Prescribing under Pressure: Parent-Physician Conversations and Antibiotics (Oxford Studies in Sociolinguistics)

Prescribing under Pressure OXFORD STUDIES IN SOCIOLINGUISTICS General Editors Nikolas Coupland Adam Jaworski Cardiff ...

Author: Tanya Stivers

17 downloads 594 Views 924KB 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

Prescribing under Pressure

OXFORD STUDIES IN SOCIOLINGUISTICS General Editors Nikolas Coupland Adam Jaworski Cardiff University

Recently Published in the Series:

Talking about Treatment: Recommendations for Breast Cancer Adjuvant Treatment Felicia D. Roberts Language in Time: The Rhythm and Tempo of Spoken Interaction Peter Auer, Elizabeth Couper-Kuhlen, Frank Müller Whales, Candlelight, and Stuff Like That: General Extenders in English Discourse Maryann Overstreet A Place to Stand: Politics and Persuasion in a Working-Class Bar Julie Lindquist Sociolinguistic Variation: Critical Reﬂections Edited by Carmen Fought

Prescribing under Pressure

Parent-Physician Conversations and Antibiotics

Tanya Stivers

3 2007

3 Oxford University Press, Inc., publishes works that further Oxford University’s objective of excellence in research, scholarship, and education. Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With ofﬁces in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam

Copyright © 2007 by Oxford University Press, Inc. Published by Oxford University Press, Inc. 198 Madison Avenue, New York, New York 10016 www.oup.com Oxford is a registered trademark of Oxford University Press All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Oxford University Press. Library of Congress Cataloging-in-Publication Data Stivers, Tanya. Prescribing under pressure: patient-physician conversations and antibiotics / Tanya Stivers. p. cm.—(Oxford studies in sociolingistics) Includes bibliographical references and index. ISBN 978-0-19-531115-0 1. Antibiotics. 2. Antibiotics—Effectiveness. Physician and patient. I. Title RM267.S75 2007 615'329—dc22 2006048256

9 8 7 6 5 4 3 2 1 Printed in the United States of America on acid-free paper

ACKNOWLEDGMENTS

A

lthough only one name appears on the cover of this book, it represents the thoughts, ideas, and contributions of many people. Some I formally acknowledge here, some appear in the list of references, but many who appear in neither place nonetheless helped me along the way by challenging me to consider yet another way of approaching the set of problems that this book brings together. I owe the biggest debt of gratitude to the people who let me study them: the nearly 800 parents and physicians I videotaped over the course of 5 years of ﬁeldwork, much of which was used for the present book. Although the research discussed here relies on videotapes of these interactions, many of these individuals (as well as others) shaped my thinking in quite different ways: parent comments in physician waiting rooms, physician comments or questions before or after parents arrived as well as during ethnographic interviews, and medical assistant comments during the course of a typical day all inﬂuenced my thinking about what happens when a parent walks through the door with a sick child. Each of the subjects in my study granted me just a small amount of time—many would not even remember participating; most would surely be shocked to ﬁnd out that their few minutes would sum to provide me with a career’s worth of data. Still, those minutes add up, and so I thank all of these anonymous individuals for their generosity. My biggest supporter from beginning to end has been John Heritage. He believed in the project from the moment I ﬂoated the idea by him, and his encouragement materially, emotionally, and intellectually was invaluable. I shudder to think how many versions of this material he has read over the years in the form of grant proposals, dissertation chapters, articles, and ﬁnally book chapters. Besides offering tireless support, he taught me how to study social interaction without losing track of either the structures through which interactants accomplish social actions or the fact that interactants have feelings, concerns, and agendas. He challenged me each step of the

vi

ACKNOWLEDGMENTS

way to be more ambitious, to be more creative in my thinking, and to consider the interplay between the microlevel interaction and the macrolevel social context. I am inﬁnitely grateful for the support, education, and contributions that he has provided. Many others to whom I owe a substantial debt of thanks were, in one way or another, connected to John. It was he who introduced me to Rita Mangione-Smith. Rita provided both data and much support for the work after qualitative analyses pointed me toward antibiotics. Rita had independently been investigating parent attitudes toward, and expectations for, antibiotics and had been surveying physicians on their perceptions. Thus, with our work combined, we could, for the ﬁrst time, see whether behaviors derived from conversation analysis could be linked with exogenous survey variables, and the merger was a success. As a student of conversation analysis at the University of California at Los Angeles, I was mentored by both John Heritage and Manny Schegloff. Then and since, Manny has consistently challenged me not to forget that whatever people are doing in a medical encounter, they are doing it Turn Constructional Unit by Turn Constructional Unit, turn by turn, sequence by sequence, as in all forms of social interaction. And time and time again, the importance of understanding each level of orderliness has proven critical to the analysis. To Manny I owe a thank-you for teaching and modeling both rigor and enthusiasm in studying interaction. Both were contagious early on, and have not waned. This book is heavily based on my UCLA dissertation in applied linguistics: Negotiating Antibiotic Treatment in Pediatric Care: The Communication of Preferences in Physician-Parent Interaction (2000). Several of the chapters are based on previously published articles. The ideas of chapter 2 previously appeared as “Presenting the Problem in Pediatric Encounters: ‘Symptoms Only’ versus ‘Candidate Diagnosis’ Presentations,” published in Health Communication (2002b). The ideas in chapter 5 previously appeared in “Parent Resistance to Physicians’ Treatment Recommendations: One Resource for Initiating a Negotiation of the Treatment Decision,” published in Health Communication (2005c). Many of the ideas in chapter 6 appeared in “Participating in Decisions about Treatment: Overt Parent Pressure for Antibiotic Medication in Pediatric Encounters,” published in Social Science and Medicine (2002a). Finally, chapter 7 is a compilation of work based partly on “Non-Antibiotic Treatment Recommendations: Delivery Formats and Implications for Parent Resistance” in Social Science and Medicine (Stivers, 2005b) and partly on Heritage and Stivers (1999), “Online Commentary in Acute Medical Visits: A Method of Shaping Patient Expectations” in Social Science and Medicine. I acknowledge Lawrence Erlbaum and Elsevier for allowing me to incorporate these articles into the present book. Bringing together the previous ideas in the form of this book was most importantly prompted by a conversation with Steve Clayman. It was his timely prodding and support that convinced me to revisit this work and draw it together. Besides prompting the writing, he generously read drafts of the chapters and provided much needed feedback, consistently pushing me to clarify my writing and, in turn, my thinking. I am also grateful to the Max Planck Institute for Psycholinguistics in Nijmegen, The Netherlands, and especially to Steve Levinson, for making the writing of

ACKNOWLEDGMENTS

vii

this book possible. The Language and Cognition Group, particularly Nick Enﬁeld, has through their own work, their way of approaching problems, and their questions, pushed me to come to terms with disciplines and ways of thinking that I had previously been hopelessly ignorant of. This sentiment wove its way into this book and led me to attempt to situate the problem and discussions of it more broadly than I otherwise would have. Much appreciation is owed to my friends and family, who have been (and will probably continue to be) subjected to ranting over the years about antibiotic overprescribing and the structures of social action. Instead of asking me to stop talking about my work, they have engaged with my ideas and, through their own stories and thoughts, prompted me to consider new analytic angles. Special thanks to Ignasi Clemente, Amy Miller, Rob McClinton, Heidi and Luella Hood, Dr. Valentine, Kathi and Milt Schmutz, Jim and Jean Stivers, and Julian and Riley Scaff. Finally, thank you to Nik Coupland, Adam Jaworski, and Peter Ohlin for their support of this book.

This page intentionally left blank

CONTENTS

1. The Miracle Drug: The Context of Modern Antibiotic Usage, 3 2. Foregrounding the Relevance of Antibiotics in the Problem Presentation, 23 3. Alternative Practices for Asking and Answering History-Taking Questions, 51 4. No Problem (No Treatment) Diagnosis Resistance, 77 5. Treatment Resistance, 105 6. Overt Forms of Negotiation, 131 7. Physician Behavior That Inﬂuences Parent Negotiation Practices, 155 8. Conclusion, 185 Appendix: Transcript Symbols, 195 Notes, 199 References, 203 Index, 219

This page intentionally left blank

Prescribing under Pressure

This page intentionally left blank

1

The Miracle Drug The Context of Modern Antibiotic Usage

The History of Medicine 2000 BC—Here, eat this root. 1000 BC—That root is heathen. Here, say this prayer. 1850 AD—That prayer is superstition. Here, drink this potion. 1920 AD—That potion is snake oil. Here, swallow this pill. 1945 AD—That pill is ineffective. Here, take this penicillin. 1955 AD—Oops . . . bugs mutated. Here, take this tetracycline. 1960–1999 AD—39 more “oops” . . . Here, take this more powerful antibiotic. 2000 AD—The bugs have won! Here, eat this root. —Anonymous

I

n 2000, primary care physicians in the United States handed out approximately 126 million prescriptions for antimicrobials (McCaig, Besser, & Hughes, 2003). Basic arithmetic shows this to be approximately 2.5 billion doses consumed by ambulatory care patients alone. Although in some respects rates of antimicrobial drug use have fallen in the recent past, the annual population-based rate of prescribing in the United States remains 461 prescriptions per 1,000 people (McCaig et al., 2003). In pediatrics, the primary care specialty with the highest rate of prescribing, 235 of every 1,000 medical visits result in an antibiotic prescription (McCaig et al., 2003). Unfortunately, many of these prescriptions are for the treatment of viral illnesses. Because antibiotics are ineffective against viruses, such prescriptions are inappropriate, and their prevalence threatens the effectiveness of antibiotics in treating bacteria that cause pneumonia, strep throat, and ear infections (Streptococcus pneumoniae), some of the most common childhood illnesses. This book asks why the problem of inappropriate antibiotic prescribing persists and seeks answers by investigating the

3

4

PRESCRIBING UNDER PRESSURE

details of interactions between pediatricians and parents in visits for children with symptoms of an upper respiratory tract infection.

History of Antibiotics The history of antibiotics is relatively short. By most accounts, penicillin was discovered by British scientist Alexander Fleming in 1928, and initial results were published less than a year later (Fleming, 1929). For various reasons, it was not until 1942 that the British and Americans began mass-producing the drug. Clinical use became widespread during World War II, when penicillin was heavily marketed to the public using what we now call “direct to consumer” advertising, such as “Thanks to penicillin, he [showing a picture of an American soldier on the ground] will come home” (Levy, 1992: 10). At this time, penicillin was widely heralded as a miracle drug. People were astonished at the ability of antibiotics to cure illnesses overnight that had previously been fatal. But the golden era of antibiotics was to be short-lived. Fleming himself noted early on that the drug required careful dosing and that bacteria mutate quickly in response to exposure. When accepting his 1945 Nobel Prize, Fleming warned of bacterial resistance. Although his concerns were primarily with underdosing (a problem that persists both through noncompliance and, primarily in developing countries, through lack of knowledge or proper antibiotic supplies), his illustration of the threat of bacterial resistance is still relevant today: Here is a hypothetical illustration. Mr. X has a sore throat. He buys some penicillin and gives himself, not enough to kill the streptococci but enough to educate them to resist penicillin. He then infects his wife. Mrs. X gets pneumonia and is treated with penicillin. As the streptococci are now resistant to penicillin the treatment fails. Mrs. X dies. Who is primarily responsible for Mrs. X’s death? Why Mr. X whose negligent use of penicillin changed the nature of the microbe. (Fleming, 1945)

As it turns out, the resistance problem is actually worse than Fleming envisioned in two ways. It turned out that even restrained use of antibiotics will generate bacterial resistance over time. Moreover, the resistance problem emerged very quickly. As early as 1946, just a few years after mass production began, when penicillin was still available without a prescription in the United States, there were reports of penicillin-resistant bacteria probably due in no small part to the misuse that Fleming had been concerned about (Levy, 1992). But at that point, new antibiotics were quickly coming onto the scene. Fast forward just 30 years to the early 1970s, and antibiotic resistance had already come to be considered a real public health threat. Strains of bacteria that cause meningitis and ear infections in children and a strain that caused gonorrhea once again proved fatal. Both had previously been treated successfully with penicillin or a derivative (Levy, 1992). At present, just over 60 years since the beginning of wide-scale antibiotic use, the growing problem of bacterial resistance to antibiotics is widely recognized as one

THE MIRACLE DRUG

5

of society’s greatest health threats (Adam, 2002; Baquero, Baquero-Artigao, Canton, & Garcia-Rey, 2002; Doern & Brown, 2004; Harbarth, Albrich, & Brun-Buisson, 2002; Jacobs, Felmingham, Appelbaum, Grüneberg, & Group, 2003; McCaig & Hughes, 1995; Neu, 1992; Reichler et al., 1992; Schwartz, 1999; Smolinski, Hamburg, & Lederberg, 2003; Whitney et al., 2000; Wise et al., 1998). As discussed in a brief introduction to a recent special issue of Emerging Infectious Diseases devoted to antimicrobial resistance, bacterial resistance promises to pose a still larger problem soon because the manufacture of new drugs is at a virtual standstill and has been since 1968 (Weber & Courvalin, 2005). This means that we are on the brink of returning to an era when common illnesses, long thought to have been conquered, may once again prove fatal. At present, illnesses caused by such bacteria are already more difﬁcult to treat (Dagan, 2000; Friedland, 1995; Watanabe et al., 2000), more expensive to treat (Gums, 2002; Holmberg, Solomon, & Blake, 1987), and result in increased mortality (Feikin et al., 2000). For all of these reasons, the problem of bacterial resistance is a paramount public health concern worldwide.

Determinants of Bacterial Resistance What lies behind the bacterial resistance problem? The answer is far from simple. A 2000 World Health Organization (WHO) report points to a number of issues, including the overuse of antibiotics in livestock (World Health Organization, 2000) and international travel that spreads resistant bacteria (Fidler, 1998; Memish, Venkatesh, & Shibl, 2003). But the biggest single factor across both developing and developed nations appears to be the very problem of misuse that Fleming pointed to in 1945 (Albrich, Monnet, & Harbarth, 2004; Harbarth & Samore, 2005). Prescribing when it is not clinically appropriate is still relatively common (Kaiser et al., 1996; Orr, Scherer, MacDonald, & Moffatt, 1993; Todd, Todd, Damato, & Todd, 1984), and this is a primary contributor to the generation of bacterial resistance (Albrich et al., 2004; Cristino, 1999; Deeks et al., 1999; Gomez et al., 1995; Nava et al., 1994; Watanabe et al., 2000). A second contributing form of misuse involves prescriptions for an inappropriate type of antibiotic (e.g., using a second-line, “stronger” antibiotic rather than a ﬁrst-line one when clinical guidelines support the latter) (Hossain, Glass, & Khan, 1982; Hui, Li, Zeng, Dai, & Foy, 1997), and this, too, has been shown to contribute to bacterial resistance (Kozyrskyj et al., 2004; McCaig et al., 2003). Although there is substantial overlap, misuse takes somewhat different forms in developing versus developed nations. Misuse in Developing Countries In developing countries, the factors that lead to misuse revolve around problems of supply and regulation. For instance, with respect to supply issues, developing countries often have difﬁculties in accessing the right medication or adequate doses of medication (Guyon, Barman, Ahmen, Ahmen, & Alam, 1994; Uppal, Sarkar, Giriyappanavar, & Kacker, 1993). It is also not always possible to gain access to diagnostic tests that would allow health practitioners important insight into the condition(s) that

6

PRESCRIBING UNDER PRESSURE

they are treating (e.g., Bosu & Ofori-Adjei, 1997; Horgerzeil et al., 1993; Mamun, 1991). Because of restricted access to information, developing countries may also suffer from a lack of physician knowledge (Igun, 1994) and a corresponding lack of patient knowledge (Braithwaite & Pechere, 1996). Regulation is also a serious problem in developing countries, where antibiotics are often available directly from pharmacies, health care providers, or roadside stalls without government restriction (Bartoloni et al., 1998). Evidence suggests that when people can self-medicate, misuse is rampant in terms of both usage for inappropriate conditions (Haak, 1988; Radyowijati & Haak, 2003; Vuckovic & Nichter, 1997) and inappropriate dosing (Indalo, 1997; Kunin et al., 1987). To drive down the cost of antibiotics, some patients purchase only one dose at a time, use insufﬁcient dosages, or truncate the course of antibiotics. Both a lack of understanding of how the drug works and the implications of such practices, as well as economic issues, contribute to this problem. These are practices that effectively educate bacteria to resist antibiotics. However, prescribing issues are quite a complex problem in developing countries because in spite of fewer antibiotic regulations, they do not necessarily have higher rates of bacterial resistance. This is probably because many people, particularly in rural areas, still have substantial barriers to access, including transportation, cost, or lack of providers in the area. For instance, India has few regulations over antibiotic prescribing but very low rates of resistance in the bacteria relevant to our discussion here, at least in rural areas (Thomas, 1999). This means that there may actually be less misuse of antibiotics in rural areas, despite the lack of regulations in place (Quagliarello, Parry, Hien, & Farrar, 2003). But as a consequence, people who should appropriately be treated with antibiotics may not receive them either. Although it is difﬁcult to fully assess the situation across developing nations, there is substantial evidence that penicillin and erythromycin resistance is an emerging problem in community-acquired Streptococcus pneumoniae across many regions of the world, even in more rural areas (see Okeke et al., 2005, for a review). Misuse in Developed Countries Developed countries generally have fewer problems with respect to physician knowledge, access to high-quality drugs (barring problems with counterfeit drugs and the like), and adequate amounts of medication. Moreover, developed countries typically regulate access to antibiotics, and many have public health campaigns in place to educate patients about antibiotics. However, existing research shows that inappropriate prescribing of antibiotics for viral infections is nonetheless common in many developed countries. In the United States, researchers and policy makers are strongly advocating more judicious prescribing practices (e.g., Bell, 2002; Belongia et al., 2001). But advocacy alone, even from national and international organizations such as the Centers for Disease Control and the World Health Organization, has not stopped doctors from inappropriately prescribing (Finkelstein et al., 2000; Gonzales, Malone, Maselli, & Sande, 2001; Gonzalez, Steiner, & Sande, 1997; Mainous, Hueston, & Clark, 1996; Mangione-Smith et al., 2004; McCaig, Besser, & Hughes, 2002; Metlay, Shea, Crossette, & Asch, 2002; Pennie, 1998). For viral colds, the

THE MIRACLE DRUG

7

prescribing rate across populations is estimated to be approximately 30%, and for bronchitis and other illnesses typically of viral origin, it is estimated to be as high as 60% (Gonzales et al., 2001). Cultural factors may also play a role. People with different cultural backgrounds may be more or less likely to visit a physician for a particular condition (Pachter, 1994). People from particular cultural backgrounds may be more likely to expect that a visit to a health care provider will result in a prescription for treatment (Radyowijati & Haak, 2003). Such expectations may, in turn, affect prescribing rates among particular ethnic and cultural groups (Froom et al., 2001; Harbarth et al., 2002; Radyowijati & Haak, 2003). Differences in cultural attitudes speciﬁcally toward antibiotics may lead to another related issue: the transportation and sales of noncontrolled antibiotics into developed countries where prescriptions are required (Mainous et al., 2005). When physicians are asked why they prescribe against clinical evidence and national guidelines, they commonly cite issues such as patient pressure (Avorn & Solomon, 2000; Little et al., 2004; Stevenson, Greenﬁeld, Jones, Nayak, & Bradley, 1999), lack of time (Little et al., 2004), and a concern with avoiding lawsuits over “missed” bacterial infections (Sargent & Welch, 2001). Whether in developing or developed nations, antibiotic misuse is unlike most types of medical errors in that it is an error that has far greater social impact than individual impact. As Avorn and Solomon observe, “Antibiotics are the only drug class whose use inﬂuences not just the patient being treated but the entire ecosystem in which he or she lives, with potentially profound consequences” (2000: 128). This was foreseen by Fleming, as noted earlier in the excerpt from his Nobel Prize acceptance speech (1945). The social consequences of misuse may be at the level of the community, as in the Fleming example, because bacteria can easily spread from children to adults within the local community, but they can also be at the regional, national, or even international level (McCormick et al., 2003). As Levy points out, bacteria do not have respect for national borders (Levy, 1992). Thus, it is possible for resistant bacteria to cross the world within 24 hours (Fidler, 1998). And travelers deﬁnitely spread bacteria around: Approximately 1,500 of every 100,000 travelers returning from developing countries bring with them an acute febrile respiratory tract infection (Steffen & Lobel, 1996). Misuse that does not result from lack of knowledge or supply problems typically pits the individual against society and thus represents a social dilemma: “situations in which individual rationality leads to collective irrationality” (Kollock, 1998).1 Antibiotic misuse by individuals is perhaps best understood as the type of social dilemma Hardin made famous in his article in Science in 1968. Hardin’s example was that of herders who collectively have access to common land for cattle grazing. Individual rationale would have it that each herder should put as many grazing cows as possible onto the land, even though the commons will be damaged as a result. To make the decision that would be collectively best would require all herders to act in a way that is not in the interest of the individual (i.e., putting fewer cattle on the land) (Hardin, 1968). This type of social dilemma—using the land for individual gain even at the cost of the collective—Hardin terms a “commons dilemma.” Typically, patients and parents of child patients do not view their illnesses as viral or bacterial but as minor or serious. A serious illness is one that they have tried

8

PRESCRIBING UNDER PRESSURE

to “wait out” or to treat but that has persisted or an illness that stands in the way of some special event for which they do not want to be sick. At least among those who believe that it is not good for human society to use antibiotics often, these cases are much like the grazing situation described by Hardin.2 For parents, the desire to use antibiotics is rational at the individual level because of their belief that antibiotics will help their children get better faster. For physicians, the desire to prescribe antibiotics is rational at the individual level because they believe that it will satisfy the parent, enable the visit to be closed, and allow both parties to move on. The desire to use antibiotics is irrational at the collective level because they expose bacteria to the drug and thereby enable them to mutate and develop resistance to the drug. In general, humans may be inclined to prioritize themselves as individuals or families over the larger community. As noted by Humphrey in his important paper on social intelligence, individuals have evolutionary reasons for prioritizing the survival of their own genes over others and thus “to do well for oneself whilst remaining within the terms of the social contract on which the ﬁtness of the whole community ultimately depends calls for remarkable reasonableness” (Humphrey, 1988). And solutions to such problems are challenging, to say the least. (See Kollock, 1998, for a discussion of strategic solutions to social dilemmas.) We will return to this in chapter 8.

Bacterial Resistance as a Global Problem Ultimately, the community that stands to suffer because of individual-level decisions is the global one. Even in countries where inappropriate prescribing is relatively low, such as The Netherlands (Melker & Kuyvenhoven, 1994; Otters, van der Wouden, Schellevis, van Suijlekom-Smit, & Koes, 2004), inappropriate prescribing does occur (Otters et al., 2004). Still, national policies clearly do make a difference, as evidenced by the broad range of rates of antimicrobial resistance and generally corresponding rates of inappropriate antibiotic prescribing across countries. Europe is particularly interesting in this respect because of the close proximity of so many countries. According to reports from the Alexander Project 1998–2000 (a continuing surveillance study that examines the susceptibility of bacteria involved in respiratory tract infections), even countries that share a border can have dramatically different rates of penicillin resistance to bacteria (Jacobs et al., 2003). Whereas Portugal has a rate of 8.2% penicillin resistance, Spain has a rate of 26.4%. Whereas Germany has a rate of 1.9%, Switzerland a rate of 8.6%, and Belgium a rate of 5.7%, France has a staggeringly higher rate of 40.5%. The same goes for nearby countries as well. And although relative to France, Belgium’s rate of 5.7% is quite good, relative to its other border country, The Netherlands, it is quite poor. The Netherlands has the lowest rate in the European Union, only 1.1%. Nearby UK has a rate of 10.9%, and growing still worse, neighboring Ireland has a rate of 24.1%. In general, the data show a generally consistent pattern between antimicrobial usage and resistance prevalence, and these patterns also appear generally consistent with outpatient antibiotic sales (Cars, Mölstad, & Melander, 2001) and antibiotic usage (Albrich et al., 2004). France is again the highest, Spain is also quite high, and Germany and The Netherlands are again very low.

THE MIRACLE DRUG

9

The Current Investigation The problem of inappropriate prescribing is clearly large and amorphous. In this book, we focus on one problem area: inappropriate prescribing of antibiotics for viral upper respiratory tract infections (URTIs) among children in the United States. There are a variety of reasons for thinking this a useful population to study. First, 65 to 70% of URTIs are viral (Wald, Guerra, & Byers, 1991). Second, according to the National Ambulatory Medical Care Survey, American pediatric patients receive two to three times more antibiotic prescriptions than any other patient group, including the elderly (Aronoff, 1996). Third, compared with adult populations, the pediatric population has been particularly resistant to efforts to alter inappropriate prescribing for viral URTIs (Belongia, Knobloch, & Kieke, 2005). Fourth, according to Alexander Project data, the United States has one of the highest rates of Streptococcus pneumoniae resistance to penicillin worldwide. Its rate of 25% surpasses its neighbor Mexico, a developing country with minimal antibiotics regulation, which has a rate of 22%. Only Spain, Japan, Israel, France, and Hong Kong (with a whopping 69.9% rate of bacterial resistance to penicillin) surpass the United States (Jacobs et al., 2003). Thus, U.S. prescribing practices for children with URTI symptoms may yield insight into the larger problems of misuse, both nationally and internationally. This book demonstrates that pediatrician-parent interactions provide a critical window into the macrolevel problem of bacterial resistance and antibiotic misuse in the United States. Close examination of such encounters reveals the impact that microlevel interactional phenomena have on diagnostic and treatment outcomes in URTI visits. This investigation will argue that while the root of misuse in developing countries is more clearly a public health issue, the root of misuse in developed countries like the United States is at least equally a sociological issue.3 Misuse in the United States Earlier we discussed various determinants of misuse in developed countries. If we move more speciﬁcally to the United States, we can look more closely at this issue. One rather obvious contributor to inappropriate prescribing is whether physicians understand the relationship between viral infections and antibiotics. Because research suggests that 89% to 97% of U.S. physicians do understand this relationship (Schwartz, Freij, Ziai, & Sheridan, 1997; Watson et al., 1999), the question remains as to why physicians continue to overprescribe antibiotics in the face of the antibiotic resistance problem. As mentioned earlier, physicians commonly cite patient and parent pressure as a reason for prescribing (Barden, Dowell, Schwartz, & Lackey, 1998; Palmer & Bauchner, 1997; Schwartz, 1999; Schwartz et al., 1997; Watson et al., 1999). In pediatrics, 50% to 70% of parents visiting report an expectation that their child will be given antibiotics (Hamm, Hicks, & Bemben, 1996; Mangione-Smith et al., 2004; Mangione-Smith, Elliott, Stivers, McDonald, & Heritage, 2006; MangioneSmith, McGlynn, Elliott, Krogstad, & Brook, 1999; Sanchez-Menegay & Stalder, 1994). However, parents’ reports of expectations are not necessarily associated with inappropriate antibiotic prescribing (Mangione-Smith et al., 2006; Mangione-Smith

10

PRESCRIBING UNDER PRESSURE

et al., 1999). And physicians are not accurate predictors of which parents expect antibiotics and which do not (Mangione-Smith et al., 2006; Mangione-Smith et al., 1999). Additionally, researchers in both adult and pediatric contexts have found that doctors’ perceptions of patients’ expectations for antibiotics have a signiﬁcant effect on whether doctors prescribe antibiotics, even in cases where they judged them to be not indicated (Britten & Ukoumunne, 1997; Cockburn & Pit, 1997; Gani et al., 1991; Hamm et al., 1996; Mangione-Smith et al., 2006; Mangione-Smith et al., 1999; Mangione-Smith, Stivers, Elliott, McDonald, & Heritage, 2003; Vinson & Lutz, 1993). Speciﬁcally, in one study there was a 25.5% increase in the probability that the physician would prescribe an antibiotic if he or she perceived the parent to expect it, controlling for a range of other issues (Mangione-Smith et al., 2006). Additionally, when physicians thought parents expected antibiotics, they diagnosed middle ear infections and sinusitis more frequently (49% and 38% of the time, respectively) than when they did not think antibiotics were expected (13% and 5%, respectively). These ﬁgures are likely to be low because of improved behavior during the study period (Mangione-Smith, Elliott, McDonald, & McGlynn, 2002). This suggests a disconnect between what parents report and how physicians perceive them. Because what physicians perceive appears to inﬂuence their behavior, and they can access parents’ expectations only through parents’ behavior, it is this that appears to be most consequential, which raises the issue of what parental behaviors lead physicians to believe that parents are looking for antibiotics. Overt parent requests for antibiotics might be expected to be the culprit, and physicians typically cite and complain about this occurring (Schwartz et al., 1997; Stevenson et al., 1999). Although overt requests and other forms of overt lobbying for antibiotics do occur (discussed in chapter 6 primarily), they are quite rare (Fischer, Fischer, Kochen, & Hummers-Pradier, 2005; Stivers, 2002a). On the other hand, this book will argue that less direct interactional behaviors also communicate pressure for antibiotic prescriptions, even if, at times, unintentionally. This book will argue that in a variety of ways, parents actively participate in the visit in ways that pressure physicians in the direction of bacterial diagnoses and antibiotic prescribing. This book will also argue that parents, even when vying for antibiotics, are oriented to this interactional work as in the physician’s domain of expertise, and thus this issue is something parents work to manage. Pressure as Parent Participation Patient participation is currently an important topic of discussion in health services research and health policy circles. Much of the emphasis from local to national levels is to encourage physicians to involve patients or parents in treatment decisions. According to the goals of Healthy People 2010, patients who participate actively in decisions about their health care can have a positive impact on national health (U.S. Department of Health and Human Services, 2000). Researchers assert that patients should, whenever possible, be offered choices in their treatment decisions (Brody, 1980; Butler et al., 2001; Deber, 1994; Emanuel & Emanuel, 1992; Evans, Kiellerup, Stanley, Burrows, & Sweet, 1987; Fallowﬁeld, Hall, Maguire, & Baum, 1990;

THE MIRACLE DRUG

11

Kassirer, 1994; Levine, Gafni, Markham, & MacFarlane, 1992). Several American medical associations now recommend that physicians overtly involve patients in their decision making. For instance, the American Cancer Society, the American Urological Association, the American Gastroenterological Association, the American College of Physicians, and the National Institutes of Health (NIH) all recommend shared decision making for decisions surrounding cancer screening (Frosch & Kaplan, 1999). Although there is recognition that not every patient wants to participate in their health care, the primary rationale fueling these recommendations is, ﬁrst, that patients generally do have the desire and are entitled to participate in treatment decisions (Blanchard, Labrecque, Ruckdeschel, & Blanchard, 1988; Cassileth, Zupkis, Sutton-Smith, & March, 1980; Emerson, 1983; Ende, Kazis, Ash, & Moskowitz, 1989; Faden, Becker, Lewis, Freeman, & Faden, 1981; Swenson et al., 2004; Thompson, Pitts, & Schwankovsky, 1993) and, second, that patients have improved outcomes when they participate in medical decision making, including satisfaction (Brody, Miller, Lerman, Smith, & Caputo, 1989; Brody, Miller, Lerman, Smith, Lazaro, et al., 1989; Evans et al., 1987), patient health (Brody, 1980; Greenﬁeld, Kaplan, Ware, Yano, & Frank, 1988; Kaplan, Greenﬁeld, & Ware, 1989; Mendonca & Brehm, 1983; Schulman, 1979), and patient mental well-being (Brody, Miller, Lerman, Smith, & Caputo, 1989; Evans et al., 1987; Fallowﬁeld et al., 1990; Greenﬁeld et al., 1988). Although the movement toward shared decision making in health care has certainly taken root in the care of chronic conditions, the issues have been far less explored in acute care. But the social factors encouraging partnership in chronic care may nonetheless also be affecting acute care encounters. First, because of the consumerist movement in health care, patients can be seen to be moving away from the guidance-cooperation models depicted as normative by Parsons in the 1950s (1951) and documented empirically in the 1970s (Byrne & Long, 1976). Speciﬁcally, as Haug and Lavin (1983) suggest, the consumer model refocuses the balance of power on the patient’s rights (as purchaser) and on the physician’s obligations (as seller) rather than on the “physician’s rights (to direct) and patient obligations (to follow directions)” (p. 213). As summarized by Roter and Hall (1992), patients, particularly younger and more highly educated ones, are becoming more likely to exhibit consumerist behaviors (Ende et al., 1989; Hibbard & Weeks, 1985; Reeder, 1972), which include having sophisticated medical knowledge, seeking information through reading, and exercising independent judgment in following physicians’ recommendations. Kravitz, Bell, and Franz (1999: 873) suggest that “patients are more than the passive recipients of doctors’ actions; they inﬂuence the clinical encounter through use of their own linguistic resources.” Second, as of 1985, drug companies have been allowed to market prescription medication directly to consumers (DTC). Promotional strategies encourage patients to act as consumers both in terms of product knowledge and with explicit suggestions to “ask your doctor if X is right for you.” As summarized by Pinto, Pinto, and Barber (1998), this shift was brought about because of pressure by drug companies who felt that managed care companies restricted their access to physicians and to patients in the sense that often only particular drugs would be covered by a given

12

PRESCRIBING UNDER PRESSURE

company. Competitiveness in the drug market also added pressure. Additionally, though, patients with the “‘activist mindset’ of many . . . baby boomers” (p. 93) also gave the movement a needed push. Once in place, the DTC move fueled still more patient activism: “patients are active decision makers for their own health needs (or the health needs of their babies) and . . . physicians are not the exclusive inﬂuencers in brand selection” (pp. 91–92). Research suggests that advertising is encouraging patients to ask about medication (Peyrot, Alperstein, van Doren, & Poli, 1998) and that physicians are likely to prescribe or consider prescribing a drug requested by a patient (Borzo, 1997). With respect to patients’ asking, Sleath, Svarstad, and Roter (1997) examine data involving patients receiving care for chronic conditions and who are prescribed a psychotropic drug. They claim that patient-initiated talk showing that they wanted a new prescription was associated with prescribing 20% of the time. More recently, Kravitz and his colleagues have shown that requests do affect prescribing rates (Kravitz et al., 2005). Moreover, there is additional evidence that patients are becoming more consumer oriented: In a survey asking how people are likely to react to being denied a requested prescription, Bell, Wilkes, and Kravitz (1999) found that 46% of respondents reported being likely to be disappointed by a denied prescription request. Additionally, 25% of all respondents reported being likely to pressure physicians. Finally, 24% of all respondents who were denied a prescription said they would be likely to seek a prescription from another physician. Related to DTC advertising is a third factor that may affect not only chronic but also acute care: A vast amount of medical information is now readily accessible to the public through the Internet (du Pré, 2000). Thus, patients are not only being inﬂuenced to be more proactive in their own health care but also being given resources with which to become more knowledgeable. Kravitz and colleagues (1999) found that the most common information requests involved questions about medications and that the most common action request was for medications. This has largely been studied with respect to chronic or serious conditions, but these factors also affect acute care. As du Pré points out, “It is no longer enough (if it ever was) to simply tell patients what to do. Empowered patients want information and the right to make their own decisions” (2000: 15). Thus, even in situations such as acute care, where physicians may feel that a more medical or disease-based approach is appropriate, there may be pressure from both patients and policy makers for patient participation in the visit. Treatment for an acute illness is typically conceptualized as something that the patient is directed to do by the physician or that the physician recommends or advises (e.g., Byrne & Long, 1976). And some research suggests that in the primary care context, doctors are much less likely to involve patients or parents in treatment decision making (Braddock, Edwards, Hasenberg, Laidley, & Levinson, 1999; Elwyn, Edwards, & Kinnersley, 1999; Tuckett, Boulton, Olson, & Williams, 1985). Generally, the treatment phase of an acute medical encounter is thought of as doctor driven (at least in the sense that doctors make recommendations and provide advice for treating the patient’s medical problem), particularly in contexts where there is a view that only one course of action is correct and thus decision making is basically straightforward (Coulter, 1997; Szasz & Hollender, 1956). Additionally, patients

THE MIRACLE DRUG

13

with acute problems are generally seeking a physician’s treatment recommendation or advice, and the physician’s recommended treatment is generally grounded in his or her medical knowledge. Therefore, patients in the acute contexts might be assumed to interact within the guidance model—not disagreeing or querying the physician’s treatment recommendation (Szasz & Hollender, 1956). And if patients were to inﬂuence acute visits at all, we might expect it to be done subtly, given the general orientation of these visits. For these reasons, only a method that examines interactions at a detailed level would be able to identify such practices. This is the method adopted in this book. Here what I show is that when we look carefully at the details of parent-physician interaction, we see that parents and physicians frequently go through a subtle but very much observable negotiation of the child’s illness. Negotiation occurs at virtually every stage in the visit, from the opening statements to the doctor to the reception of the physician’s treatment recommendation and a variety of places in between. This book will step through each of the phases of the medical visit in order to show a variety of different practices (mostly covert) that parents initiate and that can be seen to affect the diagnostic and treatment outcome of their child’s medical visit. Thus, contrary to what would be deduced from the existing literature, this book shows that even in rather doctor-centered visits, where the physician has made no real effort to explicitly involve the parent in the treatment process, parents affect the diagnostic and treatment outcomes of the visits through their interactions with the physician. Methodology Historically, there has been relatively little connection between large-scale social or public health problems and microlevel studies. Rather, these problems have been most typically investigated through large-scale surveys of either medical records or of physicians and parents. [(As one example of a survey that has generated studies relevant to this domain, see the National Ambulatory Medical Care Survey (CDC, 2004), which is a key survey in primary care for the Centers for Disease Control.)] When provider-patient interactions are examined, this most commonly involves process analysis methodology: a coding of the interaction, followed by an analysis only of the coding rather than the interaction. Debra Roter has been the leading ﬁgure in the development of this work since the pioneering research of Barbara Korsch in the 1960s (e.g., Korsch, Gozzi, & Francis, 1968). Interactions in this and similar types of approaches are coded on the basis of analyst- or literature-driven constructs (a top-down approach), whether for issues of patient participation or physician behavior, rather than codes that have emerged from an understanding of how interactions in the context of interest work. (See Roter & Hall, 1992, for a review of one of the major coding schemes in interaction studies in public health, and Roter, 2002, for a full bibliography of studies using the Roter Interaction Analysis System coding scheme.) Microlevel analytic methods such as discourse and conversation analysis have (but to a far lesser extent) taken root in the realm of health communication. These studies have illuminated important dimensions of medical interaction but, with rare exceptions (e.g., Waitzkin, 1991), the results are not generalized or are not generaliz-

14

PRESCRIBING UNDER PRESSURE

able. For instance, using a microanalytic approach to discourse, Mishler shows that physicians and patients commonly pursue very different (and, at times, conﬂicting) agendas during medical visits: the medical agenda and the “lifeworld” agenda (Mishler, 1984). West (1984) published the ﬁrst conversation analytic book on medical interaction. She wanted to understand the role language played in structuring both the social and the power relationships between physicians and patients. Todd and Fisher published a collection of discourse analytic papers dealing with the organization of medical communication (Todd & Fisher, 1993), and Heritage and Maynard’s recent collection of conversation analytic studies examines each phase of the medical visit and its organization (Heritage & Maynard, 2006). Both discourse analysis (DA) and conversation analysis (CA) take the perspective that medical interaction is, at its most basic level, still basic social interaction that is occurring in an institutional context (Drew & Heritage, 1992a). These methodologies treat social interaction as a highly structured domain where the structural underpinnings, like the structural underpinnings of a molecule, can be examined and understood. For conversation analysts in particular, social interactants accomplish social actions through language (See Heritage, 1984b, for summaries; Levinson, 1983). Thus, interactants greet each other, request things, and complain or invite through language, and the doing of these social actions is itself highly structured. One of the hallmarks of conversation analysis is that in analyzing any bit of social interaction, analysts must validate their understandings of participants’ social actions through an examination of interactants’ responses. This virtually necessitates that analysts look at interaction through the lens of the sequence (e.g., an initiating turn and a response) rather than restricting themselves to individual words, phrases, or sentences, as linguists have historically done. This methodology was particularly valuable in the present study because of the problem of understanding what parent behaviors physicians were understanding as communicating pressure to prescribe, regardless of parent intent. In examining social interaction in sequence structural terms, CA looks for patterns in the interaction that form evidence of systematic usage such that a particular turn design, for instance, can be identiﬁed as a “practice” through which people accomplish a particular social action either vocally or visibly. For example, from ordinary interaction contexts, we see practices for opening telephone conversations (Schegloff, 1968, 1972b, 1977), competing for epistemic rights over a claim (Heritage, 1998; Heritage & Raymond, 2005), or inviting another interactant to complete one’s turn at talk (Lerner, 1996). To be identiﬁed as a practice, a particular communication behavior must be seen to be recurrent and to be routinely treated by a recipient in a particular way such that it can be discriminated from related or similar practices. The signiﬁcance of these practices can be understood in terms of (1) the immediate sequences in which they occur, (2) the larger activities in which they are embedded (Heritage & Sorjonen, 1994), and (3) the overall organization of the phases in the interaction. The latter two levels of organization are particularly signiﬁcant when CA is used to analyze interaction in institutional contexts, such as medical visits, because of the general goal orientation of participants in these interactions (Drew & Heritage, 1992b).

