three Tibetan villages: An assessment of environmental, biological, and social determinants. Master’s Thesis, University of Rochester. Chisholm, J. 1993. Death, Hope, and Sex: Life-history theory and the development of reproductive strategies. Current Anthropology 34: 1–24. Chisholm, James S. 1996. The evolutionary ecology of attachment organization. Human Nature 7: 1–38.

212

References

Clegg, E. J. 1978. Fertility and early growth. In Baker, P. T. (ed.), The Biology of High-altitude Peoples. New York: Cambridge University Press, pp. 65– 115. Clifford, Terry. 1984. Tibetan Buddhist Medicine and Psychiatry: The Diamond Healing. York Beach, ME: Samuel Weiser. Cogswell, Mary E., and Yip, Ray, 1995. The inﬂuence of fetal and maternal factors on the distribution of birthweight. Seminars in Perinatology 19: 222– 240. Conlisk, Elizabeth Ann. 1987. The effect of high altitude on birthweight and mortality in Bolivia, Master’s Thesis, Cornell University. Crook, John, and Crook, Stamati. 1994. Explaining Tibetan polyandry: Sociocultural, demographic and biological perspectives. In Crook, John, and Osmaston, Henry (eds.), Himalayan Buddhist Villages: Environment, Resources, Society, and Religious Life in Zangskar. Bristol, UK: University of Bristol, pp. 735–786. Crook, John H., and Shakya, T. 1983. Six families of Leh. In Kantowsky, Detlef, and Sander, Reinhard (eds.), Recent Research on Ladakh. London: Weltforum Verlag, pp. 195–216. Cunningham, Alexander. 1854. Lad´ak: Physical, Statistical, and Historical. London: Wm. H. Allen. Cunningham, Allan S. 1995. Breastfeeding: Adaptive behavior for child health and longevity. In Stuart-Macadam, Patricia, and Dettwyler, Katherine A. (eds.), Breastfeeding: Biocultural Perspectives. Hawthorne, NY: Aldine de Gruyter, pp. 243–264. Cvejic, Elizabeth, Ades, Steven, Flexer, William, and Gray-Donald, Katherine. 1997. Breastfeeding practices and nutritional status of children at high altitude in Ladakh. Journal of Tropical Pediatrics 43: 376. Das Gupta, Monica. 1987. Selective discrimination against female children in India. Population and Development Review 13: 77–100. Das Gupta, Monica. 1995. Life course perspectives on women’s autonomy and health outcomes. American Anthropologist 97: 481–491. De Meer, Kees. 1988. Mortality in children among the Aymara Indians of southern Peru. Social Science & Medicine 26: 253–258. De Meer, Kees, Bergman, Roland, and Kusner, John S. 1993. Socio-cultural determinants of child mortality in southern Peru, including some methodological considerations. Social Science & Medicine 36: 317–331. de Onis, M., Blossner, M., and Villar, J. 1998. Levels and patterns of intrauterine growth retardation in developing countries. European Journal of Clinical Nutrition 52: S5–S15. Dennell, R. W., Rendell, H. M., and Hailwood, E. 1988. Late Pliocene artefacts from northern Pakistan. Current Anthropology 29: 495–498. Desowitz, Robert S. 1981. New Guinea Tapeworms and Jewish Grandmothers. New York: W. W. Norton. DeSweemer, Cecile, Kielmann, Arnfried A., and Parker, Robert L. 1983. Indicators of nutrition risk. In Kielmann, Arnfried A. et al. (eds.), Child and Maternal Health Services in Rural India: The Narangwal Experiment. Baltimore: Johns Hopkins University Press, pp. 126–156.

References

213

Dhonden, Yeshi. 1980a. Childbirth in Tibetan medicine. Tibetan Medicine 1: 36–40. Dhonden, Yeshi. 1980b. Embryology in Tibetan medicine. Tibetan Medicine 1: 43–48. Diamond, Jared. 1991. Curse and blessing of the ghetto. Discover, March: 60–65. Directorate of Economics & Statistics Planning and Development Department. 1987. Statistical Handbook, District: Leh 1985–1986. Srinagar, India: Government Press. Durham, William. 1991. Coevolution: Genes, Culture and Human Diversity. Stanford: Stanford University Press. Dutt, James S. 1980. Altitude and fertility: The confounding effect of childhood mortality – a Bolivian example. Social Biology 27: 101–114. Edmonston, B., and Andes, N. 1983. Community variations in infant and child mortality in Peru. Journal of Epidemiology and Community Health 37: 121– 126. Elford, Jonathan. 1994. Kumik: A demographic proﬁle. In Crook, John, and Osmaston, Henry (eds.), Himalayan Buddhist Villages: Environment, Resources, Society, and Religious Life in Zangskar. Bristol, UK: University of Bristol, pp. 331–360. Ellison, Peter T. 1994. Advances in Reproductive Ecology. Annual Review of Anthropology 23: 255–275. Ellison, Peter T. 1995. Breastfeeding, fertility, and maternal condition. In StuartMacadam, Patricia, and Dettwyler, Katherine A. (eds.), Breastfeeding: Biocultural Perspectives. New York: Aldine de Gruyter, pp. 305–345. Escudero, F., Gonzales, G. F., and Gonez, C. 1996. Hormone proﬁle during the menstrual cycle at high altitude. International Journal of Obstetrics and Gynecology 55: 49–58. Farmer, Paul. 1999. Infections and Inequalities: The Modern Plagues. Berkeley: University of California Press. Fiscella. K. 1995. Does prenatal care improve birth outcomes? A critical review. Obstetrics and Gynecology 85: 468–479. Fisher, Margaret W., Rose, Leo E., and Huttenback, Robert A. 1963. Himalayan Battleground: Sino-Indian Rivalry in Ladakh. London: Pall Mall Press. Frappell, P. B., Leon-Velarde, F., Aguero, L., and Mortola, J. P. 1998. Response to cooling temperature in infants born at an altitude of 4,330 meters. American Journal of Respiratory Critical Care Medicine 158: 1751–1756. Frisancho, A. Roberto. 1993. Human Adaptation and Accommodation. Ann Arbor: University of Michigan Press. Frisancho, A. Roberto, and Baker, Paul T. 1970. Altitude and growth: A study of the patterns of physical growth of a high-altitude Peruvian Quechua population. American Journal of Physical Anthropology 32: 279–292. Frisancho, A. Roberto, Frisancho, H. G., Albalak, R., Villena, M., Vargas, E., and Soria, R. 1997. Developmental, genetic, and environmental components of lung volumes at high altitude. American Journal of Human Biology 9: 191– 203. Gabriel, Sherif E., Brigman, Kristen N., Koller, Beverly H., Boucher, Richard C., and Stutts, M. Jackson. 1994. Cystic ﬁbrosis heterozygote resistance

214

References

to cholera toxin in the cystic ﬁbrosis mouse model. Science 266: 107– 109. Geetha, T., Chenoy, Rashna, Stevens, D., and Johanson, R. B. 1995. A multicentre study of perinatal mortality in Nepal. Paediatric and Perinatal Epidemiology 9: 74–89. Goldstein, Melvyn C. 1971. Stratiﬁcation, polyandry, and family structure in central Tibet. Southwestern Journal of Anthropology 27: 64–74. Goldstein, Melvyn C. 1976. Fraternal polyandry and fertility in a high Himalayan valley in northwest Nepal. Human Ecology 4: 223–233. Goldstein, Melvyn C. 1981. New perspectives on Tibetan fertility and population decline. American Ethnologist 8: 721–738. Goldstein, Melvyn C., Tsarong, Paljor, and Beall, Cynthia M. 1983. High altitude hypoxia, culture, and human fecundity/fertility: A comparative study. American Anthropologist 85: 28–49. Goodman, Alan H., and Leatherman, Thomas L., eds. 1998a. Building a New Biocultural Synthesis: Political-Economic Perspectives on Human Biology. Ann Arbor: University of Michigan Press. Goodman, Alan H., and Leatherman, Thomas L. 1998b. Traversing the chasm between biology and culture: An introduction. In Goodman, Alan H., and Leatherman, Thomas L. (eds.), Building a New Biocultural Synthesis: PoliticalEconomic Perspectives on Human Biology. Ann Arbor: University of Michigan Press, pp. 3–41. Government of India. 1961a. Census of India 1961, Vol. VI, Jammu and Kashmir, Part IX, Census Atlas. New Delhi: Government of India Press. Government of India. 1961b. Jammu and Kashmir District Census Handbook 4: Ladakh District. New Delhi: Government of India. Government of India. 1983. Census of India 1981. Series 8. Jammu and Kashmir. Part IVA, Social and Cultural Tables. New Delhi: Government of India Press. Grahn, D., and Kratchman, J. 1963. Variation in neonatal death rate and birth weight in the United States and possible relations to environmental radiation, geology and altitude. American Journal of Human Genetics 15: 329–352. Greene, Lawrence S. 1977, Hyperendemic goiter, cretinism, and social organization in highland Ecuador. In Greene, Lawrence S. (ed.), Malnutrition, Behavior, and Social Organization. New York: Academic Press, pp. 55–94. Greene, Lawrence S. 1980. Social and biological predictors of physical growth and neurological development in an area where iodine and protein-energy malnutrition are endemic. In Greene, Lawrence S., and Johnston, F. (eds.), Social and Biological Predictors of Nutritional Status, Physical Growth, and Neurological Development. New York: Academic Press, pp. 223–256. Greksa, Larry P., Spielvogel, H., and Caceres, E. 1985. Effect of altitude on the physical growth of upper-class children of European ancestry. Annals of Human Biology 12: 225–232. Gupta, Ranjan, Basu, Amitabha, Pawson, I. G., Bharati, Premananda, Mukhopadhyay, Barun, Mukhopadhyay, Susmita, Roy, Subrata K., Majumder, Partha P., Bhattacharya, Sanat K., Bhattacharya, Kalyan K., and Das, Swapan K. 1989. Altitude and human biology: A comparative study of Himalayan, Andean and Ethiopian data. In Basu, Amitabha, and Gupta,

References

215

Ranjan (eds.), Human Biology of Asian Highland Populations in the Global Context. Calcutta, India: The Indian Anthropological Society, pp. 1–80. Gutschow, Kim Irmgard. 1998. An economy of merit: Women and Buddhist monasticism in Zangskar, northwest India, PhD Dissertation, Harvard University. Haas, J. D., Moreno-Black, G., Frongillo, E. A., Pabon, J., Pareja, G., Ybarnegaray, J., and Hurtado, L. 1982. Altitude and infant growth in Bolivia: A longitudinal study. American Journal of Physical Anthropology 59: 251– 262. Haas, Jere D. 1980, Maternal adaptation and fetal growth at high altitude in Bolivia. In Greene, L. S., and Johnston, F. (eds.), Social and biological predictors of nutritional status, physical growth, and neurological development. New York: Academic Press, pp. 257–290. Haas, Jere D. 1981. Maternal-fetal responses to pregnancy as indicators of human adaptability at high altitude. In L’homme et son Environnement a` Haute Altitude. Paris: Seminaire CNRS/NSF. Haas, Jere D. 1990. Mortality and morbidity consequences of variation in early childhood growth. In Swedlund, Alan C., and Armelagos, George J. (eds.), Disease in Populations in Transition: Anthropological and Epidemiological Perspectives. New York: Bergin and Garvey, pp. 223–247. Haas, Jere D., Baker, Paul T., and Hunt, Edward E., Jr. 1977. The effects of high altitude on body size and composition of the newborn infant in southern Peru. Human Biology 49: 611–628. Haas, Jere D., Balcazar, Hector, and Caulﬁeld, Laura. 1987. Variation in early neonatal mortality for different types of fetal growth retardation. American Journal of Physical Anthropology 73: 467–473. Haas, Jere D., Frongillo, Edward A., Jr., Stepick, Carol D., Beard, John. L., and Hurtado, G. Luis. 1980. Altitude, ethnic and sex difference in birth weight and length in Bolivia. Human Biology 52: 459–477. Habicht, Jean-Pierre, Lechtig, Aaron, ´ Yarbrough, Charles, and Klein, Robert E. 1974. Maternal nutrition, birth weight and infant mortality. Size at Birth Ciba Foundation Symposium No 27 (NS). New York: American Elsevier, pp. 353–370. Haig, David. 1993. Genetic conﬂicts in human pregnancy. Quarterly Review of Biology 68: 495–532. Hare, E. H., Moran, P. A., and Macfarlane, A. 1981. The changing seasonality of infant deaths in England and Wales 1912–78 and its relation to seasonal temperature. Journal of Epidemiology and Community Health 35: 77–82. Harkness, S., and Super, C. M. 1994. The developmental niche: A theoretical framework for analyzing the household production of health. Social Science & Medicine 38: 217–226. Harrison, G. Ainsworth. 1966. Human adaptability with reference to the IBP proposals for high-altitude research. In Baker, Paul T., and Weiner, J. S. (eds.), The Biology of Human Adaptability. Oxford, UK: Clarendon Press, pp. 509–519. Harrison, G. Ainsworth. 1997. The role of the Human Adaptability International Biological Programme in the development of human population biology. In Ulijaszek, Stanley J., and Huss-Ashmore, Rebecca A. (eds.), Human

216

References

Adaptability Past, Present, and Future. New York: Oxford University Press, pp. 17–25. Hinde, Robert A. 1982. Attachment: Some conceptual and biological issues. In Parkes, Colin Murray, and Stevenson-Hinde, Joan (eds.), The Place of Attachment in Human Behavior. New York: Basic Books, pp. 60–76. Hirve, Siddhi S., and Ganatra, Bela R. 1994. Determinants of low birth weight: A community based prospective cohort study. Indian Pediatrics 31: 1221–1225. Hoff, Charles J., and Abelson, Andrew E. 1976. Fertility. In Baker, Paul T., and Little, Michael A. (eds.), Man in the Andes: A Multidisciplinary Study of High Altitude Quechua. Stroudsburg, PA: Dowden, Hutchinson and Ross, pp. 128–146. Hrdy, Sarah Blaffer. 1999. Mother Nature: A History of Mothers, Infants, and Natural Selection. New York: Pantheon Books. Huss-Ashmore, R., and Thomas, R. B. 1988. A framework for analysing uncertainty in highland areas. In de Garine, I., and Harrison, G. A. (eds.), Coping with Uncertainty in Food Supply. Oxford: Clarendon Press, pp. 452–468. Huss-Ashmore, Rebecca, and Thomas, R. Brooke. 1997. The future of human adaptability research. In Ulijaszek, Stanley J., and Huss-Ashmore, Rebecca A. (eds.), Human Adaptability Past, Present, and Future. New York: Oxford University Press, pp. 295–319. Jeffery, Patricia, Jeffery, Roger, and Lyon, Andrew. 1988. Labour Pains and Labour Power: Women and Childbearing in India. London: Zed Books. Jejeebhoy, Shireen J., and Rao, Saumya Rama. 1995. Unsafe motherhood: A review of reproductive health. In Das Gupta, Monica, Chen, Lincoln C., and Krishnan, T. N. (eds.), Women’s Health in India: Risk and Vulnerability. New Delhi: Oxford University Press, pp. 122–152. Jensen, Gwenn M., and Moore, Lorna G. 1997. The effect of high altitude and other risk factors on birthweight: Independent or interactive effects? American Journal of Public Health 87: 1003–1007. Justice, Judith. 1986. Policies, Plans, and People: Foreign Aid and Health Development. Berkeley: University of California Press. Kashiwazaki, H., Dejima, Y., Orias-Rivera, J., and Coward, W. A. 1995. Energy expenditure determined by the doubly labeled water method in Bolivian Aymara living in a high altitude agropastoral community. American Journal of Clinical Nutrition 62: 901–910. Kielmann, Arnfried A., DeSweemer, Cecile, Parker, Robert L., and Taylor, Carl E. 1983. Analysis of morbidity and mortality. In Kielmann, Arnfried A., et al. (eds.), Child and Maternal Health Services in Rural India: The Narangwal Experiment. Baltimore: Johns Hopkins University Press, pp. 172–214. Kishor, Sunita. 1995. Gender differentials in child mortality: A review of the evidence. In Das Gupta, Monica, Chen, Lincoln C., and Krishnan, T. N. (eds.), Women’s Health in India: Risk and Vulnerability. New Delhi: Oxford University Press, pp. 19–54. Kuhn, Alice. S. 1988. Heiler und Ihre Patienten auf dem dach der welt: Ladakh aus ethnomedizinischer sicht. New York: Verlag Peter Lang. Laderman, Carol. 1983. Wives and Midwives: Childbirth and Nutrition in Rural Malaysia. Berkeley: University of California Press.

References

217

Lamb, M. E., Thompson, R. A., Gardner, W. P., and Charnov, E. L. 1985. Infant-Mother Attachment. Hillsdale, NJ: Lawrence Erlbaum. Lang, S. D. R., and Lang, Ann. 1971. The Kunde Hospital and a demographic survey of the Upper Khumbu, Nepal. The New Zealand Medical Journal 74: 1–8. Leatherman, Thomas L. 1998. Illness, social relations, and household production and reproduction in the Andes of southern Peru. In Goodman, Alan H., and Leatherman, Thomas L. (eds.), Building a New Biocultural Synthesis: Political-Economic Perspectives on Human Biology. Ann Arbor: University of Michigan Press, pp. 245–267. Leatherman, Thomas L. 1996. A biocultural perspective on health and household economy in southern Peru. Medical Anthropology Quarterly 10: 476–495. Lechtig, Aaron. ´ 1982. Studies of nutrition intervention in pregnancy. Birth 9: 115–119. Lechtig, Aaron, ´ Delgado, Hern´an, Martorell, Reynaldo, Richardson, Douglas, Yarbrough, Charles, and Klein, Robert E. 1977. Effect of maternal nutrition on infant mortality. In Mosley, W. Henry (ed.), Nutrition and Human Reproduction. New York: Plenum Press, pp. 147–174. Lechtig, Aaron, ´ Habicht, Jean-Pierre, Delgado, Hern´an, Klein, Robert E., Yarbrough, Charles, and Martorell, Reynaldo. 1975. Effect of food supplementation during pregnancy on birthweight. Pediatrics 56: 508–520. Lechtig, Aaron, ´ Martorell, Reynaldo, Delgado, Hern´an, Yarbrough, Charles, and Klein, Robert E. 1978. Food supplementation during pregnancy, maternal anthropometry and birthweight in a Guatemala rural population. Journal of Tropical Pediatrics 24: 217–222. Leibson, C., Brown, M., Thibodeau, S., Stevenson, D., Vreman, H., Cohen, R., Clemons, G., Callen, W., and Moore, L. G. 1989. Neonatal hyperbilirubinemia at high altitude. American Journal of Diseases of Children 143: 983– 987. Leonard, William R. 1989. Nutritional determinants of high-altitude growth in Nunoa, ˜ Peru. American Journal of Physical Anthropology 80: 341–352. Leonard, William R., Katzmarzyk, Peter T., Stephen, Merrill A., and Ross, Allen G. P. 1995. Comparison of the heart rate-monitoring and factorial methods: Assessment of energy expenditure in highland and coastal Ecuadoreans. American Journal of Clinical Nutrition 61: 1146–1152. Levine, Nancy E. 1987. Differential child care in three Tibetan communities: Beyond son preference. Population and Development Review 13: 281–304. Levine, Nancy E. 1988a. The Dynamics of Polyandry. Chicago: University of Chicago Press. Levine, Nancy E. 1988b. Women’s work and infant feeding: A case from rural Nepal. Ethnology 27: 231–251. LeVine, Robert A. 1998. Child psychology and anthropology: An environmental view. In Panter-Brick, Catherine (ed.), Biosocial Perspectives on Children. Cambridge: Cambridge University Press, pp. 102–130. Lichty, John A., Ting, Rosalind Y., Bruns, Paul D., and Dyar, Elizabeth. 1957. Studies of babies born at high altitude. American Journal of Diseases of Children 93: 666–678.

218

References

Little, Michael A. 1989. Human biology of African pastoralists. Yearbook of Physical Anthropology 32: 215–247. Lubchenco, L. O., Searls, D. T., and Brazie, J. V. 1972. Neonatal mortality rate: Relationship to birth weight and gestational age. Journal of Pediatrics 81: 814–822. Lubchenco, Lula O., Hansman, Charlotte, Dressler, Miriam, and Boyd, Edith. 1963. Intrauterine growth as estimated from liveborn birth-weight data at 24 to 42 weeks of gestation. Pediatrics 32: 793–800. MacNeish, R. S., and Berger, R. P. 1970. Megafauna and man from Ayacucho, highland Peru. Science 166: 8975–8977. Main, Mary. 1990. Cross-cultural studies of attachment organizations: Recent studies, changing methodologies, and the concept of conditional strategies. Human Development 33: 48–61. Majumdar, D. N. 1960. Himalayan Polyandry. New York: Asia Publishing House. Malik, S. L., and Singh, Indera P. 1984. Ecology and human physiology in Ladak. In Lukacs, John R. (ed.), The People of South Asia: The Biological Anthropology of India, Pakistan, and Nepal. New York: Plenum Press, pp. 421–439. Mani, Anna. 1981. The climate of the Himalaya. In Lall, J. S. (ed.), The Himalaya: Aspects of Change. New Delhi: Oxford University Press, pp. 3–15. Mann, R. S. 1986. The Ladakhi: A Study in Ethnography and Change. Calcutta: Anthropological Survey of India. Manshande, J. P, Eeckels, R., Manshande-Desmet, V., and Vlietinck, R. 1987. Rest versus heavy work during the last weeks of pregnancy: Inﬂuence on fetal growth. British Journal of Obstetrics and Gynaecology 94: 1059–1067. Martorell, Reynaldo. 1980. Interrelationship between diet, infectious disease, and nutritional status. In Greene, Lawrence S., and Johnston, F. (eds.), Social and Biological Predictors of Nutritional Status, Physical Growth, and Neurological Development. New York: Academic Press, pp. 81–106. Mathai, Saramma T. 1989. Women and the health system. In Gopalan, C., and Kaur, Suminder (eds.), Women and Nutrition in India. New Delhi: Nutrition Foundation of India, pp. 247–295. Mavalankar, D. V., Trivedi, C. C., and Gray, R. H. 1994. Maternal weight, height and risk of poor pregnancy outcome in Ahmedabad, India. Indian Pediatrics 31: 1205–1211. Mazess, Richard B. 1978. Adaptation: A conceptual framework. In Meier, R. J., Otten, C. M., and Abdel-Hameed, F. (eds.), Evolutionary Models and Studies in Human Diversity. The Hague: Mouton, pp. 9–18. Mazess, Richard B. 1965. Neonatal mortality and altitude in Peru. American Journal of Physical Anthropology 23: 209–214. Mazess, Richard B. 1975. Human adaptation to high altitude. In Damon, Albert (ed.), Physiological Anthropology. New York: Oxford University Press, pp. 167–209. McClung, Jean. 1969. Effects of High Altitude on Human Birth. Cambridge, MA: Harvard University Press. McCormick, Marie C. 1985. The contribution of low birth weight to infant mortality and childhood morbidity. New England Journal of Medicine 312: 82–90.