THE MIRACLE DRUG

15

A CA approach is highly structural in orientation. To this end, CA researchers in medicine have been interested in practices of social interaction that reveal structure at different levels: the visit (i.e., overall structure), the activity (e.g., the treatment activity/phase), the sequence (e.g., the opening question and its response), or the turn (e.g., turn constructional practices). Some studies clearly ﬁt into one level, such as studies of the visit’s overall structure (for relevant work on phases, see Byrne & Long, 1976; Robinson, 2003; Waitzkin, 1991) or studies of turn design within a phase, such as Heritage and Stivers’s examination of “online comments” offered during the physical examination (Heritage & Stivers, 1999). Other studies cross multiple levels. For instance, Peräkylä examines alternative ways of designing a diagnosis delivery. This then is relevant both at the activity and the turn levels (Peräkylä, 1998). Very few conversation analytic studies have attempted to connect interactional practices to large-scale exogenous issues, whether they are relational, socioeconomic, demographic, or public health issues (but see Boyd, 1998; Clayman, Elliott, Heritage, & McDonald, 2006; Clayman, Heritage, Elliott, & McDonald, in press; Heritage, Boyd, & Kleinman, 2001; Kleinman, Boyd, & Heritage, 1997). This book represents a conversation analytic investigation of how parents and physicians communicate about children with routine upper respiratory tract infection symptoms, and it demonstrates that conversation analytic ﬁndings can offer results that bear on the large-scale public health problem of inappropriate antibiotic prescribing. Data This book draws on three data sets. First, there is a corpus of 65 videotaped encounters involving 6 pediatricians from 5 practices collected as pilot data that I will refer to as the Hillside data set. Then there is a corpus of 295 audiotaped encounters involving 10 pediatricians from 2 practices that I will refer to as the Seaside data set. Finally, there is a corpus of 522 videotaped encounters from 38 physicians in 27 practices that I will refer to as the Metro data set. All data were collected between September 1996 and June 2001 in Southern California. Children ranged in age from newborn through 16 years old, and all were accompanied by parents. Most visits involve children from 6 months to 10 years of age. All visits involve children who were being seen for routine illnesses, and approximately 98% of children had routine upper respiratory tract infection symptoms. Informed written consent was obtained from all participating parents and physicians in all samples. Only parents who could conduct the visit in English were admitted into the studies. For purposes of anonymity, in all transcripts pseudonyms replace any use of a subject’s name or other identifying information (e.g., school names). Demographic information was not collected for the Hillside data set, though I expect it represents a midpoint (socioeconomically) between the Metro and Seaside data. The Metro data involved parents who were 34 years old, on average, with a median household income of $40,000. Most of the caregivers were mothers: 86% were female. Fifty-three percent of the parents were Latino, with 28% white, 12% African American, and 7% Asian. Sixty percent of parents were high school graduates but did not have a college degree. Twenty-four percent had at least an undergrad-

16

PRESCRIBING UNDER PRESSURE

uate degree, whereas 16% had less than a high school degree. The Seaside data were, on balance, somewhat wealthier, more likely to be white, slightly older, and more likely to have a college education than the more diverse Metro sample. The demographic backgrounds of the Seaside and Metro data sets are described elsewhere in greater detail (Mangione-Smith et al., 2006; Mangione-Smith et al., 1999). Statistical information that is mentioned throughout the book refers to either the Seaside or Metro data. All data shown here were transcribed by the author according to the conventions originally developed by Gail Jefferson (see Appendix for conventions). The Role of Children in the Pediatric Visit In earlier work (Stivers, 2001), I aligned myself with other researchers of pediatric interactions against studies that fail to take into account the role of the child in the interaction (e.g., Pantell, Steward, Dias, Wells, & Ross, 1982). Here, I appear vulnerable to this very criticism. However, in my analyses of these interactions, what became clear is that children play quite a small role in the domain of treatment negotiation. Interestingly, even when children are active participants in the visit, it is rare that they perform the behaviors outlined in chapters 2 through 5 (but for a rare case, see Extract 6.1). They typically do orient to the doctor as someone who can tell them what is wrong and give medicine (see Extract 2.1 for an example). For this reason, they can occasionally, without intention, make it difﬁcult for a parent to perform a behavior that might have pushed for a bacterial diagnosis or antibiotic treatment. For instance, if a child presents his or her own problem, parents are, at least in that sequential location, blocked from presenting it in their own words: words that, as we will see, can communicate a very particular stance toward the outcome of the visit. For this reason, there will be relatively little discussion of the child’s contributions, despite the fact that in general interaction, their role is certainly important. The Social Context of These Visits Research on physician-patient interaction has been growing steadily, most signiﬁcantly since Barbara Korsch’s groundbreaking pediatric communication studies (e.g., Korsch et al., 1968). I will not attempt to review that literature here (but see Heritage & Maynard, 2006, for a comprehensive review). Instead, I will focus on the issues that are most relevant to understanding what pediatricians and parents are dealing with in their interactions involving children with routine childhood illnesses, and speciﬁcally with upper respiratory illnesses. Many American readers who have attempted to reach their primary care physician during their lunch hour, after hours, or even when the phone lines are busy are probably familiar with a common recording that physicians place on their voicemail that instructs patients to call 911 emergency or to seek help in an emergency room if their problem is life threatening. Virtually anything about which parents or patients seek assistance from their primary care doctor is something that they do not perceive as life threatening. But of course, many adults and children have health problems that they live with or manage by themselves (Dunnell & Cartwright, 1972). In the case of the visits with which we are dealing, it is likely that every child has had a similar

THE MIRACLE DRUG

17

illness in the past, and it is virtually certain that every parent has experienced an illness similar to the one the child has several times, if not on an annual or semiannual basis. Thus, parents in these visits have made a decision to seek medical care at this time. Physicians, too, are aware that each visit was not a virtual certainty (as might be expected with acute pain that results in a trip to the emergency room). Rather, visits are understood to be the result of an active consideration of alternatives. In general, this book will argue that parents seeking medical help for these routine illnesses feel they have gone beyond the point where their own expertise is sufﬁcient. Some parents may be coming speciﬁcally to get antibiotics; some may be coming because they are getting no sleep, and their child is cranky, disturbing the household, and they do not know what to do; others may want reassurance that what they have been doing is right and that there is no more to be done. In all cases, though, they have a problem that they no longer feel comfortable handling on their own. Two issues seem to inhabit these interactions: (1) the legitimacy of the visit and (2) the treatability of the child. Legitimacy In the way that adult patients present their reason for visiting the physician, they often include statements that work to show they have not rushed to the doctor at the ﬁrst sign of a problem but have waited a reasonable length of time, have come for good reason, and have attempted to manage their troubles prior to seeking medical assistance (Heritage & Robinson, 2006a). Similarly, patients work to show that they have not been overly attentive to their bodies—noticing the slightest or most minimal change—but rather are coming to the physician only with rather unusual noticings or problems (Halkowski, 2006). Heritage and Robinson (2006a) argue that there are three basic ways that patients in acute primary care encounters display their orientations to their conditions as “doctorable” or “worthy of evaluation as a potentially signiﬁcant medical condition, and worthy of counseling and, where necessary, medical treatment” (p. 58): 1. Patients routinely include in their problem presentations attributions to third parties in order to give support to their decision to seek medical assistance (i.e., physicians, spouses, friends, or acquaintances said they should see a doctor). 2. Patients routinely display “troubles resistance” both (a) in the report of their decision to visit the doctor (e.g., stating that they have waited some length of time, or that they tried over-the-counter medications) and (b) in their description of their condition (e.g., they offer objective rather than subjective evidence of their difﬁculty as severe enough to warrant the visit). For example, patients with shoulder pain will state that they cannot latch a seat belt by way of indirectly indexing the severity and doctor worthiness of their complaint, rather than describe the degree of their pain. 3. Patients rarely offer any diagnosis of their condition and furthermore orient to this as a behavior to be avoided (Gill, 1998). In this way, they defer to the physician’s knowledge for solving their medical problem.

18

PRESCRIBING UNDER PRESSURE

The ways parents communicate about their children’s conditions appear to be somewhat different. Whereas adult patients and parents share an orientation to the doctorability of the medical conditions, the pediatric context appears to have characteristics that are at variance with adult acute visits. First, parents more rarely formulate the reason for their child’s visit as based on a third party’s recommendation. Rather, this is typically reserved as a justiﬁcation for a concern either offered subsequent to the presentation or following indications by the physician that the child may not have a problem (e.g., see Extract 7.24). Second, parents typically report their decision to visit the doctor more straightforwardly and in less troubles-resistant ways than adults. For example, parents appear more willing to go to the doctor quickly on behalf of their child than on their own behalf, and they provide less justiﬁcation for this behavior. Although as Parsons (1951) pointed out, adults are normally obliged to resist the “sick role” and to make light of their troubles, “the sufferings of little children are another matter” (Strong, 1979: 204). While troubles resistance may be invoked during pediatric encounters in the form of showing, for example, that they did not rush to the doctor, the data in my corpus also support Strong’s suggestion that concerns to justify a visit to the doctor may be somewhat relaxed in the pediatric context. Third, parents’ orientations to bodily attentiveness appear to be markedly different when they are acting as caregivers rather than patients. For instance, according to Halkowski (2006), adults who are overly concerned about themselves risk being thought to be seeing the doctor in a motivated way. Halkowski further suggests that adult patients regularly show a balance between attentiveness and inattentiveness to their bodies and emergent symptoms. By contrast, parents acting as an advocate for their child appear to be more attentive. In this context, the balance normally seen in adult primary care may be recalibrated. Concern over a child’s well-being is generally seen as the sign of a good, if slightly overanxious, parent. This is supported further by data involving British health visitors and ﬁrst-time mothers (Heritage & Lindström, 1998; Heritage & Seﬁ, 1992). For example, in this data extract, the health visitor is illustrating the types of noticings and the level of detail at which noticings should be made. The parents are instructed to notice “when she smi:les” (line 9), “when . . . she’s holding her head up better” (lines 9–10), and “when she can see” (lines 11–12). (1.1) Extract from Heritage & Lindström, 1998: 404 1 2 3 4 5 6 7 8 9 10

HV: F: HV:

HV: M: HV:

.hh These uhm (1.0) are the notes that I carry arou(t) with me:, Mm hm, And I (0.2) I uh record your baby’s progress on he:re. (0.2) .hhh [So that uhm (.) I want to know when she’s [°(Oh)° doing new things when she smi:les and when she (.) .hh uh:m you know she’s holding her head up

THE MIRACLE DRUG

11 12 13

19

better: .hh I want you to notice if she: (.) .hh can see:_ ((datum continues with father volunteering information about baby’s sight).

In contrast to an avoidance of overly self-attentive behavior in adults, with children we see an orientation to both the acceptability and desirability of close bodily monitoring. Related to this argument, parents may be more sensitive to potential perceptions of negligence when they are acting as their child’s caregiver than when they are acting on their own behalf. For example, a parent may hear a doctor’s questions about her child’s health as “testing her capabilities as a mother” (Bates, Bickley, & Hoekelman, 1995; Heritage & Seﬁ, 1992). Parents have some reason to be concerned about their pediatricians’ perceptions of them. Sheridan (1994) surveyed pediatricians and family practitioners about their perceptions of the accuracy of parents’ reports of their children’s symptoms. She showed that while only 1% of parents were perceived as actually falsely reporting or inducing their child’s symptoms, 23% were perceived to be in some way misrepresenting their child’s symptoms (e.g., exaggerating them) (Sheridan, 1994). Fourth, parents appear markedly more likely to offer possible diagnoses in the pediatric context. I will analyze this practice in detail in chapter 2, but for this discussion it is important to recognize that this behavior may indicate that parents feel more entitled to have expertise over, and to participate in, the diagnosis and treatment of their child than in their own care. Speciﬁcally, a parent who is knowledgeable about childhood illnesses, symptoms, remedies, and the like is displaying “good parenting.” By contrast, an adult patient who is knowledgeable about acute illnesses may be viewed, and treated, as an “uncooperative” or “bossy” patient (Papper, 1970). Within the pediatric context, we have observed that parents, on the whole, are less oriented to (1) diffusing responsibility for seeking medical care and (2) exhibiting troubles resistance and are more willing to (3) diagnose their child’s condition and (4) be attentive to their child’s body. That said, we observed earlier that with routine illnesses parents do make a decision that this illness at this point in time requires medical attention, whereas similar illnesses at other times have not. So, although in many ways parents are under fewer constraints to legitimize their visit, there nonetheless appears to be some underlying concern that their visit be legitimate, or at least that its legitimacy be validated, and in various behaviors a concern with legitimating their visit can be observed. However, to some extent it appears that the concern is less to establish legitimacy than to have it validated. Whereas adults appear to frequently treat their visit’s legitimacy as questionable from the outset, parents appear more concerned that the physician not undermine the inherent legitimacy of their visit. This will be most clearly visible in chapter 3 on history-taking questions and chapters 4 and 5 on diagnosis delivery and treatment recommendations. One reason for this might be that parents may demand and expect more “doctoring” when they are acting as an advocate for their child rather than as a patient. Roter and Hall (1992) suggest that parents are more likely to be assertive on “another’s behalf, especially a child’s” (p. 17). They provide an example of a father who, after coming to an emergency room, “announced that he would only see a doctor who

20

PRESCRIBING UNDER PRESSURE

was a parent” (p. 17). Researchers have also found that parents are more willing to seek out information, ask questions, and voice concerns when speaking on behalf of their child (Korsch et al., 1968). And parents may have a stronger concern than adult patients about the cause of their child’s illness (Korsch et al., 1968). Treatability These issues come together in a marked way with respect to treatment decisions. Here, several key issues converge: First of all, parents have familiarity with these illnesses both in their own experience and in observing their children. Therefore, they may often feel that they know what the problem is and how it may need to be treated. Just as parents and physicians orient to greater latitude across a range of behaviors in the pediatric encounter, in contrast to the adult encounter, this may also affect parents’ willingness to explicitly or implicitly seek out treatment for their child. That is, in contrast to an adult patient, a parent may be more willing to (1) ask about treatment, (2) pressure physicians for treatment, or (3) display expertise about treatment options for their child. Second, in contrast with adult care, parents may feel additional pressures to “cure” their child. One pressure parents face in terms of sick children is the need for a quick solution to their problem. As caregivers, parents are responsible for both properly caring for a sick child in terms of keeping them at home and taking them to a doctor if needed and also in terms of getting medicine for them. In today’s society, the pressure to accomplish these things quickly has increased. As a pediatrician quoted in a newspaper article put it, “Years ago, parents might keep a child at home and just sit out an infection. Now most don’t have that luxury” (Warren, 1998). In a society where two working parents are increasingly common, a sick child poses a problem for the family in terms of both nighttime sleep and daytime care. The pressure parents feel to get their children well may understandably translate to pressure on the doctor to make them well. Thus, for parents the pressures they feel may translate to more latitude in terms of offering accounts of what they think is bothering their child, pressing the physician to treat the child, and offering opinions on how to treat the child. Third, as was mentioned earlier, in the pediatric context, physicians may feel more social pressure to “cure” a child patient than an adult patient. Insofar as children are considered more of the society’s responsibility than adults are (e.g., see Strong, 1979), pediatricians may feel self-imposed pressure to do what they can to help the child get well quickly. Furthermore, they may feel parental pressure to make the child well. In this way, the physician may feel a greater obligation to proactively address the condition of a sick child than a sick adult. In sum, from both the parent’s perspective and the doctor’s perspective, there is a pressure within the pediatric context to cure the child of an illness. As was mentioned earlier, for a variety of reasons parents come to the medical visit seeking a solution for their child’s problem. Throughout the visit, physicians must balance two issues that are, at times, competing: maintaining the legitimacy of the visit and communicating whether the child has a treatable problem or not. If the child’s problem is treatable, this is unproblematic: The legitimacy of the visit is quite

THE MIRACLE DRUG

21

easily upheld through the prescription of medication. But if the child’s illness is not treatable, as in the case of most viral colds or other infections, then this becomes a struggle. For parents, the issues are inverted: If a physician threatens the legitimacy of the visit by indicating that the condition is either not treatable or not problematic, they may do work to reestablish the legitimacy of their visit and/or advocate for the problematic and treatable nature of the child’s problem. Interestingly, what parents consider treatable is rooted in their folk model of illness. Helman notes that prior to World War II and the mass production (and availability) of antibiotics, the typical illness model viewed colds as something triggered by external sorts of causes such as being in the cold air, drafts, or getting chilled when not wearing shoes (Helman, 1978). With this model came home remedies, so few physician visits were made. By contrast, postantibiotic patients suddenly were more likely to visit physicians for colds. Arguably, this was because all forms of illness were suddenly lumped together as caused by “germs” and therefore were considered treatable (Helman, 1978). Although we are nearly 30 years past the time when Helman wrote about his suburban general practice, the basic problem is the same. A concern with treatability is different from a concern with legitimacy. A parent can orient to the visit as legitimate and to their child’s illness as treatable. Parents can respond to a physician’s recommendation against treatment as problematic either because it delegitimates the visit (e.g., “Yeah, it was my wife who called; I ﬁgured there wasn’t much you could do”) or because they feel that their child needs treatment (e.g., “Can I at least have thuh prescription an’ I’ll decide whether or not to ﬁll it in a couple days?”). However, it may not be entirely clear, to an analyst or to the physician, whether a given physician’s action is problematic because it deals a blow to the legitimacy of the visit or to the treatability of the illness. Even with respect to treatability, parents can want treatment without desiring antibiotics speciﬁcally, as will be discussed further in chapter 6. Britten and her colleagues have shown that adult patients do not always or simply want antibiotics (Britten, 1994; Britten, Jenkins, Barber Bradley, & Stevenson, 2003; Britten, Stevenson, Barry, Barber, & Bradley, 2000). Moreover, other studies show that patients and parents alike are not, across the board, less satisﬁed if they fail to get an antibiotic (Himmel, LippertUrbanke, & Kochen, 1997; Mangione-Smith et al., 2001). Much of this book will be concerned with interactional practices that are taken by the physician to be solely concerned with treatability and, even more speciﬁcally, taken to be directly indexing antibiotics. It is precisely because of this perception by physicians that we will see that parents accomplish negotiation for antibiotics.

Overview of the Book This book will examine the parent-pediatrician negotiation of legitimacy, treatability, and antibiotics in particular, as these issues emerge throughout the visit. The book is generally laid out in the order of the acute care medical visit (drawing on Byrne & Long, 1976; Robinson, 2003; Waitzkin, 1991): (1) opening; (2) establishing the reason for the visit; (3) history taking; (4) physical examination; (5) diagnosis; (6)

22

PRESCRIBING UNDER PRESSURE

treatment; and (7) closing. Chapter 2 will examine the beginning of the visit proper, when parents are frequently offered an opportunity to explain their reason for seeking medical help. This chapter will suggest that parents display rather different stances toward their children’s illness and thus toward their preferences for the visit outcome through their formulation of their reason for visiting. Chapter 3 focuses on taking the child’s illness history, during which physicians primarily ask questions and parents and children answer questions. Through the ways that physicians design their questions, they reveal their current diagnostic and treatment trajectories. We will see that although in one sense parents are put in a relatively constrained and arguably powerless position in this phase, they are nonetheless quite capable of working within these sequential and structural constraints to encourage physicians away from one diagnostic trajectory and/or toward an alternative diagnostic and treatment trajectory. Chapter 4 examines parent resistance as a response to no-problem diagnoses. This chapter argues that through the use of three different types of responses and because of the structural organization of the diagnosis, parents can take issue with the physician’s diagnosis and, at times, lead the physician to alter the diagnostic and/ or treatment trajectory from no problem and/or nontreatable to problematic and/or treatable. Chapter 5 examines parent resistance in a second environment: following a treatment recommendation. Taken together with chapter 4, this chapter shows that resistance can take quite different forms, depending on the normative sequential organization of the action it is responding to. Still, as with diagnosis resistance, treatment resistance can be observed to be a powerful tool to negotiate in favor of antibiotic treatment. Chapter 6 examines the relatively rare behavior of parents overtly lobbying for antibiotics. This chapter shows that this type of behavior can take several particular formats that can be more or less direct (and more or less coercive). It also shows that the practices, though more frequent during the treatment recommendation phase of the visit, can be offered throughout the visit. Chapter 7 shifts from the parent to the physician to explore several ways in which physicians can, through the actions they perform and their design, shape whether parents perform problematic behaviors such as diagnosis and treatment resistance. This chapter focuses on three behaviors: online commentary, formulation of the diagnosis, and formulation of the treatment recommendation. Finally, chapter 8 concludes by speculating about the issues underlying the problems of inappropriate antibiotic prescribing, particularly in developed countries but also as it contributes to the global problem of bacterial resistance to antibiotics.

2

Foregrounding the Relevance of Antibiotics in the Problem Presentation

A

s discussed in chapter 1, medical visits are generally conducted in a way that proceeds rather systematically through a series of activities, beginning with an opening and progressing through to treatment discussion and closing. To understand the variety of ways that parents inﬂuence the outcome of the visit, we will, in the course of this book, look at a number of these activities in detail. This chapter is concerned with an activity that generally occurs very early in the visit: when physicians and parents establish why the child is visiting the physician. Depending on the scheme of the medical encounter’s structural organization, this activity may be treated as the beginning of the history taking or as initiating an activity in its own right. I follow Byrne and Long (1976) in treating it as shifting to establish the reason for the visit. The question that physicians generally ask parents (e.g., “What can I do for you today?”) offers parents an opportunity to shape the course of the visit by describing their child’s problem in their own words and thus emphasize particular dimensions of the illness and de-emphasize others. With this question, physicians also provide parents with an opportunity to formulate their worries (or not), to project and tell a story about the problem (or not), and/or to offer their own speculations about the problem’s cause, all in the course of their response. When physicians solicit the problem, this represents the ﬁrst and sometimes only sequentially provided for opportunity that parents have to shape the physician’s view of the problem and directly inﬂuence the treatment decision.1 Thus, this is the obvious starting point for our study of parentphysician negotiation of treatment. 23

24

PRESCRIBING UNDER PRESSURE

Parents generally respond to physicians’ questions about the reason for their visit by using one of two problem presentation formats. The ﬁrst is simply a description of the child’s symptoms (e.g., “He has a rash all over his body”), and for this reason we term it a “symptoms only” presentation. The characteristic feature of the second format is that it includes the mention of a possible diagnosis (e.g., “We were thinking she has an ear infection because she’s been having pain”), and for this reason we term it a “candidate diagnosis” presentation. In distinguishing between these two primary forms of problem presentation, this chapter describes alternative responses by physicians that display different analyses of the parent’s stance toward the child’s illness. Speciﬁcally, in cases in which the child’s problem is presented with a symptoms-only description, parents are treated as having adopted the stance that they are primarily seeking a medical evaluation of the child. By contrast, in cases where the child’s problem is presented with a candidate diagnosis, parents are treated as having adopted the stance that they are seeking conﬁrmation of their diagnosis and seeking treatment for the illness condition. Each of these patterns will be discussed in turn in the following sections.

Background Establishing the reason for the visit is an activity that is present a large majority of the time in acute care medical visits. Even when physicians “skip” the activity and begin by starting the child’s history taking (e.g., by beginning with a history-taking question such as “So how long has the cough been going on?”), parents nonetheless commonly offer their reason for visiting. Current research suggests that around 85% of acute encounters involve the patient or parent presenting the problem (Heritage & Robinson, 2006a; Stivers, 2001, 2002b). Establishing the problem is important because it generally plays an important role in structuring the rest of the visit: Both what the problem is and how it is presented set the agenda for the visit. But the problem presentation is important for other reasons as well: A physician needs an accurate and thorough description of the patient’s problem to provide a correct diagnosis (Ong, de-Haes, Hoos, & Lammes, 1995; Pendleton, 1983). Related to this, the problem presentation allows patients to formulate their problem or concern in their own words and allows the inclusion of both biomedical and life world dimensions of the problem and its impact on the patient (Fisher, 1991; Frankel, 1984; Heritage & Robinson, 2006b; Mishler, 1984). Ruusuvuori (2000) showed that the problem presentation is internally structured. She examined several key aspects of the problem presentation, including how patients begin and end their problem presentations and their vocal and visible resources for holding the ﬂoor during their presentation. Establishing the reason for the medical visit constitutes the shift to medical business or the ﬁrst topic (Byrne & Long, 1976; Heath, 1981; Robinson, 1998). Previous interaction research has established that openings typically affect later activities in institutional interactions, including the way the issue is addressed or the remedy that is suggested. For example, the way a problem is presented to 911 emergency call takers can affect whether they agree to dispatch help immediately

FOREGROUNDING THE RELEVANCE OF ANTIBIOTICS

25

following the problem presentation (e.g., see Whalen & Zimmerman, 1987; Whalen, Zimmerman, & Whalen, 1988). Boyd (1998) has shown that the way interactions are opened can not only have interactional consequences but also affect whether the request being made is granted. She explored medical peer review telephone calls in which physician-reviewers representing a national utilization review ﬁrm call physicians who have proposed the surgical insertion of tympanostomy tubes for the management of recurrent ear infections. The reviewers, at the end of the phone call, approve or decline the surgery on behalf of the patient’s insurance company. Boyd found that the formulation the reviewer employed in moving to the business of the call was signiﬁcantly associated with whether the surgery was approved. Additionally, she found (1997) that in cases where the reviewer’s decision was negative, certain initiating formulations were associated with less interactional conﬂict. Although this research involves relating the same speaker’s actions (i.e., the speaker’s openings and decisions), it shows the importance of the opening as an activity in these contexts.

Symptoms-Only Problem Presentations The most common way that children’s problems are presented is with a symptomsonly presentation. Across the Seaside and Metro data sets, this format was used about 55% of the time (see Stivers, 2002b, regarding the Seaside data set results). The symptoms-only term underscores the fact that the problem presentation offers only a description of the problems the child is experiencing and does not attempt to identify the illness condition. If children present their own problem, it is virtually always with this format. For example, see Extract 2.1. Here a boy who is about 10 years old and his father are visiting the physician because the boy has, as he says, “red spots”. (2.1) 202 1 2 3 4 5 6 7

DOC:

O:kay: Robert. (0.5) DOC: What’s up.=h BOY: -> Uhm I have these little red s:pots all over my body. (0.5) BOY: -> An:’- we don’t know what they are: (really)

In this case, the boy ﬁrst describes his primary symptom (lines 4–5). As a response to “What’s up.” (line 3), the telling of his primary symptom displays his orientation to that symptom as being the reason for their visit. Then, after a bit of silence, he adds a second turn constructional unit (TCU) (Sacks, Schegloff, & Jefferson, 1974) that emphasizes his and his father’s (and perhaps his family’s) concern for a diagnosis of this symptom (line 7). With this second unit “An:’ - we don’t know what they are: (really)” the boy focuses on the evaluation of the spots as the reason for his visit. By contrast, the question of whether the spots are treatable (i.e., treatable with

26

PRESCRIBING UNDER PRESSURE

prescription medication) is not raised and is thus understandably left contingent on the evaluation. Here, then, the matter of diagnosis is the focus of the problem. This case is unusual because the boy explicitly indexes his and his father’s and/or family’s desire for a diagnosis of the illness. It is more common that the request for evaluation be left implicit but nonetheless to be the underlying reason for visiting. This can be seen in Extract 2.2. (2.2) 1188 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11

DOC: MOM: -> -> -> DOC: MOM: -> DOC: MOM: DOC:

And so: do- What’s been bothering her. (0.4) Uh:m she’s had a cou:gh?, and stufﬁng- stuffy no:se, and then yesterday in the afternoo:n she started tuh get #really goopy eye:[s, and every= [Mm hm, =few minutes [she was [(having tuh-). [.hh [Okay so she haso when she woke [up this morning were her eyes= [( ) =all stuck shut,

In line 1, the physician asks for the reason for the child’s visit with an open solicitation. The mother describes several symptoms in response. In lines 3–5/7 she lists a cough, a stuffy nose, and “goopy” eyes. As was the case with the symptoms offered in Extract 2.1, here, too, the mother makes no inference about the cause of the problem but simply states the symptoms as the basis for the visit. Whether the mother believes that the child’s condition is treatable is not disclosed in her problem presentation. Rather, the presentation offers only symptoms for evaluation and leaves it to the physician to determine whether and how the condition will be treated. In offering only the symptoms of the child’s illness, the parents communicate an orientation to the child’s problem as in need of evaluation but as only possibly treatable. We can see this again in Extract 2.3. (2.3) 2058 (Dr. 5) 1 2 3 4 5 6 7 8 9

DOC: BOY: MOM: -> DOC: MOM: -> MOM: -> ->

And what’s going on with you:, (2.0) (°Well-°) (0.4) .tlkh He ha:s uh: rash all over his body, Uh [huh:, [Like head to toe, (0.6) An:d uh:m he ha:s uh #fever#,=’e’s ((kid making noise)) uh hundred ‘n one today,

FOREGROUNDING THE RELEVANCE OF ANTIBIOTICS

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

DOC: MOM: MOM: -> -> DOC: DOC: MOM: -> DOC: MOM: -> DOC: MOM: -> -> DOC: MOM:

27

Mm hm:?,= =Stop it- Stop that. (Zack. Stop it.) ((to child)) (0.8) He’s had uh fever for two day:s, He’s had [uh persistent cough=for uh few weeks, [Mm hm, Uh hu[h:?, [But it w=(h)asn’t been bad enough to bring him in, Uh huh? And he’s (complai:ned) for- uhm- (0.3) (.ml[h) [two days about uh stomach:=ache_ uh: (.)stomach cramping. (1.0) .Tlkh n- n- uhm: for two days? #Yeah:. (an it started yesterday.)

In this interaction, the physician solicits the problem with an open question about the boy’s medical problem. His mother, in response, offers several symptoms. She mentions a rash (line 4), a fever (lines 8 and 13), a cough (line 14), and a stomachache (lines 19/21–22)). As in the other interactions shown thus far, the mother does not offer any theory of what is causing these problems but only details the symptoms. In this way, she treats the symptoms as problematic and as the reason for seeking medical help. For example, the mother also calls the cough “persistent”. At a point where the physician might have begun history taking, she does not, instead offering a continuer (Schegloff, 1982) (line 16). Then, with “it w=hasn’t been bad enough to bring him in,” (line 17), the mother emphasizes the gravity and doctorability of the child’s condition (Heritage & Robinson, 2006). The self-repair (Schegloff, Jefferson, & Sacks, 1977) from what was probably “wasn’t bad enough” in the simple past tense to “hasn’t been bad enough” using the present perfect also suggests a progression of his condition to the current state where he is in need of an evaluation (Bybee, Perkins, & Pagliuca, 1994). She here suggests that because of the accumulation of symptoms, the mother now feels that her son does need a medical evaluation. Although part of this evaluation may include treatment, the parent effectively remains silent on this topic, thus embodying an agnostic stance on the treatability of the child’s condition. In this section, we looked at one communication practice for outlining the reason for visiting: a symptoms-only problem presentation. When parents make use of this communication practice, they convey that their reason for visiting is to have a medical evaluation of their child’s condition and to seek advice for the management of that condition. They remain agnostic about the exact nature of the illness and its treatability. As noted earlier, this type of problem presentation was most common and is typically treated as unmarked or canonical: Both parents and physicians treat this type of presentation as doing “nothing special.” This will be discussed in more detail shortly.

28

PRESCRIBING UNDER PRESSURE

Candidate Diagnosis Problem Presentations The second way that parents regularly present their children’s problems involves mentioning a candidate diagnosis. Although many illnesses could, in principle, be presented with a candidate diagnosis, in these data this format is overwhelmingly reserved for bacterial diagnoses (e.g., ear infections, strep throat, sinusitis) or for illnesses that parents might assume to be bacterial (e.g., pneumonia, bronchitis) rather than viral (e.g., “cold” or “ﬂu”). In contrast, with symptoms-only presentations, candidate diagnoses were much less frequent: They represented only 16% of the problem presentations in a subset of the data (Stivers, 2002b).2 But despite being less frequent, this is still a relatively common type of problem presentation. As mentioned in chapter 1, this is at odds with existing research in the adult context that suggests that patients very rarely offer explicit or implied diagnoses (Gill, 1998; Heritage & Robinson, 2006a; Ruusuvuori, 2000). In a sample of 300 acute care adult encounters, candidate diagnoses were estimated to be present only 8% of the time (J. Heritage, personal communication, February 12, 2006). Reasons for the relative scarcity of this behavior may include patients’ orientations to the physician’s expertise, as well a reluctance to voice more serious diagnostic possibilities. Heritage and Robinson (2006a) suggest that patients may introduce diagnostic claims in support of the doctorability of their problem in cases where a condition has been previously diagnosed or in cases where a rather benign explanation is possible. Additionally, Gill (1998) notes that when patients offer their own theories of causation, they frame them as delicate actions, either by downgrading the certainty of their theory or by offering them speculatively. Ruusuvuori (2000) suggests that such tentative framing of a diagnostic suggestion suggests patients’ orientations to the action as stepping into “medical territory” (p. 165). Although candidate diagnoses appear to be more frequent in these data than in the adult context data, as we will see, parents in these data still orient to the action of offering a candidate diagnosis as delicate. In contrast to symptoms-only presentations, candidate diagnoses can be heard to convey a stance that the nature of the child’s medical problem is already known and thus the reason for the medical visit is primarily to seek treatment for a known condition. We can see an illustration of this in Extract 2.4. Here, in response to a problem solicitation (lines 1–2), the mother offers a candidate diagnosis (lines 4–5) and then offers the child’s symptoms as evidence for the diagnostic conclusion (lines 8 and 11). (2.4) 305 1 2 3 4 5 6 7

DOC:

Al:ri:ght, well what can I do [for you today. MOM: [(°hm=hm=hm=hm.°) MOM: -> .hhh Uhm (.) Uh- We’re- thinking she might -> have an ear infection? [in thuh left ear? DOC: [Okay, DOC: Oka:y,

FOREGROUNDING THE RELEVANCE OF ANTIBIOTICS

8 9 10 11

MOM: DOC: MOM:

29

Uh:m because=uh: she’s had some pain_ (.) [Alrighty? [over thuh weekend:(.)/(_) .h[h

As her reason for the visit, the mother offers her inference that her daughter has an identiﬁable and treatable problem (an ear infection). The claim is epistemically downgraded (e.g., with “thinking” and “might,” as well as with the strong questioning intonation). Additionally, the diagnostic claim is offered with supporting evidence. That turn begins with “because”, suggesting that what will follow is evidence for the prior inference. The observation that is provided is that the girl has had ear pain. In itself, this observation could have been offered as the reason for the visit but, placed as it is here, it is offered as an account for her candidate diagnosis. Despite the mitigation and the account, which both treat the action as delicate, the mother’s turn in lines 4–5 nonetheless asserts the existence of a known and treatable condition—an ear infection. Because this diagnosis suggests a treatable condition, it looks forward to a speciﬁc treatment recommendation—a prescription for antibiotics. A similar situation can be seen in the next example, shown in Extract 2.5. Here the physician’s question in line 1 is an initial history-taking question but in an environment where no problem presentation was solicited. The mother responds with a full problem presentation, including a candidate diagnosis. (2.5) 615 1 2 3 4 5 6 7 8 9 10 11 12

DOC:

.hh So how long has she been sick. (1.2) MOM: Jus:t (.) I came down with it last Wednesday, so she’s probably had it (0.2) DOC: °Uh huh_° MOM: (Like) over- four days? (1.0) MOM: An’ she’s been complaining of headaches. (.) MOM: -> So I was thinking she had like uh sinus -> in[fection er something. DOC: [.hhh

With her TCU initial “So” (line 10), the mother formulates her candidate diagnosis, similar to that of the mother in Extract 2.4, as an inference based on her child’s symptoms. Also similar to the mother in that extract, this mother downgrades the epistemic certainty of the diagnosis with “I was thinking”, “like”, and “er something”. In this case, the symptom of headaches precedes the conclusion offered by the mother as a candidate diagnosis: that the headaches are a symptom of an underlying sinus infection. Another way that parents can work to mitigate explicit self-diagnosis is to further downgrade the epistemic authority embodied in their formulation. A candidate

30

PRESCRIBING UNDER PRESSURE

diagnosis can, for example, be offered speculatively. An example of this is shown in Extract 2.6. Here, although the presenting concern involves the recurrence of similar symptoms, this is not a follow-up visit; rather, the child was treated for a condition previously, and the mother has scheduled a new appointment for a new problem condition, albeit similar to the last illness. (2.6) 316 1 2 3 4 5 6 7 8 9 10 11 12 13

DOC: MOM: DOC: PAT: MOM:

DOC: MOM: DOC: MOM: DOC:

Alrighty? Well- Here:=we go:! How’re you do^ing. Fine how’re you. I’m hanging in there:?, Well hi Matthew how’re you[:. [Fine, (.) .hh I brought ‘im back because ‘is- .hh He tu- we took all thuh medication but he’s been complaining of uh sore throat off ‘n o[n fer like uh week, [O:kay? .hh An’ I [didn’t (know) [(You’ll hafta) refresh my=uh: myHe [had strep. [horrible memory, ((12 lines reviewing medical history not shown))

26 27 28 29 30 31 32 33 34 35 36 37 38 39

MOM:

=But fer like thuh la:st week. Off ‘n on he- he tells me. (Not even just but) going he’ll go “Mom my throat is hurting again.” An’ I noticed it was pink.an’ I- (0.5) DOC: [Huh huh huhMOM: -> [(I-) I thought (0.5) maybe I better just- >I don’t -> know if ya know strep has secondary er anything like -> that I wasn’t sure.But he hasn’t had thuh fever er thuh nausea er anything that he’[s had before. DOC: [O:kay:, DOC: .hh [(Goo:d?,) MOM: [But I thought since t’day’s Veteran’s Day ‘n they’re off school it’d be easier fer me tuh bring ‘im t’day than-

Here, the mother states her worry as a generalized possibility: “I don’t know if ya know strep has secondary er anything like that” (lines 31–32). This boy had a diagnosed strep throat infection several weeks prior to this encounter and was prescribed antibiotics. Here, the mother suggests that strep “has secondary”. Secondary infections are bacterial infections that occur following a viral infection. A secondary throat or ear infection, for example, would normally be treatable with antibiotics. The mother’s use of “secondary” here may be the result of her confusing secondary

FOREGROUNDING THE RELEVANCE OF ANTIBIOTICS

31

infections with relapses. Relapses normally involve the return of an infection after it has appeared to go away. Regardless, what appears quite clear is that the mother is concerned that her son has a throat infection. The mother initiates the move to business by offering her reason for the visit as her son’s complaint of a sore throat, despite his having completed a full course of medication (lines 6–8). After an intervening sequence about the history of the prior illness (omitted lines), the mother reasserts her son’s symptoms by animating her son’s complaint in direct reported speech (lines 27–28). Subsequent to this, she continues with her own observation of his symptoms (line 28) and her diagnostic inference (31–33) that her son has a type of secondary infection from strep throat. Again, this claim is mitigated—this time with “I don’t know”, “ er anything”, and “I wasn’t sure.” However, even with the mitigation, the claim of infection makes treatment for that infection relevant. In this case, the mother has taken a small step away from a direct statement of a diagnostic theory by formulating it as a speculation. Mitigation of a candidate diagnosis can also be accomplished with indirection. When a candidate diagnosis is offered indirectly as, for example, a statement about past illnesses, parents regularly formulate it without mitigation (see Gill, 1998, on indirection). An example of this is shown in Extract 2.7. (2.7) P201 (Dr. 7) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

DOC: DAD: DOC: DAD:

An:- An’ what didju bring her in: for today? She had uh fever this morning, Mm hm?, .h An:d she’s complai:ned of: uh pai:n in her left ca:lf?, (.) DOC: Mm hm:?, DAD: -> And we have ha:d: (1.0) some experience in -> thuh pa:st with s:inus::=sinusitis? DOC: Mm hm? DAD: .hh A:nd it was: (.) uh lo:ng ti:me being diagnosed_=We had tuh go t’ thee emergency room, DOC: Mm hm::?, DAD: Uh::m a:nd ﬁnally thuh doctor the:re (a[t- could ﬁnd it.) This was- (1.0) ﬁve months ago= DOC: [Mm hm. DAD: or so so-.hh So she has had something in the past, <She’s had her- .hh I think her knee was hurting in thuh pa:st. DOC: O:kay. DAD: (An’) they did x-rays ( ) (0.2) DOC: Okay so she had fever: just today: it started, (.) DAD: Yeah.