References

219

McKeown, Thomas. 1976. The Modern Rise of Population. New York: Academic Press. Meyer, Fernand. 1981. Populations de haute altitude et sant´e en Himalaya. L’homme et son environnement a` haute altitude. Paris: Seminaire CNRS/NSF, pp. 97–100. Meyer, Mary B. 1977. Effects of maternal smoking and altitude on birth weight and gestation. In Reed, Dwayne M., and Stanley, Fiona J. (eds.), The Epidemiology of Prematurity. Baltimore: Urban and Schwarzenberg, pp. 81–101. Millard, Ann V. 1994. A causal model of high rates of child mortality. Social Science & Medicine 38: 253–268. Miller, Barbara D. 1981. The Endangered Sex: Neglect of Female Children in Rural North India. Ithaca, NY: Cornell University Press. Miller, Barbara D. 1984. Daughter neglect, women’s work, and marriage: Pakistan and Bangladesh compared. Medical Anthropology 8: 109–126. Miller, Barbara D. 1987. Female infanticide and child neglect in rural north India. In Scheper-Hughes, Nancy (ed.), Child Survival: Anthropological Perspectives on the Treatment and Maltreatment of Children. Boston: D. Reidel Publishing Company, pp. 95–112. Miller, Herbert C. and Hassanein, Khatab. 1971. Diagnosis of impaired fetal growth in newborn infants. Pediatrics 48: 511–522. Molbak, K., Gottschau, A., Aaby, P., Hojlyng, N., Ingholt, L., and da Silva, A. P. J. 1994. Prolonged breast feeding, diarrhoeal disease, and survival of children in Guinea-Bissau. British Medical Journal 308: 1403–1406. Moore, L. G., Hershey, Douglas W., Jahnigen, Dennis, and Bowes, W. 1982. The incidence of pregnancy-induced hypertension is increased among Colorado residents at high altitude. American Journal of Obstetrics and Gynecology 144: 423–429. Moore, Lorna G., Niermeyer, Susan, and Zamudio, Stacy. 1998. Human adaptation to high altitude: Regional and life-cycle perspectives. Yearbook of Physical Anthropology 41: 25–64. Moore, Lorna G., Young, David A., McCullough, Robert E., Droma, Tarshi, and Zamudio, Stacy. 2001a. Tibetan protection from intrauterine growth restriction (IUGR) and reproductive loss at high altitude. American Journal of Human Biology 13: 635–644. Moore, Lorna G., Zamudio, Stacy, Zhuang, Jianguo, Sun, Shinfu, and Droma, Tarshi. 2001b. Oxygen transport in Tibetan women during pregnancy at 3,658 m. American Journal of Physical Anthropology 114: 42–53. Moore, Lorna Grindlay. 1990. Maternal O2 transport and fetal growth in Colorado, Peru, and Tibet high-altitude residents. American Journal of Human Biology 2: 627–637. Mora, Jos´e O., Ricardo Sanchez, Belen de Paredes, et al. 1981. Sex related effects of nutritional supplementation during pregnancy on fetal growth. Early Human Development 5: 243–251. Nesse, Randolph M., and Williams, George C. 1994. Why We Get Sick: The New Science of Darwinian Medicine. New York: Times Books. Neuﬁeld, Lynnette, Pelletier, David L., and Haas, Jere D. 1999. The timing hypothesis and body proportionality of the intra-uterine growth retarded infant. American Journal of Human Biology 11: 638–646.

220

References

Nichter, M., and Kendall, C. 1991. Beyond child survival: Anthropology and international health in the 1990s. Medical Anthropology Quarterly 5: 195– 203. Niermeyer, Susan, Shaffer, E. M., Thilo, E., Corbin, C., and Moore, L. G. 1993. Arterial oxygenation and pulmonary arterial pressure in healthy neonates and infants at high altitude. Journal of Pediatrics 123: 767–772. Niermeyer, Susan, Yang, P., Shanmina, Drolkar, Zhuang, J., and Moore, Lorna G. 1995. Arterial oxygen saturation in Tibetan and Han infants born in Lhasa, Tibet. New England Journal of Medicine 333: 1248–1252. Norberg-Hodge, Helena. 1991. Ancient Futures: Learning from Ladakh. San Francisco: Sierra Club Books. Norberg-Hodge, Helena, and Palden, Gyelong Thupstan, eds. 1991. LadakhiEnglish; English-Ladakhi Dictionary. Leh, Ladakh: Ladakh Ecological Development Group. Norberg-Hodge, Helena, and Russell, Hazel. 1994. Birth and child-rearing in Zangskar. In Crook, John, and Osmaston, Henry (eds.), Himalayan Buddhist Villages: Environment, Resources, Society, and Religious Life in Zangskar. Bristol, UK: University of Bristol, pp. 519–532. Norboo, Tsering, and Yahya, M. 1988. Health problems in Ladakh. Voice of the Himalaya 1: 5–9. Norboo, Tsering, Yahya. M., Bruce, N. G., Heady, J. A., and Ball, K. P. 1991. Domestic pollution and respiratory illness in a Himalayan village. International Journal of Epidemiology 20: 749–757. Obert, P., Fellmann, N., Falgairette, G., Bedu, M., Van Praagh, E., Kemper, H., Post, B., Spielvogel, H., Tellez, V., Qintelas, A., and Coudert, J. 1994. The importance of socioeconomic and nutritional conditions rather than altitude on the physical growth of prepubertal Andean highland boys. Annals of Human Biology 21: 145–154. Ortner, Sherry B. 1978. Sherpas through Their Rituals. New York: Cambridge University Press. Ortner, Sherry B. 1989. High Religion: A Cultural and Political History of Sherpa Buddhism. Princeton, NJ: Princeton University Press. Osmaston, Henry. 1994. The farming system. In Crook, John, and Osmaston, Henry (eds.), Himalayan Buddhist Villages: Environment, Resources, Society, and Religious Life in Zangskar. Bristol, UK: University of Bristol, pp. 139– 198. Osmaston, Henry, and Crook, John. 1994. Sha-de: meagre subsistence or garden of eden? In Crook, John, and Osmaston, Henry (eds.), Himalayan Buddhist Villages: Environment, Resources, Society, and Religious Life in Zangskar. Bristol, UK: Bristol University Press, pp. 249–284. Osmaston, Henry, Frazer, Janet, and Crook, Stamati. 1994. Human adaptation to environment in Zangskar. In Crook, John, and Osmaston, Henry (eds.), Himalayan Buddhist Villages: Environment, Resources, Society and Religious Life in Zangskar, Ladakh. Bristol, UK: University of Bristol. PAHO. 1994. Health Conditions in the Americas. Washington DC: Pan American Health Organization. Palmer, Susan K., Moore, Lorna G., Young, David A., Cregger, Betsy, Berman, Jonathan, and Zamudio, Stacy. 1999. Altered blood pressure course during

References

221

normal pregnancy and increased preeclampsia at high altitude (3100 meters) in Colorado. American Journal of Obstetrics and Gynecology 180: 1161– 1168. Palriwala, R. 1988. Leh Development Plan 1989–2009. Leh, Ladakh: INTAC. Panter-Brick, C. 1989. Motherhood and subsistence work: The Tamang of rural Nepal. Human Ecology 17: 205–228. Panter-Brick, Catherine. 1993. Seasonality of energy expenditure during pregnancy and lactation for rural Nepali women. American Journal of Clinical Nutrition 57: 620–628. Panter-Brick, Catherine, ed. 1998. Biosocial Perspectives on Children, New York: Cambridge University Press. Patwari, A. K, Bisht, S., Srinivasan, A., Deb, M., and Chattopadhya, D. 1996. Aetiology of pneumonia in hospitalized children. Journal of Tropical Pediatrics 42: 15–20. Paul, Benjamin D., ed. 1955. Health, Culture, & Community. New York: Russell Sage Foundation. Pawson, I. G., and Jest, Corneille. 1978. The high-altitude areas of the world and their cultures. In Baker, P. T. (ed.), The Biology of High-altitude Peoples. New York: Cambridge University Press, pp. 17–45. Peter, Prince. 1963. A Study of Polyandry. The Hague: Mouton & Co. Prentice, A. M., Whitehead, R. G., Roberts, S. B. G., and Paul, A. A. 1981. Long-term energy balance in child-bearing Gambians. American Journal of Clinical Nutrition 34: 2790–2799. Prentice, Andrew M., Cole, Timothy J., Ford, Frances A., Lamb, William H., and Whitehead, Roger G. 1987. Increased birthweight after prenatal dietary supplementation of rural African women. American Journal of Clinical Nutrition 46: 912–925. Puffer, Ruth Rice, and Serrano, Carlos V. 1973. Patterns of Mortality in Childhood. Washington, DC: Pan American Health Organization. Quandt, Sara A. 1987. Methods for determining dietary intake. In Johnston, Francis E. (ed.), Nutritional Anthropology. New York: Alan R. Liss, Inc., pp. 67–84. Rabgias, Tashi. 1988. Salient features of history of Ladakh. Voice of the Himalaya 1: 1–4. Ramachandran, Prema. 1989. Nutrition in pregnancy. In Gopalan, C., and Kaur, Suminder (eds.), Women and Nutrition in India. New Delhi: Nutrition Foundation of India, pp. 153–193. Rechung, Rinpoche Jampal Kunzang. 1973. Tibetan Medicine. Berkeley: University of California Press. Rizvi, Janet. 1983. Ladakh: Crossroads of High Asia. New Delhi: Oxford University Press. Ross, J. L. 1984. Culture and fertility in the Nepal Himalayas: A test of a hypothesis. Human Ecology 12: 163–181. Sahlins, Marshall. 1968. Notes on the original afﬂuent society. In Lee, Richard B., and DeVore, Irven (eds.), Man the Hunter. Chicago: Aldine, pp. 85– 88. Sakamoto-Momiyama, Masako. 1977. Seasonality in Human Mortality: A MedicoGeographical Study. Tokyo: University of Tokyo Press.

222

References

Sander, Reinhard. 1983. Three generations in the Wanla Valley. In Kantowsky, Detlef, and Sander, Reinhard (eds.), Recent Research on Ladakh. London: Weltforum Verlag, pp. 183–194. Sangay, Thubten. 1984. Tibetan traditions of childbirth and childcare. Tibetan Medicine 7: 3–24. SAS Institute Inc. 1991. SAS Technical Report P-217 SAS/STAT Software: The PHREG Procedure Version 6. Cary, NC: SAS Institute. Scheper-Hughes, Nancy. 1992. Death without Weeping. Berkeley: University of California Press. Scheper-Hughes, Nancy, and Sargent, Carolyn. 1998. Introduction: The cultural politics of childhood. In Scheper-Hughes, Nancy, and Sargent, Carolyn (eds.), Small Wars: The Cultural Politics of Childhood. Berkeley: University of California Press, pp. 1–33. Schuler, Sidney Ruth. 1987. The Other Side of Polyandry. Boulder, CO: Westview Press. Scrimshaw, Susan C. M. 1978. Infant mortality and behavior in the regulation of family size. Population and Development Review 4: 383–403. Scrimshaw, Susan C. M. 1983. Infanticide as deliberate fertility regulation. In Bulatao, Rudolfo, and Lee, Ronald D. (eds.), Determinants of Fertility in Developing Countries. New York: Academic Press, pp. 245–266. Scrimshaw, Susan C. M. 1984. Infanticide in human populations: Societal and individual concerns. In Hausfater, Glenn and Hrdy, Sarah Blaffer (eds.), Infanticide: Comparative and Evolutionary Perspectives. New York: Aldine, pp. 439–462. Secker-Walker, R. H., Vacek, P. M., Flynn, B. S., and Mead, P. B. 1997. Smoking in pregnancy, exhaled carbon monoxide, and birth weight. Obstetrics and Gynecology 5: 648–653. Sethi, G. R., Sachdev, H. P. S., and Puri, R. K. 1991. Women’s health and fetal outcome. Indian Pediatrics 28: 1379–1393. Sever, John L., Ellenberg, Jonas H., and Edmonds, Dorothy. 1977. Maternal urinary tract infections and prematurity. In Reed, Dwayne M., and Stanley, Fiona J. (eds.), The Epidemiology of Prematurity. Baltimore: Urban and Schwarzenberg, pp. 193–196. Silbert, J. R., Jadhav, Malati, and Inbaraj, S. G. 1978. Maternal and fetal nutrition in South India. British Medical Journal 1: 1517–1518. Singer, Merrill. 1989. The limitations of medical ecology: The concept of adaptation in the context of social stratiﬁcation and social transformation. Medical Anthropology 10: 223–234. Singh, Jyoti Shankar. 1998. Creating a New Consensus on Population: The International Conference on Population and Development. London: Earthscan. Singh, Meharban. 1986. Hospital-based data on perinatal and neonatal mortality in India. Indian Pediatrics 23: 579–584. Singh, Meharban, and Paul, V. K. 1988. Strategies to reduce perinatal and neonatal mortality. Indian Pediatrics 25: 499–509. Singh, S. P. 1989. Variation in adult anthropometry, child growth and development in the Western Himalayas – A review. In Basu, Amitabha, and Gupta, Ranjan (eds.), Human Biology of Asian Highland Populations in the Global Context. Calcutta: Indian Anthropological Society, pp. 81–104.

References

223

Smith, Chery. 1997. The effect of maternal nutritional variables on birthweight outcomes of infants born to Sherpa women at low and high altitudes in Nepal. American Journal of Human Biology 9: 751–763. Spector. R. M. 1971. Mortality Characteristics of a High-altitude Peruvian Population. Master’s Thesis, Pennsylvania State University. Spielvogel, Hilde, and Vitzthum, Virginia J. 2003. Rates of fetal loss in Bolivian women indigenous to high altitude. American Journal of Human Biology 15: 286. Srikantia, S. G. 1989. Pattern of growth and development of Indian girls and body size of adult Indian women. In Gopalan, C., and Kaur, Suminder (eds.), Women and Nutrition in India. New Delhi: Nutrition Foundation of India, pp. 108–152. Srinivasan, K., and Kanitkar, Tara. 1989. Demographic consequences of low status of women in Indian society. In Gopalan, C., and Kaur, Suminder (eds.), Women and Nutrition in India. New Delhi: Nutrition Foundation of India, pp. 17–62. Sroufe, L. Alan, and Rutter, M. 1984. The domain of developmental psychopathology. Child Development 55: 17–29. Stini, William A. 1969. Nutritional stress and growth: Sex difference in adaptive response. American Journal of Physical Anthropology 31: 417–426. Stinson, Sarah. 1982. The effect of high altitude on the growth of children of high socioeconomic status in Bolivia. American Journal of Physical Anthropology 59: 61–71. Stobdan, Phuntsog. 1990. Ecology of disease in Ladakh. In Icke-Schwalbe, Lydia, and Meier, Gudran (eds.), Wissenschaftsgeschichte und Gegenwartige Forschungen in Nordwest-Indien. Dresden Forschungsstelle: Staatliches Musem fur Volkerkunde, pp. 164–170. Stuart-Macadam, Patricia, and Dettwyler, Katherine, eds. 1995. Breastfeeding: Biocultural Perspectives. Hawthorne, NY: Aldine de Gruyter. Tafari., N., Naeye, R. L., and Gobezie, A. 1980. Effects of maternal undernutrition and heavy physical work during pregnancy on birth weight. British Journal of Obstetrics and Gynaecology 87: 222–226. Thaddeus, Sereen, and Maine, Deborah. 1994. Too far to walk: Maternal mortality in context. Social Science & Medicine 38: 1091–1110. Thomas, R. Brooke. 1976. Energy ﬂow at high altitude. In Baker, Paul T., and Little, Michael A. (eds.), Man in the Andes. Stroudsburg, PA: Dowden, Hutchinson & Ross, pp. 379–404. Thomas, R. Brooke. 1998. The evolution of human adaptability paradigms: Toward a biology of poverty. In Goodman, Alan H., and Leatherman, Thomas L. (eds.), Building a New Biocultural Synthesis: Political-Economic Perspectives on Human Biology. Ann Arbor: University of Michigan Press, pp. 43–73. Thomas, R. Brooke, Winterhalder, Bruce, and McRae, Stephen D. 1979. An anthropological approach to human ecology and adaptive dynamics. Yearbook of Physical Anthropology 22: 1–46. Thomas, R. Brooke, and Winterhalder, Bruce P. 1976. Physical and biotic environment of southern highland Peru. In Baker, Paul T., and Little, Michael A. (eds.), Man in the Andes: A Multidisciplinary Study of High-altitude Quechua. Stroudsburg, PA: Dowden, Hutchinson and Ross, pp. 21–59.

224

References

Tronick, E. Z., Thomas, R. B., and Daltabuit, M. 1994. The Quechua manta pouch: A caretaking practice for buffering the Peruvian infant against the multiple stressors of high altitude. Child Development 65: 1005–1013. Ulijaszek, Stanley J., and Huss-Ashmore, Rebecca A., eds. 1997. Human Adaptability Past, Present, and Future. New York: Oxford University Press. Unger, Cynthia, Weiser, Jane K., McCullough, Robert E., Keefer, Sharon, and Moore, Lorna Grindlay. 1988. Altitude, low birth weight, and infant mortality in Colorado. Journal of the American Medical Association 259: 3427– 3432. United Nations. 2001a. 1998 Statistical Yearbook. New York: United Nations. United Nations. 2001b. 1999 Demographic Yearbook. New York: United Nations. Van Esterik, Penny. 1989. Beyond the Breast-Bottle Controversy. New Brunswick, NJ: Rutgers University Press. Victora, C. G., Smith, P. G., Vaghan, J. P., Nobre, L. C., Lombard, C., Telxeira, A. M. B., Fuchs, S. M., Moreira, L. B., Gigante, L. P., and Barros, F. C. 1988. Inﬂuence of birth weight on mortality from infectious diseases: A casecontrol study. Pediatrics 81: 807–811. Vitzthum, V. J. 1989. Nursing behaviour and its relation to duration of postpartum amenorrhoea in an Andean community, Journal of Biosocial Science 21: 145–160. Vitzthum, V. J. 1992. Infant nutrition and the consequences of differential market access in Nunoa, ˜ Peru. Ecology of Food and Nutrition 28: 45–63. Vitzthum, V. J. 2001. The home team advantage: Reproduction in women indigenous to high altitude. Journal of Experimental Biology 204: 3141–3150. Vitzthum, V. J., Ellison, P. T., Sukalich, S., Caceres, E., and Spielvogel, H. 2000a. Does hypoxia impair ovarian function in Bolivian women indigenous to high altitude? High Altitude Medicine and Biology 1: 39–49. Vitzthum, V. J., Spielvogel, Hilde, and Caceres, Esperanza. 2000b. Prospective longitudinal study of menstrual patterns in an agropastoralist Aymara population. American Journal of Physical Anthropology, Supplement 30: 312. Vitzthum, Virginia J., and Wiley, Andrea S. 2003. The proximate determinants of fertility in populations exposed to chronic hypoxia. High Altitude Medicine and Biology 4: 125–139 Vobecky, Jitka S. 1986. Nutritional aspects of preconceptional period as related to pregnancy and early infancy. Progress in Food and Nutrition Science 10: 205–236. Weinstein, R. S., and Haas, Jere D. 1977. Early stress and later reproductive performance under conditions of malnutrition and high altitude hypoxia. Medical Anthropology 1: 25–54. Weisner, Thomas S. 1987. Socialization for parenthood in sibling caretaking societies. In Lancaster, Jane B., Altmann, Jeanne, Rossi, Alice S., and Sherrod, Lonnie R. (eds.), Parenting across the Lifespan: Biosocial Dimensions. New York: Aldine de Gruyter, pp. 237–270. Wilcox, Allen J. 1993. Birthweight and perinatal mortality: The effect of maternal smoking. American Journal of Epidemiology 137: 1098–1104. Wilcox, Allen J. 2001. On the importance – and the unimportance – of birthweight. International Journal of Epidemiology 30: 1233–1241.

References

225

Wiley, Andrea S. 1992a. Adaptation and the biocultural paradigm in medical anthropology: A critical review. Medical Anthropology Quarterly 6: 216–236. Wiley, Andrea S. 1992b. High Altitude Adaptation and Maternal-Infant Health: Neonatal Characteristics and Infant Survival in the Himalaya. PhD Dissertation, University of California, Berkeley and San Francisco. Wiley, Andrea S. 1994. Neonatal and maternal anthropometric characteristics in a high-altitude Himalayan population. American Journal of Human Biology 6: 499–510. Wiley, Andrea S. 1998. The ecology of low natural fertility in Ladakh. Journal of Biosocial Science 30: 457–480. Wiley, Andrea S. 1999. Demographic transition in Ladakh. Paper presented at the American Anthropological Association Annual Meeting, Chicago, IL. Wiley, Andrea S. 2002. Increasing usage of prenatal care in Ladakh: The roles of cultural and ecological factors. Social Science & Medicine 55: 1089– 1102. Wiley, Andrea S., and Carlin, Leslie E. 1999. Demographic contexts and the adaptive role of mother-infant attachment: A hypothesis. Human Nature 10: 135–161. Wiley, Andrea S., and Pike, Ivy L. 1998. An alternative method for assessing early mortality in contemporary populations. American Journal of Physical Anthropology 107: 315–330. Wilkinson, Richard G. 1996. Unhealthy Societies: The Afﬂictions of Inequality. New York: Routledge. Wilson, Jane M., Campbell, M. J., and Afzal, M. 1990. Heights and weights of children in Ladakh, north India. Journal of Tropical Pediatrics 36: 271– 272. Windham, G. C., Eaton, A., and Hopkins, B. 1999. Evidence for an association between environmental tobacco smoke exposure and birthweight: A metaanalysis and new data. Paediatric and Perinatal Epidemiology 13: 35–57. Wood, J. W., Johnson, P. L., and Campbell, K. 1985. Demographic and endocrinological aspects of low natural fertility in highland New Guinea. Journal of Biosocial Science 17: 57–79. Wood, James W. 1994. Dynamics of Human Reproduction. New York: Aldine de Gruyter. World Health Organization. 1977. International Classiﬁcation of Disease. Geneva: WHO. Yip, Ray. 1987. Altitude and birth weight. Journal of Pediatrics 111: 869–876. Zamudio, S., Palmer, S. K., Droma, T., Stamm, E., Cofﬁn, C., and Moore, L. G. 1995. Effect of altitude on uterine artery blood ﬂow during normal pregnancy. Journal of Applied Physiology 79: 7–14. Zamudio, S., Droma, T., Norkyel, K. Y., Acharya, G., Zamudio, J. A., Niermeyer, S. N., and Moore, L. G. 1993. Protection from intrauterine growth retardation in Tibetans at high altitude. American Journal of Physical Anthropology 91: 215–224. Zamudio, Stacy, and Moore, Lorna G. 2000. Altitude and fetal growth: Current knowledge and future directions. Ultrasound in Obstetrics and Gynecology 16: 6–8.

226

References

Zeitlin, M. F., Ahmed, N. U., Beiser, A. S., Zeitlin, J. A., Super, C. M., and Guldan, G. S. 1995. Developmental, behavioural, and environmental risk factors for diarrhoea among rural Bangladeshi children of less than two years. Journal of Diarrhoeal Disease Research 13: 99–105. Zhiman, An, et al. 1982. Palaeoliths and microliths from Shenja and Shuanghu, Northern Tibet. Current Anthropology 23: 493–499.

Index

Abortion (see fetal death) 15, 105, 172 Acclimatization (see also adaptability) 26–27 Activity patterns (see also workloads) during pregnancy 4–6, 76, 77, 89, 150, 198–199 during pregnancy, questions about 76–77 effect on birthweight 91–96, 99–100, 150, 161–164, 180–182, 192, 198 seasonal variation in 5, 38, 49, 91–92, 94–95, 129, 140, 181–182, 202 Adaptability (see also adaptation) 7–8, 10 history of studies 7–8, 26–27 intrapopulation variation in 3, 4, 30–31, 105–106, 185–186 modes of 13–14 relationship to biocultural approaches 7–8 Adaptation (see also natural selection; evolutionary theory) 10, 13–15, 192–193 and culture 13–14 and population history 4, 11, 25–26, 150–151, 185–186 as compromise 14, 19–21, 34, 186–189 biocultural 7–8, 9, 192–193 birthweight as a measure of 15, 27, 33–34, 105–106, 150–151, 160–161, 178, 185–186

genetic, among high-altitude populations 25–27, 185–186, 206 health and 7–8, 15 measures of 14–15, 185 relevance for health policy 23–24, 178–179, 196–197 Age (see maternal characteristics; gestational age) pattern of infant mortality, in Ladakh 6, 116–119, 151, 153–154, 182, 183–184 pattern of pediatric hospital admissions 122–123, 134–135, 183 Agriculture competes with reproduction 6, 186–189 decline of, in Ladakh 41–43, 190–192, 201–203 house gardens in Ladakh 49, 53, 188 production of food 4, 48, 49–50, 79, 181, 184 seasonality of, in Ladakh 5, 38, 49, 91–92, 94–95, 129, 140, 181–182, 202 subsistence in Ladakh 4, 42, 49–50, 79 women’s contributions to 5, 50, 52–53, 76–77, 95, 186–187 Allostasis 11 Altitude (see high altitude) Amchi (see also Tibetan medicine) 58–69, 102–103, 141–142, 207