32

PRESCRIBING UNDER PRESSURE

Here, the father mentions that they have “some experience in thuh pa:st with s:inus::=sinusitis?” (lines 8–9). Although he does not directly state, “I think she has a sinus infection,” the father nonetheless communicates his belief that his daughter’s condition may be sinusitis. Whether direct or indirect, in each of these examples the parent does two primary things: (1) describes one or more symptoms that the child is experiencing and 2) offers an inference about the underlying diagnosis that is producing the symptom(s). As discussed earlier, when parents formulate their reason for the visit with a symptoms-only problem presentation, they make no claims about the treatability of their child’s symptoms. They formulate their child’s medical problem as, in the ﬁrst instance, in need of a physician’s evaluation. However, when parents formulate their child’s problem with a candidate diagnosis, they adopt a stance that their child’s condition is medically problematic and in need of prescription treatment— overwhelmingly antibiotics.

Implied Candidate Diagnoses The candidate diagnoses discussed thus far offer examples that are clearly articulated. However, parents and physicians alike orient to “implied candidate diagnoses” in very similar ways. Thus, ultimately I will argue that candidate diagnoses include both explicit and implied varieties. In this section, though, we will discuss implied candidate diagnoses separately. Implied candidate diagnoses were less frequent than the explicit candidate diagnoses, only 10% of the cases in a subset of the data (Stivers, 2002b). This type of problem presentation represents a hybrid of the two practices outlined thus far. On the one hand, it involves the presentation of symptoms only; however, the symptoms are highly speciﬁc. Their speciﬁcity appears to imply a particular bacterial condition. When compared with many formulations of symptoms, these “diagnosis implicative symptoms” involve a ﬁner level of detail and speciﬁcity, or what Schegloff terms “granularity,” than is typical (Schegloff, 1972a, 2000). The ﬁrst feature suggests that these problem presentations could be classiﬁed with other symptoms-only presentations because they involve no actual articulated diagnosis. But the second feature suggests that when a parent offers diagnosis implicative symptoms they are displaying their stance that the child has the implied condition and is thus in need of treatment for that condition. In particular, they canonically employ a level of “technical speciﬁcity” that is relevant for a health professional rather than for an ordinary recipient (e.g., mentions of the color of nasal discharge or the color of spots in the throat) and is for that reason understood to imply a particular diagnosis. For example, a parent might mention that her child has a “barky” cough to index croup, she might mention green nasal discharge to index sinusitis, or she might mention white or yellow spots on the child’s throat to index strep throat. When parents mention symptoms such as these, the stance they take is similar to that taken up with a direct candidate diagnosis: They treat the symptoms as both medically problematic and treatable. One type of evidence that offering these speciﬁc sorts of symptoms works to index particular infections is that physicians treat them as conveying a concern about

FOREGROUNDING THE RELEVANCE OF ANTIBIOTICS

33

a particular diagnosis. As an example, we can look at Extract 2.8, in which the mother asserts that she saw yellow spots on her daughter’s throat (lines 10/12), which regularly index strep throat. Although the physician does not reject strep throat speciﬁcally, he displays his orientation to the parent as having implied a diagnosis in his formulation of lines 17–18/20–21. He treats a diagnosis of “blisters” or “cold sores” as a position that contrasts with that of the parent. This is accomplished particularly with his mention of “actually” as a preface to his identiﬁcation of the “spot” as “blisters” (line 17). “Actually” marks the ﬁnding as counter to what was previously offered (Clift, 2001; Schegloff, 1996). (2.8) 1126 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

MOM: DOC: MOM: DOC: DOC: MOM: DOC: MOM: DOC: GIR: DOC: DOC: MOM: DOC: MOM: DOC:

MOM: DOC: DOC: DOC: MOM: DOC:

And I- s- she was complaining about her #throa:t.# Nkay:, an’ she had uh fever last night?, (.) Uh::- (.) uh little bit. so I- I kept plying her with Tylenol just to help [#her throat pai:n.# [>Okay,< Sure. (0.2) -> And then uh- I looked down her throat yesterday-> last ni:ght, an’ I could see thuh yellow:_ ^Okay. -> #spo:[t so:. ((trails off)) [.hh Well open up rea::l big. let’s take uh look an’ (say-) say #”Ah:::[:::.”=hh [Ah::::=hh .hh (0.5) °Yeah:.° You know actually what those a:re °pr=h° .hh are primarily blisters back there. Yea:h? It’s almost like she’s got cold sores in thuh back of ‘er throa:t. (Oh:[::.)/(Aw:::.) [And u:sually that’ll go along with this just being viral. (.) [Really.= [#er-# =Y:eah. .hh => One ‘v thuh teachers told me it might be stre:p => so:[:_ [.mlk Yeah we are starting to see some strep so I’m gonna culture just in case .hh she’s got both going on at the same ti:me but- .hh when you

34 35 36 37 38 39

PRESCRIBING UNDER PRESSURE

MOM: MOM:

see: (you know)/(any uh) those #uh:# (thuh)/(that) white stuff you see back there is- is really not: like pus pus but it’[s ya know like she’s got blisters n’ [Oh yeah:_ Oh:::.

That the parent’s original diagnosis implicative symptom was designed to imply a bacterial candidate diagnosis is made explicit in line 30. Here, the mother identiﬁes the diagnosis of concern as strep and further asserts an account for this concern: “One ‘v thuh teachers told me it might be stre:p.” Her turn ﬁnal “so” retroactively casts her suspicion of the spots as having been related to this teacher-offered candidate diagnosis (Raymond, 2004). Typically the implied diagnosis is not brought to the surface of the interaction, but in this case we have clear evidence that the parent’s mention of spots earlier in the encounter was an indirect way of conveying her worry about strep. The candidate diagnosis that was implied is attributed to a third party, distancing the mother from the diagnosis that was previously discounted (Clayman & Heritage, 2002). Another illustration is offered in Extract 2.9. Following some detailing of symptoms (earlier in data not shown and here in lines 8–11/13), the physician, in overlap, shifts into a joking examination of the girl’s stomach. The mother returns to the symptoms and the problem presentation as the physician’s joking talk is reaching completion. (2.9) 1050 (Dr. 1) ((just following some joking about responses to DOC’s initial inquiry “What’s up.” . . . “the sky”)) 1 2 3 4 5 6 7 8 9 10 11 12 13

DOC: MOM:

MOM: GIR: DOC: MOM: DOC: MOM:

And what else. (2.2) Tell thuh doctor what did you told me this morning.= When I was brushing=uh (.) your hair. (0.5) What=do you have. (.) Uh tummyache.=[h [.h She’s had (uh) fever for three [days she’s had a cold off an’= [Lemme feel y- ((move to examine girl’s stomach)) =o:n for about (three) days.

((14 lines not shown DOC begins exam feeling child’s stomach - joking)) 28 29 30 31 32

MOM: -> I thought I saw the little white (.) dot[s, DOC: [.h There was one little sp:o:t_ but it didn’t look -too ba:d. MOM: -> Because sh- there’s strep throat goin’ around -> [in her class an:- an’ I can’t seem to get rid of=

FOREGROUNDING THE RELEVANCE OF ANTIBIOTICS

33 34 35 36

DOC: MOM: DOC: MOM:

35

[Yeah: w=this (.) co:ld an’_ .h she’s beenTurn your hea:d, really high fevers.

Here the mother offers another symptom—this one a diagnosis implicative symptom (line 28). When the mother says that she saw “the little white (.) dots,” with the deﬁnite article “the”, she conveys that these dots are speciﬁc and have a previously established meaning. Additionally, this symptom hearably indexes a diagnosis of strep throat. In response, the physician rejects the implicit claim that the dots are problematic (lines 29–30). Subsequently, the mother makes explicit that the diagnosis she was alluding to with the diagnosis implicative symptom was strep throat (line 31). But this more overt stance toward her daughter’s illness as treatable is not displayed until the physician rejects the parent’s less direct conveyance. The candidate diagnosis offered in lines 31–32 hearably accounts for her prior statement. The physician also treats this as an account by accepting the turn at ﬁrst possible completion with “Yeah.” However, both statements (lines 28 and 31–32, respectively) appear to convey the same diagnostic theory. A ﬁnal example is shown in Extract 2.10. Here, as part of the narrative presentation of her daughter’s problem, the mother states that “she started with uh little clear ﬂui:d on: uh:m h tlk Saturday. . . . And then- by yesterday it turned- gree:n,” (lines 6–7/15/17). With her use of “started”, she projects that there has been some change. Additionally, “clear” suggests that the change may be in terms of color because clear ﬂuid is nonproblematic, and typically a problematic formulation would be simply a “runny nose” or a “lot of drainage” rather than the naming of a color. (2.10) 1046 (Dr. 1) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

DOC: MOM:

Oka::y, so:, let’s see what’s doin’ he:re?=hh We:ll, Erin:, thuh ﬁrst up to bat here, (0.2) she: uhm (.) ^she’s been ac[ting prettyDOC: [.hh DOC: Come clo[se to me (Er,) MOM: [pretty happy but- .hh she started with uh little clear ﬂui:d on: uh:m h tlk Saturday. (.) MOM: running out of her no:se_ DOC: (Who[o hoo) ((whistled)) MOM: [and draining into ‘er throat_ DOC: I think there’s uh bird in ‘er ear:.= MOM: =#huh hu[h# ((throat clear)) DOC: [Did=you hear tha:t? MOM: -> And then- by yesterday it turnedGI?: #Hu::h hu[h# ((cough)) MOM: -> [gree:n, DOC: Kay her ears look perfect.

36 19 20 21 22 23 24 25

PRESCRIBING UNDER PRESSURE

MOM: MOM:

Okay:, .h And=uh .h it’s mostly at night when- it drains dow:n [it’s:DOC: [Yeah_ MOM: -> An’ I’ve had uh sinus infec[tion, DOC: [Okay. Open up your mouth real wide-

As the mother continues her narrative, she does, in fact, assert that the color changed to green. With this level of technical speciﬁcity, the characterization indexes a diagnosis of a sinus infection. That this is the mother’s design is made explicit when, in her next turn, the mother states that she has had a sinus infection. The implication is that the mother has experienced similar symptoms and thus believes that green discharge can be a symptom of sinusitis. That the mother’s turn in 23 is designed to be connected to her earlier diagnosis implicative symptom is partly carried by the “An’”, which connects it back to what was said previously. This helps the turn to be heard for its ramiﬁcations for the daughter rather than as a discrete unrelated announcement. Implied candidate diagnoses can be seen to have a resemblance to both symptoms-only presentations and articulated candidate diagnosis. However, in this section we have seen that parents appear to use them to index particular diagnoses rather than suggesting them outright or explicitly with articulated candidate diagnoses. In this way, they appear to be displaying the stance that their child has a given condition and that they are seeking treatment for that condition. Combined, then, in over a quarter of cases, parents are identifying their reason for visiting in a way that physicians treat as seeking antibiotic treatment. Primary evidence for this claim is found in the way physicians respond to the two different types of problem presentations.

Responses to Problem Presentations When parents offer a symptoms-only presentation, physicians treat them as embodying a stance that their child has a doctorable condition for which they are seeking medical evaluation. By contrast, when parents offer a candidate diagnosis, physicians treat them as embodying a stance that their child has not only a doctorable condition but also a treatable one and, further, that parents are lobbying for antibiotic treatment. So by using a candidate diagnosis rather than a symptoms-only problem presentation, parents initiate a negotiation of their child’s treatment in favor of antibiotics. Responding to Symptoms-Only Presentations As noted earlier, symptoms-only presentations are the most common type of problem presentation and are oriented to as the unmarked format for presenting the problem.3 This section focuses on two ways in which physicians respond to symptoms-only presentations preparatory to a contrast with physicians’ responses to candidate diagnoses. First, physicians typically move from a symptoms-only presentation directly

FOREGROUNDING THE RELEVANCE OF ANTIBIOTICS

37

into an investigation of the child’s problem, which, as has been noted, is the most common trajectory for a medical visit (Byrne & Long, 1976; Robinson, 2003; Waitzkin, 1991). This may mean a move directly to physical examination (as seen in Extract 2.1) or (most commonly) a move into history taking (as shown in Extracts 2.2 and 2.3). But what is critical is that the physician does not in any case in these data take issue with parents about the symptoms they describe. This suggests that physicians treat symptoms-only presentations as making an investigation of the patient’s problem the most immediately relevant next activity. Moreover, physicians typically formulate their subsequent diagnoses as direct, positively formulated announcements. That is, they offer the diagnosis without an orientation to a previously implied or articulated diagnosis, thereby treating an explanation of the problem as the primary task set by the parent’s problem presentation. Both of these features are illustrated in Extract 2.11. (2.11) 1188 (Dr. 3); [problem presentation shown previously in 2.2] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

DOC:

And so: do- What’s been bothering her. (0.4) MOM: Uh:m she’s had a cou:gh?, and stufﬁng- stuffy no:se, and then yesterday in the afternoo:n she started tuh get #really goopy eye:[s, and every= DOC: [Mm hm, MOM: =few minutes [she was [(having tuh-). DOC: [.hh [Okay so she haso when she woke [up this morning were her eyes= MOM: [( ) DOC: =all stuck shut, MOM: Yeah but- Well actually during thuh middle of the ni:ght [she woke u[:p_ and they we[re stuck shut n’_ DOC: [Okay, [Okay_ [Okay_ 1-> An’ how about fever. Any fever at all? ((33 lines of history taking/examination not shown))

49 50 51 52 53 54 55 56 57 58 58 60 61

DOC: 2-> 2-> 2-> MOM: DOC: DOC: 2-> 2->

Basically she’s mov- i- she’s: :y’know> kinda: developed the co:ld an’ respiratory thing that’s goin’ arou:nd. [Uh huh, [.hh An’ it’s moved into her eyes, so she’s got like #uh:# pink eye or conjunctivitis. .hh and so thuh: cou:gh, and the stufﬁness I would treat symptomatically with uh cough an’ cold medicine like Pediaca:re, Dimetapp, whatever:. DOC: .hh And then I’m gonna give you some eyedrops to put in her eyes_ MOM: Okay?, ((DOC continues on to detail dosage))

38

PRESCRIBING UNDER PRESSURE

Here, we ﬁrst see that at arrow 1 the physician moves from establishing the reason for the child’s visit directly to taking the patient’s history. Second, at arrows 2, when the physician delivers his diagnosis, it is simply asserted rather than framed as rejecting an alternative, denying the parent’s theory, or conﬁrming it. It states that the condition is a “cold” and “pink eye”. In lines 55–60, the physician outlines his treatment recommendation for the two conditions. This, too, is formulated straightforwardly as a proposal. Like the problem presentation, the diagnosis and treatment recommendation are offered in an unmarked way, suggesting that they are providing only an evaluation and advice on treatment. This is further supported by the example shown as Extract 2.12. As with Extract 2.11, here, too, the physician moves directly from establishing the reason for the visit into history taking (arrow 1). (2.12) 2058 (Dr. 5); [problem presentation shown previously in 2.3] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

DOC:

And what’s going on with you:, (2.0) BOY: (°Well-°) (0.4) MOM: .tlkh He ha:s uh: rash all over his body, DOC: Uh [huh:, MOM: [Like head to toe, (0.6) MOM: An:d uh:m he ha:s uh #fever#,=’e’s ((kid begins noise)) uh hundred ‘n one today, DOC: Mm hm:?,= MOM: =Stop it- Stop that. (Zack. Stop it.) ((to child)) (0.8) MOM: He’s had uh fever for two day:s, He’s had [uh persistent cough=for uh few weeks, DOC: [Mm hm, DOC: Uh hu[h:?, MOM: [But it w=(h)asn’t been bad enough to bring him in, DOC: Uh huh? MOM: And he’s (complai:ned) for- uhm- (0.3) DOC: (.ml[h) MOM: [two days about uh stomach:=ache_ uh: (.) stomach cramping. (1.0) DOC: 1-> .Tlkh n- n- uhm: for two days? MOM: #Yeah:. (an it started yesterday.) ((48 lines of history taking and exam not shown))

74 75 76 77 78

DOC: MOM: MOM: DOC: DOC: 2->

.Tlkhh You want to [know what you ha:ve? [HisHis chest and his genital:s are the reddest, #Yeah:.#=h He’s got scarlet #fever:#.

FOREGROUNDING THE RELEVANCE OF ANTIBIOTICS

39

After the history taking and physical examination (data not shown), the physician moves to offer her diagnosis (shown in line 78). Similar to Extract 2.11, here, too, the physician formulates the diagnosis afﬁrmatively and straightforwardly. This sequence begins quite early in the physical examination. The physician’s turn in line 74 is hearably a preannouncement (Terasaki, 2004) addressed to the boy with “you.” This may indicate that the forthcoming news is delicate or unusual. However, the mother does not orient to the physician’s turn as initiating a pre-sequence. Rather, she does some additional work to assert the problematic nature of her child’s condition by offering an additional problematic symptom (lines 75–76). In this way, the mother may be treating the pre-announcement as preceding the full investigation of the boy. In line 77, the physician offers minimal agreement with the mother’s turn before moving directly to her diagnostic assertion that the boy has “got scarlet #fever:#.” It is also notable that the physician has now shifted from addressing the boy to addressing his mother (evidenced by her reference to the boy using the thirdperson pronoun “he”). In looking at these cases, what we can see is that when parents use a symptomsonly problem presentation formulation, doctors treat them as taking a stance toward their visit as legitimate and to their child’s condition as doctorable, but doctors do not treat parents as making any claims about the child’s treatability. Parents in these encounters speciﬁcally orient to the diagnosis, leaving treatment “to the physician.” There is no explicit orientation to whether the condition is in need of treatment. And physicians in these encounters routinely treat these parents as primarily seeking an evaluation of their child’s illness. In these cases, then, parents are not observably vying for any type of treatment, nor are they treated by physicians as doing so. Earlier in this section, I suggested that the symptoms-only presentation may be the unmarked way of presenting a child’s problem. In addition to it being the most common format, evidence supporting this is that physicians respond to these presentations with an unmarked diagnosis delivery: Physicians routinely move from establishing the reason for the child’s visit into an investigation of the problem, and they routinely offer their diagnoses and treatment recommendations as simple, straightforward announcements (i.e., not apparently responsive to, in the sense of conﬁrming or disconﬁrming, any particular previous diagnostic theory). In the straightforwardness of their formulation, these diagnosis announcements appear to be doing “nothing special.” Responding to Candidate Diagnosis Presentations If symptoms-only presentations are the way parents show themselves to be doing “nothing special,” candidate diagnoses convey the reverse: that something special is being done. Physicians display this analysis in the way that they respond to candidate diagnoses—whether suggested or implied—by (1) treating conﬁrmation or disconﬁrmation of the parent’s proposed diagnosis as relevant and (2) orienting to the relevance of antibiotic treatment. The two most common locations where physicians address parents’ candidate diagnoses are immediately after the presentation—particularly with disconﬁrmations or challenges—and during the diagnosis phase. We will look at each of these contexts.

40

PRESCRIBING UNDER PRESSURE

Candidate Diagnosis Uptake: Just Following the Presentation As was just discussed, following symptoms-only problem presentations, physicians typically move directly into history taking or physical examination. In these data, there are no cases of physicians challenging the existence of a parent-reported symptom such as ear or throat pain, congestion, or a runny nose. By contrast, if a parent presents a candidate diagnosis, the physician may counter that diagnosis then and there. In the Seaside data set, this occurred 19% of the time (Stivers, 2002b). Thus, physicians treat the two types of presentations quite differently. As an illustration, we can return to the case shown earlier as Extract 2.5. After the mother presents her daughter’s problem and offers a candidate diagnosis of a sinus infection, the physician moves to counter the proposed diagnosis. (2.13) 615 [shown earlier in Extract 2.5] 10 11 12 13 14 15 16

MOM: -> So I was thinking she had like uh sinus in[fection= DOC: [.hhh MOM: -> =er something.= DOC: => =Not necessarily:, Thuh basic uh: this is uh virus basically:, an’=uh: .hh (.) thuh headache seems tuh be:=uh (0.5) pretty prominent: part of it at ﬁr:st uh: (0.2) .hh

The physician’s turn in lines 13–16 is clearly responsive to the mother’s candidate diagnosis at lines 10/12. In the ﬁrst TCU of line 13, although slightly mitigated, the physician rejects the mother’s assertion as unlikely. The forcefulness of the counter is partly carried by its being latched to the mother’s turn in line 12. Although the physician’s ﬁrst TCU does not completely rule out a sinus infection, in the second TCU he asserts that “this is uh virus basically:,”. This offers an alternative diagnosis unequivocally and thus strongly rejects the mother’s candidate diagnosis. The third TCU suggests that the headache is part of this viral condition, accounting for one of the symptoms that the mother stated had led her to her own candidate diagnosis, thus rejecting not only the mother’s conclusion but also her logic. Unlike the way physicians respond to symptoms-only presentations, after candidate diagnoses, physicians are more likely to take up candidate diagnoses, whether to contest or support them. A similar example can be seen in Extract 2.14. In this case, the mother presented her candidate diagnosis as: “>I don’t know if ya know strep has secondary er anything like that I wasn’t sure. But he hasn’t had thuh fever er thuh nausea er anything that he’s had before” (shown earlier in 2.6). Following the completion of a somewhat extensive joking sequence (data not shown), the physician moves to address the mother’s diagnosis. (2.14) 316 [full presentation shown in 2.6] 31 32

MOM: -> [(I-) I thought (0.5) maybe I better just- >I don’t -> know if ya know strep has secondary er anything like

FOREGROUNDING THE RELEVANCE OF ANTIBIOTICS

33 34 35 36

41

-> that I wasn’t sure.But he hasn’t had thuh fever er thuh nausea er anything that he’[s had before. DOC: [O:kay:, DOC: .hh [(Goo:d?,) ((16 lines not shown including joking about BOY having day off but not MOM))

53 54 55 56 57 58 59 60 61 62 63 64 65 66 67

DOC: DOC: => => => DOC: => => MOM: DOC:

DOC: MOM: DOC:

££O:kay:,££ .hh Well:, (.) o:ne good thing is: that- uhm (0.5) strep infections:- respond really well tuh amoxicillin. .hh so wh:ile he may not have strep any more (.) he could still have- uh viral process going on, he could still have just residual sore throa:t, .h[h dry weather kind of things, .hhh [(°Okay.°) Uhm: besides having an actual infection so we can always look at those issues, .hh an then if you want we can also just retest his throat. (.) An’ make sure there’s no more strep there too. (.) Well (you it) kinda depends on what you- what you [think. [Mkay,

The mother framed her candidate diagnosis in a way that allowed both agreement and disagreement. That is, on the one hand, she suggested evidence that her son no longer has strep (“hasn’t had thuh fever er thuh nausea” in lines 33–34). But she still speculated about strep as a diagnostic possibility (“I don’t know if ya know strep has secondary er anything”). With this possible diagnosis on the interactional table, the physician addresses both dimensions of the mother’s presentation. In line 54, he takes an inbreath, prefaces his turn with “Well” stretches the “Well:,” and then delays the turn further with a micropause. All of these features are common in dispreferred turn formats (Heritage, 1984b; Pomerantz, 1984). This turn design projects that the physician is headed toward disafﬁliation with the mother’s candidate diagnosis. But the physician also frames his response as in agreement with the mother through his “wh:ile he may not have strep any more” (line 56). This works to maximize the appearance of agreement with the mother’s problem presentation. The physician also goes on to validate the mother’s reason for visiting: “he could still have- uh viral process going on, he could still have just residual sore throa:t, .hh dry weather kind of things,” (line 58). In this way, the physician both counters the mother’s candidate diagnosis of strep and validates her reason for coming (by referencing alternative causes of the sore throat). Finally, in lines 61–62, the physician offers another sort of response; he offers to retest the boy’s throat to make sure that he no longer has any strep. When a candidate diagnosis is implied, physicians also routinely act responsively. Here, we can see an example both of a disconﬁrmation and of an orientation to the physician’s orientation that the candidate diagnosis was looking forward to antibiotic treatment. In Extract 2.15, the mother presents her child’s symptom of

42

PRESCRIBING UNDER PRESSURE

nasal discharge by speciﬁcally mentioning the color as problematic. She says, “it’s gotten- it was green” (line 4). (2.15) P110 (Dr. 1) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

DOC: GIR: MOM: -> DOC: MOM: -> -> DOC: MOM: -> DOC: => => MOM: DOC: => MOM: DOC: => MOM: DOC: => => DOC: => MOM:

You’re sick_ well what’s u:p. (1.1) I don’t kno[: [w, [B[etween yesterday and toda:y, she-= [How- hh =.hh ya know it’s (this-)/(j’s-) nasal crap an’ it’s gotten it was gree[:n.=it was [uh= [[Nkay:, =really uh beauti[ful color (yesterdange ) [.hh Okay well just because it’s green [it doesn’t [(doesn’t mean [an-) (I kn-) [ma- mean it’s bacterial. (Right right right right.) There’s a [who:le new thing about: uh: .h - sinus [I know. an’ everybody’s saying we’d so we’ve been trying very hard not tuh put kids on antibiotics if we can avoid it, Ri:ght. right.

We can see evidence that the physician hears the mother’s mention of the green color as indexing a bacterial sinus infection if we look at his virtual rejection of the diagnosis with “well just because it’s green it doesn’t ma- mean it’s bacterial.” (lines 10–11/13). Although no direct candidate diagnosis is offered here, the physician treats the particular formulation of symptoms as clearly implying one, and in that way he is enabled to reject it in a way that does not emerge in response to symptoms-only formulations. Moreover, following the rejection of the mother’s implied candidate diagnosis, the physician goes on to foreshadow his own unwillingness to prescribe antibiotics (lines 14/16–19). With this, he displays his understanding that the mother’s candidate diagnosis was not only seeking conﬁrmation or disconﬁrmation but also working to advocate for, and thus to initiate a negotiation in favor of, antibiotic treatment. In this section, we have examined immediate responses to candidate diagnosis formulations. Although no instances of a conﬁrmation or disconﬁrmation of a symptoms-only presentation occurred in these data, such responses immediately following a candidate diagnosis do occur. By responding to candidate diagnoses in this way, physicians treat these presentations as involving diagnostic and treatment implications that are not implied by symptoms-only presentations. This case is made stronger when we observe that such responses occur not just immediately but are even

FOREGROUNDING THE RELEVANCE OF ANTIBIOTICS

43

more commonly addressed later in the encounter, when there is a candidate diagnosis problem presentation. Candidate Diagnosis Uptake: During the Counseling Phases The second primary area where physicians can be seen to directly address parents’ candidate diagnosis presentations is in the counseling phases, when they offer their ﬁnal diagnosis and treatment recommendation. In these locations, physicians work to show that the diagnosis and treatment recommendations they are providing are being offered in light of the earlier candidate diagnosis. This occurred 71% of the time in a subset of these data (Stivers, 2002b). As an example, see Extract 2.16. In this case, the mother earlier offered a candidate diagnosis that her daughter has a sinus infection. In line 42, the physician is completing his examination of the girl and offering his diagnosis. (2.16) 615 [shown earlier in 2.5 and 2.13] 10 11 12

MOM: -> So I was thinking she had like uh sinus in[fection= DOC: [.hhh MOM: -> =er something.= ((29 lines of history taking and examination not shown))

42 43 44 45 46 47

DOC: => .hh Uh: (1.4) Let’s see (now ) (1.1) I think uh: I => don’t think she h:as: uh: sinus infection,Have you noticed uh lot of (0.2) heavy drainage:?, (0.2) MOM: Yeah she’s been:: (.) When she does cough she coughs up (the-)/(th’t) (.) gree:n, (1.0) that mucus stuff?

When the physician begins his diagnosis, he appears to be headed for a diagnostic assertion with “I think” but, initiating repair on this turn beginning, changes tack and instead designs a diagnosis that is responsive to the mother’s problem presentation. He does this by disconﬁrming her candidate diagnosis with “I don’t think she h:as: uh: sinus infection” (lines 42–43). Additionally, the formulation the physician uses to disconﬁrm the diagnosis is the very one the mother used: “sinus infection.” This may seem inconsequential, but in the next case (Extract 2.17), we can see that the father uses a different expression for the same diagnosis (this time “sinusitis”) in his candidate diagnosis in lines 8–9. When the physician reaches her own diagnosis, it is this formulation that she returns to. (2.17) P201 (Dr. 7) [full problem presentation shown in 2.7] 8 9 10

DAD: DOC:

(1.0) some experience in thuh pa:st with s:inus::=sinusitis? Mm hm?

((254 lines of history taking and examination not shown))

44 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280

PRESCRIBING UNDER PRESSURE

GIR: DOC:

Ah::[:=hhh [Just a teeny teeny teeny bit. See in thuh back [there, DAD: [Uh huh, (.) DAD: [Uh huh, DOC: [But nothing too #ba:d,# so that might be ea:rly or- .hh uh:m:=but ^otherwise :her ears look great.= =She’s not< having uh lot uh mucus or stuff. You usually get- i- bad ear infectio:ns_ .hh a know -after you get uh lot of co:ld, => I=don’t=know if that’s been=’er history in thuh => pa:st but- .hh uh lotta times you’ll get sinusitis => or ear infections after a lot of mucus up here, => and right now she’s - pretty clear_ it seems like it’s mostly the fever,

At this point in the visit, the physician has completed most of her examination and is, across lines 265–270, inspecting the girl’s throat and perhaps inviting the father to look with her at the girl (with “See in thuh back there,”). The physician mentions both sinusitis and ear infections as conditions that can occur after a cold but asserts that “right now she’s - pretty clear_” as well. The initial assertion (line 273) that “She’s not> having uh lot uh mucus or stuff.” appears to be mentioned as support for her claim that there is no sinus infection. Here, as in Extract 2.16, the physician uses the characterization of the illness that was used by the parent to disconﬁrm it as a diagnosis. We can see another illustration of a diagnosis that is delivered as a disconﬁrmation subsequent to an earlier candidate diagnosis in Extract 2.18. Here, the physician returns to the child’s ears and presents his diagnosis as regretfully disconﬁrming (lines 65–66/68). (2.18) 1017 (Dr. 1) ((Simpliﬁed; BRO is the child patient’s brother)) 1 2 3 4 5 6 7 8

MOM: BRO: BRO: DOC: BRO: DOC: MOM:

He- no[::. he’s thuh- he’s got the ear infection. [He’sHe- he (los-) He’s SICK. You ^think so? Yeah. He’s sick. [#Oh.# Well I can see he’s not smi:[ling, C[(Hm_) ((laugh)) [Kind of listless.

((53 lines of problem presentation, history taking and examination not shown)) 62 63 64

DOC:

.hh °Yeah #Say° “Ah:-”
FOREGROUNDING THE RELEVANCE OF ANTIBIOTICS

65 66 67 68 69 70

45

DOC: => hh=£Wish (we) could s(h)ay h(h)e h(h)ad an e(h)ar => i(h)nfection butMOM: ( [I don’t know what_ Yeah:.) DOC: => [£I don’t see: it.£ (.) MOM: Go^od.

The physician’s diagnosis is clearly retrieving the mother’s prior candidate diagnosis of an ear infection by repeating it here as something he “wishes” he could say. Additionally, in formulating his diagnosis in the negative “£I don’t see: it.£” (line 68), he treats his ﬁnal diagnosis as one that is, by design, disconﬁrming the mother’s candidate diagnosis in second position rather than asserting a diagnosis in ﬁrst position.4 In the counseling phases, as earlier in the visit, physicians treat direct and implied candidate diagnoses as being basically functionally equivalent. So, just like direct candidate diagnoses, implied candidate diagnoses are also routinely conﬁrmed or disconﬁrmed during the physician’s ﬁnal diagnosis. In the following case (Extract 2.19), the father implies a diagnosis of swimmer’s ear when he mentions “He’s been swimming a lot,” (line 6) after having mentioned ear pain. (See Gill, 1998, for a discussion of patients’ diagnostic explanations in this format.) Although this case differs slightly in that this is not the symptom, note that the symptom of “eara:che” is here being explained by a fact of swimming, although the diagnosis is not here stated. This is quite similar to stating a noticing of green discharge or pus on the tonsils without asserting a diagnosis of sinusitis or strep throat. ( 2.19) 1189 (Dr. 2) 1 2 3 4 5 6 7 8

DOC: DAD:

DAD: DOC: DAD:

Well Charlie’s got an eara:che. #A[w::.# [Well- YeahHis- ba- it’s bothering him la- lot of swimming. (0.5) He’s [been swimming a lot, an’ then he went= [Okay. =to thuh snow.

((30 lines not shown; history taking and physical examination)) 39 40 41 42 43 44

DOC: DOC: DOC:

Let’s peek at=your ea:r. (3.0) .hh so=what:=h=°ow:.° (0.5) Well he does not have a swim ear: but he does have a middle ear infection,

In this case, after the physician has examined the child, he says, “Well he does not have a swim ear: but he does have a middle ear infection,” (lines 43–44). With this formulation, the physician treats the parent as having implied a diagnosis of swim-

46

PRESCRIBING UNDER PRESSURE

mer’s ear to explain his son’s ear pain. Thus, the physician does not simply assert the ear infection but ﬁrst rejects the implied diagnosis. Here, the disagreement is evidenced by the “Well” preface (Pomerantz, 1984) and the rejection component “does not have a swim ear:”. The physician only then provides the positive diagnosis. This is marked as contrastive both with the negative diagnosis and with the “but”. Additionally, with the second “does” the physician can be heard to support the father’s claim that his son has ear pain. The stress marks the second component as contrastive with the negative and suggests that although the parent was wrong on one count (with the implied diagnosis), he was right in his recognition of a medical problem. There is also evidence that when parents present their children’s problems with a candidate diagnosis, conﬁrmation-disconﬁrmation is treated as relevant in the course of the visit. For instance, in Extract 2.20, when a physician fails to address the parent’s candidate diagnosis, the parent reinvokes it as a question (in lines 75–76). (2.20) 1141 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12

DOC:

So: what’s goin’ on he:re. (1.2) MOM: He’s got uh:- (0.2) tlk (They’re kinda-) He stayed out of school on Monday:, DOC: [Uh huh, MOM: [(w-) MOM: With kind of #uh-# low grade fever an’- (.) uh crummy no:se_ an’ now he’s complaining about -> ears_
73 74 75 76 77

DOC:

Yeah I think he’s got thuh bug that’s goin’ around right now: [anMOM: => [(Oh) you don’t- He doesn’t have uh => infection? DOC: I don’t think so. Not yet.