227

228

Index

Amchi (cont.) usage of, during pregnancy 100–101, 199 usage of, for infants 141–142 Anemia (see also hemoglobin; malnutrition) 61, 160 among infants 138 among north Indian women 169, 176 during pregnancy in Ladakh 82, 90–91, 94, 169, 200 effect on birthweight 90 Andes 3, 15–17 antiquity of habitation 25, 150–151 biocultural studies in 15–16 birthweight in 28–29, 32–33, 147–151 birthweight, compared to Ladakh 147–148, 151, 185 breastfeeding patterns in 154–155 child care patterns in, 154 child growth and development in 15, 30–32, 150 infant/neonatal mortality in 34, 151–156 maternal characteristics in 148, 149–150 reproductive problems in, among Spanish 3, 35 Anthropometry (see also neonates; maternal anthropometry; methods) 73–75, 82–83, 108 APGAR score 71 Aridity 37, 41, 135 Asphyxia 161, 173 Attachment 130–131, 133, 189 childcare patterns and 132, 189 of Ladakhi infants and mothers 131–133, 189 Autonomy in relation to attachment 131–133, 189 individual, and household membership 51, 61, 132, 188–189, 190, 206

Baker, P. T. 17, 26, 34 Barley (tsampa) 4, 31, 49, 56, 90, 129, 137, 155, 163, 190 Beall, C. M. 4, 25–26, 27, 151–156 Bias (see also sample characteristics) gender, in childcare 133–134, 142–144, 164, 165, 184 in anthropological ﬁeldwork 62–64 in birth outcome analyses 108–109, 117 in sample 63, 77–81 Bilirubin (see jaundice) 71, 110, 133 Biocultural perspectives (see also adaptability, adaptation) 7–8, 9, 192–193 as one story about reproductive health in Ladakh 24 biocultural ethnography 17, 21–23, 62–63 body in 9–11 culture in 8–10, 13–14 on health 6–7, 15, 57–58 on health in the Andes 15–16 on reproductive health in Ladakh 5–6, 16–19, 63–64, 186–189, 192–203 population in 8–11 Biomedicine (allopathy) 1, 6 facilities in Ladakh 58–60, 100 interventions to improve birth outcome 140–141, 197, 198–201 Leh hospital 52, 58–59, 101, 140, 171, 198 resources for infant health 140, 141 usage during pregnancy 1, 68, 100, 103, 170–175, 181 Birth (see childbirth; fertility) age at ﬁrst 100, 169, 170 seasonality of, in Ladakh 116–120, 121 Birth control 109, 144, 165, 170, 193, 202, 203, 208 Birth order (see also parity) and infant mortality 115–118, 142, 164–169

Index effect on birthweight 87–88, 93, 117–118, 157 Birth outcome (see birthweight, neonatal mortality, infant mortality) Birth records (see also reproductive histories) from Leh hospital 69, 115–116, 206 Birthweight 71, 72–74 among Sherpas in Nepal 30, 92, 147–157, 159–160 and socioeconomic status 1, 19, 72, 106–108, 150–154, 181–201 and socioeconomic status, in Ladakh 18, 95–96, 97–98, 115, 150, 156, 180–181, 186 as a measure of adaptation 15, 27, 33–34, 105–106, 150–151, 160–161, 178, 185–186 as a predictor of infant mortality in Ladakh 6, 111–115, 151–154, 179, 186 as a predictor of infant mortality, in general 17, 27, 32, 72, 105, 107, 171–173 as link between maternal and infant resources 17, 72, 106, 107 at high altitude 4, 27–30, 146–147 blood pressure in pregnancy, effect on 91 Buddhist-Muslim differences in Ladakh 98–100, 115, 181–192 comparison of Ladakhis and migrants in the Andes 147–149, 150–151, 185 comparison of Ladakhi and Han Chinese in Tibet 157, 161 comparison of Ladakhi and North Indians 167–173, 174 comparison of Ladakhi and Sherpa 147–158, 166 comparison of Ladakhi and Tibetan 157–166, 185

229 comparison of Ladakhis and Tibetans in Ladakh 96–98, 157–159, 181–185 determinants of, in general 6, 71–73, 86–87, 107 determinants of, in Ladakh 86–100, 101, 149–151, 179–182, 197–200 diet during pregnancy and effect on 76, 88–91, 150, 157, 180–182, 198–199 distribution, in Ladakh sample 83, 84 effect of hypoxia on 4, 27–30, 32, 73, 105–106, 124, 174, 175, 182, 185–186 ethnic differences, in Ladakh 84, 85–96, 115, 181 gestational age, effect on 73, 87, 92, 180 in Ladakh sample 5–6, 82, 83–85, 111, 146, 182 in relation to disease vulnerability 34, 111, 121, 155, 175 in the Andes 28–29, 32–33, 147–151 in Tibet 29–30, 147–157 interventions to increase, in Ladakh 197–200 link to infant mortality modiﬁed at high altitude 4, 32–34, 111, 151–153 low (see low birthweight) maternal education, effect on 96, 115, 192 maternal weight as a predictor of 86–95, 198 maternal workloads, as a predictor of 91–96, 99–100, 150, 161–164, 180–182, 192–198 nutrient supplementation, effect on 88, 198 of migrants in Ladakh 97, 98, 185 of migrants to high altitude 4, 29, 98, 146–149 parity, effect on 87–88, 93, 117–118, 157

230

Index

Birthweight (cont.) pregnancy-induced hypertension, effect on 28, 91, 182, 198 prenatal care, effect on 101, 181, 200 seasonality in 94–95, 120, 181–182 sex differences in, at high altitude 29, 149, 157 sex differences in, in general 29, 73–83, 84–85, 148 sex differences in, in Ladakh 84, 85, 149, 157, 180 smoking, effect on 73–99, 150–157 BCG and polio vaccination 125 Blood pressure (see also hypertension, pregnancy-induced hypertension) 15, 76 during pregnancy in Ladakh, effect on birthweight 91 Body as product of evolutionary history 10–11 in biocultural approaches 9–11 Bon religion 45–46 Bowlby, J. 130–131, 133, 189 Breastfeeding 128 among Andean populations 154–155 among Tibetans 163–164 frequency of, in Ladakh 129, 137–138, 154 in North India 173–174, 176 in relation to women’s work 129–137, 155–163, 183, 198 of Ladakhi infants 129, 137–138, 154–155, 176, 183 termination of 139 Buddhism 2, 47, 131 and women’s status 5, 52 in Ladakh 43–45, 51–52, 131 Buddhist–Muslim conﬂict in Ladakh 47–49, 57, 79, 90, 98, 181, 190 Calories, in Ladakhi diet 5, 18, 60–61 Capillarization as an adaptation to hypoxia 26

Carbon monoxide (CO; see also smoking, pollution) 18, 99, 126, 136, 191 Cardiovascular system, at high altitude 26, 30 Census of India 77–78, 80, 206 Cesarean section 1 Chang-Tang plateau 43 Chen, L. C. 17–19, 106–107, 179 Chest circumference (see also lung volume) 74, 97, 111–112 as a predictor of infant mortality 114 of Ladakhi neonates 74, 97, 111–112 Childbirth in Ladakh 53–66, 78–79, 81, 131, 175 in Leh hospital 63–66, 69, 81, 118–132, 175, 199 in North India 173 Childcare 2–6, 129–130, 133 and attachment, in Ladakh 132, 189 ﬂuidity of 53–55, 132, 139–140, 184, 189 gender biases in 133–134, 142–144, 164, 165, 184 health care usage for infants 133, 134–140, 174 in Himalayan populations 134, 143, 163–165 Child health (see also growth and development) and infectious disease, in Ladakh 136, 137 at high altitude 15, 30–32, 150 in Ladakh 136, 137, 150, 159 Child mortality 116, 163 deaths in Leh hospital 67, 116–118 in Himalayan populations 116, 163 in Ladakh 5, 118, 135 in Leh hospital 67, 116–118 proximate determinants of 19, 20, 106–108, 178–179 Childhood, anthropology of 17

Index Children’s work 55, 143, 144, 165, 192, 207 Christians 43 Cold (see also hypothermia) 3, 12, 30, 37, 135, 154, 155–156 concern about, in Ladakh 75, 125, 164, 183 seasonality of, in Ladakh 2–4, 40–41, 126, 127–135, 183– 189 Colorado Rockies antiquity of habitation 25 birthweight in (see also PIH) 29, 33, 198 Contraception 109, 144, 165, 170, 193, 202, 203, 208 Cox regression (hazard analysis) 109–110, 112–115, 208 Cultural ecology in Ladakh research 18, 39–40, 184–185, 186 Culture and adaptation 13–14 and reproduction 2 and reproductive health 6 in biocultural analyses 8–10, 13–14 Dairy products consumption by infants (see also supplementation) 129, 137, 155, 163, 183, 198 consumption during pregnancy 5, 76, 94 in Ladakhi diet 4, 49, 60 Darwin, C. 10 Death (see also neonatal, infant, child, maternal mortality) emotional response to 131–132 Diarrheal disease (see gastrointestinal infection) Diet and household resources 5, 168–169, 187–188 at high altitude 31 calories in, in Ladakhi 5, 18, 60–61

231 change during pregnancy, cultural beliefs about 2, 73, 159–160, 168–169 during pregnancy, variation by parity 94 in Ladakh 55–57, 60–61 in pregnancy, beliefs about in Ladakh 5, 89–90, 181, 187–188 in pregnancy, effect on birthweight 76, 88–91, 150, 157, 180–182, 198–199 in pregnancy, questions about 76–77 of infants (see breastfeeding, supplementation) protein in, in Ladakhi 5, 60 recent changes in 190–191 seasonal variation in 4, 31–38, 55–56, 64, 89, 94–95, 138, 181–182 supplementation in pregnancy, effect on birthweight 88, 198 Disease deﬁnition of, in medical anthropology 11 Dogra invasion 46 Domestic pollution (see also pollution; carbon monoxide) 18, 99, 126–127, 182–200 and respiratory disease 99, 126–127, 135, 156, 183–189 compromising infant health 99, 126–127, 135, 156, 183–189 Dowry 51, 134, 165–166, 167 Ecology of infancy 17, 19 in Ladakh 2, 39–41 Education as a determinant of birth outcome 96, 115, 192 of women and their husbands in sample 76–80, 82 EGA (see gestational age) 71, 76–83, 114, 149

232

Index

Eggs consumption during pregnancy 76–88, 89 Emotions (see also attachment) in marriage 51 in relation to childbearing 6, 57, 67, 131–132 Employment and household absenteeism 52–53, 95–96, 163, 181, 190 men’s 52, 53, 79–80, 95–96, 190 nonagricultural 52–53, 79–80, 95–96, 157, 181–190, 192–201, 208 women’s 46–47, 52, 80, 190 Environment (see also high altitude; Ladakh) deﬁnition of, in biocultural analyses 12–13 relation to health 7, 12–13 social (see socioecology, stratiﬁcation, poverty) Ethnicity in relation to birth outcome in Ladakh 84, 85–96, 115, 181 of women in the sample 76–77, 79 Ethnographic human biology 19–21, 24 Evil eye (mik-ya) 101–103, 108, 125 Evolutionary theory 7–10, 182 and biocultural approaches 7–8, 9, 192–193 Family (see also household, marriage) rise of nuclear, in Ladakh 190 Family planning 109, 144, 165, 170, 193, 202, 203, 208 Farmer, P. 23, 193–194, 195 Fat deposition (see also skinfolds) fetal, at high altitude 28, 85 Fertility 15–17, 105 at high altitude 35–37, 38 in Ladakh (see also contraception) 5–6, 37, 186, 202–203 Fetal death 15, 105, 172 at high altitude 32–35, 36–37, 149, 152, 161

in Ladakh 37, 67, 82, 115–116, 118, 199 Fetal growth (see also birthweight; Ponderal Index; neonates) 32, 72, 85, 86 affected by work during pregnancy 91–96, 99–100, 150, 161–164, 180–182, 192–198 affected by diet during pregnancy 76, 88–91, 150, 157, 180–182, 198–199 at high altitude 4, 27–30, 37, 91, 149 in Ladakh 86–92, 114, 149, 154, 174, 182–187 Fever 14, 154 Fieldwork anxiety about 69–70 in Ladakh 21, 61–63, 70 reﬂexivity about, in human biology 21, 61–62 Fitness (see reproductive success) 4–13, 27, 105 Food taboos for pregnant women 5, 89–90, 181, 187–188 Formula feeding (see also supplementation; breastfeeding) 128, 129 among Ladakhi infants 129 Gastrointestinal infection as a cause of infant morbidity 121–124, 135–137, 163, 183–201 as a cause of infant mortality 35, 119, 135–137, 155, 161–163, 183–201 at high altitude 31–32, 35, 60, 155 gender differences in 119, 122–123, 176, 184, 189 seasonality of 126, 155, 183, 184 Gender and bias in childcare 133–134, 142–144, 164, 165, 184 bias in health, in North India 133–134, 166–167, 168

Index desire for sons or daughters 142–143, 144, 165 differences in health, in Ladakh 5, 122–123, 138–139 differences in infant morbidity 122–123, 134 differences in infant mortality 34, 111–112, 114–115, 161–174 differences in infanticide 53, 154 Gene ﬂow 10–11, 26 Gene pool 1, 8, 49 Genetic adaptation to high altitude 25–27, 185–186, 206 drift 10–11, 26 Gestational age 71, 76, 83, 114, 149 as a predictor of birthweight 73, 86, 92, 180 by last menstrual period 76, 83 effect on infant mortality 33, 111–114, 182 Goiter (see also iodine) 16, 61 Goldstein, M. C. 37, 51, 53, 134, 165, 186 Growth and development 10–15, 32, 187 in relation to socioeconomic status 15, 30–31, 150, 156, 207 in the Andes 15, 30–32, 150 in the Himalaya 157, 163 of Ladakhi children 136, 137, 150, 159 Gur-gur cha 53–56, 90–91, 137 Haas, J. D. 4, 28, 83–85, 90, 114, 147–149, 150 Han Chinese birth outcome of, in Tibet 4, 29, 157 birthweight, compared with Ladakhis 157, 161 Hazard analysis (Cox regression) 109–110, 112–115, 208 Head circumference 71, 74, 111–112, 149 Health as a measure of adaptation 7–8, 15

233 as a relative state 11–12 biological indices of 7 deﬁnitions of 11–12 in Ladakh 5, 60–61 policy, and biocultural analysis 23–24, 178–179, 196–197 problems, encountered by pregnant women in Ladakh 77, 100, 178, 203 Health care (see also biomedicine, Tibetan medicine) 57–60 seasonality in biomedical care usage 120, 140 usage, during pregnancy 100–104 usage, for infants 133, 134–140, 174 Health interventions medical anthropological perspectives on 193–195, 196–197 to decrease neonatal and infant mortality in Ladakh 23–24, 193–203 to improve reproductive health in Ladakh 23–24, 193–203 to increase birthweight in Ladakh 197–200 to reduce negative effect of hypoxia on birth outcome 140–141, 197, 198–201 Health policy (see also health interventions) and adaptation 23–24, 178–179, 196–197 different perspectives on high neonatal mortality in Ladakh 193–195, 201–202 Height of Ladakhi children 136, 137, 150, 159 of mothers in sample 73–82, 83, 84–159 Hemoglobin (see also anemia) 12–15, 26, 27, 76, 169 effect on birthweight 90 of mothers in sample 83, 90, 157, 159

234

Index

High altitude (see also hypoxia) adaptation to 26–27, 105 average birthweight at 4, 27–30, 146–147 cardiovascular system 26, 30 child health at 15, 30–32, 150 deﬁnition of 25 diet at 31 ecological conditions at 3, 25 fertility at 35–37, 38 fetal death at 32–35, 36–37, 149, 152, 161 fetal growth at 4, 27–30, 37, 91, 149 gastrointestinal infections, prevalence at 31–32, 35, 60, 155 history of habitation at 25–26, 150–151, 160–161, 185–186 hypoxia at 3, 25 infant mortality rates at 32–35, 152, 161, 166 infectious disease patterns at 25, 31, 32, 137, 155, 163–176 menstruation at 3, 36 ovulatory function at 36, 37 oxygen saturation among infants, at 35, 126, 127, 197 oxygen saturation at 27 population genetic characteristics at 25–26, 185–186 populations living at 17, 25–26 pregnancy-induced hypertension, at 28, 91, 182 reproduction at 3–4, 27, 35–37 respiratory infection at 31–32, 35, 135 Himalaya 2, 17–21, 51–62 child growth and development in 157, 163 childcare patterns in 134, 143, 163–165 infant mortality in 161–166 populations living in 2, 18, 156–166 Hookworm 31, 169, 176

Hospital age pattern of pediatric admissions 122–123, 134–135, 183 birth records 69, 115–116, 206 causes of infant mortality at 119–121 in Leh 52, 58–59, 101, 140, 171, 198 neonatal admission diagnoses 122, 123 pediatric admissions, causes of 124–138 pediatric records 69, 116 usage of, for birth 63–66, 69, 81, 118, 132, 175, 199 Household absenteeism, due to employment 52–53, 95–96, 163, 181, 190 and production of health 18–19, 21, 50, 106, 125 as adaptive locus 15, 16 as developmental niche 18, 125, 184 as locus of production and reproduction 19, 185, 186–189 distribution of resources within 18–19 dynamics 50–53, 125, 190 gardens 49, 53, 188 membership, and individual autonomy 51, 61, 132, 188–189, 190, 206 resources, and diet 5, 168–169, 187–188 structure 50, 52, 76, 190, 207 Human biology 17 ethnographic 19–21, 24 Huss-Ashmore, R. 12–14, 16 Hypertension during pregnancy 28, 91, 198–200 during pregnancy, at high altitude 28, 91, 182 during pregnancy, effect on birthweight 28, 91, 182, 198 in Ladakh 191 related to salt consumption 61, 91

Index Hypothermia (see also cold) 120–126, 127, 161–164, 183 Hypoxemia (see also oxygen saturation) 35, 126, 127, 197 Hypoxia (see also high altitude) 3–4, 41, 106, 189, 193 adaptation to 26–27, 105 and physical work 4–5, 6, 94, 182 at high altitude 3, 25 effect on birthweight 4, 27–30, 32, 73, 105–106, 124, 174, 175, 182, 185–186 effect on human biology 3, 4–5, 25, 26–27 effect on reproduction 3–4, 27, 35–37 genetic adaptations to 25–27, 185–186, 206 interaction with respiratory disease 34–35, 124–176, 183 interventions to reduce effect on birthweight and neonatal mortality 140–141, 197, 198–201 population variation in response to 27, 105–106, 146–147, 150–151 social factors inﬂuence variation in response to 3, 4, 30–31, 105–106, 185–186 ventilation as an adaptation to 26, 27–29 Illegitimate births 132–133, 143, 164 Illness deﬁnition of, in medical anthropology 11 India birthweight in, compared to Ladakh 167–173, 174 census of 77–78, 80, 206 conﬂict with China 46, 205 conﬂict with Pakistan 46, 63–202, 205 gender bias in health 133–134, 166–167, 168

235 government investment in Ladakh 21, 46–47, 190–202 infant mortality, compared to Ladakh 169–170, 171–177, 195–196 Ladakh, integration into 18, 49, 166, 176–177, 189–190, 191–192 map of 40 maternal characteristics, compared to Ladakh 168, 174 poverty and reproductive health 166–167, 168–169 reproductive health in 166–174 Indian migrants to Ladakh 79, 96 birthweight of 97, 98, 185 mothers, characteristics of 98 Indian military Ladakh Scouts 52, 79–94, 95 presence in Ladakh 21, 47, 60, 190–202 Individualism subsumed to household 51, 61, 132, 188–189, 190, 206 Indus River 41, 42, 43 Inequality (see also stratiﬁcation; socioeconomic status; women’s status; gender) 1–8 Infant health (see infant morbidity) Infant morbidity affected by mobility 135, 139, 184, 195 anemia 138 caused by gastrointestinal infection 121–124, 135–137, 163, 183, 201 caused by respiratory disease 135–136, 138–139, 155, 183, 200–201 causes of hospital admissions 124–138 diet and (see breastfeeding; supplementation) gender differences in 122–123, 134 malnutrition as a cause of 119, 121–123 seasonality of 123

236

Index

Infant mortality at high altitude 32–35, 152, 161, 166 birthweight, effect on in general 17, 27, 32, 72, 105, 107, 171–173 birthweight, effect on in Ladakh sample 6, 111–115, 151–154, 179, 186 birthweight, effect on modiﬁed at high altitude 4, 32–34, 111, 151–153 causes of, at Leh hospital 119–121 compared with other Himalayan populations 161–166 compared with the Andes 151–156 deaths in the hospital 67, 116–118 deaths in the study 110–111 gastrointestinal disease, as a cause of 35, 119, 135–137, 155, 161–163, 183, 201 gender differences in 34, 111–112, 114–115, 161–174 gestational age, effect on 33, 111–114, 182 global patterns of 107, 123–124, 195–197, 208 in North India 169–170, 171–177, 195–196 in the Andes 34, 151–156 Ladakhi perspectives on infants at risk 113 pattern of, in Ladakh 6, 116–119, 151, 153–154, 182, 183–184 policies to reduce, in Ladakh 23–24, 193–203 Ponderal Index, effect on 111, 113–114, 182 protein-energy malnutrition as a cause of 119, 121–123 proximate determinants of, in general 19, 20, 106–108, 178–179 proximate determinants of, in Ladakh 179–184, 193–195 rate (IMR) 195–197, 207, 208 rates, in Ladakh 5–6, 60, 116–117, 186, 187–196

respiratory disease, as a cause of 35, 111, 119–120, 155, 161–163, 183, 200–201 seasonality of 115–116, 119–121, 147, 183 septicemia, as a cause of 119, 127, 175, 182–183, 194–200 Infanticide 133, 154 gender differences in 53, 154 Infectious disease (see also respiratory, gastrointestinal infection; septicemia) 1, 11–12, 31–33, 73, 106–135, 137–140, 187–191 among Ladakhi children 136, 137 at high altitude 25, 31, 32, 137, 155, 163, 176 gender differences in 119, 122–123, 176, 184, 189 seasonality of 126, 155, 183, 184 Inheritance 51, 52, 142–143, 185 Interventions (see health interventions) Interviews with mothers 75–77, 108 Iodine (see also goiter) 15–16 Islam (see religion; Muslims) Jammu-Kashmir 2 Jaundice, neonatal 71, 110, 133 Kargil (western Ladakh) 43, 77 Kashmir 2 India-Pakistan conﬂict over 46, 63–202, 205 road, connecting Ladakh 43, 46 Kerosene poisoning among children 140, 191 Labor (see children’s work; work, workloads, childbirth) Lactational infecundability 35, 105–137 Ladakh as a political entity 2, 46 average birthweight in 5–6, 82, 83–85, 111, 146, 182

Index Buddhism in 43–45, 51–52, 131 comparative perspectives on 17–18, 22–23, 39–40, 85, 146–177 cultural ecology perspectives on 18, 39–40, 184–185, 186 culture 2 ecology 2, 39–41 geography 2, 39–43 health in 5, 60–61 history of 45–49 infrastructural development in 21, 46–47, 190–202 integration into modern India 18, 49, 166, 176–177, 189–190, 191–192 language in (Ladakhi) 63, 65 map 40–43, 205 migration into 45, 47 population distribution 41–43 population growth in 6, 47, 51, 185–193 population origins 45, 150, 185, 206 rainfall 2, 41 religious conﬂict in 47–49, 57, 79–90, 98, 181, 190 road access to 43, 46 seasonality in climate 2–4, 40–41, 126, 127–135, 183–189 tourism 21, 41–43, 47, 95, 159, 190–192 urbanization 47, 191–192 Ladakh Scouts 52, 79, 94, 95 Lahdol, Dr. Tsering 63, 68–69, 81–129, 132–140, 171, 190–199, 200–202, 207 Lama 51–52, 57, 101, 125, 133–141 Landholdings 49, 76–99, 190 and birth outcome 89, 95, 115, 180–181, 192 LEDeG 201 Leh (capital of Ladakh) 41–43, 45–47, 78–79, 140, 190, 191 District 77–78 District, Hill Council status 49