In line 9, the mother offers a candidate diagnosis of an “infection”. This is responded to during the physical examination (data not shown), but in the ﬁnal diagnosis, the physician neither conﬁrms nor disconﬁrms it until the parent requests that conﬁrmation (lines 75–76). Here, the actual diagnosis is given with “I think he’s got thuh bug that’s goin’ around right now:” (lines 73–74). In response—in fact at ﬁrst possible completion—the parent treats her primary concern as remaining unaddressed and pursues conﬁrmation of that diagnosis (lines 75–76). Although the physician’s turn was not apparently designed to be complete at that point (both intonationally and with the presence of a cutoff following “an’”), the mother still begins her turn, coming in immediately upon possible grammatical completion, to ask about whether the child has “uh infection?” (Sacks et al., 1974). Here, she uses the same formula-

FOREGROUNDING THE RELEVANCE OF ANTIBIOTICS

47

tion she used previously, which helps display her action as requesting conﬁrmation of a diagnosis she had offered earlier in her problem presentation rather than asking about something new. It is only at this point that the physician actually disconﬁrms the diagnostic possibility. This example suggests that parents actively monitor physicians’ diagnoses for the way they address previously suggested diagnoses. Interestingly, this appears to be the case even when physicians display attending to the parent’s concerns earlier in the visit. Here, the physician had attended to the parent’s concern of infection during the physical examination, but the mother did not treat this as sufﬁcient and still pursued the physician’s conﬁrmation-disconﬁrmation in the ofﬁcial diagnosis phase. I have claimed that the use of a candidate diagnosis represents one of the earliest forms of parent behavior that can systematically inﬂuence the treatment outcome by conveying pressure for antibiotic treatment. Earlier, we saw that the relevance of antibiotics following a candidate diagnosis could be observed immediately following the problem presentation (Extract 2.15). Physicians also display an orientation to the relevance of antibiotics following a candidate diagnosis later on in the visit, such as in the diagnosis or treatment phase. For example, see Extract 2.21. (2.21) 305 ((shown earlier in Extract 2.4; Extract 2.21 begins just following the physical examination)) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

DOC:

.hh So: it would loo:k hh like she is:=uhm (.) prob’ly ﬁghting some (.) viral: upper respiratory kinda stuff, DOC: .hh More on thuh left than on thuh right, which c[an account for some pain maybe, MOM: [Okay. DOC: .hh Uhm:=hh Ears are- hh I mean .hh there’s not even uh lot of wax in her ears. Her ears are prett[y clea:n. MOM: [( ) DOC: I mean [they look s- sec- exceptional_ .hh (°ya know.°) MOM: [(Great.) MOM: °Yeah: [uhm° DOC: [For uh kid [her age. MOM: [huh huh huh .h[hh She loves tuh have= DOC: [Good job. MOM: =her ears cleaned. hzuh hu[h huh .hhh huh huh huh DOC: [Well- (.) fantastic (cuz) DOC: (they’ve be-) you guys are doing uh great jo:b, .hhh MOM: (M[m.) DOC: [Uh:- I would tell you though I don’t hhh (.) I -> don’t see anything that requires like antibio:tics er anythi:ng, but certainly sympto[matic treatment might= MOM: [Mm. DOC: =be in order, DOC: .hh MOM: Okay.

48

PRESCRIBING UNDER PRESSURE

Similar to Extracts 2.16–2.19, here the physician formulates his diagnosis as responsive to the mother’s candidate diagnosis: First, he conﬁrms that the girl has an illness. This is carried with the full form “is . . . ﬁghting” and the additional stress on “is” that works to conﬁrm. (See Stivers, 2005a, for a discussion of this practice in immediately subsequent position.) Second, he accounts for the pain that the mother mentioned previously as evidence for her candidate diagnosis, noting that there may be more infection on the left than the right, “which can account for some pain maybe,” (lines 3–4). Here, the physician’s use of the word “pain” ties back to the mother’s own use of “pain” in her candidate diagnosis. Third, the physician disconﬁrms the candidate diagnosis in that he speciﬁcally targets the ears to note that there is not “even” wax, which suggests that there was something else being searched for, and neither the searched-for item (i.e., infection) nor the more minimal wax could be found. Finally, when the physician begins his treatment recommendation in line 19, he formulates this as responsive also. Although antibiotics had not been explicitly raised previously in this visit, the physician frames this recommendation as responsive. This is accomplished in part by his raising them at all. Other potentially relevant medications are not ruled out, so the raising of this treatment is signiﬁcant and displays an orientation to their signiﬁcance for the visit. Additionally, the use of “I would tell you though” suggests that this is part of his response to the mother. He has provided a conﬁrming response in that the child is ﬁghting an illness. Here, he is providing the counterpart, disconﬁrming a need for antibiotics. The “though” carries much of the weight in establishing the utterance as contrastive with the position taken by the mother in the earlier problem presentation. Delivering the treatment recommendation in these ways suggests the physician’s understanding that the parent was oriented to antibiotics as the appropriate treatment for the illness she said she believed the child had—an ear infection—communicated through her use of a candidate diagnosis early in the visit. Quantitative evidence also supports the claim that physicians perceive candidate diagnoses as treatment implicative. For the Seaside data, when parents used a candidate diagnosis, physicians were 5.23 times more likely to report having perceived the parent as expecting antibiotics (p >.05) (Stivers, Mangione-Smith, Elliott, McDonald, & Heritage, 2003). The same pattern emerged in the replication of the ﬁrst study with the larger and more ethnically and socioeconomically diverse Metro data (Mangione-Smith, Elliott, Stivers, McDonald, & Heritage, 2006). But parents may not always be using candidate diagnoses with the motivation to advocate for antibiotics. Parent use of candidate diagnoses was not signiﬁcantly associated with parents’ reporting an expectation for antibiotic treatment (Mangione-Smith et al., 2006; Stivers et al., 2003). We have seen that physicians treat symptoms-only problem presentations as making relevant only an investigation and evaluation of the child’s problem. By contrast, we have seen that physicians treat candidate diagnoses as inviting conﬁrmation or disconﬁrmation and that physicians often respond directly to parents’ candidate diagnoses. I have shown that by offering a candidate diagnosis in the problem presentation, parents adopt a stance that their child’s illness is medically problematic and treatable. Physicians’ responses support this analysis. Speciﬁcally, the responses to candidate diagnoses are generally conﬁrmatory or disconﬁrmatory and often men-

FOREGROUNDING THE RELEVANCE OF ANTIBIOTICS

49

tion antibiotic treatment. Physicians treat parents as in search of conﬁrmation that their own diagnosis was correct and in terms of treatment for that condition. As we saw in Extract 2.18, physicians may even be somewhat apologetic if they are unable to conﬁrm a candidate diagnosis. By contrast, in cases where parents present their child with a symptoms-only problem presentation, physicians routinely offer their diagnoses as straight afﬁrmative statements.

Discussion This chapter has outlined the two primary ways that parents present their children’s problems during the reason for the visit phase of the encounter. We saw that these practices convey parents’ alternative stances toward the child’s problem in terms of its doctorability and treatability. Symptoms-only problem presentations are treated as the unmarked type of presentation and display a stance that parents are, ﬁrst and foremost, seeking an evaluation of their child. By virtue of having come to the physician, they claim a need for medical assistance but make no claims about diagnosis or treatment. In contrast to the actions of presenting symptoms only, presenting a candidate diagnosis pushes forward across the physician’s medical judgment by anticipating this judgment—whether straightforwardly or more obliquely—thereby making treatment directly relevant. The fundamental claim of this chapter is that through the way they present their children’s problems, parents effectively initiate a negotiation of the treatment their child will receive at the visit’s end. When parents put a candidate diagnosis onto the interactional table, physicians must then contend with it. When that diagnosis is for a bacterial infection, then by extension they must also deal with the relevant treatment: antibiotics. This behavior is associated with physicians perceiving parents as expecting antibiotics. As mentioned in chapter 1, physicians who perceive a parent to expect them to prescribe antibiotics for an illness are more likely to inappropriately prescribe antibiotics (Mangione-Smith et al., 2006; Mangione-Smith, McGlynn, Elliott, Krogstad, & Brook, 1999). So, the use of this behavior is extremely consequential both for the interaction and for the treatment decision, even if parents’ use of it may be motivated by a range of different reasons. In this chapter, we observed the interactional consequentiality of the different presentation formats. We saw that physicians deal with candidate diagnosis presentations as entering into their domain of expertise. This is true both when physicians conﬁrm and when they disconﬁrm the parent’s candidate diagnosis. When the parent conﬁrms, the physician claims, from second position, to have been the person to make such a determination. Conversely, and when the parent was incorrect, physicians commonly state that antibiotics were not necessary. Parents typically did not bring up antibiotics in these contexts, but physicians were nonetheless treating other parent behaviors as indexing antibiotics and a parent preference for them. At the same time parents, even while using behaviors that apply pressure for antibiotics, show their own orientation to and respect for medical authority and the physician’s primary rights over that domain. In part, this is evidenced through the infrequency of candidate diagnoses and the contrasting prevalence of symptoms-only problem

50

PRESCRIBING UNDER PRESSURE

presentations. But it is also evidenced by the accounts provided and the downgrading used when parents do begin to tread into the medical domain. This chapter offers strong evidence that parents affect prescribing decisions in ways that might be quite unexpected. Traditionally, parent and patient participation has been thought to (1) require physician invitations, (2) occur during the treatment phase of the visit, and (3) involve some direct statement of preferences. With problem presentation formats, none of these elements is present, and yet parents can nonetheless be observed to shape the outcome of the prescribing decision by shaping how the physician sees the child’s illness and by suggesting what sort of stance they have toward the illness. As mentioned earlier, these data suggest that parents may not always (or necessarily) intend their problem presentation format to have the impact that it does. Whereas to a physician a bacterial diagnosis indicates antibiotics, this is only one thing that it may be associated with for a parent. Parents may be more concerned to underscore the legitimacy of their visit, and one resource for accomplishing this might be suggesting that the reason for the visit is because the parent thinks the illness is treatable with medicine. Thus, it is not the medicine that they want per se but rather a cure. Another possibility is that parents may be concerned that their child is seriously ill. Although physicians may understand that viral and bacterial illnesses can involve similar discomfort and severity of illness in terms of the illness experience, parents seem to generally believe that viral illnesses are minor whereas bacterial illnesses are serious. Their model of illness is somewhat at odds with physicians’ in this respect. Thus, mentioning a concern about a bacterial illness could be a practice through which parents underscore their concern for their children. In all cases, parents seem to view their use of candidate diagnoses as marked, delicate, and rather restricted in use. But even if delicate, rarely used, and differentially motivated, when used, candidate diagnoses represent the behavior that most commonly ﬁlls the ﬁrst sequentially provided-for opportunity to inﬂuence the diagnostic and treatment outcome of the medical visit.

3

Alternative Practices for Asking and Answering History-Taking Questions

A

fter a reason for the child’s visit has been established, physicians typically begin taking the child’s illness history. During this activity physicians generally ask questions about what sorts of symptoms the child has had, and prior treatment. In contrast to the reason for the visit phase, the history-taking phase is generally driven by very particular physician questions that make relevant quite constrained answers from parents. The structure of the history-taking activity does provide parents with an opportunity to participate, but their participation is arguably rather constrained because questions are generally quite focused. This is different from the reason for the visit phase, during which parents were answering a very unconstrained question about why they were there that allows them the interactional space to begin a brief narrative about the illness, detail the symptoms, or report on their concerns. Given all of this, parents might be thought to have few if any resources for participating in this activity at all, let alone shaping the diagnostic and treatment outcomes of the visit. This chapter shows that substantial negotiation over the child’s condition does occur during this phase of the visit. Through the ways physicians design their questions, they display their stance toward the child’s illness as problematic or not and, by implication, forecast whether it will be treatable. Similarly, parents have resources both for displaying their own stance toward the illness and for pushing physicians toward or away from particular diagnostic and treatment trajectories, as understood through physicians’ questions, in the way that they respond to questions.

51

52

PRESCRIBING UNDER PRESSURE

In this chapter, I ﬁrst discuss the principles that underlie the way physicians design their questions and then offer evidence that parents hear history-taking questions as implicating diagnosis and treatment. I then discuss two resources parents have for shaping the diagnostic-treatment trajectory during this phase of the visit. Finally, I demonstrate that physicians treat these practices as ways of negotiating the diagnostic and treatment outcome of the visit.

Background History taking is generally characterized as a series of doctor-initiated questionanswer sequences (Beckman & Frankel, 1984; Boyd & Heritage, 2006; Byrne & Long, 1976; Mishler, 1984). However, these sequences are best understood as part of larger courses of action concerned with both gathering information and initiating a “differential diagnosis” of the patient’s condition (Athreya & Silverman, 1985). Medical students are taught to begin considering diagnostic possibilities as soon as they read about the patient’s symptoms in a chart or hear about them during the encounter (Athreya & Silverman, 1985; Cohen-Cole, 1991; Greenberger & Hinthorn, 1993). Textbooks that offer prescriptive techniques for taking patients’ histories typically suggest questions to ask in the face of given symptoms, with the idea being that each question should rule out a possible diagnosis or move a step closer to ruling one out. For example, if a child is having abdominal pains, one textbook suggests inquiring into factors that either aggravate or relieve the symptoms. Answers to such questions do not simply provide more information about the condition; they also indicate whether the problem is inﬂammatory or spastic in nature (Athreya & Silverman, 1985). So, each answer to a history-taking question furthers the physician’s progress toward a particular diagnosis. When we examine interactional data, we see that it is not only doctors who treat questions as part of a diagnostic and evaluative process. Drew (2006) showed that in calls to a British “on call” doctor after hours, callers treat doctors’ history-taking questions as a time during which the doctor is determining the urgency of the patient’s needs and whether a home visit is necessary. Often, callers treat history-taking questions as “an opportunity to embellish their initial accounts, in order to convince the doctor of the seriousness or urgency of the condition” (p. 423). Additionally, he found that callers often volunteer further details of new or unrelated patient symptoms. These are often presented following a no-problem answer to a question about symptoms. Through the design of history-taking questions, physicians reveal much about their presuppositions and, in turn, about their stances toward patients’ illnesses. As Clayman and Heritage note with respect to news interviews, there are not really any truly neutral questions (2002). Researchers in the medical context have noted that the design of history-taking questions is consequential. At the most basic level, researchers of provider-patient communication have discriminated between “closed-ended” and “open-ended” questions (Byrne & Long, 1976; Mishler, 1984) and examined the relative merits of these alternative designs (Beckman, Markakis, Suchman, & Frankel, 1994; Mishler, 1984; Roter & Hall, 1992). In particular, the narrower the range

ALTERNATIVE PRACTICES FOR ASKING AND ANSWERING

53

of appropriate responses, the more a question limits the patient’s participation (Roter & Hall, 1992). This may include the way that a question can exclude the patient’s “life world” perspective and personal experience (Mishler, 1984). Boyd and Heritage (2006: 163) observe that questions (arguably, particularly yes-no questions) “unavoidably establish agendas, and embody presuppositions, and preferences” concerning patient responses (See also Heritage, 1984a, 2002; Koshik, 2002; Lindström, 1995; Pomerantz, 1984). They further argue that, in the context of comprehensive history taking (i.e., the kind of history taking that occurs in annual checkups as opposed to acute care visits) (Bates, Bickley, & Hoekelman, 1995), the design of doctors’ questions is guided by two principles: “optimization” and “recipient design.” Optimization involves designing questions to “prefer” or to facilitate “no-problem” or “prosocial” responses. (See Heritage, 1984; Pomerantz, 1984, for further information about structural preference.) For example, even in the case of a patient in late middle age, a history-taking question is structured as “Is your father alive?” rather than “Is your father dead?” (Boyd & Heritage, 2006: 165). This principle of optimization is departed from, they argue, only “for cause.” So, a question that is not optimized in its design is marked and conveys that something out of the ordinary is being done with the question. The second principle, the principle of recipient design, means that questions should be ﬁtted to some matter the patient has raised where an optimized question would be inappropriate. (For discussions of the broad principle of recipient design, see Sacks & Schegloff, 1979; Sacks, Schegloff, & Jefferson, 1974; Schegloff, 1972a.) Thus, although asking if patients’ parents are alive reﬂects the principle of optimization, if patients are relatively old or have said things that suggest that their parents are not living, then physicians normally reﬂect this information in their question design. Failure to do so will raise questions about how attentive they have been to the patient. A major piece of evidence for these claims is that recipients of questions respond differently, depending on their design. It is only a small step from here to the observation that through the design of recipient responses, recipients do substantial interactive work to align or disalign with the questioner’s stance toward the question. For instance, Heritage (1998) has shown that people who respond to a question with an “Oh”-prefaced response convey that they held a position on the topic prior to the inquiry being made, and Raymond (2003) has shown that when recipients of yes-no questions provide “non type conforming” answers, they resist a question’s agenda or the “terms” (e.g., the presuppositions) of the question. With respect to the medical context, Stivers and Heritage (2001) have shown that in comprehensive history taking, patients may not be as conﬁned by question agendas as was previously thought. Rather, they may implement various forms of sequence expansion, including narratives, to offer unsolicited information. In these ways, we see that patients have multiple resources for “maneuvering” even in a seemingly narrow answering space. This maneuverability is particularly important for conveying the patient’s own stance toward the illness—particularly its doctorability and treatability. Thus, during the history-taking activity of interest to us, doctors and parents alike have a variety of resources for displaying their relative stances toward the child’s problem and thus toward each other. We will see that through question asking

54

PRESCRIBING UNDER PRESSURE

and responding, parents and physicians once again negotiate what kind of problem the child has: both whether it is doctorable and whether and how it is treatable.

Problem Presumption: Principles of Question Design in Acute Care The principles underlying question design in the acute care context are, of course, related to the principles underlying question design in the comprehensive or routine care context (Boyd & Heritage, 2006). By virtue of structural preference, yes-no history-taking questions are inevitably tilted toward either a “problem” or a “no-problem” response (Boyd & Heritage, 2006). For instance, in Extract 3.1, the question “Are you eating?” prefers a yes answer through its use of an unmarked interrogative construction (Pomerantz, 1988), whereas the formulation “Any vomiting or diarrhea?” prefers a no answer because of the addition of an unstressed negative polarity item “any” (Boyd & Heritage, 2006; Horn, 1989; Koshik, 2002). Negative polarity items reverse the structurally preferred answer from an afﬁrmative (e.g., following “Vomiting?”) to a negative.1 But both of these questions can be said to presume no problem in these health domains because the preferred answer to the question is a “no-problem” answer. (3.1) 206 [girl presents with a sore throat] 1 2 3 4 5 6 7 8 9 10 11 12

DOC: 1-> Are you eating? (0.2) PAT: Y:eah:, (0.5) DOC: ( ate, (0.2) Okay.) DOC: 2-> .hh Any vomiting er diarrhea? (.) PAT: Mm[:_ [((PAT looks to MOM)) MOM: That wasn’t diarrhea honey, DOC: No, Okay. So no vomiting, no diarrhea, MOM: (Mm mm.)

In acute care, the principle of optimization may, at times, conﬂict with concerns of both visit legitimacy (Heritage & Robinson, 2006a) and condition treatability because questions that are optimized are built for responses indicating that a symptom is not problematic. When a question is designed to presume wellness, parents may understand the physician as displaying a stance that the child’s condition is neither serious nor in need of treatment. But a physician who designed all history-taking questions to display a no-problem presumption might be heard as questioning the likely existence of symptoms and hence the legitimacy of the child’s visit altogether. Conversely, a physician who designed routine questions about the child’s general health to prefer problem answers could be heard to convey a stance that the child’s

ALTERNATIVE PRACTICES FOR ASKING AND ANSWERING

55

illness was quite serious. Questions about unmentioned problems or routine background questions are generally designed to prefer a no-problem response. Evidence for this can be seen both in Extract 3.1 and again in Extract 3.2. (3.2) 308 [girl presents with stuffy nose, sore throat, and a blister in her mouth] 1 2 3 4 5 6 7 8 9

DOC: -> .hh Any fever at home? at all? MOM: No^:.= DOC: =Okay, MOM: [NoDOC: -> [No vomiti:ng, er[: MOM: [No:_ DOC: Any other problems like that huh, (1.5) MOM: N:o.

The doctor’s question at line 1 prefers a no (and a “no-problem”) answer through the use of the polarity marker “any” (Boyd & Heritage, 2006; Quirk, Greenbaum, Leech, & Svartvik, 1972). Similarly, the doctor’s request for conﬁrmation in line 5 (continuing into 7) prefers a no answer in that it solicits a conﬁrmation of a negatively formulated assertion. This suggests that similar to comprehensive history taking, as discussed by Boyd and Heritage (2006), acute care history-taking questions are guided by two partially competing principles. Like comprehensive history taking, there is a similar orientation to the principle of optimization: Physicians display a presumption that if the child had a fever, vomiting, or other problems like that, they would have been mentioned. Questions about these symptoms are generally designed to embody a presupposition that they do not exist because they were not mentioned and are generally optimized to prefer a no-problem response. However, unlike the comprehensive history-taking context, in the acute care context, physicians are aware that there is at least one problem that prompted the visit. It is not simply an issue of recipient design because this underlies optimization in this context as well. Rather, physicians also typically design some of their questions with an eye toward the principle of “problem attentiveness.” This principle asserts that physicians should design questions about dimensions of the child’s condition that have been mentioned or implied to be problems in such a way as to prefer a problem response in order to display that they do, in fact, regard these issues as problematic. These two guiding principles are very much in line with the general conversational principles Levinson covers in his theory of generalized conversational implicature (2000). The design of physicians’ history-taking questions in the acute care context juxtaposes the Q (quantity) and I (informativeness) principles. The Q principle states both that speakers should provide the strongest possible statement of knowledge that they can and that, most relevant for us here, recipients will assume that what is stated by their interlocutor is the strongest possible statement or description.2 The I principle states that speakers should say as little as necessary to achieve communicative ends. The corresponding recipient’s corollary assumes this and therefore

56

PRESCRIBING UNDER PRESSURE

allows recipients to assume the richest description possible, consistent with what is taken for granted. The practices described here are very much an interactional version of the dilemma described by Levinson because doctors are provided only a very small amount of information about the child, and yet this information is clearly utilized in terms of the design of history-taking questions from that point forward. And, as mentioned earlier, doctors appear oriented to the assumption that if the parent did not mention particular symptoms, they are not likely to exist (Q principle), and if particular symptoms were mentioned, then questions broadly in line with those symptoms should be designed to presuppose a problem (I principle). Moving from the theoretical principles to interaction principles, these issues broadly correspond to the interactional constraints surrounding the two principles described earlier: optimization and problem attentiveness. A clear illustration of this can be seen in Extract 3.3, where a toddler-age girl has come in for a runny nose. The questions beginning in line 3 reﬂect the principles discussed by Levinson because in their design the physician displays her own orientation to the parent’s not having mentioned symptoms as indicating that the child has no other symptoms: nothing else, no fever, no vomiting, no diarrhea. (3.3) 211 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

DOC: DAD: DOC: MOM: DAD: DOC: DOC: MOM: DAD: DOC: DOC: DAD: MOM: DOC: DOC:

She has uh runny nose, °#Ye:[:s.=h#° [Okay anything el:se? That’[s (°it.°) [N::ah.=h No:?, (.) No fever? [No:. [No. No fever, no [coughing so far, [Okay. No cough:, Okay:?, No vomiting, no diarrhea?, N:o. °No.° (°nothing.°) Okay.

When physicians design questions to be problem attentive, they underscore the legitimacy of the child’s visit by displaying themselves to assume a problem. We can see this in the design of the two questions in Extract 3.4. (3.4) 206 1 2 3

DOC: -> Okay:? .hhh Uh:=uhm Are you coughing? (0.2) PAT: Yeah:.

ALTERNATIVE PRACTICES FOR ASKING AND ANSWERING

4 5 6 7 8 9 10

57

DOC:

Mkay:, (0.5) DOC: -> Uh runny nose? (1.0) PAT: #m# Like every once in uh while it’ll start running. (8.0) DOC: O::kay.

Here, both inquiries are designed to prefer a yes answer that conﬁrms the existence of problems. Both use the positively formatted question design, with the second question acting as a “second” in that series, thereby also utilizing that structure (Heritage & Sorjonen, 1994). Whereas the questions in this example have to do with identifying the existence of particular symptoms, we can also observe the same principle in questions about the quality, quantity, or duration of a particular symptom. An illustration of the ﬁrst two is shown in the questions in Extract 3.5. (3.5) 505 [presented with worsening cold and lethargy] 1 2 3 4 5 6 7 8 9 10 11

DOC: -> Has he been coughing uh lot? (0.2) MOM: .hh Not uh lot.=h[h DOC: [Mkay:?, MOM: But it- it <sound:s:> deep. (1.0) MOM: An’ with everything we heard on tee v(h)ee=hhhh £we got sca:re.£ DOC: -> Kay. (An fer i-) It sounds deep? (.) MOM: Mm hm.

The questions asked in lines 1 and 9 are both designed to prefer yes answers and thus to conﬁrm the existence of a problem. In the ﬁrst case, the question is afﬁrmatively designed. In the second, the request for conﬁrmation uses a similarly afﬁrmative assertion with rising intonation (as noted with “?”) and stress on “deep” to indicate what is being questioned. The design of history-taking questions in these acute care visits can be seen as guided by the principles of optimization and problem attentiveness. This is an important component to our argument because through an orientation to these two principles, parents are able to ascertain the physician’s stance toward their child’s illness and can therefore negotiate treatment. The way a question is designed unavoidably conveys the physician’s stance toward the child’s condition (or some aspect of it) as problematic or nonproblematic. When questions are built as part of a larger activity of history taking, the presence of a trajectory toward a problem or a no-problem diagnosis is even more visible. Thus, understanding how questions are designed is important for our larger argument.

58

PRESCRIBING UNDER PRESSURE

The Diagnosis Implicativeness of History-Taking Questions One primary mechanism through which physicians reveal their stances toward a child’s illness is their question design. A major reason is that questions are understood to be directly linked to or in the service of the activity of diagnosing. Earlier, we reviewed medical textbook literature stating that physicians should use history taking to begin differential diagnosis. But not only in theory is the activity of history taking in the service of diagnosis; physicians and parents treat history-taking questions as beginning the diagnosis process. This becomes visible both in the way physicians design their questions and in the way parents respond to them. Although parents and children are not likely to recognize all of the diagnostic implications of any given question, they nonetheless commonly display an orientation to history-taking questions as being part of a differential diagnosis. For example, in Extract 3.6 a grandmother attending the visit with her son and grandchildren responds to a physician’s history-taking question and then negates the diagnosis that would have been implicated, had the question been answered afﬁrmatively and which the physician’s trajectory of questions was likely to have been pursuing. (3.6) 506 1 2 3 4 5 6

DOC:

Does he have uh history of wheezing? (0.5) GMA: No. (.) GMA: -> He doesn’t have asthma. DOC: (Okay,)

The grandmother ﬁrst answers only the question that was asked (line 3), but then, after a micropause (line 4), she asserts that her grandson does not have asthma, thereby treating the prior question as part of a differential diagnosis for asthma. The doctor receipts this assertion with “(Okay,)” which acknowledges the grandmother’s understanding of the implicit link between the question and the diagnosis without overtly conﬁrming it. Besides recognizing the link between questions and diagnoses, insofar as the question at line 1 could initiate a series of questions to investigate whether the child has asthma, the grandmother’s “He doesn’t have asthma.” works to shut down that diagnostic trajectory. Even young children can observe that questions are being asked in the service of diagnosing. We can see this in Extract 3.7, where an 8-year-old girl orients to the diagnosis-implicative nature of history taking by relating a possible diagnosis teasingly offered by her parents to the doctor. (3.7) 517 1 2

DOC:

Does it itch? ((begins to dry hands with paper towel)) (2.1)

ALTERNATIVE PRACTICES FOR ASKING AND ANSWERING

3

DAD:

4 5 6 7 8 9 10 11 12 13 14 15 16 17

PAT:

59

(M-)Yea[:h?,

[Sort of. (0.2) DOC: °Mka:y,° (.) DOC: An’ where’d it start off. (0.5) DAD: (In thee=uhm) (.) thuh ba:ck? DOC: Mka:y?, (12.5) PAT: -> My mah- .hh my mommy an’ daddy said I’m allergic to chocolate. (.) DAD: -hhh ((laugh)) DOC: (hhh) ((laugh))

That the girl performs this action at this juncture during history taking displays her orientation to diagnosis as a process being accomplished partially through history taking, even though the possible diagnosis is neither reported nor received in a serious way.

The Treatment Implicativeness of History-Taking Questions Even treatment can be implicated through history-taking questions. Although the relevance of treatment is generally introduced through an orientation to diagnosis, evidence of the treatment implicativeness of history-taking questions is also visible in the ways parents and physicians deal with history taking. And parents hear the stance revealed by doctors’ questions as relevant to whether the child will ultimately be treated. Although we will be exploring the practices through which parents negotiate with physicians in the next section, here I want only to observe that treatment is oriented to by both parents and physicians as linked to the problem versus no-problem nature of the condition, and the stance physicians and parents take to the child’s condition begins early. We can see this quite clearly in Extract 3.8. Here, the mother provides a problematic response (line 2) to a question designed to prefer a no-problem answer and then upgrades this response with “quite uh little bit.” Interestingly, this upgrade is revised to a downgrade in response to a repair initiation at line 5 (see line 6). It is following the physician’s receipt of this that the parent goes on to assert, ﬁrst, that she is a relatively troubles-resistant parent (Heritage & Robinson, 2006a; Jefferson, 1988) (lines 9–10/12/14–16) and, most relevant for this analysis, that “we’re doing everything that I think we’re supposed to be doing” and it is still not getting better. (3.8) 161307 1

DOC:

Is she uh (0.2) eating okay?

60 2 3 4 5 6, 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

PRESCRIBING UNDER PRESSURE

MOM: MOM: DOC: MOM: DOC: MOM: DOC: MOM: DOC: MOM:

DOC: MOM:

MOM: DOC:

No, I would say her appetites down. (.) [quite uh little bi[Pardon, I would say her appetite’s down a little. °Yeah.° (0.3) She just- You know I don’t like to bring her- iI don’t typically bring my kids in (.) you [know [Yeah. drop of a [hat_ .hh but it’s been goin’ on about= [(Right.) = ten days now and it’s: (.) I mean I’m- we’re doing everything I think we’re supposed to [be doing and it’s not getting any better. [Yeah. (Yeah ) (7.5) °Actually now that I’m thinking about it she was sick before Thanksgiving_ (0.8) So: yeah ^it’s probably been about ten days;° (°I always °) Is the cough worse at night?

The mother’s response here appears defensive in the sense that she appears to hear the question as indexing the doctor’s stance toward her daughter’s condition as not terribly problematic. In this position, she invokes their failure to resolve or manage the girl’s condition through home remedies (lines 14–16). So history-taking questions are heard as consequential for both diagnosis and treatment outcomes. Parents also sometimes use history-taking questions to bring up very speciﬁc treatment options. In Extract 3.9, the parent brings up antibiotics. Earlier in this encounter, the parent presented the child’s condition as a cold with congestion and headaches. Just prior to this extract, the physician asked about previous sinus infections, and the parent conﬁrmed that the child has suffered from this in the past. Here, they are in the middle of extensive history taking. The parent re-raises sinus infections as part of her response to a question about when they had previously met this physician, who is not their regular pediatrician. The way that she refers to the prior infection with “last .h sinus £infection.” treats this illness as a sinus infection, though this is very indirect because it is embedded in the reference to the prior. But “last” is relevant only with respect to “next,” which she here conveys is her stance. (3.9) 2069 (Dr. 9) 1 DOC: ^Oh I- I knew I’d seen you guys before_ I didn’t 2 realize it was- ((sniff))/(0.2) 3 MOM: Yeah, it’s [(been uh while) 4 DOC: [>was< back in No[vember. 5 MOM: [( ) [Ye:s. I think it=

ALTERNATIVE PRACTICES FOR ASKING AND ANSWERING

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

DOC: MOM DOC: MOM: MOM: DOC: DOC: MOM: DOC: MOM: -> MOM: -> -> DOC: MOM: -> DOC: MOM:

61

[Yeah:.=h =was -his last .h sinus £in[fection. huh huh [huh [Yeah. [Yeah, .hh (2.0) So- November. So it’s four months uh, Yieah:. (8.0) Did we get an x ray then? (.) No [(there was no_ Hm mm. [Or did we(6.5) It=did clear up though. (.) With=that- (1.0) I think it was like twenty days or whate[ver it was. [Uh huh,=hh of antibiotics it cleared up. (3.2) And thuh headaches went away? Yeah.

Moreover, in response to a question that pursues information about a previous sinus infection (line 14), the parent mentions that the condition resolved with antibiotics (lines 19/21–22/24). Although the physician’s question “Did we get an x ray then?” (line 14) suggests a pursuit of how the condition was diagnosed rather than how it was treated, the parent, after a no answer to this question (line 16) and after a long silence, instead goes on to provide information about the treatment that was given rather than the diagnosis. With this turn, the mother suggests that the antibiotics worked to cure a condition similar to the one her daughter has now. Thus, with this turn, the parent implies that they are seeking similar treatment. Important for this point in our analysis, parents orient to the relevance of information about treatment during history taking because it may well shape the physician’s stance toward the child’s condition as problematic and treatable when they reach that phase of the visit. Physicians also treat the way parents design their responses as having direct treatment implications. This is illustrated in Extract 3.10. After the parent has taken a stance that the child’s condition is getting worse and is therefore problematic (line 4), the doctor, in her response, addresses the concern as lobbying for antibiotics (lines 11–12). (3.10) 119 1 2 3

DOC: MOM: DOC:

.hh And this has been going on for about uh week?, .hh Seven day:s, Okay:, [(°yeah.°)

62 4 5 6 7 8 9 10 11 12 13 14 15

PRESCRIBING UNDER PRESSURE

MOM: DOC:

[But thuh cough is (wearing) wor:se. Getting worse. (.) DOC: °Okay. Well we’ll take uh-° I hear- actually heard it out there. (.) DOC: .hh (Yep-)/(eh-) .h I heard your cou^gh. DOC: .hh Uh:m sometimes coughs certainly can hang on -> for uh while. We’ll look an’ see whether or not she -> needs something that needs an antibiotic. DOC: .hh Uh:m (.) very often (yuh) get uh col:d an’ then thuh cough sorta kinda hangs o:n.

Although the parent does conﬁrm the physician’s question about the illness’s duration (line 2), she does it with a nonconforming answer: She uses an alternative formulation rather than the more minimal alternative “yes” (Raymond, 2003). Additionally, in her expansion (line 4), the physician hears her to be treating her daughter’s illness as more problematic than the physician had treated it. But rather than dealing with this in terms of diagnosis, she deals with it in terms of treatment. Finally, even if antibiotics are not overtly broached at this stage in the visit, treatment can nonetheless be seen to be relevant in relation to the questions. We saw this in the previous example with the physician’s comment that “We’ll look an’ see whether or not she needs something that needs an antibiotic.” We have seen that the diagnostic and treatment import of history-taking questions and responses are understood by both parents and physicians. In terms of negotiation, this has important implications for what parents do in their responses to history-taking questions. A physician who takes a stance toward the condition as nonproblematic is not likely to recommend prescription treatment and, in particular, not antibiotic treatment. This stance is potentially problematic for parents because, at a minimum, it threatens their own judgment that the child was ill and therefore threatens the legitimacy of their visit. Parents have resources that are deployed systematically to deal with these issues and through which they negotiate their child’s treatment and, by extension, their diagnosis and visit legitimacy.

Parent Resources for Negotiating Physicians’ Diagnosis and Treatment Trajectories Although physicians’ history-taking questions are quite constraining in terms of the relevant next action, as we have already begun to see, parents can still do interactional work to break out of those constraints and, for instance, interactionally push physicians away from the stance they conveyed through a question and/or even toward an alternative stance. As we have already seen, these stances have clear implications for diagnosis and treatment. This section examines two interactional resources for accomplishing this and thus for accomplishing visit outcome negotiation: “mentioning additional symptoms” and “mentioning possible diagnoses.”

ALTERNATIVE PRACTICES FOR ASKING AND ANSWERING

63

Mentioning Additional Symptoms Parents are frequently confronted with a dilemma in responding to physician questions: They must respond in a way that supports a no-treatment outcome despite this being potentially contrary to their interests and beliefs. One way that parents manage this interactional dilemma is to steer physicians away from a no-problem diagnosis by mentioning additional symptoms as part of their question response. This is not a common practice. In the Seaside data, it was observed to occur in the history-taking context 9% of the time. Although not terribly frequent, the behavior still represents yet another interactional resource parents have for initiating a negotiation of the visit outcome with the physician. We can see an example of a parent mentioning an additional symptom in Extract 3.11 (shown earlier as 3.5). In line 1, the physician asks a question that, contrary to the other questions just prior (lines 1–11), prefers a problem response. But the parent provides a dispreferred, nonconforming (Raymond, 2003), and no-problem answer when she disconﬁrms that it is “not uh lot.” However, following that, she offers a dimension of the illness that is problematic: the quality of the cough: “But it- it <sound:s:> deep.” This works to shape the trajectory of the physician’s diagnosis in two ways: First, it pushes the physician away from her present direction by introducing a new dimension of the illness. By providing “problem-implicative” information at this point in the interview, the parent provides the physician with an alternative diagnostic path. Second, the practice invites pursuit and sequence expansion (see Schegloff, 2006) that would move the trajectory in a different direction. Evidence for this is provided in the physician’s reaction: Although she had proposed closure of the sequence following “Not uh lot.” with “Mkay:?,” (line 16), following the additional symptom presentation, she initiates a line of further questioning. (More on physician uptake of additional symptoms will be discussed at the end of this chapter.) (3.11) 505 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

DOC: BOY: DOC: DOC: DOC: DOC: DOC: MOM: DOC:

Anything hurt? (1.5) No:_ (.) Your ears don’t hurt? (0.3) ((BOY shakes head laterally)) Your throat doesn’t hurt? (0.6) ((BOY shakes head laterally)) Your tummy doesn’t hurt? ((extra rise at the end)) (0.5) ((BOY shakes head laterally)) No:? (0.2) Has he been coughing uh lot? (0.2) .hh Not uh lot.=h[h [Mkay:?,

64 17 18 19 20 21 22 23

PRESCRIBING UNDER PRESSURE

MOM: -> But it- it <sound:s:> deep. (1.0) MOM: An’ with everything we (heard) on tee v(h)ee=hhhh £we got sca:re.£ DOC: Kay. (An fer i-) It sounds deep? (.) MOM: Mm hm.