237 hospital (Sonam Norbu Memorial) 52, 58–59, 101, 140, 171, 198 population growth in 47, 191–192 sanitation in 47, 136–137, 191–192, 201 Leh Nutrition Project (LNP) 60, 138–140 Length at birth 28–29 at high altitude 28–29 in Ladakh 74–84, 111–114, 149 Leonard, W. R. 3–4, 15, 30–31, 94 Levine, N. E. 51, 134–143, 163–165 Lhaba/lhamo 6, 26, 58, 69 usage during pregnancy 100, 200 usage for infants and children 141 Lhu 49, 136, 201 Life expectancy 12, 205 Little, M. A. 15–17, 26, 185 Logistic regression compared with hazard analysis 110, 207 Low birthweight (LBW) 32, 72, 111, 167, 168–171, 197 effect on mortality, in general 17, 27, 32, 72, 105, 107, 171–173 effect on mortality at high altitude 33–34, 113, 153, 161 effect on mortality, in Ladakh 6, 111–115, 151–154, 179, 186 frequency at high altitude 29, 33, 147, 148–158 frequency in Ladakh 6, 83, 84, 113, 138, 147, 148, 174, 182 Lung volume (see also chest circumference) among high altitude populations 26, 27 Malaria 31, 137 Malnutrition 1, 30–31, 32, 60, 106, 195 as a cause of infant morbidity 119, 121–123 as a cause of infant mortality 119, 121–123

238

Index

Malnutrition (cont.) among Ladakhi children 136, 137, 150, 159 among mothers in Ladakh (see also maternal anthropometry) 5–6, 33, 161 Man in the Andes 17 Marriage (see also polyandry) 2 age at 100, 169, 170 in Ladakh 37, 50–51, 52 Maternal anthropometry 73–75, 82–83, 108 compared with Andean women 148, 149–150 compared with North Indian women 168, 174 compared with Tibetan and Sherpa women 157–160, 161 differences between Muslims and Buddhists 98–100 ethnic variation in, in Ladakh 97, 98 height 73–82, 83, 84, 159 skinfolds 73–83, 84, 97, 149, 157, 159–160 weight 73–82, 84–97, 149–157, 159–160, 180 Maternal characteristics age 72–76, 82, 157–159 age at ﬁrst birth 100, 169, 170 age at marriage 100, 169, 170 of migrants in Ladakh 98 of North Indian women 168, 174 parity 72–76, 81, 149, 157–159 Maternal factors predicting birthweight, in Ladakh 86–100, 101, 149–151, 179–182, 197–200 related to birthweight, in general 6, 71–73, 86–87, 107 Maternal health 5–6, 33, 161 link to infant health and survival 6, 72, 107, 115, 197 Maternal mortality 67 McKeown, T. 194 Measles 121–122, 137 vaccination 137

Meat consumption during pregnancy 5, 76, 88, 89 in Ladakhi diet 49, 56, 57 Medical anthropology 62 and health policy 193–195, 196–197 and human adaptability 7–23 Menarche at high altitude 35–36 Menstruation at high altitude 3, 36 last menstrual period (LMP) 76–83 Methods 24-hour dietary recall 76–77 analysis of mortality 108–110 diet and activity questions 76–77 interviews with mothers 75–77, 108 maternal and neonatal anthropometry 73–75, 82–83, 108 reproductive histories 76, 115–118 Migrants birth outcomes of, in Ladakh 97, 98, 185 birthweight of, at high altitude 4, 29, 98, 146–149 descend to low altitude for pregnancy 3–4 into Ladakh 79, 96 maternal characteristics of, in Ladakh 98 Spanish, at high altitude 3, 35 Military employment (see men’s employment; Ladakh Scouts) Indian, in Ladakh 21, 47, 60, 190–202 Millard, A. V. 19–20, 106–108, 179 Milk (see dairy products) Miscarriage (see fetal death) 15, 105, 172 Moore, L. G. 3–4, 25, 26–27, 90, 91, 147–151, 157

Index Morbidity (see also infant morbidity, child health) causes of, in Ladakh 60–61 Mortality (see also fetal, infant, neonatal, perinatal, postneonatal, child mortality) causes of, in Ladakh 60, 192 maternal 67 Muslims 43, 45 birthweight of, in Ladakh 98–100, 115, 181, 192 conﬂict with Buddhists, in Ladakh 47–49, 57, 79, 90, 98, 181, 190 diet and workloads during pregnancy of, in Ladakh 99–100, 181 maternal characteristics of, in Ladakh 98–100 Mutation 10, 11, 26, 37, 38 Naming of infants 125, 184 Natural selection (see also adaptation, evolutionary theory) 10–13, 186 Neglect of infants 132–134, 154, 164 Neonatal health 125–134 hospital admission diagnoses 122, 123 Neonatal mortality as a proportion of infant mortality 6, 116–119, 151, 153–154, 182, 183–184 compared with North India 169–170, 171–177, 195–196 compared with other Himalayan populations 161–166 compared with the Andes 151–156 gender differences in, in Ladakh 117, 182 in Ladakh 6, 112–118, 161, 182–183, 186–193 in the Andes 34, 151–156 in the Leh hospital 67, 116–118 policies to reduce, in Ladakh 23–24, 193–203 policy perspectives on 193–195, 201–202

239 Ponderal Index, effect on 111–114, 182 predicted by birthweight in Ladakh 6, 111–115, 121, 151–154, 182 seasonality of 115–116, 119–121, 147, 183 Neonates anthropometry of 73–75, 82–83, 108 chest circumference 74, 97, 111–112 estimated gestational age of 71, 76–83, 114, 149 head circumference 71, 74, 111–112, 149 length 74, 84, 111–114, 149 Ponderal Index 86, 93, 150 range of variation in 71–72 size of, determinants of 6, 71–73, 86–87, 107 skinfolds (fatness) 71, 74, 111–112, 149 weight (see birthweight) 5–6, 82, 83–85, 111, 146, 182 Nepal 2, 62, 79, 134–143, 163–165 Nepalis in Ladakh 98 Newborns (see neonates) Norberg-Hodge, H. 5, 18, 51, 185 Nubra Valley 43 Nutrition (see also malnutrition; protein-energy malnutrition) in Ladakh 5, 60–61 Nyingka 100–101 Occupation in relation to infant mortality 115 in the sample 76–79, 96 Osmaston, H. 4, 18, 49, 50, 61, 185 Ovulatory function at high altitude 36, 37 Oxygen 3, 12–18, 26–30, 94, 126–136, 197–198 and biological function 3, 4–5, 25, 26–27 ﬂow to fetus 4, 28, 29–34, 91, 94, 161, 182

240

Index

Oxygen (cont.) saturation, among high altitude infants 35, 126, 127, 197 saturation, at high altitude 27 Parasitic infection among Ladakhi children 136, 137 Parity as a predictor of birthweight 87–88, 93, 117–118, 157 in relation to infant mortality 115–116, 118, 142, 164–169 of women in sample 72–76, 81, 149, 157–159 Patrilineage 50–53, 143 PEM (see protein-energy malnutrition) Perinatal mortality (see also fetal death and neonatal mortality) 117, 118, 121, 154, 161, 171, 200, 205 Placenta morphology and function at high altitude 28 Plasticity (see also adaptability) 7–8, 10 Polyandry 5, 19, 37, 51, 185–186 and women’s status 5, 51, 164 Pollution 37 air, in Leh area 135–136, 191 domestic 18, 99, 126–127, 182–200 outdoor, and infant health 135–136, 191 water contamination 128–136, 137, 163, 184, 191 Ponderal Index (PI) 85–86 comparison of Ladakhis with other populations 148, 149, 150 effect on neonatal/infant mortality risk 111–114, 182 of Ladakhi newborns 86, 93, 150 sex differences in 86 Population deﬁnition of 8–9 growth, in Ladakh 6, 47, 51, 185, 193

history and adaptive status 4, 11, 25–26, 150–151, 185–186 history, at high altitude 25–26, 150–151, 160–161, 185–186 in biocultural approaches 8–11 origins, in Ladakh 45, 150, 185, 206 variation in response to hypoxia 27, 105–106, 146–147, 150–151 Postneonatal mortality (see also infant mortality) 6, 34, 119–134, 135–142, 151–153, 156, 183–184 Postpartum infecundability 35, 105, 137 seclusion 108, 125, 173–175, 183 Potatoes 4, 31, 49, 65, 155 Poverty 1, 15, 16, 201 and reproductive health, in north India 166–167, 168–169 biology of 7–8, 9 Pre-eclampsia (see pregnancy-induced hypertension) Pregnancy (see also fetal growth, birth outcome) and women’s workloads, in Ladakh 4–6, 76, 77–89, 150, 198–199 as a time of dietary change, cultural beliefs about 2, 73, 159–160, 168–169 complications of, at high altitude 28 descent to low altitude, by migrants 3–4 diet during, beliefs about in Ladakh 5, 89–90, 181, 187–188 diet during, in Ladakh 76, 88–91, 150, 157, 180–182, 198–199 diet during, variation by parity 94 hypertension in 28, 91, 182 placental function and morphology at high altitude 28 problems encountered during 77, 100, 178, 203

Index studies of dietary supplementation and effect on birthweight 88, 198 usage of health care during 100–104 views of, by Ladakhis 100–103, 181, 187–188 Pregnancy-induced hypertension (PIH) 28, 91, 198–200 effect on birthweight 28, 91, 182–198 frequency at high altitude 28, 91, 182 Prematurity (see also gestational age) 32, 33–205 Prenatal care 1, 68, 76, 100, 101–103, 157, 170, 171–175, 181 effect on birthweight 101, 181–200 Progesterone levels in high altitude populations 36 Protein in Ladakhi diet 5, 60 Protein-energy malnutrition (see also malnutrition) among Ladakhi children 136, 137, 150, 159 as a cause of infant morbidity 119, 121–123 as a cause of infant mortality 119, 121–123 Proximate determinants analysis 19, 195 historical component added 19, 179 of fertility 35 of infant mortality in Ladakh 179–184, 193–195 of infant/child mortality, in general 19, 20, 106–108, 178–179 Rainfall 2, 41 rdun 125 Reﬂexivity in human biological research 21, 61–62

241 Religion Buddhism in Ladakh 43–45, 51–52, 131 Buddhist–Muslim conﬂict 47–49, 57, 79, 90, 98, 181, 190 Buddhist–Muslim differences in birth outcome 98–100, 115, 181, 192 Buddhist–Muslim differences in diet and workloads 99–100, 181 Buddhist–Muslim differences in maternal characteristics 98–100 distribution, among women in sample 76–77 Reproduction (see also fertility, pregnancy, birth outcome) 1–2, 15, 187 at high altitude 3–4, 27, 35–37 competing with production in Ladakh 6, 186–189 culture and 2 population variation in 1–2 Reproductive ecology 17 Reproductive health 5–6, 16–19, 63–64, 186–189, 192–203 comparative perspectives on 17–18, 22–23, 39–40, 85, 146–177 conducting research on, in Ladakh 21, 61–63, 70 factors inﬂuencing, in Ladakh 1–2 in North India 166–174 policies to improve, in Ladakh 23–24, 193–203 Reproductive histories 76, 115–118 Reproductive success 4–13, 27, 105 Research questions 17, 21–23, 62–63 Resources (see also socioeconomic status, poverty) access to, and health 12–13 distribution, within the household 18–19

242

Index

Respiratory infection 18, 34, 60, 191 and infant health 135–136, 138–139, 155, 183, 200–201 as a cause of neonatal/infant mortality 35, 111, 119–120, 155, 161–163, 183, 200–201 as a cause of pediatric hospital admission 121–123, 124 at high altitude 31–32, 35, 135 gender differences in 119, 122–123, 176, 184, 189 interactions with domestic pollution 99, 126–127, 135, 156, 183–189 interactions with hypoxia at high altitude 34–35, 124, 176, 183 seasonality of 126, 155, 183, 184 Respiratory Syncitial Virus (RSV) 127–128 Rockies (see Colorado) Salt (see also gur-gur cha) and hypertension 61, 91 in Ladakhi diet 56–61, 91, 191 Sample characteristics 21, 73–74, 77 biases in 63, 77–81 education 76–80, 82 ethnic composition 76–77, 79 occupation status 76–79, 96 religion 76–77 representativeness of Ladakhi population 63, 77–81 residence patterns 77–79, 80–81 sex ratio in 81, 149 socioeconomic 77–81 Scheper-Hughes, N. 17, 108, 194, 201 Seasonality at high altitude 38 in biomedical health care usage 120, 140 in births 116–120, 121 in birthweight, in Ladakh 94–95, 120, 181–182 in climate in Ladakh 2–4, 40–41, 126, 127–135, 183–189

in food availability and diet 4, 31–38, 55–56, 64, 89, 94–95, 138, 181–182 in infant morbidity 123 in infectious disease 126, 155, 183, 184 in neonatal and infant mortality 115–116, 119–121, 147, 183 in workloads 5, 38, 49, 91–92, 94–95, 129, 140, 181–182, 202 Septicemia as a cause of neonatal mortality 119, 127, 175, 182–183, 194–200 Sex (see also gender) differences in infant mortality 34, 111–112, 114–115, 161–174 differences in infant/child morbidity 122–123, 134 differences in birthweight 29, 73–83, 84–85, 148 differences in birthweight, at high altitude 29, 149, 157 Sexually transmitted disease (STDs) 37 Shamanism (see also lhamo/lhaba) 6, 26, 58, 69 Sherpa populations birthweight among 30, 92, 147–157, 159–160 pregnancy and birth outcome, compared with Ladakh 147–158, 166 maternal characteristics of 157–160, 161 Skinfolds of mothers 73–83, 84–97, 149–157, 159–160 of neonates 71, 74, 111–112, 149 Smoking (see also carbon monoxide; domestic pollution) and birth outcome 73, 99, 150, 157 Socioecology (see also stratiﬁcation, poverty, household) 2–6, 8, 125 Socioeconomic status (see stratiﬁcation, sample characteristics)

Index and growth in the Andes 15, 30–31, 150, 156, 207 and health 12, 15–16, 194–195, 201–202 and variation in birth outcome 1, 19, 72, 106–108, 150–154, 181–201 and variation in birth outcome, in Ladakh 18, 95–96, 97–98, 115, 150, 156, 180–181, 186 inﬂuences responses to hypoxia 3, 4, 30–31, 105–106, 185–186 South America (see Andes) 3, 15–17 Spanish in the Andes, problems with reproduction 3, 35 Srinagar (see also Kashmir) Stillbirths (see fetal death) 37, 67, 82, 115–116, 118, 199 Stratiﬁcation (see also poverty, inequality) and health 12, 15–16, 194–195, 201–202 in relation to birth outcome in Ladakh 18, 95–96, 97–98, 115, 150, 156, 180–181, 186 and growth in the Andes 15, 30–31, 150, 156, 207 Stressors 3–10, 25 adaptation to 7–8, 14 Subsistence (see agriculture) Supplementation of Ladakhi infants 129, 137, 155, 183, 184 studies of dietary supplementation and birthweight 88, 198 24-hour dietary recall 76–77 Tamang pregnancy among 160, 161 Tetanus 127, 173, 175, 207 vaccination against, during pregnancy 76, 127, 170 Thomas, R. B. 12–13, 14–16 Tibet (see Himalaya) 2, 46, 98 history of habitation 25, 160

243 Tibetan populations (see Sherpas, Tamang, Himalaya, Nepal) adaptations to hypoxia 27 average birthweight among 29–30, 147–157 birth outcome, compared with Ladakhis 157–166, 185 birthweight of, in Ladakh 96–98, 157–159, 181–185 breastfeeding patterns among 163–164 child growth and development in 157, 163 childcare patterns in 134, 143, 163–165 infant mortality in 161–166 maternal characteristics of 157–160, 161 migrants to Ladakh 45, 46, 79 mothers, in Ladakh 97–98 plateau 2, 39–40 workloads of women in 157, 159–160 Tibetan medicine 6–7, 15, 57–58 and infant health 141–142 usage during pregnancy 100–101, 199 views on pregnancy 101–103, 207 Tourism contributes to social stratiﬁcation 192, 202 in Ladakh 21, 41–43, 47, 95, 159, 190–192 Traditional Birth Attendants (TBAs) 103–104, 170–173, 199 TransHimalayan (Tibetan) plateau 2, 39–40 Typhus 31 Urbanization 47, 191–192 Urinary tract infection 99 Uterine blood ﬂow 4, 28, 29, 34, 91, 94, 161, 182 UV (solar) radiation 3, 25–26, 37, 41

244

Index

Vaccination BCG and polio 125 measles 137 tetanus toxoid, during pregnancy 76, 127, 170 Ventilation as an adaptation to hypoxia 26, 27–29 Vitzthum, V. J. 17, 35, 36, 154–155, 208 Water as a limiting resource in Ladakh 41, 186, 205–206 contamination of 128, 136, 137, 163, 184, 191 Weaning in relation to child health 139 of Ladakhi children 139 Weight (see also birthweight) of mothers in the sample 73, 82, 84–97, 149–151, 159–160, 180 Well-being 11, 13–14, 27 Women (see maternal anthropometry, characteristics) employment of 46–47, 52, 80, 190 health of, linked to infant health 6, 72, 107, 115, 197 role, in household 53–55, 93–94 workloads of, during pregnancy 4–6, 76, 77–89, 150, 198–199 Women’s status and health 5, 53, 188–189 in Ladakh 5, 53–55, 188 related to Buddhism 5, 52 related to polyandry 5, 51, 164 Work children’s 55, 143, 144, 165, 192, 207

under hypoxic conditions 4–5, 6, 94, 182 value of women’s productive work 5, 186–189 Workloads 1, 4, 5, 53 beliefs about during pregnancy, in Ladakh 5, 89–90, 181, 187–188 conﬂict between production and reproduction 6, 186–189 during pregnancy and household dynamics 5, 93–96, 169, 180–181, 186–189 during pregnancy, questions about 76–77 effect on birthweight 91–96, 99–100, 150, 161–164, 180–182, 192–198 in relation to infant feeding 129–137, 155–163, 183, 198 of Himalayan women 157, 159–160 of primiparous women 53, 93–94, 169, 186–188 seasonal variation in 5, 38, 49, 91–92, 94–95, 129, 140, 181–182, 202 social factors contributing to variation in 18, 95–96, 97–98, 115, 150, 156, 180–181, 186 women’s contributions to agriculture 5, 50, 52–53, 76–77, 95, 186–187 women’s during pregnancy 4–6, 76, 77–89, 150, 198–199 World Health Organization (WHO) 6, 72, 85, 163, 174 Zangskar 41–43, 77

Other books in the series (continued from page iii )

8 Margaret Lock, Allan Young, and Alberto Cambrosio (eds.), Living and Working with the New Medical Technologies: Intersection of Inquiry 11 Janis H. Jenkins and Robert John Barrett (eds.), Schizophrenia, Culture, and Subjectivity: The Edge of Experience 12 Andrea S. Wiley, An Ecology of High-Altitude Infancy: A Biocultural Perspective

The series also includes a group of theme books that are designed as course material for advanced undergraduate and graduate students and that synthesize emerging scholarship from relatively new subﬁelds or reinterpret the literature of older ones. Published theme books are 9 Daniel Moerman, Meaning, Medicine and the “Placebo Effect” 10 Susan Reynolds Whyte, Sjaak van der Geest, and Anita Hardon, The Social Lives of Medicines

An Ecology of High-Altitude Infancy: A Biocultural Perspective (Cambridge Studies in Medical Anthropology)

Menopause: A Biocultural Perspective (Studies in Medical Anthropology)

The Transmission of Chinese Medicine (Cambridge Studies in Medical Anthropology)

Medical Anthropology in Ecological Perspective: Fifth Edition

Human Biology: An Evolutionary and Biocultural Perspective

Human Biology: An Evolutionary and Biocultural Perspective

Medical Anthropology in Ecological Perspective: Fifth Edition

Epidemiology and Culture (Cambridge Studies in Medical Anthropology)

Cultural Anthropology: An Applied Perspective

Human Ecology: Biocultural Adaptations in Human Communities (Ecological Studies)

Cultural Anthropology: An Applied Perspective

Biological Anthropology An Evolutionary Perspective

Cultural Anthropology: An Applied Perspective

The Ecology of Freshwater Molluscs (Cambridge Studies in Ecology)

Chemical Ecology of Vertebrates (Cambridge Studies in Ecology)

The Ecology of Freshwater Molluscs (Cambridge Studies in Ecology)

Key Topics in Landscape Ecology (Cambridge Studies in Landscape Ecology)

Key Topics in Landscape Ecology (Cambridge Studies in Landscape Ecology)

Spatial Pattern Analysis in Plant Ecology (Cambridge Studies in Ecology)

Key Topics in Landscape Ecology (Cambridge Studies in Landscape Ecology)

Key Topics in Landscape Ecology (Cambridge Studies in Landscape Ecology)

Living and Working with the New Medical Technologies: Intersections of Inquiry (Cambridge Studies in Medical Anthropology)

Riparian Landscapes (Cambridge Studies in Ecology)

Human Senescence: Evolutionary and Biocultural Perspectives (Cambridge Studies in Biological and Evolutionary Anthropology)

Neanderthals and Modern Humans: An Ecological and Evolutionary Perspective (Cambridge Studies in Biological and Evolutionary Anthropology)

Western Diseases: An Evolutionary Perspective (Cambridge Studies in Biological and Evolutionary Anthropology)

Ecological Versatility and Community Ecology (Cambridge Studies in Ecology)

A New Ecology: Systems Perspective

The Social Life of Things: Commodities in Cultural Perspective (Cambridge Studies in Social and Cultural Anthropology)

The Social Life of Things: Commodities in Cultural Perspective (Cambridge Studies in Social and Cultural Anthropology)

Global Ecology in Human Perspective

An Ecology of High-Altitude Infancy: A Biocultural Perspective (Cambridge Studies in Medical Anthropology)

Menopause: A Biocultural Perspective (Studies in Medical Anthropology)

The Transmission of Chinese Medicine (Cambridge Studies in Medical Anthropology)

Medical Anthropology in Ecological Perspective: Fifth Edition

Human Biology: An Evolutionary and Biocultural Perspective

Human Biology: An Evolutionary and Biocultural Perspective

Medical Anthropology in Ecological Perspective: Fifth Edition

Epidemiology and Culture (Cambridge Studies in Medical Anthropology)

Cultural Anthropology: An Applied Perspective

Human Ecology: Biocultural Adaptations in Human Communities (Ecological Studies)

Cultural Anthropology: An Applied Perspective

Biological Anthropology An Evolutionary Perspective

Cultural Anthropology: An Applied Perspective

The Ecology of Freshwater Molluscs (Cambridge Studies in Ecology)

Chemical Ecology of Vertebrates (Cambridge Studies in Ecology)

The Ecology of Freshwater Molluscs (Cambridge Studies in Ecology)

Key Topics in Landscape Ecology (Cambridge Studies in Landscape Ecology)

Key Topics in Landscape Ecology (Cambridge Studies in Landscape Ecology)

Spatial Pattern Analysis in Plant Ecology (Cambridge Studies in Ecology)

Key Topics in Landscape Ecology (Cambridge Studies in Landscape Ecology)

Key Topics in Landscape Ecology (Cambridge Studies in Landscape Ecology)

Living and Working with the New Medical Technologies: Intersections of Inquiry (Cambridge Studies in Medical Anthropology)

Riparian Landscapes (Cambridge Studies in Ecology)

Human Senescence: Evolutionary and Biocultural Perspectives (Cambridge Studies in Biological and Evolutionary Anthropology)

Neanderthals and Modern Humans: An Ecological and Evolutionary Perspective (Cambridge Studies in Biological and Evolutionary Anthropology)

Western Diseases: An Evolutionary Perspective (Cambridge Studies in Biological and Evolutionary Anthropology)

Ecological Versatility and Community Ecology (Cambridge Studies in Ecology)

A New Ecology: Systems Perspective

The Social Life of Things: Commodities in Cultural Perspective (Cambridge Studies in Social and Cultural Anthropology)

The Social Life of Things: Commodities in Cultural Perspective (Cambridge Studies in Social and Cultural Anthropology)

Global Ecology in Human Perspective

Recommend Documents