Mentions of additional problematic symptoms can be found in both answers to individual questions and in environments where a series of questions (and their responses) may be taken into account. For instance, in Extract 3.12, the mother provides a series of no-problem responses to questions. The ﬁrst two answers are to optimized questions. The third question is hearable as beginning to search for a possible problem, but once again, the parent provides a no-problem answer. And in fact, the mother goes on to further support the physician’s no-problem trajectory with an additional no-problem statement (arrow 4). However, immediately after (line 24), she shifts position, stating, “So that’s why it didn’t concern me too much” and which overtly displays her orientation to the prior sequences as aligned with a no-problem trajectory. In line 25, the mother continues her turn by moving toward the event that brought her to the medical visit, this time providing problem-implicative information—that she saw something in her daughter’s throat. This, like the previous instance, is prefaced by the conjunction “but,” which treats the action to follow as contrastive with the prior. (3.12) 308 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

DOC: 1-> .hh Any fever at home? at all? MOM: No^:.= DOC: =Okay, MOM: [NoDOC: 2-> [No vomiti:ng, er[: MOM: [No:_ DOC: Any other problems like that huh, (1.5) MOM: N:o. MOM: Thank g(h)oodn(h)ess [(huh huh,) DOC: [N:kay:, DOC: 3-> .hh What abou:t uhm as far as her appetite, eating wi:se, MOM: Still has uh good appeti:te? DOC: Thuh appetite is [goo:d? MOM: [(Yeah:,???) DOC: hh=nkay, so thisMOM: 4-> [An’ she still (have) energy tuh pla:y, DOC: [(Thuh) sore throat-

ALTERNATIVE PRACTICES FOR ASKING AND ANSWERING

20 21 22 23 24 25 26 27 28 29

MOM: DOC: DOC: DOC: MOM: -> => => DOC: MOM: => DOC:

65

[ T u h r i d e h e r bicy]cle: ‘n[Still has energy tuh play:?,] Nkay, .hh Uhm: So that’s why it didn’t concern me too much but I:- I looked at it this morning: an’ I thought I saw (.) Something i[n there. [something in there. [Yeah:. [Mkay,

Similar to Extract 3.11, the mother pushes the physician away from an entirely noproblem trajectory toward an investigation of a possibly problematic dimension of the child’s illness: in this case, something visible inside the child’s throat. Yet another type of environment where we see parents using this practice is exempliﬁed in Extract 3.13. Here the pattern is similar to the examples in Extracts 3.11 and 3.12, but it follows not a question to which a no-problem answer must be given but after the doctor has indicated a shift away from illness-speciﬁc history and to more general questions about whether other family members are sick. The mother perhaps hears this question as indicating that the physician has completed an investigation of the illness and with her action redirects the physician back toward a problem trajectory. (3.13) 2053 (Dr 8) 1 2 3 4 5 6 7 8 9 10 11 12 13

DOC: -> Mlk Uh:m: how ‘bout this #morning.# Has he thrown up? MOM: No. Not since uh:- i- it must have been uh combination of the medicine and the [(milk). DOC: [#an’ thuh mi(h)lk,=hh huh huh huh huh (h)o(h)kay, DOC: -> .hhh £O:kay.£ Is anyone else at home sick?=hh= MOM: =No. (0.5) MOM: => But thuh temperatu:re- w- went up every time: #uh:# (0.7) after taking the Tylenol [after a while it goes up. DOC: [Uh huh,

In this case, after providing a no-problem answer at lines 3 and 9, the mother then goes on to offer further problem-implicative information—that her son’s temperature can be kept down only with the use of Tylenol. In their own responses, physicians show that they hear parents to be pushing them away from no-problem trajectories when they respond to their questions with the presentation of additional symptoms. We can see evidence of this in Extract 3.14.

66

PRESCRIBING UNDER PRESSURE

(3.14) 119 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

DOC: .hh And this has been going on for about uh week?, MOM: .hh Seven day:s, DOC: Okay:, [(°yeah.°) MOM: -> [But thuh cough is (wearing) wor:se. DOC: Getting worse. (.) DOC: °Okay. Well we’ll take uh-° I hear- actually heard it out there. (.) DOC: .hh (Yep-)/(eh-) .h I heard your cou^gh. DOC: .hh Uh:m sometimes coughs certainly can hang on for uh while. We’ll look an’ see whether or not she needs something that needs an antibiotic. DOC: .hh Uh:m (.) very often (yuh) get uh col:d an’ then thuh cough sorta kinda hangs o:n.

Here, the physician has requested conﬁrmation that the duration of the girl’s condition is “about uh week?,” a characterization that does not treat this as either a long or a short period of time. However, the mother’s response “Seven day:s,” is hearably an upgrade. Although technically “Seven days” is merely an alternative formulation of “uh week”, this formulation’s precision works to emphasize the condition’s long duration and therefore treats the illness’s duration as problematic. Moreover, this answer fails to conform with the terms of the question (Raymond, 2003). The physician nonetheless proposes closure of the sequence with “okay:,” after which the mother offers the additional and problematic symptom that the cough is getting worse. In this case, it is not that the physician goes on to do a further inquiry, but she addresses the import of the mother’s action in several ways. First, the doctor assures the mother that she “actually heard it” (that is, the cough), so she knows whether to be concerned about it or not (lines 7–8). Second, she addresses the mother’s assertion that the cough is a problem by suggesting that, in fact, it may not be: “sometimes coughs certainly can hang on for uh while.” (lines 11–12). Third, the doctor states that she will “look an’ see” (line 12). Finally, returning to whether the cough should, in fact, be treated as a problem, she suggests that it may or may not be something “that needs an antibiotic” (lines 12–13). Her account is just following: “very often (yuh) get uh col:d an’ then thuh cough sorta kinda hangs o:n.” (lines 14–15). Thus, across this turn, the physician addresses a range of issues embedded in the mother’s turn and treats the mother as having conveyed a stance that the child’s cough is problematic. We have looked at one resource parents have for pushing physicians away from a no-problem diagnostic and treatment trajectory during history taking: the mentioning of additional symptoms. It works because it offers a symptom as problematic at a point where only something unproblematic has been raised. This effectively encourages physicians to consider an alternative diagnostic trajectory, and it invites

ALTERNATIVE PRACTICES FOR ASKING AND ANSWERING

67

sequence expansion, which, if done, does place physicians on an alternative diagnostic trajectory. Mentioning Alternative Possible Diagnoses Another communication practice that parents use following a question-answer sequence involving a “no-problem” answer is to mention an alternative possible diagnosis. This, too, is not frequent, occurring 12% of the time in the Seaside data. These diagnoses are not typically offered in quite the same way that they are offered in the problem presentation position (i.e., as “candidate diagnoses” as discussed in chapter 2). They are only rarely overtly stated as the parent’s diagnostic theory. Instead, they are more typically offered as justiﬁcations for their child’s condition being problematic when their question response is otherwise aligned with a no-problem stance toward the child’s illness. What they do, though, is to propose an alternative diagnosis to be considered, and this is quite similar to candidate diagnoses. How overtly this is done certainly varies, as we will see. As an initial example, we can look at Extract 3.15. (3.15) 313 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

DOC: 1-> How has he been today.Has he [eaten anything t’day? DAD: [(uh t-) DAD: Uhp- uh:=l:little bit_ (he di:d.??) (0.2) DOC: 2-> Did he throw it up today? DAD: Uh: no he hasn’t. (0.2) DOC: 3-> °Ok:ay.°=Has he had fever at home today? DAD: Uh:m, (0.2) No No. DOC: (2.0) DAD: We give him some Tylenol for thuh fever, DOC: When was tha:t. DAD: Uh:m, last ni:ght, DOC: 4-> But today he’s had no fever? DAD: t- Today (we h-) no:. (.) DOC: O[kay. DAD: [Yeah no fever. (2.1) DOC: 5-> So no fever today. So he seems okay today. (.) DAD: Yeah seems tuh be okay. DOC: (°h[Mkay.°??) DAD: => [It’s just that: I wanna make sure that (1.0) (might not have uh) i[nfection er something ( ) DOC: [O:ka:y,

68

PRESCRIBING UNDER PRESSURE

The doctor asks a series of history-taking questions that receive no-problem answers (questions 1–5). At line 11, the father offers information that might problematize the previous answer. If the boy did not have a fever because it had been brought down with Tylenol, then he could potentially have misanswered the doctor’s question. However, because the Tylenol was given the previous evening, this is ruled out, and the line of questioning (at line 14) underscores the child’s present no-problem status. This is still further underscored when, at arrow 5, the doctor requests conﬁrmation that “he seems okay today.” prefaced with a summative marker “So” (Raymond, 2004). Here, the father ﬁrst conﬁrms (line 22) but then continues his turn with a contrastive “It’s just that:” and an alternative possible diagnosis: that he has an infection. The alternative diagnosis works to push the physician toward a conceptualization of the illness that is at odds with the prior line of questioning and indicates the parent’s own stance toward the child as not as healthy as he might seem. This alternative diagnosis defends the father’s decision to seek medical care. Alternative possible diagnoses are sometimes delivered quite explicitly. In Extract 3.16, the grandmother’s mention of a diagnosis is speciﬁc: throat infection and ear infection. But both are still offered as general possibilities and not as strong theories, and there are two possibilities. The mention of more than one diagnostic possibility underscores that it is not a single theory being advocated, but rather it is a class of problematic diagnoses that is being pushed by the grandmother. (3.16) 506 1 2 3 4 5 6 7 8 9 10 11 12 13 14

DOC: GMA: GMA: -> -> GMA: -> GMA: -> GMA: -> -> ->

Has he ever needed uh breathing treatment? (.) No. (7.0) No but what he gets is like throat infections an’ ear infections. (0.4) Frequently. (0.4) An’ he (gets with thuh high fever.) (0.5) So that’s why I=uhm (1.5) we decided to bring ‘im in because (0.4) with thuh temperature. (.) (I know ) something’s gonna be wrong.

The grandmother ﬁrst answers a history-taking question with a no-problem answer. After 7 seconds of silence, the grandmother again offers a “No” and then goes on to contrast this with the type of illnesses her grandson has had trouble with in the past. This conveys the grandmother’s stance that the physician may be pursuing a less productive diagnostic trajectory by suggesting an alternative possible diagnosis (i.e., ear/throat infection) to the one being pursued by the physician (i.e., allergies/ asthma). The grandmother’s reuse of “No” and the connective “but” work to tie her

ALTERNATIVE PRACTICES FOR ASKING AND ANSWERING

69

turn in lines 5–6 back across the lengthy silence to her turn in line 3. In this way, her concerns are voiced as an expansion of her prior answer and hence as “part of” a history-taking question-answer sequence rather than as an assertion of a candidate diagnosis. Single, speciﬁc diagnoses are, occasionally, offered by parents during history taking. We can see an example of this in Extract 3.17. Here the mother responds to a history-taking question about the duration of her son’s fever (line 1) with an answer that might suggest she rushed to the doctor too quickly (line 3). The question embodies a presupposition that the child has a fever and orients to this as a problematic symptom. However, the mother’s response treats the answer she gives as inadequate to evidence a problematic illness, and perhaps as potentially undermining the legitimacy of her medical visit (Heritage & Robinson, 2006a). (3.17) 2067 (Dr. 7) 1 2 3 4 5 6 7 8 9 10 11

DOC: BOY: MOM: DOC:

How long has he had uh fever for? Ah [bah buh[Just since last ni:ght_ Mm=hm::?, (0.2) MOM: -> But his brother an’ his sister have ear infection. DOC: Mm:, okay. MOM: So:, (.) DOC: .h MOM: I’m uh li[ttle ( -) DOC: [Uhm has he been sick at all:?

The mother displays her orientation to 1 day as a minimal and relatively insigniﬁcant length of time with her use of “just”. But following the doctor’s receipt of this (line 4), the mother adds a further component to her turn with “But his brother an’ his sister have ear infection.” This offers a possible diagnosis as an account for seeking medical care within what is normally a relatively unproblematic duration of time (less than 24 hours). This addition also indexes her primary concern—that her son may also have an ear infection, and infection stands as a proxy for “taking antibiotics” as well. Extract 3.17 and the following Extract 3.18 both show cases where the diagnostic theory approaches that shown in the reason for the visit phase. Although alternative diagnoses are rarely presented as overtly during history taking as they are during the problem presentation, in these last two cases we can see that parents do approach this degree of directness. Here, in response to a question about coughing, the mother (following an exchange with her daughter in lines 3–4) offers a no-problem response (line 5): “No cough.” (3.18) 2074 (Dr. 9) 1 2

DOC:

An:d coughing too?, (0.8)

70 3 4 5 6 7 8

PRESCRIBING UNDER PRESSURE

MOM: GIR: MOM: DOC: MOM: -> ->

Uh little bit or no coughing. #No:?# No [cough. [No cough. [Okay. [.hh for strep, Thuh results came in last Tues:day.

Subsequent to the physician’s receipt of the answer (line 6), the mother offers a possible diagnosis (strep throat) that pushes the physician away from a line of inquiry regarding coughing and toward one about a throat infection. Considering both mentions of additional symptoms and mentions of alternative possible diagnoses, we can see that they do slightly different but related interactional work. Whereas mentions of additional symptoms push doctors away from a particular trajectory of questioning, mentions of alternative possible diagnoses in this same sequential environment actively instruct physicians as to which trajectory of inquiry they should pursue in the service of conﬁrming or disconﬁrming this diagnosis. Thus the latter pushes them toward an alternative diagnosis (and, by extension, treatment) trajectory. This is precisely how these behaviors can be understood as negotiation tools. The best evidence for this is the way that physicians tend to respond to these actions. They consistently treat these behaviors as pressuring them to consider an alternative diagnosis and treatment and often as speciﬁcally indexing a desire for antibiotics.

Physician Responses to Parent Actions When physicians ask questions, in addition to conveying particular presuppositions, they also often display that the questions were asked either as part of a relatively standard anticipatable series of questions or, alternatively, as a follow-up to some prior comment by the patient. Heritage and Sorjonen (1994) discuss the difference between two similar types of questions in the context of health visitor–parent interaction: what they term “routine or agenda-based questions” and “contingent inquiries.” They observe that agenda-based questions typically embody a “forward movement within a larger sequence,” and the use of an agenda question following a previous question-answer sequence treats the prior response as “unproblematic” (p. 11). By contrast, contingent questions emerge in environments in which there is some “unexpected” or “problematic” response to the prior question, where the inquiry sustains the topical focus of the preceding question-answer sequence, and where it is “recognizably produced as . . . contingent in character, rather than as ‘anticipated’” (p. 11). Similarly, Byrne and Long (1976) discussed questions that relate a given symptom to a particular time, place, or activity as “supplementary” questions (p. 34). They discuss the importance of these questions in making a diagnosis. The contingent or supplementary quality of a question may also be suggested by its placement within a given sequence. Whereas the previous question types show, in Heritage and Sorjonen’s terms, a “forward movement” through an agenda of questions (1994: 11),

ALTERNATIVE PRACTICES FOR ASKING AND ANSWERING

71

doctors also sometimes expand a question-answer sequence by asking further questions about a particular symptom as a postexpansion sequence (Schegloff, 2006). When they do this, they fail to progress in the same way that a new agenda-based question would. Halts in progressivity are important here because they represent one way that physicians treat a symptom as problematic. We can see an example of this type of question and response in Extract 3.19 (shown earlier as 3.11). The doctor asks a question about the amount of coughing (line 13). In response, the parent offers an additional symptom about the quality of the cough in line 17, as we discussed earlier. (3.19) 505 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

DOC: 1-> Has he been coughing uh lot? (0.2) MOM: .hh Not uh lot.=h[h DOC: [Mkay:?, MOM: But it- it <sound:s:> deep. (1.0) MOM: An’ with everything we heard on tee v(h)ee=hhhh £we got sca:re.£ DOC: a-> Kay. (An fer i-) It sounds deep? (.) MOM: Mm hm. DOC: b-> Like uh barky cough? MOM: .hh (1.1) Uhhhm=hhh It sounds very:=uhm (.) (I don’t know:=wwlike:) (0.2) It sounds- (2.5) Tlk .hh Tlk Not like that like: DOC: [Not (barky.) MOM: [Like when someone has bronchitis that it sounds ( ) DOC: Okay. DOC: c-> Does he sound like uh dog er uh seal barking? MOM: No. DOC: Okay. (0.5) DOC: .hh Not bark(y).=hh (0.9) DOC: °Okay.°

In lines 19–20, the mother explains that this quality to the cough caused her (and presumably her husband or family) concern because of what had been on television lately. At line 21, the doctor further expands the sequence by asking another question, this time one that builds off the second component of the mother’s turn. In this way, the physician initiates postexpansion of the sequence and pursues the mother’s additional symptom, rather than progressing to a next agenda item. This shows that

72

PRESCRIBING UNDER PRESSURE

the mother’s effort to move the doctor away from her existing trajectory and toward an alternative trajectory was successful. This sequence is pursued until that line of inquiry is exhausted (lines 34–38). The effectiveness of additional symptom mentions is also evidenced by cases where a physical examination is generated by the additional symptom mention. We can see this in Extract 3.20 (shown earlier as 3.12). Here, the mention of an additional symptom is in lines 25–28. This immediately generates a new sort of inquiry (lines 30–32) and just after this, a physical examination (line 36). (3.20) 308 24 25 26 27 28 29 30 31 32 33 34 45 36

MOM:

DOC: MOM: DOC: DOC:

MOM: DOC: MOM: DOC:

So that’s why it didn’t concern me too much but I:- I looked at it this morning: an’ I thought I saw (.) Something i[n there. [something in there. [Yeah:. [Mkay, .hh Uhm °tl° but it hasn’t hurt her enough that she has had trouble swallowing for example. She complain:ed of that: today. Is that right? Yeah. => Let’s see. Sit up (straight,)

What these cases show is that in the question-answer sequence, parents and physicians can push and pull to negotiate the doctor’s orientation toward the illness. Beyond this, we can observe that what is truly being negotiated is whether the child will be treated. Physicians understand parents who are using these practices to push them away from particular inquiry or diagnostic trajectories and/or toward alternative trajectories to be negotiating the diagnostic and treatment outcome of the visit. We can see evidence for this in Extract 3.21 (shown earlier as 3.16). The grandmother offers an alternative diagnosis (or rather two: throat and ear infections). The physician’s initial response is to claim that a “regular cold” can also be accompanied by a fever. By mentioning “regular cold” in this environment, the physician does several things: (1) she shows that this was, in fact, the trajectory she was going down because she defends the diagnosis; (2) she treats the grandmother as proposing a counterdiagnosis to the one she was pursuing through her inquiries; and (3) when she mentions “it’s just uh virus.” this is a shift from “cold” to a category of diagnosis that is directly related to whether antibiotics are prescribed. This suggests that the doctor hears antibiotics as having been made relevant by the grandmother. (3.21) 506 5 6

GMA:

No but what he gets is like throat infections an’ ear infections.

ALTERNATIVE PRACTICES FOR ASKING AND ANSWERING

7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

GMA: GMA: GMA:

GIR: DAD: GIR: DOC:

DOC:

DOC: GMA: DOC:

73

(0.4) Frequently. (0.4) An’ he (gets with thuh high fever.) (0.5) So that’s why I=uhm (1.5) we decided to bring ‘im in because (0.4) with thuh temperature. (.) (I know ) somet[hing’s gonna be wrong. [Daddy. [Stop that. [( ) -> Well actually uh regular cold can give you uh high -> fever for: three day:s. (0.4) -> As long as they act okay then actually it’s- it: -> (0.8) may go away by itself, (it’s) just uh virus. (2.0) => Uh:m, an’ you said he’s had uh lotta ear infections? (.) In thuh pa:st yes. °Okay.° An’ when was thuh last time he had one?

Finally, note, too, that at the double arrow the physician returns to questioning and does so in pursuit of one of the issues mentioned by the grandmother. Thus, there is evidence that the behaviors are both successful in terms of moving the physician to a pursuit of alternative trajectories of inquiry and are understood as working toward this end. This case shows the physician explicitly orienting to the diagnosis as under negotiation. In Extract 3.22, we see that the physician treats antibiotics as being negotiated (and pushed for) by the parent. (3.22) 119 1 2 3 4 5 6 7 8 9 10 11 12 13 14

DOC: MOM: DOC: MOM: DOC:

.hh And this has been going on for about uh week?, .hh Seven day:s, Okay:, [(°yeah.°) [But thuh cough is (wearing) wor:se. Getting worse. (.) DOC: °Okay. Well we’ll take uh-° I hear- actually heard it out there. (.) DOC: .hh (Yep-)/(eh-) .h I heard your cou^gh. DOC: .hh Uh:m sometimes coughs certainly can hang on -> for uh while. We’ll look an’ see whether or not she -> needs something that needs an antibiotic. DOC: .hh Uh:m (.) very often (yuh) get uh col:d an

74

PRESCRIBING UNDER PRESSURE

Here, following the parent’s mention of an additional problematic symptom (i.e., the worsening cough), the physician goes on to offer a no-problem explanation for this (lines 11–12) and explicitly treats the parent as lobbying for antibiotic treatment (lines 12–13). Physicians respond very similarly when the parent’s action has been to offer an alternative possible diagnosis. We can see an example of this in Extract 3.23. (3.23) 2074 (Dr. 9) 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

MOM:

.hh for strep, Thuh results came in last Tues:[day. DOC: [(^Oo.) Okay. MOM: So: (0.5) Uhm: h(h) .h(h)h £Just giving you that piece of information.£ DOC: Yeah. (.) DOC: -> It could very possibly. see strep with uh runny nose. DOC: h- Uhm_ .lkh ^Bu:t (.) if you’ve been exp^osed tuh strep, an’ exposed to uh co:ld I suppose you could get them both at thuh same: -ya know at thuh same ti:me, DOC: .h[h MOM: [And my son didn’t have a fe[ver ei[ther. DOC: [.mlk [Really, MOM: So:_ DOC: #Okay.#

In line 7, the mother announces an alternative possible diagnosis relative to the prior questioning trajectory. This is receipted as news in line 9 but is not pursued at that point. The mother pursues a fuller response from the physician with “£Just giving you that piece of information.£”. With this turn, the mother orients to her previous possible diagnosis as a departure from the norms of doctor-parent interaction. She does this through her “smile voice” (noted in the transcript with the “£”). As Haakana described, patients often mark their actions as delicate through the use of such resources (Haakana, 2001). With this action, though, she also creates a context for the doctor to elaborate on his response—“a post response pursuit of response” (Jefferson, 1981). The physician initially receipts the information with “Yeah.” but then does go on to address the diagnostic implications (“It could very possibly.”). She draws on her previous line of questioning to defend not going toward a strep throat diagnosis. (The mother had previously said the child had a runny nose, and this would be counter to a standard symptom list for strep throat.) What this shows is that physicians can resist a parent’s pressure to move toward a particular diagnostic trajectory, but they still feel the pressure and reﬂect this either by resisting or by succumbing to it. It is interesting, too, that later on in the treatment phase of the medical visit there is a statistical association between parent mentions of additional symptoms and

ALTERNATIVE PRACTICES FOR ASKING AND ANSWERING

75

the way physicians talk about their treatment recommendation. Following a statement of additional symptoms, physicians were more likely to offer parents delayed prescriptions (i.e., prescriptions that they start if the child fails to improve) or “contingency plans” (i.e., offers to prescribe following a phone call if particular circumstances follow, such as if the fever reaches a particular level) (Mangione-Smith et al., 2001). In the Seaside data, if a statement of additional symptoms was present in the visit, physicians exhibited these types of “concessionary” treatment recommendation behaviors 24% of the time. By contrast, in the absence of any mention of additional problematic symptoms, these concessionary behaviors were present only 9% of the time (chi2 (1) = 7.67; p = 0.006). Although unlike candidate diagnoses, mentions of additional problematic symptoms were not signiﬁcantly associated with physicians perceiving parents to expect antibiotics, there is still both qualitative and quantitative evidence that the behavior is understood as applying pressure to the physician to treat the child with antibiotics.

Discussion In this chapter, we examined how physicians design their history-taking questions and how parents respond. We observed that through the details of how these question-answer sequences are produced, physicians and parents negotiate whether the child’s illness (or speciﬁc symptoms) is problematic. Although it is rare that the history-taking activity involves any overt discussion of whether antibiotics should be prescribed for a child (but see Extract 6.9), it is quite common for physicians and parents to take part in a relatively covert verbal “push-pull” over their respective stances toward the child’s illness and its treatability. And through a negotiation of whether some aspect of an illness is problematic, the patient’s diagnosis and treatment are also being negotiated. History taking may seem an unlikely environment for negotiation to occur, but physicians and parents treat history-taking questions as part of a diagnostic trajectory. Whether or not parents are always able to determine what trajectory a physician’s questions may be part of could be seen as less important than that they understand physicians to be on a trajectory. If a parent understands the physician to be pursuing a diagnosis, even if they do not know which diagnosis, they may still be able to assess whether it is a problem-implicative one that will ultimately yield treatment for their child. In cases where a no-problem diagnostic or treatment trajectory appears to be projected by the questioning trajectory, parents make use of interactional practices to push physicians away from this trajectory and/or toward an alternative one. We examined two practices: mentioning additional symptoms and mentioning alternative possible diagnoses. Both of these practices are solutions to an interactional dilemma: Parents must offer a “no-problem” answer to a history-taking question, which potentially undermines their own position that the child has a problem. Thus, at this juncture, when parents extend their no-problem response in order to mention either the additional symptom or a possible diagnosis, we observed that this pressures physicians to adjust their stance toward the child’s illness by inviting sequence expansion regarding the new information (the symptom or the diagnosis).

76

PRESCRIBING UNDER PRESSURE

The two practices are slightly different in terms of the action that they accomplish. Whereas the additional symptoms pressure physicians by pushing them away from their prior questioning trajectory (usually a “no-problem” trajectory), alternative possible diagnoses afﬁrmatively push physicians toward another trajectory (usually one that involves a bacterial diagnosis with corresponding antibiotic treatment). Thus, although both practices are a form of lobbying, the latter practice can be seen to be somewhat stronger. Finally, we observed that physicians treated these behaviors as lobbying for a problem trajectory even when the parent did not offer an alternative. That is, even when only additional problematic symptoms are offered, physicians treat parents as lobbying for a problematic diagnosis and often for antibiotics in particular (recall 3.22, for example). As in earlier work, here I argue against the idea that in the history-taking activity the patient is “imprisoned within courses of action that are overwhelmingly undertaken at the doctor’s initiative” (Stivers & Heritage, 2001: 178). Although this portrayal is common (Byrne & Long, 1976; Fisher & Todd, 1983; Mishler, 1984; Waitzkin, 1991; West, 1984), particularly with respect to yes-no types of questions (Mishler, 1984; Roter & Hall, 1992), social interaction is much more ﬂexible than it is often credited to be. Parents, as shown in this chapter, hear the consequences of their responses relative to the question trajectory environment and exploit the sequential opportunity to add to their turn, and thus they do work that can, and frequently does, affect not only subsequent actions interactionally but also diagnostic and treatment outcomes later in the visit. This chapter also illustrates a second opportunity that parents and physicians utilize in covertly negotiating the treatment outcome of the visit. Whereas the primary way that the candidate diagnosis problem presentation works to negotiate is by asserting the parent’s stance that the child’s condition is a problem that can and should be treated, during history taking, parents react to physicians’ stances by either afﬁliating with or resisting physicians’ stances toward the child’s condition. And further, they make use of turn constructional resources to “manage” the direction physicians proceed with the visit. History taking leads to physical examination and quite often substantially overlaps it. Both activities represent an investigation of the child’s problem. However, during the physical examination, two actions relevant to our discussion are common: (1) additional inquiries that investigate the problem and (2) statements about what the physician is observing. We do not deal with the physical examination in its own right here, but we can readily understand that the ﬁrst type of action will provide parents with a similar response environment to the environment discussed here. The second is similar to the diagnosis environment. I will therefore look at both “online” and ﬁnal diagnoses in the next chapter. Online comments will be discussed as a distinct physician behavior in chapter 7.

4

No Problem (No Treatment) Diagnosis Resistance

U

ntil the diagnosis phase, covert forms of parent pressure typically work to encourage the physician in a particular diagnostic direction on the basis of parents’ own inferences about where the physician is heading. A diagnosis of a child’s condition transforms the interactional context because negotiations of what the diagnostic or treatment outcome will be must now contend with a diagnostic “result” and its implications. This chapter will examine parent resources for dealing with a no-problem diagnosis and its corresponding no-treatment outcome. In particular, we will examine parent resistance to the physician’s diagnosis as a form of pressure. Diagnosis resistance, like the other covert behaviors we have examined, is not terribly frequent. In the Seaside data, it was observed 17% of the time (Stivers, Mangione-Smith, Elliott, McDonald, & Heritage, 2003). But as we will see, it is nonetheless another consequential interactional resource in the negotiation of diagnostic and treatment outcomes. In this chapter, we will discuss what constitutes resistance and how it works. Diagnoses of the type typically given in the context of upper respiratory tract infection symptoms are rather transparently antibiotics (or nonantibiotics) implicative. But because the treatment recommendation has typically not yet actually been made, physicians are still subject to pressure for a particular outcome. Furthermore, even the diagnosis can be delivered as more or less ﬁnal. For instance, some diagnoses are delivered following a complete physical examination, whereas others are offered “online” during an otherwise in-progress physical examination activity (Heritage & Stivers, 1999; Stivers, 1998). Although most of my comments about diagnoses will be for diagnoses that follow a physical examination, the claims appear to hold for 77

78

PRESCRIBING UNDER PRESSURE

both online and ﬁnal diagnoses; in either situation, a treatment recommendation is pending and will be forthcoming, even if what may be immediately subsequent is more physical examination.

Background Heath (1992) and Peräkylä (1998) have shown that patients rarely respond to diagnoses at all or tend to offer only minimal acknowledgments (e.g., “mm hm” or “uh huh”). One analysis of why this would be the case is that diagnoses fall within the physician’s domain of expertise and authority. As Heritage argues, “Diagnostic reasoning is an activity based on special knowledge possessed and controlled by the profession of medicine” (2006: 85). Because the physician and layperson are so far apart in their relative degree of insight into the diagnosis, when physicians deliver a diagnosis, they are at the point in the visit where they are most able to “deploy their cultural authority to deﬁne the nature of the patient’s problem” (Heritage, 2006: 85). Because of this discrepancy of insight, patients are at risk of being only marginal participants in the diagnostic process. Additionally, patients may not generally respond to diagnoses because of the implications of the diagnosis for treatment. Patients exhibit a general orientation to diagnoses as preliminary to treatment recommendations (Peräkylä, 1998; Robinson, 2003). Peräkylä found that patients are more likely to respond to physicians’ diagnoses when physicians explain their diagnostic reasoning. Both Heath and Peräkylä identify alternative ways of formulating the diagnosis that signiﬁcantly affect the patient’s degree of responsiveness, and both also suggest that patients are more likely to resist no-problem diagnostic evaluations. Across these studies, there is general agreement that patient and parent responses to diagnoses are interactionally marked. This means that patients and parents who do respond are treated by physicians as doing something special. Despite this, the diagnosis delivery still offers a further locus for the negotiation of what kind of problem the child has and whether it is treatable. Physicians communicate their stance directly through the diagnosis, and parents have that announcement as an opportunity to communicate their own stance: that their child’s condition is problematic. Although they may be marginal participants in the clinical dimension of viewing signs, listening to symptoms, and triangulating the diagnosis, they are not marginal participants in terms of inﬂuencing the diagnosis and the treatment. When physicians offer their diagnoses, the category of treatment that will be suggested is usually projectable either as prescription treatment (most commonly antibiotics) or symptomatic treatment (overwhelmingly over-the-counter medication). Thus, in this chapter I argue that parent resistance to no-problem diagnostic evaluations is a communication behavior—like candidate diagnoses, mentioning additional symptoms, and mentioning alternative possible diagnoses—that displays parents’ stances toward their children’s conditions as problematic and, in many cases, as in need of treatment. One issue that underlies this is the acceptability of the diagnosis for the parent. As Helman (1978: 125) points out, “No diagnosis would be acceptable to patients, it

NO PROBLEM (NO TREATMENT) DIAGNOSIS RESISTANCE

79

appears, unless it was to a large extent consonant with their world view, and particularly with their interpretation of illness.” What we may further see is that a diagnosis that does not correspond with a treatment that is consonant with the patient’s model of their child’s illness is equally problematic.

A Preference to Progress to Treatment Recommendation In considering how diagnoses are responded to, we should keep in mind that parents bring children to the physician as the expert: the one with the cultural authority to give advice. Parents do not position themselves as experts, so when the physician reaches the diagnosis phase of the visit, parents are hardly in a position to corroborate the diagnosis. These issues support an interactional structure where the physician’s diagnosis receives minimal if any response. On top of these issues of cultural authority and expertise, there is the treatment dimension. One of the most common reasons parents seek help from physicians is that they feel they cannot successfully self-manage their child’s illness. Thus, they are seeking a solution. This then provides yet another sort of pressure to keep response to the diagnosis at a minimum in order to facilitate progress to the treatment phase. This pressure for progressivity is, of course, very consequential for parent involvement in this phase of the visit, including making it difﬁcult for them to participate. This is true, ﬁrst, because there is no structurally provided opportunity for parents to respond to the diagnosis: No action is “conditionally relevant” (Sacks, Schegloff, & Jefferson, 1974; Schegloff, 1968), and there is no other form of pressure for response. Conversely, there is pressure against responding because diagnoses are actions that can be understood as part of a larger structural organization for which treatment is the next activity (Byrne & Long, 1976; Robinson, 2003); any action that responds to a diagnosis (and particularly any action that itself makes relevant a responsive action) impedes the progressivity of the physician to the treatment recommendation and thus may be seen as a dispreferred action in this context. Empirical support for the preference for progressivity in this context is offered in two ways: First, neither physicians nor parents treat nonresponse or minimal response (e.g., acknowledgments) as impediments to progressing to the treatment recommendation. When physicians deliver diagnostic comments during the physical examination or offer ofﬁcial diagnoses following the completion of the examination, they display no orientation to seeking acceptance or even acknowledgment from parents. That is, physicians do not wait for parents. Rather, they move directly into treatment recommendation (in the case of ofﬁcial diagnoses) or to the next stage of the physical examination (in the case of online diagnoses). Similarly, parents typically offer no, or only minimal, acknowledgment of diagnoses (Heath, 1992; Heritage & Stivers, 1999; Peräkylä, 1998). Conversely, virtually any action that initiates a sequence in this environment is treated as resistant to the diagnosis, even if the action does not appear, on its face, to be treating the diagnosis as problematic. Thus, what appears to be problematic about such actions is, in large part, that they delay the physician’s progress to the treatment recommendation.

80

PRESCRIBING UNDER PRESSURE

Physicians typically progress immediately from delivering the diagnosis into treatment without waiting for the parent. This can be seen in Extract 4.1, where the parent does not respond to the physician’s diagnosis of a cold, bronchitis, and “no ear infection” (lines 1–2). (4.1) 1056 (Dr. 2) 1 2 3 4 5 6 7

DOC: -> He’s got uh good co:ld, little bronchitis_ -> <no ear infection. DOC: .h So I- Since you’re traveling I will wanta put him on some medi- me- byu:h: medication but_ (0.2) DAD: I kinda [ﬁgured that. DOC: [Yeah. DOC: Cause she has uh little tendency to get into trouble.

Despite the lack of parent uptake to the diagnosis, the physician moves directly from the diagnosis into the treatment recommendation (line 3). This suggests that a lack of parent uptake following the diagnosis is not treated by the physician as problematic. If a response was due, in line with other interactional research, we would expect to see accounts for the diagnosis, a backing down from it, and other reactions (Heritage, 1984b; Pomerantz, 1984; see chapter 5). A very similar case is shown in Extract 4.2. In this case, there is even a small silence following the diagnosis in line 1, but the physician continues through a description of the infection and then progresses directly into treatment (line 6). (4.2) 2002 (Dr. 6) 1 2 3 4 5 6

DOC:

.hhh Uh:m his- #-# lef:t:=h ea:r=h, is infected, -> (0.2) DOC: .h is bulging, has uh little pus in thuh ba:ck,=h DOC: Uh:m, an’ it’s re:d, DOC: .hh So he needs some antibiotics to treat tha:t,

Parents do sometimes minimally acknowledge diagnoses. We can see this in Extract 4.3. (4.3) 1183 (Dr. 1) 1 2 3 4 5 6

DOC:

Well I think what’s happened is is that she ha:s this: uh- (.) .h ear infection in her left ear?, MOM: -> [Mm:. DOC: [And we’ll put her on some medicine and she’ll [be ﬁne. MOM: [Okay.

But we can still observe that the physician does not indicate that acceptance or acknowledgment following the diagnosis delivery is required or even being sought.

NO PROBLEM (NO TREATMENT) DIAGNOSIS RESISTANCE

81

In fact, although the mother does receipt the physician’s diagnosis of an ear infection with a minimal acknowledgment “Mm:.” (line 3) (Gardner, 1997), the physician was already beginning his treatment recommendation, so the two occur in overlap. Thus, the acknowledgment was not being treated as a necessary prerequisite to his progressing to this next activity. Although sometimes, in response to diagnoses, parents offer more substantive acknowledgments such as “Okay,” as we have seen, it is more typical for parents to offer either no acknowledgment (e.g., 4.1, 4.2) or only minimal acknowledgment (e.g., Extract 4.3) in response. This is important because it suggests that physicians do not treat parent uptake of the diagnosis as mattering for whether and how they proceed. This, at least potentially, makes it difﬁcult for parents to be truly involved in this phase of the visit in that, unlike the other two phases we examined, here there is no sequentially provided-for opportunity for them to respond. But parents do exploit the “transition space” (i.e., the space between the conclusion of one TCU and the beginning of a next, not necessarily a silence) in this location to initiate actions that delay the physician’s progress to the treatment recommendation. The remainder of this chapter will focus on these actions as parents’ primary means of diagnosis resistance and thus of diagnosis-treatment outcome negotiation in this phase of the visit.

Diagnosis Resistance As we will see in the next chapter, what constitutes “resistance” is very much dependent on the sequential context and what action is due next, if any. Resistance to the diagnosis takes the form of initiating a new sequence in a context where nothing more is relevant and where progress to the treatment recommendation activity appears preferred. Resistance generally involves calling into question or disafﬁliating with the physician’s diagnostic evaluation. This can be accomplished with three different sorts of sequence-initiating actions: newsmarks (Heritage, 1984a; Jefferson, 1981), questions about symptoms, and questions about the diagnosis. In all three cases, the action initiates a new sequence in an environment where movement to the next activity is preferred. As we will see, most forms of sequence expansion—newsmarks or full questions—are built to prefer reconﬁrmation of the physician’s previously stated position. However, despite this preference, these actions are clearly designed and treated as resistant because they obstruct the progress of the course of action that is under way. The turn design of the resistant actions suggests that parents generally work to mitigate the resistant action. This maintains respect for the physician’s medical authority while nonetheless conveying a position of disafﬁliation with him or her. Newsmarks According to Heritage and Seﬁ (1992: 390), newsmarks (e.g., “Really?” and “It is?”) “treat the prior talk as ‘news’ and, to varying degrees, promote further informings.” In these data, I argue that their work in promoting further informing is disaligning because this defers a relevant next move to the treatment recommendation (in terms

82

PRESCRIBING UNDER PRESSURE

of the overall structural organization of the visit). As an initial illustration of this, we can look at Extract 4.4. The ﬁrst point at which the diagnosis is possibly complete is not responded to at all. The physician then reviews a couple of the relevant symptoms and signs, and the mother receipts the second one (that her child’s throat “does look uh little bit re:d” in lines 3–4) with what Heritage and Seﬁ (1992) term a “marked acknowledgment.” She responds with “Right.” (line 5). This portion of the diagnosis is conﬁrmatory of the parent’s position that the child’s condition is problematic, and this is particularly carried by the physician’s stress on and use of the “do”-auxiliary (“does”). (See Stivers, 2005a, for a discussion of modiﬁed repeats in an immediately subsequent position.) (4.4) 104 1 2 3 4 5 6 7 8 9

DOC:

.hh Uh:m, I think probably he’s- ya know has uh little viral col:d, His nose is uh little stuffy, DOC: .hh Uh:m (.) tl=His throat does look uh little bit re:d [but it doesn’t look like any[t h ing] much,=^Yeah, MOM: -> [Right. [Really,] DOC: An’ I think thuh redness is really -again like I say from dripping down thuh ba:ck, DOC: .hh Uh:m his chest is perfectly clear. There’s nothing in his lungs at all:.

However, when the physician then suggests that this redness is not actually signiﬁcant in line 4 (“but it doesn’t look like anything much,”), which is a no-problem diagnosis, the mother receipts this with the newsmark “Really,”; this constitutes diagnosis resistance. It initiates a new sequence and further discussion of the news that the child does not have anything serious: In response, the physician conﬁrms with “^Yeah,” a prosodically strong turn design, and then an expansion of her theory of the etiology of the redness that further downplays the seriousness of the boy’s condition. The sequence initiated by the newsmark is brief. But the physician goes on to further justify the diagnosis with supporting physical examination ﬁndings. This is built, with the “and” preface (Heritage & Sorjonen, 1994), as a continuation of the previous turn and thus not as a response to the newsmark. It is nonetheless notable that justiﬁcations for diagnoses are common following resistance. Sometimes the response is even more minimal than that shown in Extract 4.4. In Extract 4.5, the physician answers the question posed by the newsmark and then immediately moves on to the next unrelated physical examination component. (4.5) 1150 (Dr. 3) [the physician asked about the ears earlier, and the mother reported the child to be pulling at them indicating that they might be causing her pain] 1 2 3

DOC: Her ears are ﬁne. MOM: -> Are they? A[h=hhh DOC: [°Yeah.° (‘t’s) open up real big say “ah:::,”

NO PROBLEM (NO TREATMENT) DIAGNOSIS RESISTANCE

83

The physician provides an online report (Heritage & Stivers, 1999) that the child’s ears are ﬁne. In response to this, the mother offers a partial questioning repeat “Are they?” (line 2), which initiates a new sequence. In response, the physician offers a minimal and, in fact, quiet “°Yeah.°” (line 3) before moving on to the next segment of the child’s physical examination. Although this is an online diagnostic evaluation rather than a ﬁnal summative diagnosis, it indicates that there is no problem. Moreover, insofar as this is the very symptom that prompted the visit, it is unlikely that in other examinations the physician will ﬁnd additional problems. In this example, as in the previous one, we see that a newsmark does not necessarily engender a lengthy sequence, but it still initiates a sequence in an environment where the physician is simply progressing through the physical examination activity. In still other cases, physicians respond to newsmarks by revising their just-offered diagnosis. We can see this in Extract 4.6. Here, in response to the parent’s implied candidate diagnosis of yellow spots in the girl’s throat, the physician examines the throat. In his online diagnosis, he suggests that contrary to the parent’s implied theory, these are “primarily blisters back there.” (lines 1–2). Like the previous example, the spots are stated as the reason for the visit, so a no-problem evaluation of this symptom is tantamount to a no-problem summative diagnosis. (4.6) 1126 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12

DOC:

°Yeah:.° You know actually what those a:re °pr=h° .hh are primarily blisters back there. MOM: -> Yea:h? DOC: It’s almost like she’s got cold sores in thuh back of ‘er throa:t. MOM: (Oh:[::.)/(Aw:::.) DOC: [And u:sually that’ll go along with this just being viral. (.) MOM: -> [Really.= DOC: [#er-# DOC: =Y:eah.

In response, the mother offers a strongly question-intoned “Yea:h?” that calls the online diagnosis into question and is thus a ﬁrst instance of diagnosis resistance. This engenders further talk, though here it is not obviously responsive. Rather, in this position, the physician offers a slightly redone version of his prior diagnosis, leaving it equivocal as to whether it was designedly responsive to the “Yea:h?” or more of the diagnosis offered earlier. Either way, the physician, in lines 7–8, goes on to suggest that the diagnosis is “viral.” The move to labeling the diagnosis as viral may be part of the physician’s effort to strengthen his diagnostic position by shifting from a report of his examination ﬁndings to a diagnosis. In response, the mother offers a second newsmark, “Really.” which again initiates a new sequence, this time responded to with a reconﬁrmation, “Y:eah.” (line 12).

84

PRESCRIBING UNDER PRESSURE

I have asserted that newsmarks are the most minimal form of resistance. This concept is quite parallel to that laid out by Schegloff, Jefferson, and Sacks (1977): In cases of other initiated repair, there is a “natural ordering based on their relative ‘strength’ or ‘power’ on such parameters as their capacity to ‘locate’ a repairable” (p. 369). In line with this, Drew (1997) has shown that “open class repair initiators” such as “Huh?” or “What?” communicate to the recipient that there is a repairable within the prior turn but leave it to the speaker to locate the precise trouble source. Similarly, a newsmark following a diagnosis does not specify which aspect of the diagnosis is being resisted but nonetheless communicates some problem with it. Physicians do not necessarily “do” much with newsmarks besides conﬁrm the diagnosis. Often a “Yeah” or other single TCU response closes that sequence. But physicians may subsequently continue on their prior course of action with more informing, and this is arguably engendered by the newsmark-initiated sequence. However, even with nothing further, a reconﬁrmation sequence still both delays the physician’s progress to treatment recommendation and projects the possibility of more explicit disagreement. This is in contrast with “uh huh” or “mm hm,” which pass on the opportunity to do something more and treat a course of action as in progress (Schegloff, 1982). This is how we can analyze these newsmarks as accomplishing resistance and thus as a resource for negotiating the diagnostic and treatment outcome. At times, it is more explicit, as in the physician’s response to the parent’s newsmark in Extract 4.7. During a physical examination of a girl who presented with complaints of ear pain and coughing, the physician offers an online no-problem evaluation of the girl’s cough (line 5). Because it is still during the physical examination, it is arguably not a ﬁnal diagnosis, but it nonetheless is diagnostic and projects no treatment. To this, the parent offers a newsmark “No?” that resists the physician’s evaluation. (4.7) 1075 (Dr. 1) [presented with complaints of ear pain and coughing] 1 2 3 4 5 6 7 8 9

DOC:

Let me hear your cough. Can you cough for me?, (.) DOC: Go “.h #huh#” GIR: #huh huh huh#=.h DOC: She’s not too ba:d_ DAD: -> No?= DOC: =really:,=h right no:w, DOC: I don’t:=think she needs an antibiotic_ [uh: .hh DAD: [Okay.

In response to this resistance, the physician suggests, “I don’t:=think she needs an antibiotic_”. The physician’s introduction of treatment discussion in an environment where he had yet to provide a ﬁnal diagnosis appears to be an “early” move to treatment. And the move to antibiotics in particular appears to treat the mother’s “No?” as projecting disafﬁliation with his projected no-treatment evaluation. The physician’s strong assertion in line 8 (strong in part because of its placement) may display his lack of preparedness to overtly negotiate over antibiotics (though, of course, that is precisely what appears to be going on covertly).

NO PROBLEM (NO TREATMENT) DIAGNOSIS RESISTANCE

85

Another piece of evidence that supports the claim that newsmarks project disagreement and are used to negotiate the visit outcome is that following a newsmark and reconﬁrmation of the diagnosis, parents relatively commonly escalate to more explicit forms of diagnosis resistance. This can be seen in what follows in several of the cases we have looked at in this section. For example, if we return to the case shown in 4.4, we can see that following the sequence expansion engendered by the newsmark “Really,” the physician goes on to further account for the boy’s red throat and reconcile that with her no-problem diagnosis (lines 6–7), asserting that the redness is “from dripping down the ba:ck,”. This is further followed by the report of other null ﬁndings: a clear chest and no fever. In response, the mother offers more full-scale resistance: a statement of concern about her son’s moods that seems to be in line with her stance that her son has a treatable problem (lines 13–14). (4.8) SG104 [shown earlier in 4.4] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

DOC: DOC: MOM: DOC: DOC: MOM: DOC: MOM: -> DOC: MOM: -> DOC: -> MOM: -> DOC:

.hh Uh:m, I think probably he’s- ya know has uh little viral col:d, His nose is uh little stuffy, .hh Uh:m (.) tl=His throat does look uh little bit re:d [but it doesn’t look like any[thing ] much,= [Right. [Really,] =^Yeah,An’ I think thuh redness is really -again like I say from dripping down thuh ba:ck, .hh Uh:m his chest is perfectly clear. There’s nothing in his lungs at all[:. [Okay, .hh An’ I- ya know (you see) thuh fevers have gone dow:n, .hh uh:m_ I was just concerned cuz he’s been so cranky an’ I thought well [there must be something= [^Well:=botherin’ ‘im [that I can’t: [Well (that) could be. I mean-= =see_ What will happen.
((37 lines of ear anatomy explanation not shown)) 57 58 59 60 61 62 63 64 65 66

DOC:

[.hh [An- An- An’ it can: #i:#=ya know an’ then it can open up again.=If he were older he might say to you my ears’re popping. (.) MOM: Mm hm, DOC: -> Ya know an’ that just tells you it’s more of -> uh .h eustachian tube dysfunction. (.) MOM: [Oh:. DOC: -> [An’ that can bother them.

86 67 68 69 70

PRESCRIBING UNDER PRESSURE

MOM: Yeah.= DOC: -> =°That can bother them.° DOC: -> .hh But he has no ear infections.=e- His ear is[( ) an’ (.) looks just (ﬁ:ne.)

In contrast with newsmarks, which frequently receive minimal reconﬁrmation, after fuller scale resistance, physicians more commonly and more explicitly either defend their diagnosis or back down. In Extract 4.8, in response to the mother’s second resistant move (lines 13–14/16/18), the physician concedes that there may be something bothering him but then returns to her diagnosis as she suggests that this is “eustachian tube dysfunction” (line 63) and reasserts that “he has no ear infections” and “His ear is . . . looks just (ﬁ:ne.)” (lines 69–70). Here, the physician’s response to fuller scale resistance is to maintain her diagnosis, though it is not done immediately. The exact way that parents escalate varies. For instance, in Extract 4.9 the parent shifts from a newsmark to questioning the diagnosis, a stronger form of diagnosis resistance that we will discuss in more detail later. (4.9) 1150 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

DOC: Her ears are ﬁne. MOM: -> Are they? A[h=hhh DOC: [°Yeah.° (‘t’s) open up real big say “ah:::,” (.) GIR: A[h::, DOC: [(What=you’re tryin’ tuh say) “ah::::.” DOC: .hh “Ah::::#:::.#”=h DOC: Yeah? DOC: .h So unfortunately everything_ (.) .h I think she’s probably jus’ got kind of the ﬂu thing (in dow).= MOM: =Okay:, (.) MOM: -> So you don’t think there’s:- it’s in thee ear. Her ea:rs look perfectly ﬁne, so there’s no ear thing starting there. and so .hh I think she’s prob’ly got thuh same viral thing that -everybody else’s had in thuh house.

In response to the initial resistance “Are they?” the physician continues his physical examination and then offers an ofﬁcial diagnosis (lines 9–10). Although initially the mother offers a provisional-sounding acceptance of this with “Okay:,” (line 11), she then goes on to offer fuller scale resistance of the diagnosis with a request for conﬁrmation that the girl does not have “thee ear thing.” (lines 13–14). By asking

NO PROBLEM (NO TREATMENT) DIAGNOSIS RESISTANCE

87

about her daughter’s ears, the mother maintains a position that the girl may have an ear infection. Here, the resistance is formulated as an explicit question that makes relevant a response from the physician. The physician maintains and strengthens his position that the child has a nontreatable illness despite the mother’s concern that her daughter has an ear infection, but this still reﬂects that a negotiation of the diagnosis is underway. Yet a stronger form of resistance is to assert an alternative diagnosis. We can observe this pattern in Extract 4.10 (shown earlier as Extract 4.6). Following two newsmark-initiated sequence expansions and two rounds of physician response, the mother upgrades from this minimal form of sequence expansion to a stronger form of resistance: offering an alternative diagnosis in line 22. (4.10) 1126 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

MOM:

And then uh- I looked down her throat yesterdaylast ni:ght, an’ I could see thuh yellow:_ DOC: ^Okay. MOM: #spo:[t so:. ((trails off)) DOC: [.hh Well open up rea::l big. let’s take uh look an’ (say-) say #”Ah:::[:::.”=hh GIR: [Ah::::=hh DOC: .hh (0.5) DOC: °Yeah:.° You know actually what those a:re °pr=h° .hh are primarily blisters back there. MOM: -> Yea:h? DOC: It’s almost like she’s got cold sores in thuh back of ‘er throa:t. MOM: (Oh:[::.)/(Aw:::.) DOC: [And u:sually that’ll go along with this just being viral. (.) MOM: -> [Really.= DOC: [#er-# DOC: =Y:eah. DOC: .hh MOM: -> One ‘v thuh teachers told me it might be stre:p so:[:_ DOC: [.mlk Yeah we are starting to see some strep so I’m gonna culture just in case .hh she’s got both going on at the same ti:me but- .hh when you see: (you know)/(any uh) those #uh:# (thuh)/(that) white stuff you see back there is- is really not: like pus pus but it’[s ya know like she’s got blisters n’ MOM: [Oh yeah:_ MOM: Oh:::.

88

PRESCRIBING UNDER PRESSURE

The mother’s alternative diagnosis of strep throat both claims visit legitimacy and underscores her stance that her daughter needs treatment. More locally, this puts the physician in a position where he must deal directly with this diagnostic possibility. In this section, we have examined several instances of when diagnostic information is responded to with a newsmark. I have argued that these newsmarks do two primary things: First, they initiate a new sequence by making relevant at least conﬁrmation or disconﬁrmation of the diagnosis. Second, they disalign with the activity in progress and in doing so project disafﬁliation with the diagnosis and/or the treatment implications of that diagnosis. The claim is that both through the disruption of the physician’s progressivity to the next activity and through the projection of possible disagreement, the use of newsmarks following these diagnoses constitutes resistance. This was supported both in the way that physicians buttress their diagnoses and even their treatment recommendations in the face of newsmarks and also in the way that, commonly, full-scale resistance follows newsmark resistance sequences where no physician backing down is visible. Questioning an Examination Finding The second primary method of initiating a sequence and delaying a physician’s progress toward the next activity following diagnosis is to question one of the examination ﬁndings. Questioning the physician in this way is a stronger way of resisting than the use of a newsmark. Whereas the newsmark “merely” seeks reconﬁrmation of what the physician has just stated, questioning an examination ﬁnding identiﬁes a problem area explicitly. Moveover, it is an area that is speciﬁcally in the physician’s domain of expertise and thus more strongly projects disagreement with the physician. An initial example of this type of resistance was shown in Extract 4.9. Another example is shown in Extract 4.11, where the physician makes a general statement that there has been a lot of “cold and stuff going arou:nd,” implying that this is what this boy has. She gives her diagnostic conclusion that the boy does not have a problem: “aside from #thuh:# .hh thuh lotta mucus an’ stuff that he #ha:s,# °he- n- he sounds ﬁ°::ne,” (lines 3–5). In response to this no-problem diagnosis, the mother inquires about the results of the just preceding lung examination—whether he has chest congestion (line 6). (4.11) 2081 (Dr. 8) [just following lung examination; parent presented BOY as having cold symptoms and a bad cough] 1 2 3 4 5 6 7 8 9

DOC: BOY: DOC:

There’s been uh lotta cold and stuff [going arou:nd,=h [( ). but- n- .h ya know:, aside from #thuh:# .hh thuh lotta mucus an’ stuff that he #ha:s,# °he- n- he sounds ﬁ°::ne, His chest isn’t congested. Is his [cheDOC: [Yeah::, h=it’s just all of his:: up here: actually not really lin:- down in his lu:ngs,

NO PROBLEM (NO TREATMENT) DIAGNOSIS RESISTANCE

89

Questioning the examination ﬁnding is like responding with a newsmark in that both initiate new sequences and thus extend the diagnosis activity by making relevant a response. In this case, what we see is that in response to the parent’s question, the physician conﬁrms that there is no chest congestion and then goes on to more fully defend her evaluation that there is no problem. We previously observed with newsmarks that they normally are designed to prefer conﬁrmation that there is no problem. This is true here as well. In both cases, the physician is asked to reassert that there is no problem. This is a similar phenomenon to that discussed in the previous chapter: If there were a problem, the parent orients to it as something that would have been mentioned. In this case, the original question is designed to prefer a no-problem answer, but the mother immediately attempts to revise the preference design of the question (the shift from a negative declarative to an afﬁrmative interrogative design). And this may reﬂect her own difﬁculty in requesting reconﬁrmation that the symptom she thinks is problematic is not actually problematic. Arguably, then, the resistance in these questions is not so much the design of the questions as their position: that they are asked here at all. Because the preference within this activity is for parents not to intervene, the sheer presence of this kind of question, which seeks evidence and/or justiﬁcation for the physician’s diagnosis, is disaligning with the in-progress course of action. There is support for the idea that questions about examination ﬁndings are treated as a stronger form of resistance than newsmarks. Following questions about symptoms more commonly than following newsmarks, physicians go on to provide additional justiﬁcation for their diagnoses. In Extract 4.11, when the physician continues, she positively asserts that it is “up here: actually”, thereby rejecting the mother’s idea of chest congestion with “not really l- in:- down in his lu:ngs,” (lines 8–9). Here, the physician responds to the parent’s resistance by ruling out an alternative problematic condition—lung congestion—which would probably require medical treatment. When physicians offer accounts for their diagnoses, this reﬂects their understanding of a parent’s inquiry as in some way doubting or problematizing the diagnosis: a domain of knowledge normally treated as resting solely with the physician. Thus, the claim that parents are treated by physicians as lobbying is not evidenced solely by cases where physicians back down from their diagnosis. Rather, cases where physicians defend their position are equally good evidence that they are treating the interaction as involving a negotiation. One way that progressivity to the next activity is visibly delayed is that physicians sometimes back up into another phase of the visit and redo their examination and/or diagnosis. In Extract 4.12, the physician can be observed to offer what is normally done as an online comment during the physical examination (“Her ea:rs look perfectly ﬁne,” in line 17) and then moving to the diagnosis for the second time (notice that the ﬁrst time is in lines 9–10). And there is a built-in account for this diagnosis in that it is tied to “everybody else” in the house. (4.12) 1150 (Dr. 3) [shown earlier in 4.9] 1 2

DOC: Her ears are ﬁne. MOM: -> Are they? A[h=hhh

90 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

PRESCRIBING UNDER PRESSURE

DOC:

[°Yeah.° (‘t’s) open up real big say “ah:::,” (.) GIR: A[h::, DOC: [(What=you’re tryin’ tuh say) “ah::::.” DOC: .hh “Ah::::#:::.#”=h DOC: Yeah? DOC: .h So unfortunately everything_ (.) .h I think she’s probably jus’ got kind of the ﬂu thing (in dow).= MOM: =Okay:, (.) MOM: -> So you don’t think there’s:- it’s in thee ear. Her ea:rs look perfectly ﬁne, so there’s no ear thing starting there. and so .hh I think she’s prob’ly got thuh same viral thing that -everybody else’s had in thuh house.

The mother’s question “you don’t think there’s:- it’s in thee ear.” in line 13 is already, as discussed earlier, an upgrade on a prior resistant move in line 2. With the question, the mother directly queries the physician’s examination of the girl’s ear. Although it is designed to request conﬁrmation that there is no problem, by its sheer presence as an inquiry, it invites a revision of his prior statement. Following her resistant inquiry, the mother offers an account for that question with “
NO PROBLEM (NO TREATMENT) DIAGNOSIS RESISTANCE

91

hold have had viral infections, the physician also may inhibit further challenges by the mother. Next, by suggesting that the illness is “thuh same viral thing that -everybody else’s had” (lines 19–20), the physician invokes what is known in common with the mother—the epidemiology of the illness of other family members—and associates it with the condition of this child. Finally, whereas “ﬂu” projects no treatment by virtue of its being a viral illness, “viral” removes this intervening step, thus making a stronger claim that the illness is not treatable. In responding to this mother’s resistant move, the physician thus deploys a range of resources to defend his own position that the child has a no-problem condition. But note that these are all defenses and treat the mother’s inquiry as problematic, and thus as pressure. Like newsmarks, questions of physical examination ﬁndings that are not met with a backing down by the physician also are at times pursued. We can see this in Extract 4.13. Here the physician offers an online comment that is both designedly responsive to the mother’s concern of an ear infection and tantamount to offering a diagnostic evaluation. This diagnostic comment is shown in line 1. In line 2, the mother resists by calling the online ﬁnding into question through a repeat of it. (4.13) 1083 (Dr. 1) [presented with a concern of ear infection] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

DOC: We^:ll, his ears are oka:y.= MOM: -> =His ears are o[kay. DOC: [They’re not in[fected. MOM: [SHOW ‘IM where you [think it hurts. DOC: [Wh- where does it hurt in your ear.=h (.) BOY: Ear. DOC: That one? (0.5) DOC: °Okay.° DOC: .h[h MOM: [He told ‘em at school it was #hurting.# (0.4) DOC: .h #I mean it’s not even like=h [maybe uh= MOM: [Nothing. huh? DOC: =^little bit infected, DOC: .h Does it hurt when I go like tha:t?,=h BOY: Mm- yeah:, DOC: Oh does it, h DOC: Lemme look. (2.0) MOM: °You can’t see anything?,° DOC: .hh hh .hh Mkay:, we’ll look at the rest.

In response, the physician restates his diagnosis as an upgrade: He speciﬁes his previous “okay” as an explicit evaluation that the ears are “not infected” (line 3), which

92

PRESCRIBING UNDER PRESSURE

more explicitly rules out a treatable condition. Peräkylä (1998) has shown that patients are less likely to resist diagnostic evidence to which they have no access. The mother in Extract 4.13 will have difﬁculty resisting the physician’s claim that the ears are not infected because it is a type of claim to which she has no epistemic access. In response, the mother asks her son to restate the location of his symptoms (lines 4–5). This further resists the diagnosis because it now solicits consideration, if not actual reinvestigation, of the child’s symptoms. And it suggests that if there is no infection, the physician must reconcile the symptoms her son is experiencing with this lack of cause. Subsequently, the physician investigates the boy’s pain, ﬁrst by asking him about it (lines 6 and 9). When the mother resists still further with additional outside evidence that her son reported the symptom at school, the physician restates his diagnosis, this time more strongly. The strength here is done both with prosody and with the negation of any small possibility of infection: “not even like=h maybe uh ^little bit infected,” (lines 15/17). But before completion, the mother is resisting further with “Nothing. huh?” (line 16). It is notable that, similar to previous cases, while the form of this request for conﬁrmation prefers a no, it is launched in the middle of an assessment that offers exactly this. Again, then, by the sheer presence of this inquiry, the mother pressures the physician to revise his assessment. And responsively, the physician investigates the child’s symptoms: He both inquires about them (line 18) and reexamines the ear (line 22–23). Finally, following still further inquiry from the parent (line 23), the physician does not speciﬁcally respond to her but rather addresses the issue by announcing a move to continue a thorough physical examination (line 24). In this section, we have examined parent resistance that takes the form of a query of a child’s symptom or a physician’s examination ﬁnding. I have argued that this practice is stronger than a newsmark in terms of resisting the diagnosis. In each case, the physician’s response to the question has not been a simple conﬁrmation but has instead ruled out an alternative more problematic condition (4.11), restated the diagnosis (4.12), or further investigated the child’s complaints (4.13). And these types of responses are very common, unlike the more minimal responses to newsmarks. In all cases, physicians either buttress and defend their position or open themselves up to revising their position (by moving toward a diagnostic concession or by restarting the examination). This suggests that both parties—physicians as well as parents—are negotiating the diagnosis and treatment outcome of the child’s visit. Calling into Question the Physician’s Diagnostic Inference The ﬁnal type of diagnosis resistance to be discussed here involves the parent calling into question the physician’s diagnostic evaluation. Of the three types of diagnosis resistance, this is the strongest because it questions a medical evaluation of the child’s symptoms and signs: a domain over which the physician is normally treated as having sole responsibility and epistemic ownership. As an initial example of this practice, we can look at Extract 4.14. Here, the parent’s resistant action requests conﬁrmation that the condition the physician characterized as involving a “whee:ze” is not “bronchitis . . . (or anything,)” (line 6). Although

NO PROBLEM (NO TREATMENT) DIAGNOSIS RESISTANCE

93

designed to prefer disconﬁrmation and offered cautiously (with the addition of “or anything,” at the end of the turn), the mother nonetheless questions the physician’s diagnosis as involving a wheezing condition. The physician’s announcement that “she’s- just got uh little bit of uh whee:ze_” may initially sound incomplete, but it is, from the parent’s perspective, enough to display a clear diagnostic alternative to her own concern of bronchitis. (4.14) P202 (Dr. 8) 1 2 3 4 5 6 7 8 9 10 11 12 13 14

DOC:

I think you’re all do:ne_ I can try giving you some medicine for that- (.) - uh:m=h tlk for thuh cou::gh, Is- she’s- just got uh little bit of uh whee:ze_ an:’ sometimes- uhm .h[h MOM: -> [(It’s) not: bronchitis though (or anything,) DOC: No::.=h Uhm_ .h Bu:t- after they have uh cold sometimes they can have this kind of uh wheezy cou:gh, DOC: -> Uhm, .hh (.) an:d I think probably:=your husband smoking inside the house is contributing to it so: he should really smoke outside, It makes:[their: uhm=h lungs more=h sensiti:ve=huh yeah. MOM: [°I know:.° ( big issue )

In her response to this resistant action, the physician ﬁrst conﬁrms the negatively formulated question about the candidate diagnosis with “No::.” (line 7). Second, she reasserts her observation that the cough is wheezy, but this time it is embedded in a larger diagnosis of a “cold” (lines 7–9). By adding that smoking is a contributing factor (lines 10–13), the physician may be understood to be not only defending her diagnosis but also addressing the mother’s difﬁculty in understanding a wheezing cough—often associated with allergies or irritants. Thus, the physician provides some insight into what might underlie the wheezing condition—the father’s smoking. This analysis is in lines 12–13, where the physician accounts for her explicit linking of smoking and wheezing. In this case, though, the basic practice is the same as we have seen: a sequence-initiating question is responded to ﬁrst by a disconﬁrmation and then a justiﬁcation of the diagnosis, which further displays the physician’s analysis that her judgment had been called into question. Thus, as with the other forms of sequence expansion, we have evidence that questioning the diagnosis is heard as resistant and that it is working to negotiate the diagnostic outcome of the visit. The precise way that parents question the physician’s diagnosis can vary. Whereas in Extract 4.14 the parent offered an alternative possible diagnosis as a question, in Extract 4.15, the parent offers an account for why she thought the illness was problematic, which adopts a stance that is at odds with the one the physician has adopted through his diagnosis that she has “an acute gastroenteritis:,” (line 1) that “Doesn’t look like it’s too signiﬁcant,” (line 4). In terms of the mother’s stance that

94

PRESCRIBING UNDER PRESSURE

her child’s condition is problematic, this is a blow to the legitimacy of her visit and certainly treats the illness as no problem. In response, the mother ﬁrst accepts the diagnosis with “°Okay:” (line 6); however, this is prosodically built as preliminary to something more, and she goes on to state her concern with the amount of vomiting her daughter had the previous evening (lines 6/8–11). In the way it is framed as a concern, the mother is hearably defensive. Also, this symptom was earlier stated as the primary reason for the visit, so her raising it here is for the second time. Therefore, particularly with its placement just following the diagnosis, she hearably calls into question the physician’s diagnosis of the condition. Additionally, this construction builds her daughter’s illness as a signiﬁcant one whether to justify her child’s condition, ensure adequate medical investigation of her child, or present evidence for the treatability of that condition. (4.15) 1059 (Dr. 2) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

DOC: GIR: DOC: GIR: MOM: -> DOC: MOM: -> -> MOM: -> DOC: MOM: -> -> DOC: MOM: -> DOC: MOM: -> DOC: -> -> MOM: DOC: -> -> MOM: DOC: -> MOM: DOC: MOM: DOC: ->

So it looks like she has an a[cute gastroenteritis:, [( ) ((about lollipop)) Doesn’t look like it’s too sig[niﬁcant, [( ) °Okay: th-° .hh th- I was con[cerned cause uh[(And expect ‘er-). Usually: if- she:- when she=if she vo:mits: she(0.8) it- it doesn’t last as long or as o:ften. (.) La[st night (was -) [(‘ts-) jus- jus’ uh few throw ups an’ that’s thee end of it. #Yeah [thee-# [Yea:h_ Like I sa[y she was probably at least=h_ [(It) was uh little uIt was at least ten [ti:mes. [(s- s-) [more like: ten to twe:lve (yeah.) [Sh #uh:#=yeah. well she had very signiﬁcant=uh: (5.5) Signiﬁcant throw-up, but i:t=uh: [(Yeah [ .) [.hh [there wasn’t uh whole lot of bile in it so she’s not obstru- I don’t (wanta th=[say) [No. .h[ likely be ob[structed #it_# [No. [No it was just once that [I saw [that. [.mlh [Or: uh°. (2.8) °But=uh:=h° (0.5) Okay: every- An’ everything checks out ﬁ:ne,

NO PROBLEM (NO TREATMENT) DIAGNOSIS RESISTANCE

95

In response, the physician upgrades the mother’s evaluation of the vomiting with “very signiﬁcant” (line 20). In this way, he works to reconcile his diagnosis of “acute gastroenteritis:,” his assessment of “not too signiﬁcant,” and the mother’s narrative about her daughter’s illness experience. He then accounts for his evaluation by articulating part of the rationale that had previously been unavailable to the parent, though surely utilized by the physician: “there wasn’t uh whole lot of bile in it” (line 23), thus that there is no bowel obstruction. Finally, following the mother’s agreement that there was not in fact much bile, the physician reinvokes his earlier diagnosis by restating his examination ﬁndings: “An’ everything checks out ﬁ:ne,” (line 31). That the physician is reinvoking this here is also marked with the self-initiation of repair that appears to be primarily in order to insert the “And” preface that ties this back to the diagnosis he was providing prior to the mother’s resistance. In contrast with the prior two examples, here the physician not only maintains his diagnosis but also responds to the parent’s resistance by ruling out an alternative, medically treatable, and much more problematic condition. Although the child’s diagnosed condition might be treatable by, for example, intravenous ﬂuids for dehydration, here the physician orients to the girl as not in need of medical treatment. Another way that parents question the diagnosis is related to that shown in 4.15: Parents can offer up symptoms that are not easily reconciled with the diagnosis offered by the physician. This is shown in Extract 4.16. In this example, the girl’s visit is due to a rash. As the physician is examining her, she asks the question shown in line 1. In line 6, she offers an explanation for the condition—an informal evaluation but still potentially understandable as a diagnosis that is coordinated with a shift from examining the girl to writing information in the chart. (4.16) 517 1 2 3 4 5 6 7 8

DOC: DAD: DOC: DOC: DAD:

Any new soaps, new detergents, Uh::, (1.0) I don’ know. [Could be new soap. [((DOC moves from GIRL to chart)) Mm:. (.) That could be thuh rea[son. ((looks at DAD)) [That could be it?, (2.0) ((DOC nods))

((31 lines not shown talking about what soaps they have; physician writes in patient’s chart)) 40 41 42 43 44 45 46 47 48

DAD: GIR: DAD: GIR: DAD: DOC: DAD: DOC:

Maybe no- no soap for uh week I guess?, Because= =hh= =[ma=[°hh° Use [Dove. [.hh hh .hh Use Dove?, Yeah. (.)

96 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

PRESCRIBING UNDER PRESSURE

DOC: GIR: DAD:

[Thuh whi[te Dove. [(wah??) [O:kay. (2.0) DAD: -> But how come it’s not showing up on her: hand. DOC: -> That’s kinda wei:rd. ((moves from chart to GIRL)) DAD: Yeah_ (.) DOC: .hh= DAD: =I mean [cuz she (did) DOC: [I- That’s thuh ﬁrst thing I would do is [tuhDAD: [Okay. (0.2) DOC: (.hh) (Turn this way lemme) peek in thuh back.

Following an extensive detailing of the use of soaps in the family household, the father jokingly proposes a solution of “no soap for uh week I guess?,” (line 40). In response, the physician offers a serious recommendation “Use Dove.” (line 45). After acceptance of this recommendation (line 51), the father resists this diagnostic explanation for his daughter’s condition with a query about why, if it were a soap allergy, the rash would not be on his daughter’s hands as well, where she almost certainly was exposed to the soap. We have seen that in response to diagnostic resistance, physicians consistently respond by taking up the resistance and either defending or backing down. In this case, we see this again. In response to the father’s resistant query, the physician both acknowledges the apparent contradiction and moves back into a physical examination of the girl across line 54. After more examination, she recasts her earlier unmitigated recommendation (to replace the new soap with Dove soap) as “thuh ﬁrst thing I would do”. This downgrades her earlier recommendation from one that was ﬁnal to one that is a ﬁrst step. Thus, although she does not completely revise her diagnosis, she does still back down from it. In addition to negotiating the diagnosis, physicians also routinely treat diagnosis resistance as related to the treatment outcome. Although we will discuss the treatment implications in the next section, here I want to draw the connection between the two. We can see this in Extract 4.17, where, following no indication of a problem, the mother initiates a sequence with a request for reconﬁrmation of the ﬁndings (line 1). As is typical, this inquiry is designed to prefer a “no”, and this pattern is again present in her follow-up newsmark “No?” and then her question of the diagnosis itself: “not even earache or.” (line 7). (4.17) 302509 1 2 3

MOM: MOM:

He- he [doesn’t have any[(thing)? [((head shake)) [((Doc washing hands))

NO PROBLEM (NO TREATMENT) DIAGNOSIS RESISTANCE

4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

DOC: DOC: MOM: MOM: DOC: DOC: DOC: MOM: DOC: MOM: DOC: MOM: DOC:

97

[No. [((head shake)) No? Not- not even earache or. No. Ears are ﬁne, the throat’s ﬁ:ne? (.) Her che:st sounds okay, Okay, It’s just maybe a viral (.) uhm_ Yeah cause he’s been coughing a ^lot, a lo:t, and then=[I start(ed) giving him Robitussin,= [Mm hm, =but I want(ed) to make sure he doesn’t have throa:t or ear infection,= =No. When did you start- giving the Robitussin.

In response to the inquiry about the diagnosis, the physician disconﬁrms the alternative diagnosis and goes back over the boy’s physical examination ﬁndings (lines 9 and 11). These ﬁndings are accepted (line 12), but following a diagnosis (line 13) that suggests possible infection (though viral and thus untreatable), the diagnosis is resisted once again, ﬁrst by offering a problematic symptom (line 14) and then with an account for the mother’s inquiries (lines 17–18) and stating that she was concerned about a throat or ear infection. Like the other turns, this one is disconﬁrmed (line 19), but here we see that the physician moves to what treatment was successful, thereby displaying her understanding that treatment may be one of the parent’s underlying issues and thus what the resistance is motivated by. This example also offers further evidence that parents treat types of resistance as ordered in strength, such that newsmarks often precede inquiries about physical examination ﬁndings, which, more often than not, precede inquiries about diagnoses. In this section, we have examined a third type of sequence expansion for resisting the physician’s diagnosis: calling the diagnosis into question. Like questioning examination ﬁndings, this form of sequence expansion also typically engenders fullform responses to resistance, including a restatement of the diagnosis (4.14, 4.17), ruling out a more serious condition (4.15), accounting for the diagnosis (4.17), and further investigation (4.16). In all cases, we have observed that these practices are treated as resisting the diagnosis and as working to negotiate the diagnostic outcome of the visit.

Implications of Diagnosis Resistance for Treatability Parents resist physician no-problem diagnoses for a range of reasons. Generalizing from the cases we have, there appear to be three different but closely related motivations: First, parents may see a no-problem diagnostic evaluation as threatening the legitimacy of their visit. This invalidation is potentially embarrassing for a parent

98

PRESCRIBING UNDER PRESSURE

because an unwarranted visit is possibly viewed as resulting from poor judgment as to the severity of the child’s illness. Alternatively, parents may resist such an evaluation because they are concerned that the physician may have failed to notice a more serious diagnosis and believe that their child truly is sicker than the physician’s diagnosis reﬂects. A third motivation is that parents may feel that their child needs treatment and a no-problem diagnosis threatens the prospect of this outcome. A concern with treatability may in turn be motivated by a desire for a restful night, a need to return to work, a belief that the child is seriously ill, or a direct desire for antibiotics (which may itself be motivated by one of the previous issues). Sometimes a parent’s resistance may not clearly index just one of these three underlying motivations, and of course, these issues are often intertwined. I will argue that although each of the motivations outlined here is at times evident in a parent’s talk, physicians routinely understand parent resistance as indexing a desire for treatment for their child. In what follows, I show qualitative evidence for this, but quantitative evidence also exists. For instance, in the Seaside data, if there is no parent resistance, physicians report perceiving parents as expecting antibiotics only 7% of the time, whereas when parents have resisted a diagnosis, this jumps up to 20% of the time ( p<.05) (Stivers et al., 2003). And in a multivariate model, if parents resisted a viral diagnosis, the physician was 2.73 times more likely to report perceiving them as expecting antibiotics ( p<.001) (Stivers et al., 2003). But as discussed in chapter 1, it is possible for a physician to misinterpret a behavior as indexing a desire for treatment when it actually represents a concern about the legitimacy of the visit or a more serious condition. There is some support for this in the quantitative data as well, because there was no association between parents resisting a diagnosis and having reported an expectation for antibiotics. Although parents may have a variety of concerns and motivations that inform whether and how they resist, physicians typically treat diagnosis resistance as a mechanism for treatment negotiation. This is illustrated in the case shown previously in Extract 4.9. The way the physician formulates the diagnosis following parent resistance reﬂects his orientation to the treatment implications of the diagnosis without saying anything about the treatment recommendation. (4.18) 1150 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12

DOC: MOM: DOC:

Her ears are ﬁne. Are they? A[h=hhh [°Yeah.° (‘t’s) open up real big say “ah:::,” (.) GIR: A[h::, DOC: [(:What=you’re tryin’ tuh say) “ah::::.” DOC: .hh “Ah::::#:::.#”=h DOC: Yeah? DOC: -> .h So unfortunately everything_ (.) .h I think she’s -> probably jus’ got kind of the ﬂu thing (in dow).= MOM: =Okay:, (.)

NO PROBLEM (NO TREATMENT) DIAGNOSIS RESISTANCE

13 14 15 16 17 18 19 20

MOM: DOC: DOC:

99

So you don’t think there’s:- it’s in thee ear.
Speciﬁcally, in lines 9–10 the physician prefaces his diagnosis with “unfortunately”. This treats the girl’s illness of “just . . . kind of the ﬂu thing” as bad. But the only negative thing about this illness is the lack of treatment for it, so it is to this that the physician’s turn design appears to be oriented. Not only do physicians treat diagnosis resistance as treatment implicative but also they treat parents as seeking a bacterial illness (and the corresponding antibiotics treatment). Evidence is shown in Extract 4.19. The physician’s diagnosis was offered a bit earlier as “I think probably he’s- ya know has uh little viral col:d,” (Extract 4.8; lines 1–2). The parent then resisted the diagnosis with a newsmark “Really,” and subsequently resisted the diagnosis again with “I was just concerned cuz he’s been so cranky an’ I thought well there must be something botherin’ ‘im” (Extract 4.8; lines 13–14). At line 69, the physician is returning to her diagnosis. She has still not offered a treatment recommendation. (4.19) 104 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89

DOC: ???: MOM:

DOC: MOM: DOC:

MOM: DOC: DOC: DOC: DOC:

.hh But he has no ear infections.=e- His ear is[( ) an’ (.) looks just (f[i:ne.??) [#hu:h .h huh# [Yeah I just thought- I thought- Cuz thuh fever (was gone) I said “^well” .hh I don’t need tuh take him (‘n) then he was so cranky this [morning (I thought we[ll: ya know)= [Yeah:. [^Well (see) =[ I better take him because_ ] =[(you listen)_Things can change.^] If all of uh sudden his fever(s) start going up agai:n? .hh He starts developing other symptoms, please let us know:. (.) [(Well yeah:.) [(But)/(What) I think what it’s doing now is resolving. .hh An- #e-# tincture of ti:me. (0.4) is basically usually what you need with uh co:ld. -> .h An’ viruses can start thuh same way that bacterial -> infections can.=They can be just as sick?

100 90 91 92 93

PRESCRIBING UNDER PRESSURE

DOC: DOC: DOC:

.hh They can have high fevers? (.) You know all that sort of thing. .hh What is important are lots of ﬂuids,

Just as the physician is returning to her no-problem diagnosis, the parent offers justiﬁcation for bringing her son in for the visit (lines 72–75/76). Then, at line 84, the physician suggests a prognosis for the illness—that it is resolving and will continue to do so. With this, she offers a nonantibiotic treatment recommendation (“tincture of ti:me.”), which clearly suggests a self-limiting and self-healing process. But before offering a positive formulation of her treatment recommendation, she asserts that “viruses can start thuh same way that bacterial infections can.=They can be just as sick?” (lines 88–89). Here, the physician does several things. First, with “viruses” she indexes the boy’s current illness, which has earlier been diagnosed as a “viral col:d”. Next, by comparing viruses with bacterial infections, she implies that this is the relevant contrast and treats the parent as having been concerned about a bacterial infection by suggesting that the way the boy’s illness started is the way a bacterial illness would; she states that they “can start thuh same way”. This formulation also supports the mother’s position that her son is quite sick through her mention of “They can be just as sick?” (line 89). Importantly, the physician raises bacterial infections here in an environment where the parent had resisted a no-problem diagnosis. In this way, the physician conveys her understanding that the resistance was in search of a bacterial diagnosis. Additionally, the physician works to address the doctorability dimension of the resistance by suggesting that a viral illness may be equally problematic in experiential terms because patients can be equally sick. With the mother’s reraising some justiﬁcation of her reason for visiting, the two issues converge. Similarly, returning to the case shown in Extracts 4.6, we can see a second case where the viral-bacterial distinction is raised in the face of diagnosis resistance. Again, this offers evidence that physicians understand this to be being negotiated with parent resistance. (4.20) 1126 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12

MOM:

And then uh- I looked down her throat yesterdaylast ni:ght, an’ I could see thuh yellow:_ DOC: ^Okay. MOM: #spo:[t so:. ((trails off)) DOC: [.hh Well open up rea::l big. let’s take uh look an’ (say-) say #”Ah:::[:::.”=hh GIR: [Ah::::=hh DOC: .hh (0.5) DOC: °Yeah:.° You know actually what those a:re °pr=h° .hh are primarily blisters back there. MOM: -> Yea:h? DOC: It’s almost like she’s got cold sores in thuh

NO PROBLEM (NO TREATMENT) DIAGNOSIS RESISTANCE

13 14 15 16

101

back of ‘er throa:t. MOM: (Oh:[::.)/(Aw:::.) DOC: -> [And u:sually that’ll go along with this just -> being viral.

In the face of diagnosis resistance to the physician’s diagnosis (line 11), the physician suggests as part of his response that the symptoms are part of a “viral” condition. Speciﬁcally, the physician orients to the parent as having implied that the condition was bacterial with the rejection component of the diagnosis in lines 15–16. When the physician says “that’ll go along with this just being viral.”, he implicitly contrasts viral with another category that would be more signiﬁcant. This is done primarily through the use of “just”, which treats the given diagnosis as less signiﬁcant and possibly less treatable. Like other parent behaviors that are understandable as pressuring the physician for an antibiotic prescription, diagnosis resistance can affect treatment decisions and thus visit outcomes. Extract 4.21 shows evidence of a treatment recommendation being changed in the face of diagnosis resistance. In this case, the physician offers her diagnosis in lines 1–3. As part of this diagnosis, the physician offers a mitigated nonproblematic ﬁnding (“uh little bit of ﬂuid on thuh right si:de,” [line 2]). In response, the parent agrees with the physician about the ﬂuid (line 4). Before the completion of the physician’s next TCU, the parent goes on to assert that the right ear is a recurrent problem (lines 4–5). Although the physician appeared to be offering her report of “uh little bit of ﬂuid” as part of an overall no-problem evaluation, the parent’s agreement with it treats it as a problematic ﬁnding with which to agree. In addition, by acknowledging the physician’s turn unit by unit, she may be conveying incipient speakership (Jefferson, 1983) and, in this position, an indication of forthcoming resistance. (See Clayman & Heritage, 2002, for a related phenomenon in news interviewee responses to interviewer questions.) (4.21) 2057 (Dr. 8) [Mom mentioned a concern of ear infection earlier] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

DOC:

.h Yeah. her ears aren’t infected. <She has uh little bit of ﬂuid on thuh right si:de, ‘n I thi[nk that’s probably whe[re_ MOM: -> [Yeah. [(it’s) -> (just) thuh one that always (dr ). DOC: Yea:h, DOC: .hh Uh:m,=h .mlkh So it’s been going really nonstop for two weeks, (.) DOC: You think,=[h MOM: [Uhm, well it- on and off: (as I say:.) (I mean she would like uh little bit.) DOC: Uh huh:, h DOC: -> .h But when you do:n’t it- [seems to relapse? MOM: [( medicate her

102 16 17 18 19 20 21

PRESCRIBING UNDER PRESSURE

an’ (.) Yeah. ) DOC: -> #An’ it comes back, okay.# DOC: .h Yeah:. I think she may have got uh #s:inus infection,=h DOC: Uh:m_ (2.0) When was the last time she took antibiotics:. ((11 lines of discussion of types of antibiotics not shown))

33 34 35

DOC: MOM:

.hh Okay. I’m gonna put ‘er on Amoxicillin. which she’s been on before probably::, Okay,

On the surface, lines 4–5 do not appear to be resistant. But they take up a portion of the physician’s turn that, although problematic, was being designed to support a no-problem diagnosis. Moreover, it was a portion of the turn not designed for response. In this way, the mother conﬁrms a problem that the physician had not designed her turn to identify as such. It is in this rather indirect way that the mother’s turn accomplishes opposition to the physician’s diagnostic evaluation. And the physician treats her as opposing a no-problem diagnosis insofar as she shifts from listing nonproblems and minor problems to querying the parent further. Following the mother’s resistance, the physician agrees with “Yeah.” and then begins to summarize previous talk in the form of a request for conﬁrmation with “So it’s been going really nonstop for two weeks,” (lines 7–8) and further pursuing response with the “You think,” (line 10). In this case, the sequence of questions and answers across lines 7–17 eventuates in a change in her diagnosis from what, across the physical examination, appeared to be headed for a no-problem evaluation to “I think she may have got uh #s:inus infection,=h” (lines 18–19). This change in diagnosis implies a corresponding change in treatment because, although treatment had not yet been recommended, the treatment projected by the previous diagnostic trajectory was an over-the-counter remedy. By contrast, the treatment for sinusitis is usually antibiotics (line 33). In this case, the treatment is clearly negotiated, most obviously because the physician changes her diagnosis and corresponding treatment in the face of resistance. Her questions in lines 7–8 and 14 appear designed to foreshadow a treatment recommendation that would be in line with prescription treatment and a bacterial diagnosis rather than in line with the no-problem ﬁndings she has reported throughout her examination. Finally, in support of the claim that diagnosis resistance is a type of pressure for antibiotic treatment in particular, we can look to Extract 4.22. In this case, the parent’s diagnosis resistance is an overt lobbying for antibiotics. The physician offers his diagnosis in lines 3/5. The mother receipts that with overt pressure for antibiotics (which will be the focus of chapter 6). She suggests that the diagnosis is in line with previous physicians who suggested “the only way he got rid of his cold is tuh keep him on antibiotics for like twenty- uh: for: twenty one d^ays.” (lines 12–13).

NO PROBLEM (NO TREATMENT) DIAGNOSIS RESISTANCE

103

(4.22) 1016 (Dr. 1) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

DOC: BOY: DOC: MOM: DOC: MOM: DOC: MOM: BOY: MOM: BOY: MOM: DOC:

So:- let me hear your cough. Cough. #huh huh huh huh# So now- i- he has [this[He has uh cou:gh, co:ld, (Here-) He has uh cough, he has uh col:d, .h [Kay hold thi[s up. ((to BOY)) [.hh [An:d ya know i- i- thuh doct[ors in= [Ohhhha= =[in Sand City f- ﬁgured the only way] he got= =[ a a h h h = f h h h h ] =rid of his cold is tuh keep him on antibiotics for like twenty- uh: for: twenty one d^ays. Well ^sometimes that’s an answer, only because_ thuh reason for that i:s is that- sometimes they have uh sinus infection.

Here, unlike previous examples, the mother takes an overt (rather than covert) position in favor of antibiotics, but it is still an instance of diagnosis resistance—just a very strong form of it. In response, the physician suggests that such a treatment would require a corresponding bacterial diagnosis—a sinus infection. The physician’s diagnosis had been a “cold” (line 5). But notice that the doctor is amenable in his response to the resistance: “Sometimes that’s an answer,”. This example demonstrates that parents are oriented to diagnoses for their treatment implications. At least sometimes, parents exploit this opportunity to either indirectly pressure physicians for antibiotics by resisting a diagnosis that does not correspond to such treatment or, though less common, directly push for antibiotics in this location, where they initiate rather than respond to the treatment recommendation. And here it is clear that the parent is concerned with receiving antibiotics as a solution to her child’s problem. This section has demonstrated that regardless of the variety of motivations that may underlie it, parent resistance to the diagnosis is consistently understood by physicians as a way of lobbying for antibiotic treatment. Primary evidence was that in response to diagnosis resistance, physicians typically either succumb to the pressure or ﬁght it. Either way, they enter into a negotiation of it with parents.

Discussion This chapter has argued that once a physician offers a diagnosis—whether online or ofﬁcial—parents are no longer in a position of encouraging physicians to explore a particular diagnostic and/or treatment trajectory. Rather, if they are to effect change in the treatment trajectory and visit outcome, they must contend with the diagnosis presented by the physician. The primary mechanism through which parents negoti-

104

PRESCRIBING UNDER PRESSURE

ate treatment outcomes in this sequential context is through querying the physician. I argued that questioning works because the action of diagnosis delivery does not make response conditionally relevant. Doctors and parents share an orientation to the diagnosis as within the sole knowledge domain of the physician. Both of these issues make parent uptake of the diagnosis dispreferred, and the structural organization of the visit and the subsequent activities also make actions that delay visit progress dispreferred. So sequence-initiating actions were shown to be a primary resource for parents to resist no-problem (and thus no-treatment) diagnoses. In particular, this chapter outlined three primary types of sequence initiations that parents use, which increase in strength: newsmarks, inquiries about physical examination ﬁndings, and inquiries about diagnoses. Each of these makes relevant sequence closure and thus, at the very least, minimally delays the physician’s progress in the visit. Evidence supported an analysis of these inquiries as resistant: Parents frequently escalate from quite minimal requests for conﬁrmation to quite explicit challenges of the diagnosis if they encounter inadequate physician response to their initial resistance (e.g., failures to return to earlier examination ﬁndings or to a reconsideration of the diagnosis). Moreover, physicians routinely respond defensively to inquiries by justifying their diagnoses, accounting for them, and/or backtracking to history taking or physical examination ﬁndings. Diagnosis resistance was shown to be connected to the issues raised in chapter 1: legitimacy and treatability. At times, either or both of these issues may drive diagnosis resistance. On the one hand, as in the last case, parents can resist a diagnosis apparently in the service of obtaining antibiotic treatment. On the other hand, they may also resist for other reasons. Despite this, resistance appears to be treated by physicians as primarily about treatability and thus, even in cases where it was not intended to be “pressure” for antibiotic treatment, it may engender a shift in diagnosis or ﬁnal treatment outcome. This chapter documents a third phase in the visit, where negotiation of the diagnosis and treatment occur. Although some practices permeate different phases (e.g., the offering of particular diagnostic possibilities), other practices are related to the phase itself. In the reason for the visit phase, the practices had to do with how to answer a question. Question answering was also in play during the history-taking phase. By contrast, in the diagnosis phase, parents are not being offered any “slot” to speak. When they seize the opportunity by electing to speak after the diagnosis is offered (Sacks et al., 1974), they perform a rather assertive behavior, even though it is still quite indirect. Thus, although none of these behaviors is terribly frequent, and all are covert, we can nonetheless see a consistent pattern emerging: that parents have and make use of interactional resources at every step in the medical visit and thereby inﬂuence the diagnostic and treatment outcome of the visit.

5

Treatment Resistance

I

n previous chapters, the argument has been that antibiotic treatment is being negotiated through behaviors that are not directly connected to treatment. Early in the visit, the parent is offered an opportunity to suggest an initial direction that the physician should follow in the investigation of the problem, as well as in the subsequent diagnosis and treatment. If parents offer a candidate diagnosis that their child has, for example, an ear infection, this suggests that the physician investigate a trajectory of conﬁrming or disconﬁrming this diagnosis and offering treatment in line with that diagnosis. In the history-taking phase, parents rely on the physician’s questions to assess what diagnosis-treatment trajectory the physician is working toward and rely on the design of their responses to history-taking questions to push the physician away from that trajectory or toward an alternative trajectory. But there, too, we saw that this had implications for antibiotic treatment. In the last chapter, antibiotic treatment was observably being lobbied for through resistance to no-problem diagnoses. In most visits, parents are not offered an opportunity to simply state their treatment preference, and in the rare cases where physicians ask, parents seem taken aback and do not tend to respond, even if in other ways they appear to have preferences. Despite this, parents shape the treatment outcome through the behaviors we have seen thus far. But the phase where parents can most strongly affect their child’s treatment is during the treatment recommendation phase because, in contrast to the other phases, here what is being negotiated becomes relatively more explicit than elsewhere in the visit. The treatment recommendation phase of the visit is also critical because it represents the second and ﬁnal structurally provided-for opportunity for parents to inﬂuence the treatment decision. In this chapter, I will show that this is both because at this point the decision must be made and because, unlike diagnosis 105

106

PRESCRIBING UNDER PRESSURE

deliveries, both physicians and parents treat treatment recommendations as a domain of shared, though not equal, epistemic rights and responsibilities. In this chapter, I ﬁrst examine the types of responses that parents give to treatment recommendations and build the case that parents display that they have rights to accept the proposed treatment recommendation. Second, I demonstrate that, in the absence of parent acceptance, both physicians and parents display that mutual agreement is required before the activity of treatment recommendation can progress to closure: Physicians actively pursue parent acceptance, and parents who do not receive a concession shift from passive to active displays of resistance to the treatment recommendation until consensus is reached. Third, insofar as parent acceptance is required as a condition of closing the treatment recommendation activity, parent resistance can lead almost directly to concessions and modiﬁcations of the physician’s treatment recommendation.

Treatment Recommendations: Proposals That Make Relevant Acceptance Unlike diagnoses, treatment recommendations are oriented to by parents and physicians alike as proposals that normatively require parent acceptance for the physician to progress to the next phase of the visit—closing the encounter. (See Byrne & Long, 1976; Robinson, 2003; Waitzkin, 1991, for a discussion of acute medical encounter activity structure.) Evidence for this comes from multiple sources. We will discuss several types of evidence here. Parent Responses to Treatment Recommendations One type of evidence for the argument that treatment recommendations make conditionally relevant response is the way that parents respond to treatment recommendations. This is particularly striking when parent responses to treatment recommendations are compared with their responses to diagnosis deliveries. Broadly, both activities—diagnosis delivery and treatment recommendation—involve the physician imparting medical knowledge to the parent. For this reason, we might expect that they would be responded to by parents in rather similar ways. This is not the case. Instead, as discussed in chapter 4, parents and physicians alike treat diagnoses as within the physician’s domain of expertise, whereas they treat treatment recommendations as a domain of shared expertise: Parents typically respond to treatment recommendations, and acceptance is, as will be shown in later sections, oriented to as relevant. In this way, parents are treated as having an important role in the treatment decision. We can see the contrast between how parents respond to diagnoses and treatment recommendations in Extract 5.1. (5.1) 2002 (Dr. 6) 1 2

DOC:

.hhh Uh:m his- #-# lef:t:=h ea:r=h, is infected, -> (0.2)

TREATMENT RESISTANCE

3 4 5 6 7 8 9

107

DOC:

.h is bulging, has uh little pus in thuh -> ba:ck,=h DOC: -> Uh:m, an’ it’s re:d, DOC: .hh So he needs some antibiotics to treat tha:t, DAD: => Alright. DOC: Mka:y, so we’ll go ahead and treat- him:
Having just completed her examination of the child, the doctor here explains the child’s diagnosis (lines 1–5). Although the doctor comes to possible turn completion most notably at the end of line 1 but also at the end of line 4 and at the end of line 5, the parent does not respond. By contrast, after the physician offers her treatment recommendation in line 6, the father accepts this with “Alright.” immediately upon possible completion of that TCU, and this is particularly notable because the intonation on the two TCUs is quite similar. The physician then announces, as an outcome of this, that “we’ll go ahead and treat- him:” and then moving to what type of antibiotic to prescribe. A very similar acceptance token to that used in Extract 5.1 is shown in Extract 5.2. Although here the diagnosis is responded to, we can still see a big difference in the type of uptake the two announcements receive. The mother receipts the doctor’s diagnosis of an ear infection with “Mm:.” (line 3). This token offers only minimal acknowledgment of the diagnosis (Gardner, 1997). (5.2) 1183 (Dr. 1) 1 2 3 4 5 6

DOC:

Well I think what’s happened is is that she ha:s this: uh- (.) .h ear infection in her left ear?, MOM: [Mm:. DOC: -> [And we’ll put her on some medicine and she’ll [be ﬁne. MOM: [Okay.

The parent’s response to the treatment recommendation is “Okay.” (line 6). This token—particularly with ﬁnal intonation—accepts the doctor’s recommendation. The consistency with which acceptance is offered to treatment recommendations and the consistency with which minimal or no response is offered to diagnoses suggest that parents hear treatment recommendations to make uptake relevant, whereas they do not apparently hear this for diagnoses. In this case, the parent’s two different receipt tokens offered in close proximity provide good evidence that parents orient to diagnoses and treatment recommendations as actions that make relevant different sorts of responses. Parent Rights to Respond Parents not only respond to treatment recommendations regularly but also act as though they have a right to accept these recommendations through the way they

108

PRESCRIBING UNDER PRESSURE

design their responses. This provides a second type of evidence that treatment recommendations make acceptance relevant. We can see this exempliﬁed in Extract 5.3. Here, the parent’s ultimate acceptance of the treatment recommendation is a full form agreement: “Let’s do that.” (line 15). At line 1, the doctor offers his ﬁndings during the chest examination of the child as “uh little congested in his che:st,” and appears to be moving into the ﬁnal diagnosis of the patient with “Yeah I think-” (line 2). At this point, the mother takes issue with the physician’s mitigation of “congested” and asks a question about her son’s more severe morning congestion (lines 5/7). When the physician moves to his treatment recommendation (line 11), the recommendation is offered in an unequivocal manner with “we hafta”; however, the mother nonetheless displays her orientation to it as a proposal to be accepted or rejected in line 15 with “Let’s do that.” (5.3) 1120 (Dr. 1) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

DOC: DOC: MOM: DOC: MOM: BOY: MOM: DOC:

Well he sounds uh little congested in his che:st, Yeah I [think[Now it’s little:. [Yea:h. [.hh But why is it that- in thuh [morning= [#HUH huh# =it’s just [so:[Well that’s because a:ll drips down the back of his throat, MOM: Uhhh ((sigh))/(0.2) DOC: I think we hafta put him on >an antibiotic<. #Eh: he just has- with his ea:r, BOY: Duh! DOC: an’ he has like a bronchitis in his chest, MOM: -> Let’s do that. DOC: Bu:t=he’ll be ﬁ:ne.=h

It is not just that she treats the proposal as something to be accepted. The mother’s formulation “Let’s do that.” is also stronger than “Okay.” or “Alright.”. “Let’s” explicitly treats a decision about her child’s treatment as shared. Although acknowledgment tokens such as “Okay.” and “Alright.” accept the treatment (Heritage & Seﬁ, 1992), the design of acceptance turns provides evidence that parents orient to the relevance of their stance toward the treatment recommendation. Notably, the parent does not acknowledge the diagnoses of an ear infection (noted earlier and indexed here with “with his ea:r,”) and bronchitis, despite their being just prior to the parent’s turn. In fact, her response “Let’s do that.” addresses only the action of putting him on antibiotics.1 Although sometimes diagnostic evaluations are acknowledged with “Okay”, they are not routinely treated as proposals for acceptance or rejection. By contrast, treatment recommendations are routinely accepted with objects such as “Okay.” or “Alright.”; “Let’s do that.”; “That’s ﬁne.”; and assessments such as “Good”. 2 This

TREATMENT RESISTANCE

109

acceptance is treated by parents and physicians alike as relevant after the provision of the treatment recommendation. Physician Pursuits of Parent Acceptance That parents routinely accept physicians’ treatment recommendations but not diagnoses is one form of evidence that treatment is understood as a domain of joint responsibility and that parents participate in treatment decisions in a way that they do not participate in diagnosis deliveries. Further evidence lies in physicians’ pursuits of acceptance when none is forthcoming. If there is a normative constraint that makes parent acceptance of the treatment proposal relevant, then, in addition to active resistnace to the proposal, passive withholding of acceptance will also constitute resistance to the proposed treatment. Passive resistance was found by Heritage and Seﬁ (1992) in the context of community nurse visits to ﬁrst-time mothers. They showed that such resistance to health visitor advice involved “unmarked acknowledgments” such as “mm hm,” or “yeah,”. These objects, they argue, “do not acknowledge or accept that talk as advice” and thus “do not constitute an undertaking to follow the advice offered” (p. 395). And following these unmarked acknowledgments, health visitors commonly pursue fuller parent uptake. In the present data, we see an orientation to silence and unmarked acknowledgments as similarly withholding acceptance of the proposed treatment. First, when parent acceptance is not forthcoming following a physician’s treatment recommendation (i.e., “passive resistance”), physicians typically pursue the parent’s acceptance, treating it as noticeably absent. Moreover, they do not move out of treatment recommendation (i.e., they do not initiate activity closure). These behaviors offer evidence that physicians orient to parent acceptance as normatively required. Pursuit of parent acceptance takes several formats, including offering a rationale for the treatment recommendation, offering evidence for the underlying diagnosis, returning to the examination ﬁndings, and offering the parent a concessionary future action. An example of a physician pursuing parent acceptance when none is forthcoming is shown in Extract 5.4a. (5.4a) 2043 (Dr. 8) ((BRO is older brother)) 1 2 3 4 5 6 7 8 9 10 11 12

DOC: MOM: DOC: DOC: DOC:

MOM: DOC: DOC:

(Is) it’s not infect:e:d, [There’s- uhm no ﬂuid or= [Mm. =anything, .hh An’ his lungs are completely clea:r_ Uhm_ (0.5) An’ he’s not- breathing very fa:st, or har:d, Uhm_ .mlkh So I think he’s just on his road to recovery:_ he just needs: another_ .h prob’ly another week or so to get rid of thuh cou:gh completely, -> (°Mm hm,°) => Just lots of=ﬂui:ds, Uhm he was prob’ly uh little bit dizzy cuz he was:°he had fever and he prob’ly hadn’t drank enough

110 13 14 15 16 17 18

PRESCRIBING UNDER PRESSURE

=> at thuh ti:me, prob’ly.° DOC: => .h Uh:m_ .h So lots to dri::nk, DOC: and then uhm .mlk if he gets- - fever agai:n, (0.2) thou:gh uhm .h in thuh next two or three => day:s, .h uhm_ (0.2) we may need to see him ba:ck, DOC: in case he-n- does come down with something secondarily,

In line 1, the doctor offers a diagnostic evaluation that continues across lines 3–5. The parent minimally acknowledges this in line 2. At line 7, the doctor begins to detail her treatment recommendation: doing nothing for another week. The parent again offers only a minimal acknowledgment (line 9). In response, the physician expands her treatment recommendation by adding a second recommendation: ﬂuids. In contrast to her ﬁrst recommendation, this proposes something the parent can do. But here, too, the mother withholds acceptance. The physician next offers an account that supports her treatment recommendation to offer ﬂuids (line 10) by proposing that the symptom the parent reported as problematic was the result of dehydration, but the mother again passes on an opportunity to accept the proposal. The physician here pursues agreement by restating her treatment recommendation with “So lots to dri::nk,” (line 14). By redoing the treatment recommendation, she overtly renews the relevance of parent acceptance. When acceptance is, once again, not forthcoming, the physician slightly modiﬁes her proposal. Here, she suggests what the parent can do if the child fails to improve—the parent can bring the child back (line 17). Finally, in line 18 after the parent has, once again, passed on the opportunity to respond, the physician concedes that the boy may need different treatment in the future if he should “come down with something secondarily,”. Each of the physician’s moves works to elicit parent acceptance of the existing treatment recommendation and thus displays the physician’s orientation to the relevance of parent acceptance. Physicians pursue acceptance in a wide range of ways. In Extract 5.5a, the physician returns to earlier phases of the visit in order to pursue acceptance. The father withholds acceptance of the physician’s recommendation to just “watch i:t?”.3 (5.5a) P201 (Dr. 7) 1 2 3 4 5 6 7 8 9 10 11 12 13

DOC: Unfortunately like most viruses we have to watch i:t? DOC: -> .hh becau:se- you know- she (can)/(could) have uh fever:: for another few days, and nothin’ el:se. -> (.) DOC: and jus- an’ be ﬁ:ne, DOC: .hh Or else if she got uh fever an’ got wor:se, and: started limping actually at that time we’d probably need ‘er tuh come ba:ck, DOC: -> .hh But at this moment since there’s no swelli:ng?, or there’s no: .hh you know <nothing else, th-uh most important thing t’ do is tuh watch her. DOC: -> .hh So we’ve had a fe:w people right no:w that have had-

TREATMENT RESISTANCE

14 15 16

DAD:

111

uh few of our kids are having tlk .h fever:s, for a few days, and not much other symptom:s. So can she go to preschool now?

In this interaction, the physician has diagnosed the girl with a viral infection. In line 1, she suggests that the best course of action is “to watch i:t?”. The father neither acknowledges nor accepts this recommendation, despite even the rising intonation. The physician expands her recommendation in lines 2–3 with an account for it: that the girl could easily have no other symptoms. The father does not accept this either (line 4). The physician here shifts to a discussion of a future plan. Similar to the doctor in Extract 5.4a, in lines 6–8, the physician suggests when the parent could reasonably return for another medical evaluation. Here, too, the father withholds acceptance. In lines 9–10, the physician returns to her previous physical examination ﬁndings as further support for her diagnosis. Typically, when physicians retreat to previous activities, including restarting a verbal or physical examination or restating diagnostic ﬁndings, physicians then proceed again through the remaining activity phases back to treatment recommendation. This is similar to what we observed in chapter 4, and this action occurs here, too. Having retreated to diagnostic ﬁndings, the physician next restates her treatment recommendation to watch and see (lines 11–12). By restating the treatment recommendation, the physician—similar to the physician in Extract 5.4a—renews the relevance of the father’s acceptance. Yet here, too, none is forthcoming. In lines 13–15, the physician offers a more generic rationale for her diagnosis—that several other children are having similar symptoms. Still, there is no acceptance from the father, though he initiates another sequence with a question about whether the girl can return to school. Yet more types of pursuits can be observed in Extract 5.6. Here, we can see the physician pursuing by repetition, the use of lists, and the use of overt requests for acceptance. At this point in the encounter, the physician has completed an in-ofﬁce throat culture and is waiting for the results. She begins her treatment recommendation with suggestions that are irrespective of these culture results. Throughout this explanation, the parent says very little. At each single arrowed line, there is an opportunity for the parent to respond to the physician’s recommendation: Acceptance is a relevant action. But in each case, the parent does not offer acknowledgment, let alone acceptance. (5.6) 2020 (Dr. 6) 1 2 3 4 5 6 7 8

DOC: DOC:

#Mkay:::.# so::,=h (0.5) Tlk=.h Let’s see: what=thuh results of this i:s,=h while we’re waiting for tha:::t, DOC: .h So no matter what the result i:s, h she does ha:ve uh:m hh redness in ‘er throa:t, an’ looks like she has pharyngitis, <whether it’s from bacterial -> or from virus, DOC: -> .hh So:: uhm I want her to do mouthwashes?,

112 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49

PRESCRIBING UNDER PRESSURE

DOC: -> .h Gargling at ho:me?, DOC: -> Really deep gargling. (.) All the way back. => #Aghghghgh.# All thuh way back of thuh throat. okay:?, DOC: -> .hh Do it as many as- time as you can. (.) DOC: -> Three:_ four times uh day. Especially after eating. => Mkay, DOC: -> .h That clears it out an’ that makes it feel better. Mkay,=you can do it with salt water:, you can do it -> with Sco:pe, DOC: -> .hh whatever mouthwash: ﬂavor that she likes. DOC: -> .hh So lets do tha:t, DOC: => .hh Give ‘er uh soft die:t?, Mkay:, Don’t give her anything heavy, nothing oily:, -> French fries, (.) fried chicken_ hamburgers, DOC: => .hh Nothing spicy.=h for uh couple days. Okay:, DOC: .h Cuz it’s gonna hurt every time she swallows those -> kind uh stuff. DOC: -> .hh Let’s give ‘er lots of liquids at ho:me, (0.6) DOC: -> .hh Give ‘er: water, jui:ce, whatever she wants to drink.=h DOC: -> Ice cream is okay:, That will make her feel better:, DOC: -> .h Popsicles, (.) DOC: -> That makes you feel better, DOC: => .h Mkay:?, DOC: -> .h Maybe some mashed potatoe::s, you know -> (so)/(it’s uh) soft diet. as uh general. (.) DOC: => Yogur:t, things like that. Nkay:, DOC: -> .hh Uh:m_ and you’re just gonna have to rest. (.) DOC: You know?, (.) DOC: She’s gonna have to rest. MOM: Yeah.= DOC: =No more running arou:nd an’- (.) ya know staying -> up la:te, an’ things like that. DOC: .h You’re just gonna have=t’ take lots of na:ps, -> an’ re:st, throughout thuh weekend. DOC: => .h Mkay:, ((Doc moves to look at rapid strep culture))

The physician pursues acceptance of her recommendations for mouthwashes (line 8), a soft diet (line 21), liquids (line 27), and rest (line 39). We can see this in several ways. First, similar to Extract 5.4a, she provides accounts for her recommendations (e.g., lines 16, 25, 30, and 33). She also restates her treatment recommenda-

TREATMENT RESISTANCE

113

tions (e.g., lines 10–11, 35–36, 43, and 47–48). Third, she adds additional treatments (lines 21, 27, and 39). Fourth, she uses rising intonation at the end of TCUs, such as in lines 8, 9, and 21, which has been shown in other environments to be a resource for securing uptake (Sacks & Schegloff, 1979; Schegloff, 1996). That these locations were designedly in pursuit of acknowledgment can be seen, for example, in the doctor’s repeat of lines 8 and 9 in line 10 and the respeciﬁciation with “All the way back.” also in line 10. There is still further pursuit in line 11, ﬁrst with the demonstration of gargling and second with the redoing, yet again, of “All thuh way back of thuh throat.” and then with a more direct request for acceptance with “okay:?,” Similarly, through the physician’s use of three-part lists, she also hearably invites the parent’s uptake because these lists project completion and have been shown to be strongly designed for recipient uptake (Atkinson, 1984; Heritage & Greatbatch, 1986; Jefferson, 1990). For example, at the end of line 19, the doctor reaches the third item of her projected three-part list and thereby implicates conﬁrmation. A similar list is in line 29, but as before, the parent does not offer any uptake. The physician actively pursues the parent’s acceptance through other means. For example, in the double arrowed lines, the doctor pursues acceptance with various forms of “okay”. The physician also switches from addressing the mother to addressing the child (see lines 33 and 39). This change appears to be designed to elicit acceptance, even if that is from the child.4 And in line 41, the physician pursues a response with “You know?” But it is not until line 44, after multiple pursuits and a change in addressee back to the mother, that the mother even minimally agrees with the doctor’s treatment recommendation of rest. In this section, I have shown that physicians work diligently to elicit parent acceptance, if it is not forthcoming, before closing the activity of recommending treatment. We saw that their pursuits of acceptance include extending the activity with accounts, returning to prior activities such as diagnostic ﬁndings in support of the treatment recommendation, offering additional recommendations, pursuing acceptance with rising intonation, or, more explicitly, with variations on “Okay?”. A parent’s failure to accept constitutes withholding acceptance. Combined with the prior two sections, we now have substantial evidence that both parents and physicians treat the treatment recommendation as an activity that requires parent acceptance prior to moving forward with the visit. This is very different from the orientation shared by parents and physicians to the diagnosis delivery. This is important because it means that parents have additional resources for negotiating the treatment. Speciﬁcally, whereas with the diagnosis, resistance could only take the form of actively impeding the physician’s progress to treatment through the initiation of a new sequence, in the treatment recommendation, parents can impede the physician’s progress to visit closure through inaction. So when parents fail to accept the treatment, this can be analyzed as treatment resistance as well. We might best understand this resistant behavior as “passive treatment resistance” following Heritage and Seﬁ (1992). As we have already observed, when parents passively resist a treatment recommendation, physicians pursue acceptance. What we can also observe is that in pursuing acceptance, physicians act as though they are in a negotiation. Thus, passive resistance is yet another resource parents have for pressuring physicians for treat-

114

PRESCRIBING UNDER PRESSURE

ment. This argument relies on a normative structure of treatment recommendations to suggest that even “doing nothing” in a particular sequential environment can be a consequential form of participation and can affect treatment decisions. Further evidence of this is shown in the next section.

Active Resistance Part of the evidence that withholding acceptance is resistant is from the parent’s side. If physicians do not alter their treatment recommendation in the face of passive resistance, parents routinely shift from passive to “active” resistance. Active treatment resistance closely parallels diagnosis resistance because it includes an action that questions or challenges the physician’s treatment recommendation, including proposals of alternative treatments. These actions make relevant a response by physicians, and this feature differentiates active resistance from passive resistance. Heritage and Seﬁ (1992) found that this upgrading pattern was present in their advicegiving sequences as well. Sequences that included unmarked acknowledgments culminated in a “more overt expression of resistance” (p. 402).5 An example is shown in Extract 5.4b. We saw the ﬁrst component of the treatment recommendation activity in Extract 5.4a. Here, following the physician’s indication of what sort of symptoms would cause her to review the child’s case for treatment (lines 15–18), the parent shifts from passive to active resistance of the physician’s treatment recommendation. (5.4b) 2043 14 15 16 17 18 19 20 21 23 23 24 25 26 27 28 29 30 31 32

DOC:

.h Uh:m_ .h So lots to dri::nk, and then uhm .mlk if he gets- - fever agai:n, (0.2) thou:gh uhm .h in thuh next two or three day:s, .h uhm_ (0.2) we may need to see him ba:ck, in case he-n- does come down with something sec[ondarily, MOM: -> [(See c- cuz-) what I was worried about I [(would’ve)/(wouldn’t)= DOC: [Mm hm, MOM: =normally_ (0.9) DOC: m- Bring [him in, MOM: [interpreted [this as a co- ya know= DOC: [Mm hm, MOM: =uh thing that [would run its course but- (.) this= DOC: [Mm hm?, MOM: =guy had thuh same thing and wound up on antibiotics cuz he got an infection. MOM: .hh[h DOC: [Whe:[re. MOM: [How can I prevent that. from happening.

TREATMENT RESISTANCE

115

In this example, the mother is clearly concerned that her son’s condition will become worse and that he will need further treatment if not given treatment now. This concern is not articulated until the doctor has provided more and more details about her recommendations for future action—understandable as pursuing uptake from the mother. It is only at this juncture that the mother explains “what I was worried about” (lines 19–20). Framed in this way, her turn is formulated as an account: arguably an account for her prior passive resistance. The mother here actively resists the recommended treatment through the juxtaposition of her announcement that her other son is on antibiotics for an infection with her inquiry about how to prevent this child from having to deal with a similar infection. This announcement, and further the inquiry, challenges the physician’s suggested treatment of watchful waiting and ﬂuids. That parents consistently upgrade from passive to active resistance in the face of a failure by physicians to modify their treatment recommendation is evidence that passive resistance is, from the parents’ perspective as well, a form of treatment negotiation. Active resistance, then, is a much stronger form of treatment negotiation. In the following example, the mother once again began her resistance passively and then here upgrades to active resistance (lines 27–28). This case is particularly striking because the active resistance she moves to is an explicit inquiry about antibiotic treatment. Extract 5.5b follows Extract 5.5a shown earlier. The physician is, in line 25, returning to her previous ﬁndings, having just responded to a parent question. (5.5b) P201 ((8 lines omitted following 5.5a)) 25 26 27 28 29 30 31

DOC:

She: doesn’t have anything right no:w, any symptoms of mucus or vomiti[ng that’s contagious. DAD: => [Are you gonna give her ana- antibiotics? DOC: Yeah- uh No: I don’t have anything tuh treat right now for antibiotics. Her ears look really goo:d, .hh she has no sign of bacterial infection right no:w?,

The resistant inquiry: “Are you gonna give her ana- antibiotics?” makes relevant an answer, but physicians typically respond to such inquiries by not only responding to the question but also treating the actions as lobbying for antibiotics. In this way, the father’s question rather directly challenges the physician’s own recommendation. (We will discuss this further in chapter 6.)

Antibiotic Negotiation The social norm that treatment recommendations require parent acceptance has the consequence that active or passive resistance of a treatment recommendation puts the physician in a position of either working to “convince” a parent to accept the proposed treatment recommendation or offering the parent concessions—either pos-

116

PRESCRIBING UNDER PRESSURE

sible or actual. In this way, treatment resistance can be seen as yet another resource for initiating a negotiation of the treatment decision. This is the most overt type of negotiating behavior we have seen, because through treatment resistance, parents take a position against the treatment they are being offered rather than merely urging physicians toward or away from a particular diagnostic or treatment trajectory. Parents usually resist over-the-counter, nonantibiotic treatment plans. Across the whole of these data, there were very few cases where parents could be understood to be resisting antibiotics. In most cases, the parent’s position is not fully on the surface of the interaction. However, it may be best exempliﬁed by looking at an initial interaction where it is brought to the surface of the interaction. Here, after the physician offers his position against antibiotics in line 4, the father resists (lines 6/10/12/14/17–18/20/23/25/27). (5.7) 322803 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

DOC: DAD: DOC: DAD: DOC: DAD: DOC: DAD: DOC: DAD: DOC: DAD:

->

-> -> -> ->

DOC: DAD: -> -> DOC: DAD: -> DOC: DOC: DAD: -> INF: DAD: -> DOC: DAD: -> DOC: DAD: DOC: DAD

I th:ink from what you’ve told me (0.2) that this is pro:bably .h uh kind of (0.2) virus infec[tion, [Uh huh, (0.4) th:at I don’t think antibiotics will ki:ll, (0.2) Well[Thee other[( ) >Go=ahead_< Yeah. .hh ( ) I had it- I had thuh symp[toms [I understand. Three weeks ago. [Right. [.hh An:d I’ve been taking thuh over the counter cough [( ) [(Good_) Uh s- ( ) coughing syrup, Nothing take away .hh Especially my sor- my [th- my throat was real= [Mm hm, =sore [for (awhile- et- that) w:eek. [Uh huh, °Right,° an:d (.) I start taking thuh antibiotic (0.5) eh he ((cry)) Yesterday. Right, And it (.) seemed to take care of the problem. [(Well) that’s why we’re doin’ a throat [culture. [( ) [Yeah. [is TUH SEE if they need antibiotics. [( ) Yeah yeah.

TREATMENT RESISTANCE

32 33 34 35 36 37 38 39 40 41 42 43 44 45

DOC DAD DOC: => => INF: => DAD

DOC DAD: DOC: DAD

117

(0.2) Cause
In lines 23/25/27, the father states that antibiotics solved his own illness, thereby implying that they would be helpful for his two sons who are ill with “the same thing” (as he mentioned earlier in the encounter). In response, the doctor ﬁrst explains that this is a possibility, which is why he performed a throat culture to test for the strep bacteria. But in lines 35–36 and 38, the doctor then goes on to offer to give the antibiotics against his medical judgment if the parent insists. I want to note three things about this case: First, the physician overtly treats the father’s participation, indeed his acceptance of the nonantibiotic recommendation, as both relevant and important. Second, he treats the father’s narrative about his own experience with antibiotics as applying pressure for antibiotics. In fact, he treats it as just one step short of “insistence.” We can see this when he says “if you (.) absolutely insist” (line 35), where he treats absolute insistence as having not yet occurred. However, as a condition that is being discussed, the physician further conveys his understanding that this is the behavioral trajectory the father has been on. Third, the physician here overtly acknowledges the impact of parent pressure: If the parent continues to lobby for antibiotics, he will provide them in spite of the fact that they would, in his opinion, be ineffective and thus inappropriate. More typically, the negotiation of treatment between parents and physicians is not brought to the interactional surface, but parent resistance can nonetheless be seen to initiate a negotiation, as evidenced by physician responses. For instance, we can return to a case shown earlier in Extract 5.5a. In the previous section, we observed that the father here shifts (line 27–28) from passive to active resistance. Now I would like to focus on how the physician responds to this shift. (5.5c) ((8 lines omitted following 5.5a)) 25 26 27 28 29

DOC:

She: doesn’t have anything right no:w, any symptoms of mucus or vomiti[ng that’s contagious. DAD: => [Are you gonna give her ana- antibiotsics? DOC: Yeah- uh No: I don’t have anything tuh treat right now

118 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

PRESCRIBING UNDER PRESSURE

-> for antibiotics. DOC: -> Her ears look really goo:d, DOC: -> .hh she has no sign of bacterial infection right no:w?, DOC: -> .tlkh and that’s (what) she’d get antibiotics fo:r. DOC: .hh So uh lotta times you can start out with uh virus like uh co:ld, (.) .h an:d if you- it goes on for uh while #uh:# bacteria (should) set in you can get uh secondary -> bacteria infection? and that’s when you need antibiotics. DOC: .hh #But- y-# otherwi:se: since she doesn’t have any source of an antibi- of uh bacterial infection?, -> that=uh=we just watch her. (0.3) DOC: right now. in too:. DOC: -> .h And they’re not draining or any[thing. DAD: => [It just means that=> ya know if she gets another fever we hafta bring her ba:ck, DOC: .hh Well what I’ll do is she might still get uh fever: in thuh next couple uh #da:ys.# because: .h that’s th’ way viruses wor:k?, you can have- you know (how have=you) if you have uh co:ld, you can get a fever for uh few da:ys? .hh And tha:t Since she’s o:lder:, .h if something’s #uh# she (would) com[plain ( ) thuh symptom,= GIR: [( ). DOC: =then she would need tuh come back. DOC: .hh But what you ca:n do i:[sGIR: [Guess what.

At the beginning of this segment in the interaction, the father has not yet accepted the physician’s proposal of no prescription treatment, and in overlap with the physician’s reassertion that there is nothing really problematic wrong with his daughter, the father asks about antibiotics—an alternative treatment proposal and a strong form of resistance (other examples of overt negotiation were shown in Extract 5.4b and 5.7). In response, the physician ﬁrst answers the question (line 29). She then goes on to account for her answer, stating that there is nothing to treat at this point (lines 29–30). The father does not offer acceptance. Next, similar to how physicians respond to passive resistance, the physician restates a physical examination ﬁnding (line 31). Following no uptake again, the physician restates her diagnosis of no bacterial infection (line 32). The father still does not accept, and the physician expands her treatment recommendation, further ruling out the need for antibiotics (line 33). Again, there is no acceptance. At this point, the physician shifts to a scenario where treatment would be warranted. In this way, she intimates that she may make a concession in the future. However, there is still no parent acceptance, and in lines 38–40, the physician restates her treatment recommendation.

TREATMENT RESISTANCE

119

Following 0.3 seconds of silence, the physician returns to restate additional physical examination ﬁndings (lines 42–43) in the face of no parent acceptance. This is further expanded in line 44. Here, the father again actively resists (lines 45–47). Up to this point in the encounter, the physician has been working to secure parent acceptance of the current treatment recommendation to just watch the girl for a bit longer with the intimation that if things changed, she would be willing to treat her with antibiotics. However, she has not achieved parent acceptance, and in fact, her work has been met with increasingly stronger parent resistance. Here, the physician frames her response as a concession with “Well what I’ll do is.” This does not reach completion before the girl initiates a sequence that the physician takes up. After the physician closes the sequence with the girl, she returns to offering a contingency plan (Mangione-Smith et al., 2001)—a concession to the parent—that he could call rather than coming back in (lines 65–70). The physician also intimates (in line 70 with “talk to us and see:_”) that the physician might be able or willing to take another course of action over the phone or as a result of the phone call, further suggesting concession to the father’s pressure. But she also maintains her stance in favor of the current treatment recommendation to watch the girl. (5.5d) ((6 lines of conversation with GIRL not shown)) 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85

DOC: -> DOC: -> DOC: DOC: -> DOC: -> DOC:

-> DOC: -> -> DAD:

.hh Uhm: mlk but usually what you can do is if over thuh next few days she might still get a fever of (a hundred an’ two) you can give us a call if you’re concer:ned. .h And if it goes on more than tha:t, .hh she might need tuh come in #but:# .h most uh thuh time you can just uhm call us and talk to us and see:_= =If she has uh new symptom, breathing difﬁculty:?, .hh if she had ear draina:ge, if she s:- did start tuh limp, then we would say she does need tuh come in. .hh But for uh child her a:ge, you c’n get fever for uh few day:s, an:’ as long as she looks this goo:d, an’ no other symptoms, .hh ya know we just- we’ll watch her. So like if she got uh fever this afternoo:n that doesn’t mean she needs tuh come in right away:_ What I would do is like you did: this morning, .hh give ‘er some Tyleno:l, If she .h looks great like thi:s then it’sshe’s probably still just ﬁghting off thuh virus. (0.3) Mka:y:? (0.8) °Mkay.°

In lines 71–73, the physician suggests that only certain circumstances would require the parent to return to the ofﬁce. The parent still does not accept the treatment proposal. In lines 74–76, the physician restates her treatment recommendation that “we just- we’ll watch her.” However, this restatement still does not engender acceptance.

120

PRESCRIBING UNDER PRESSURE

Following this, she also redoes her treatment proposal to include a recommendation of something the parent can do to be more proactive in lines 79–806 and once again restates her diagnosis. This action is met with 0.3 seconds of silence. At this point, the physician overtly pursues acceptance with a heavily question-intoned “Mka:y:?” and, after a substantial delay, receives a quiet but minimal acceptance from the father (line 85). This case provides evidence that when physicians face treatment resistance from parents, they orient to this as initiating a negotiation of the treatment recommendation and work to secure parent acceptance. Although this physician did not ultimately modify her treatment recommendation from no antibiotics to antibiotics, she could nonetheless be seen to be making concessions to the parent, including suggesting that the parent call if the child got worse and suggesting that Tylenol would work to bring down the fever. Therefore, he would not need to return to the ofﬁce. The negotiation of antibiotics can also be observed quite clearly in Extracts 5.8a–c. Here, as part of her diagnosis, the doctor denies sinusitis7 (lines 1–2) and then moves into her treatment recommendation beginning in line 4. The mother endorses the doctor’s recommendations ﬁrst by inquiring about a decongestant that is in the general category of Sudafed, which was recommended by the doctor (line 8). (5.8a) 2015 (Dr. 9) 1 2 3 4 5 6 7 8 9 10

DOC:

^Ya know, I probably_ (0.5) wouldn’t call it sinusitis right now. (0.5) DOC: Uhm- h- What I would do: is keep up with thee uhm h over thuh counter- you know maybe like children’s Sudafe:d or something like that to help with thuh: thuh congestion in her nose. MOM: -> [Now shu- we should (be giving) her uh deconges[tant. DOC: [.hhh [hhh DOC: Yiea:h, I think that would probably help. ((35 lines of talk about different types of decongestant not shown))

46 47 48 49 50 51 52 53 54 55 56 57 58

DOC: MOM: DOC: DOC:

MOM: DOC: MOM:

.hh Uh- but certainly if thuh fever goes up higher than just this low grade ninety nine or uh hundred. [Okay, [ . h h Uh:m tlk (.) this doesn’t seem to be going away. =[YOU’RE NOT SEEING it on her throat or anything. (.) ((Possible head shake by DOC))

TREATMENT RESISTANCE

59 60 61 62 63 64 65 66

MOM: DOC: MOM: DOC: MOM: DOC: MOM: DOC:

121

Oka[y. [Uh uh. [Okay. [ .hh Uhm .h but: if ^it’s persisting: into next week. Ok[ay. [I mean_ Then I think we should see her ba:ck. [Okay. [I mean- and think about °sinusitis.°

Following the parents’ endorsement of the proposed treatment, the physician appeared to complete the treatment recommendation and to have moved into plans for future action: a move into closings.8 This can be seen here in line 46–47/49–53. However, in line 54, the mother offers a slightly premature “Right.” that may be working toward speaker transition (similar to “Yeah” as discussed by Jefferson, 1983), and then immediately upon possible grammatical completion, the mother initiates a turn of active resistance (line 54). Here, she requests conﬁrmation of the physician’s previous diagnostic ﬁndings, thus asking the physician to retreat into an earlier activity (something we have seen in previous cases). In overlap, the physician shifts from her previous straightforward plan for future action to competitively addressing the mother as resistant. The competitiveness of the physician’s talk is shown in that it is substantially louder than her other talk. Additionally, her second TCU “an’ I’M NOT SEEING ANYTHING.” restates her examination ﬁndings. Although we cannot conﬁrm whether the parent responds visibly (e.g., with a nod or facial expression), after a micropause, the parent vocally accepts this with “Okay.” The physician returns to a future plan of when to see the girl back and further addresses the parent’s resistant action by stating that she might consider “sinusitis” at that time (line 66). Following this, the doctor offers another type of concession to the parent—that they could consider doing an x-ray of the sinuses. But rather than resulting in full acceptance, the mother’s resistance to the current line of diagnosis and treatment escalates. Although she accepts the physician’s position of not liking to x-ray children (line 71), she then goes on to offer a brief narrative about her older daughter, who was also without the classic sinusitis symptom of heavy nasal drainage but who apparently had a severe infection. (5.8b) 67 68 69 70 71 72 73 74 75

DOC:

DOC: MOM: DOC: DOC: DOC:

.hh We can always get- just uh plain x ray: of thuh sinuses. An’ sometimes that’s helpful whenon these equivocal things. get an x ray.< She’s not really tender, .hh [over her sinuses either.

122 76 77 78 79 80 81 82 83 84 85

PRESCRIBING UNDER PRESSURE

MOM: MOM: -> -> DOC: MOM: -> DOC: MOM: -> DOC: MOM: -> DOC:

[Yeah. My older daughter- I brought her in: This was two years ago an’ they s[aid “well we don’t think it-” [Uh huh, We- she had no:ne uh thuh symptoms of sinusitis. [Uh huh:, [(ever.) An’ she had headaches for uh yea:r? Yeah. An’ when they ﬁnally x rayed her she was [totally blocked. [just socked in,

The mother’s narrative is, like the previous examples of treatment resistance, positioned late in the counseling phase of the encounter and, furthermore, is positioned after several recommendations which the mother has endorsed. The mother’s narrative conveys her position that she would like an x-ray of her daughter’s sinuses because in the past the x-ray revealed sinusitis with her older daughter. Like the situation the mother is in at this point in the encounter, in her narrative she relates doctors telling her “we don’t think it-” (line 78), which appears on its way to denying sinusitis (precisely what this physician has done earlier in the encounter). Although this position is embedded in her narrative, it is nonetheless conveyed. This type of resistance is primarily focused on the diagnosis and ways to detect sinusitis. So unlike some of the other types of active treatment resistance we have seen, here a symptomatic treatment recommendation is being resisted through a challenge to the underlying diagnosis. In 5.8c, the doctor responds by dealing with both the mother’s position in favor of an x-ray and her use of this as a vehicle for resisting a lack of prescription treatment. First, the doctor agrees with the mother’s narrative as plausible (line 86). She then provides an account for a lack of drainage and having a sinus infection with “it’s so blocked” (line 90). However, she also asserts that the decongestants may allow drainage to begin. The doctor’s turn effectively disagrees with the mother’s position in favor of an x-ray. But her opposition is embedded. She provides a condition under which she does recommend x-rays, and the contrast is built through the use of the additional modal “will,” along with its contrastive stress (line 88). Additionally, the doctor focuses her turn on how the treatment she has recommended may help: It may help allow the drainage to begin, if in fact the mother is right about her daughter’s condition. So this is another way that the physician displays her understanding that she and the parent are still in negotiation of the ﬁnal treatment decision. (5.8c) 86 DOC: => Yeah:. 87 DOC: => .h That’s why especially- in kids (>who’re<) complaining 88 => of headaches an’ things like that I will get an x ray. 89 MOM: Right.

TREATMENT RESISTANCE

90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110

123

DOC: => ^Because uh lotta times it’s so blocked ‘at n(h)othing’s => dr(h)aining [ou(h)t. MOM: [Ri:ght. [Right. DOC: => [ . h h And what you may ﬁnd (.) => Her nose doesn’t look particularly swollen er anything => inside but you may ﬁnd if you- consistently give ‘er => thuh decongestants for uh couple days, .hh that it opens => up: thuh passages t’ thuh sinuses. DOC: => .hh An:’ [an’ you’ll start se[eing thuh stuff coming ou:t. MOM: [( ) [seeing it. MOM: [Okay. DOC: => [ . h h An’ then- I mean I would: at this point if she => comes back in on Monday h=er- or Tuesday and stuff=> I would have prob’ly uh lower- since it’s been: (.) => continuing for all [this time_ .h]h to start her on= MOM: [Right. right.] DOC: => =some antibiot[ics. MOM: [Right. DOC: => <But I hate to- tuh put her on if she doesn’t really => nee:d it. MOM: -> ^Okay. That’s ﬁne.

The doctor also offers a possible concession in that she states that she would be willing to start the girl on antibiotics if the condition were to persist into the next week (lines 101–104/106). The doctor’s response also addresses the mother’s action as resisting her treatment recommendation. Speciﬁcally, she again denies the need for antibiotics at this time (lines 108–109). Having previously outlined her treatment recommendation and having moved from treatment recommendation into recommendations for other future action, such as when to bring the girl back to the ofﬁce, this action is speciﬁcally a return to her treatment recommendation. As such, it is hearably responsive to the parent’s resistance. The mother then accepts the doctor’s decision with “^Okay. That’s ﬁne.” (line 110). This acceptance is the most full acceptance provided throughout this phase of the encounter. Although “^Okay.” alone might have been equivocal as a move to accept and close the sequence (especially given a context where “okay” has been used repeatedly at various junctures in the discussion), “That’s ﬁne.” is much stronger as an acceptance of the doctor’s position, and further as taking a position of closing the sequence. This example again shows an elaborate negotiation of the treatment recommendation, including two concessions by the physician: (1) the offer to do an x-ray to conﬁrm the mother’s concern that it is sinusitis and (2) the physician’s offer of antibiotics if the condition is not better by the following Monday or Tuesday. The latter concession especially appears to work, and the mother shifts from “Okay” and “Right”—acknowledgments that had been offered previously to a fuller form acceptance: “^Okay. That’s ﬁne.” (line 110).

124

PRESCRIBING UNDER PRESSURE

Treatment Change as a Response to Resistance As with diagnosis resistance, the most extreme form of physician concession is to alter the treatment recommendation from no antibiotics to antibiotics. Although this is relatively rare, that it happens at all provides strong support for the power of treatment resistance as an interactional resource for negotiating treatment outcomes and, more generally, the orientation to parent acceptance of treatment proposals as required. An instance is shown in Extracts 5.9a–d. In lines 1–2 of 5.9a, the physician recommends against antibiotics, but the parent does not accept. The physician expands her treatment recommendation against antibiotics in line 3 with an increment (Schegloff, 2001). The parent does not accept this either. The physician then afﬁrmatively states that she would like to treat the girl’s eyes and give her “some decongestant” (lines 4–5). She provides a rationale for that recommendation in lines 6–7. The parent continues with passive resistance and then initiates active resistance during the 68 lines of talk that I have not shown here. (5.9a) 2019 (Dr. 6) 1 2 3 4 5 6 7

DOC: -> DOC: -> DOC: -> -> DOC:

.hh So: uh:m a- at this time I don’t wanta commit ‘er to: antibiotics. Like two weeks, or three weeks, or whatever:? .h I thi:nk I’ll go ahead and treat her for the eye:s?, an’ I wanta give her some decongestant. So that would, suck out all that, um, secretions?=

((68 lines including passive and active resistance not shown)) 76 77 78 79 80 81 82 83 84 85 86 87

MOM: DOC: MOM: DOC: MOM: DOC: MOM: DOC: DOC: MOM: DOC:

But anyway she’s had low-grade temp [(an’ uhm), [Mm hm. (1.1) just really hasn’t been hersel:f. It’s- it’s- It’s:= =M[m hm. [(ya know)/(even) more than: uhm (1.5) thee eye thi:ng. Uh huh:,
In lines 76/78, she asserts that her child simply is not “well”—that she is sick. The implicit claim appears to be that her daughter is “sicker” than the doctor’s treatment recommendation would suggest. This is pursued further and more explicitly in line 80, where she says “it’s more than thee eye thi:ng.”. This case further evidences that treatability can be separated from concerns of visit legitimacy: Here, the eye infection provides visit legitimacy but does not deal with the parent’s concern for a solu-

TREATMENT RESISTANCE

125

tion to the child’s other symptoms and her desire for antibiotics. In lines 82–83/86, the mother suggests that normally she is very “troubles resistant” (i.e., not a mother who rushes her child to the doctor). Although this is often associated with threats to visit legitimacy, here the implication is that her daughter’s illness is more serious than the doctor’s treatment recommendation would suggest and appears to be more of a means to address treatability. We can see the next component of the interaction in Extract 5.9b. Here, the physician begins a turn that is more concessionary in its design. She ﬁrst agrees with the parent with “Yeah” (line 87) and then with “I mean: if you wa:nt ya know-”, which begins to frame her forthcoming response as a concession to what the parent wants. Note that the parent has not yet stated anything that she wants or expects explicitly, but she has (1) passively resisted the physician’s treatment recommendation by failing to accept it and (2) actively resisted the treatment recommendation by implying that her child is sicker than the doctor is prepared to recognize. (5.9b) 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107

MOM: DOC: MOM: DOC: DOC: MOM: DOC:

DOC: -> DOC: -> -> DOC: -> -> DOC: -> ->

[Cuz it’s such a big deal to come here [( ) [Yea:h,=h I mean: if you wa:nt ya know- I mean she looks.= =Can I at least have thuh prescription an’ I’ll decide whether or not to ﬁll it, i[n a couple day:s, [.tlk For the antibiotics[:? [Ye[ah. [Uh::m_ I really don’t like to do tha:t, because: I mean .hh She doesn’t look: like she has sinusitis:. Ya know?, (.) Uhm, if you really wanta be su:re we can go ahead and take: x rays to make su:re if it’s really opacify:, .hh cause unnecessary treatment for sinusitis: she can get resistant to uh lot of those antibiotics?, uh lot of those bugs. I mean. .hh An:d it’s- it’s not really good for her:. (1.0) So:: we try to minimi:ze ya know- treatment until it’s really necessary. (.)

The concessionary frame is abandoned in favor of a less concessionary “I mean she looks.” that, given the no-problem physical examination that preceded this discussion, is likely to be heard as headed for another no-problem evaluation. This would be inconsistent with prescribing antibiotics. At this point, the mother’s strongest form of treatment resistance comes: an overt request for antibiotics in lines 89–90. The mother’s request not only calls into question the treatment recommended so far but

126

PRESCRIBING UNDER PRESSURE

also speciﬁcally challenges the physician’s assertion earlier in Extract 5.9a that she does not want to commit the girl to antibiotics at this point. The mother’s request “Can I at least have thuh prescription” orients to the prescription as a minimal form of action and implies that it is signiﬁcantly less than actually treating the child. This is accomplished largely with “at least”. The second unit of her turn “an’ I’ll decide whether or not to ﬁll it in a couple day:s,” claims some measure of discretion (i.e., that she would not immediately ﬁll the prescription and give her child antibiotics), as well as claiming that she would have the knowledge to determine whether and when to ﬁll the prescription. The doctor denies her request in lines 94–96 but offers the parent a concession: They could perform an x-ray that would potentially clarify whether the child should appropriately be treated for sinusitis. In addition, the physician cites the inappropriateness of treating this condition with antibiotics and the general need to avoid inappropriate prescribing as an account for her recommendation against antibiotics. The mother fails to accept either the physician’s rejection of antibiotics or the concession. At each arrowed line, the mother passes on an opportunity to accept the physician’s recommendation. The mother continues to actively resist across the next stretch of interaction. Here, after the doctor again returns to outline a situation in which she would concede and prescribe antibiotics (if the girl “looks really -ba:d,”), the mother asserts that her daughter never looks bad (lines 110/112). She then claims that her daughter is not herself, thus implying, again, that her daughter is sicker than the physician is recognizing and further justifying her seeking of antibiotics. (5.9c) 105 106 107 108 109 110 111 112 113 114

DOC:

DOC: MOM: DOC: MOM: DOC: MOM:

So:: we try to minimi:ze ya know- treatment until it’s really necessary. (.) You know of course if she’s s- you know looks really -ba:d, [then I’ll go ahead. [(see she ne-) she never looks: ba:d. I mean [she can be really [Mm hm:, sick and she never looksMm hm[:, [You know: I’ve taken her in here with:

((20 lines not shown: examples of girl not acting sick)) 135 136 137 138 139 140 141 142 143

MOM: MOM: DOC: MOM: DOC: MOM: DOC: MOM:

[And plus it’s her (t=her:) uhm (0.6) tlk (0.4) Uh:hm_ (0.5) °What’m I tryin’ t’ say:_° Emotionally. (I [mean she’s been) .hh (0.8) t- you know more ‘n more= [Mm hm:, =tire:[d, [Mm [hm:, [And more ‘n mo:re (.) upset easily_ [an’ stuff: [Mm hm, over thuh past couple weeks, [an’ it’s- it’s just been building=

TREATMENT RESISTANCE

144 DOC: 145 MOM: 146 DOC:

127

[Mm hm:, =an’ building an’ bui[lding. [Mm hm.

Throughout this chapter, I have argued that both parents and physicians are oriented to the treatment recommendation as an activity requiring agreement between the parent and the physician. We have seen a variety of evidence for this claim. In this case, we have seen the physician pursue agreement in several ways, including offering accounts for her treatment recommendations and offering an alternative course of action—the x-ray. Throughout, the parent has been unyielding in her dissent, ﬁrst passively resisting and ultimately overtly requesting an alternative type of treatment. In the ﬁnal extract that I will show, the physician works to close the activity after what is now over 150 lines of discussion of the treatment recommendation. Note that if the mother had agreed readily to the treatment following the recommendation shown in Extract 5.9a, this activity might have closed within just a few lines. At this point, the physician offers yet another concession—a willingness to talk to the girl’s regular physician (lines 149; 152–153; 160–161). (5.9d) 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165

MOM: DOC: DOC: MOM: DOC: MOM: DOC:

MOM: DOC: DOC:

DOC:

DOC: MOM: MOM:

=an’ building an’ bui[lding. [Mm hm. .tlkhh Who: usually sees her. Doctor Hilton. .hh Uh:m lemme call him an’ see what he uhm says.= =Oh is h[e around (today?) [Okay? I don’t know if he’s arou:nd but I’ll=lemmme try to call him. .hh because: uh:m_ He’s not [( ). [Tlk I really don’t want to treat ‘er. (0.5) Uhm but then I’ve only seen her ﬁrst time. This is my ﬁrst time seeing her so I really don’t know how she (.) you know i:s, .hh So let me call ‘im an’ see: what he sugge:st, .h An’ the:n we’ll go from there. (.) [Does that sound okay? [°Okay.° Sure, if you [can (reach) him £it sounds great.£

Even here, after proposing to call the child’s regular doctor, the mother resists when the physician reraises her treatment recommendation in line 155. The mother still does not accept this (line 156). But when she proposes, as an alternative, that she

128

PRESCRIBING UNDER PRESSURE

will “see what he suggest,” in line 159 and make a decision at that point (line 160), even though the mother does not accept this immediately, she does offer acceptance in line 163. The doctor pursues more explicit acceptance in line 163 with “Does that sound okay?” Then the mother more fully accepts (albeit provisionally) in line 165 with “Sure, if you can (reach) him £it sounds great.£”. Ultimately, the physician cannot reach the girl’s regular doctor, and she ends up prescribing for the girl, despite having diagnosed only conjunctivitis, having explicitly rejected a sinusitis diagnosis, and having repeatedly expressed a desire not to treat the girl with antibiotics (mentioned again in line 155 here). Similar to other concessions that physicians offer, this one is offered at a point when the parent has both passively and actively resisted the proposed treatment. In this case, the physician worked to convince the parent of a nonantibiotic treatment recommendation but was entirely unsuccessful. Despite the physician’s strong position against prescribing, she is pressured through normative constraint that she must secure parent acceptance of the treatment recommendation in order to close this activity. When that is not forthcoming, the physician alters her recommendation to obtain the required acceptance. Although this case offers a rather extreme and overt example of the negotiation process and the possible outcome such a process can yield, that the outcome is negotiated is, as we have seen, not so unusual. In fact, there is no case in these data where a parent fails to accept the treatment and a physician nonetheless proceeds to activity and visit closure. Treatment resistance, like other behaviors we have looked at in this book, is not terribly common. Across analyses of both the Seaside and Metro data, parents resist 19% of nonantibiotic treatment recommendations (Mangione-Smith, Elliott, Stivers, McDonald, & Heritage, 2006; Stivers, Mangione-Smith, Elliott, McDonald, & Heritage, 2003). But as we have observed, qualitatively we can see that this behavior can have an important inﬂuence on the diagnostic and treatment outcome of the visit. The reason appears to be that the behavior, like others we have looked at, places pressure on the physician for antibiotics and physicians respond to that pressure in various ways. At times, they defend themselves, but at other times they make concessions or even alter their treatment recommendation. Quantitative analyses also support this link. Parents are signiﬁcantly more likely to resist the diagnosis when the diagnosis is viral (21% vs. 6%; p <.001) (Stivers et al., 2003). And physicians were signiﬁcantly more likely to perceive parents as expecting antibiotics if they resisted their treatment recommendation (MangioneSmith et al., 2006; Stivers et al., 2003). In an analysis of the Metro data, physicians were 24% more likely to perceive parents as expecting antibiotics if they resisted than if they did not resist the treatment recommendation. Although any single behavior we have looked at may be present less than 10% of the time, what we have been able to see is that there are many opportunities for parents to inﬂuence physicians’ diagnoses and treatment recommendations. In the Seaside data, there was some form of negotiation in a full 62% of total cases. In the Metro data, there was some form of negotiation in 42% of cases.9 A substantial number of cases in the data involve interactions where the illness will have been pronounced bacterial and treatable early on in the visit. Thus, a rate of 40–60% is quite substantial. Additionally, most behaviors were not associated with each other. Across

TREATMENT RESISTANCE

129

both the Seaside and Metro data, only diagnosis resistance and treatment resistance were associated with each other (p <.01). Thus in general, these behaviors do not represent the same parents escalating across the visit. However, within a particular type of resistance, we can observe the same parents escalating (e.g., during diagnosis or treatment resistance as shown in chapters 4 and 5). At the same time, though, we see that parents are rather reticent in their use of such behaviors. We might not expect to see this, given that unlike the diagnosis or history-taking phase, parents are normatively required to be part of the treatment decision by accepting it. However, even in this context, parents remain largely covert in their resistance and very rarely make any overt statements about their treatment preferences. Their inﬂuence is much more by pushing physicians away from a nonantibiotic form of treatment.

Discussion This chapter has shown that unlike diagnoses or any other activity in the medical visit, treatment recommendations are oriented to by doctors and parents alike as a domain of shared responsibility and epistemic rights. This was evidenced ﬁrst by the regularity with which parents accept treatment recommendations; second, through the observation that physicians pursue parent acceptance, if it is not forthcoming, prior to initiating closure of the medical visit; and third, through the observation that parents typically move from passive resistance to active resistance if physicians do not alter their treatment recommendation. Parents can exploit this normative structure to affect the treatment decision or outcome of the visit because by failing to accept the physician’s recommendation, they effectively block the physician from being able to close the visit. When they actively resist the treatment recommendation, this blocks the physician from closing even more extremely because physicians must also deal with the initiation of new sequences, much like after diagnosis resistance, except that physicians must not only close these sequences but also secure parent acceptance: at minimum, a two-step process. When parents resist a nonantibiotic treatment, physicians systematically treat this as indexing a desire for antibiotics. They treat parents as having been in search of antibiotics, and thus as “not getting what they want.” Oftentimes, this appears to be what is at issue for the parents, as in Extract 5.9. But the more general issues of treatability and of legitimacy may also be involved. As was the case with most of the other behaviors, treatment resistance was not associated with parent reports of an expectation for antibiotics (Mangione-Smith et al., 2006; Stivers et al., 2003). Based on both this evidence and what we see interactionally, it appears that although parents who resist a nonantibiotic treatment recommendation are often resisting speciﬁcally to lobby for antibiotics, they may also be resisting what represents a denial of a solution to their problem: a child who cannot sleep, a child who cannot go to school, a day care provider who will not accept the child until treated, or a mother who needs to return to work but is at home with the child until he or she is better. Thus, some parents may be resisting the lack of their child’s treatability and not only the lack of antibiotics. Another possibility is that some parents who resist a nonantibiotic treatment

130

PRESCRIBING UNDER PRESSURE

recommendation may be resisting the delegitimization of their visit—effectively a judgment against their decision to seek medical care. And as discussed in previous chapters, these issues can be conﬂated: A doctor’s judgment that a child is not in need of treatment may imply that there was no need to “waste time” on a medical visit. It is often virtually impossible to tease apart these issues in real-time interaction. But physicians often fail to address either of the latter dimensions while focusing nearly exclusively on the former issue. This chapter showed several instances of more overt forms of negotiation. Although this behavior is relatively rare, it does sometimes occur. When it does, it is yet another way that parents inﬂuence the visit. In general, such overt negotiation happens through one of the other practices and therefore represents only a very strong form of that behavior (e.g., a strong form of treatment resistance). In the next chapter, we will examine overt forms of negotiation and how physicians respond to them.

6

Overt Forms of Negotiation

T

his book has thus far been devoted to showing how parents play a role in the diagnostic and treatment outcomes of their children’s medical visits through covert communication practices. But as we have seen in the discussion of practices of mentioning additional symptoms, diagnosis resistance, and treatment resistance, parents do occasionally talk about antibiotics overtly. When they do this, they are quite explicitly lobbying physicians to prescribe antibiotics, whereas when they perform these behaviors without speciﬁcally mentioning antibiotics, they lobby physicians more implicitly. In this chapter, I examine four main ways that parents overtly lobby for antibiotics and show evidence that parents and physicians alike treat antibiotics as under negotiation. Unlike the other practices that are largely conﬁned to a single phase of the interaction, overt lobbying for antibiotics occurs in virtually all phases of the medical encounter.

Background Although overt parent lobbying for antibiotic treatment is unusual, it does occur, and by examining cases in which it is involved, we gain useful insights into the parentphysician negotiation process that hold across cases where the negotiation is entirely covert. In Extract 6.1, the parent and physician have already been in negotiation over antibiotics. The parent actually already has an antibiotic prescription from another physician for a prior illness. But that physician recommended against ﬁlling the prescription unless the child signiﬁcantly worsened. Here again, the physician has indicated that the child is basically okay, but the parent continues to bring up problems with the child. Apparently, she is looking not just for a prescription (which she has) 131

132

PRESCRIBING UNDER PRESSURE

but for the physician’s support of her using the medication. I show this case here because it offers evidence that physicians commonly ﬁnd themselves in opposition to parents on the topic of antibiotics, as conveyed by his statement that “NOW you’re on my si(h)de.” (6.1) 151205 1 2 3 4 5 6 7 8 9 10 11 12 13

DOC: DOC: MOM: DOC: MOM: MOM: DOC: MOM: DOC: DOC:

Well she- she could have the same germ, (0.4) I mean_ °Okay° (0.2) But=uhm (.) she’s ﬁghtin’ it off pretty well_ [(°I’d say.°) [Okay. Well I just don’t like to give antibiotics if we don’t have to.= =Okay well good. [°just=uh:m° Well I HAVE- I= [huh th=huh =NOW you’re on my si(h)de.=huh

In this case, as noted earlier, the parent had just resisted the treatment recommendation by the physician and raised problematic symptoms. At line 1, the physician is attempting to address these issues and implying that no treatment is recommended with “But=uhm (.) she’s ﬁghtin’ it off pretty well_” (line 6). In these data, it is rare that parents indicate that they dislike or do not want antibiotics (But compare Britten, Stevenson, Gafaranga, Barry, & Bradley, 2004). Usually when such a statement is made, it is actually still part of a negotiation, as in “I hate to use antibiotics but. . .” thereby underscoring the unique need for them in the present illness context (see Extract 6.8). The parent’s position with respect to antibiotics here appears to be very much this, because she has continued to press for endorsement of antibiotics. Here, the physician allows insight into the battle he is engaged in by noting that “NOW you’re on my side” (in contrast with earlier). This statement reveals precisely what we see through a close analysis of covert cases: that physicians commonly feel that they are in battle with parents over issues of antibiotic prescribing. Thus, the argument throughout this book, that parents and physicians are frequently in tacit negotiation over antibiotics, is completely transparent in this case. Although parents do push physicians, it is not the case that they do so relentlessly. Rather, they show considerable self-restraint. Earlier, I argued that this restraint reﬂects their own orientation to the domains of diagnosis and treatment outcomes as primarily the physician’s domain of authority and that pressure intrudes into this domain. In fact, there is an even more fundamental way in which pressure is problematic because it represents a threat to the physician’s negative face: his or her freedom of action (Brown & Levinson, 1987). In the world of acute medicine, the physician is typically allowed (and typically takes) the lead in what should be done for the patient, even in these more modern days of consumer medicine. When parents

OVERT FORMS OF NEOGTIATION

133

bring a child to the physician, they come as the layperson to the expert and imply an acceptance of these roles (Freidson, 1970a, 1970b; Parsons, 1951; Starr, 1982). When they advocate for a particular outcome, they are vulnerable to being sanctioned for violating this implicit agreement about doctor-patient roles. Moreover, they may worry about the consequences of being understood by doctors to be intruding on their medical expertise and authority in relation to prescribing decisions. We can see these issues emerge in Extract 6.2. Here, the child collaboratively completes his father’s problem presentation by adding a candidate diagnosis (line 4). The father downgrades his epistemic authority to make such a diagnosis across line 6 with “I believe that it: might be I don’ know.” In spite of this, the physician’s question “Now where did you:- guys have that diagnosed at.” treats the parent as out of line for having ventured a diagnosis on his own if it had not been veriﬁed by a health practitioner. (6.2) 313 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

DOC:

=I’m sorry tuh interrupt. Now: we can go ahead(.)/(-) (1.0) DAD: Uhm, (.) On Friday (they)/(we)/(he) had uh ( -) uhm PT1: (ear infection.) ((running water as DOC washes his hands)) DAD: (ehb ah- I-) I believe that it: might be I don’ know. DOC: [An ear infection? DAD: [( ) DAD: Uh huh(.)/(,) (.) DAD: So: th’t [that’s what I want tuh make sure that his= DOC: [Kay:, DAD: =ear is okay. DOC: O[ka:y, DAD: [He was c- complaining of pain an:- .hh #uh:m# he (couldn’t m- uh:=sleep,) #u# b’cuz uh thuh pain. DOC: (That) was on Friday? DAD: That was (from)/(on) Friday. DOC: -> Now where did you:- guys have that diagnosed at. (1.0) DAD: Uh: hh We don’t have (it)/(-) diagnosed that it’s infection [(we just-) he just was complaining (uhr-) DOC: [Oh:. Okay. DOC: =of pain.

Although physicians often let candidate diagnoses pass without explicit comment or sanction, parents nonetheless treat these actions as vulnerable to physicians’ sanctions for treading into their territory. In short, parents and physicians alike treat any type of movement into the domain of medicine (whether diagnostic or treatment related) to be somewhat of an intrusion into the doctor’s territory. It is likely for

134

PRESCRIBING UNDER PRESSURE

this reason that parents only rarely and in special circumstances overtly advocate for antibiotic treatment. We might expect that although we see a general respect for physician expertise in the covert behaviors, once parents are prepared to lobby for antibiotics overtly, such deference would vanish. But this is not the case. In the next section, we examine evidence that, even in cases involving overt lobbying, parents remain reticent to enter into the physician’s domain of authority.

Evidence for Parent Reticence Perhaps the most obvious evidence for the claim that parents are reticent to overtly lobby for antibiotics is the infrequency of this type of behavior. Only 8% of the cases in the Metro and Seaside data involved any direct advocacy for antibiotics at all. And even when this occurs, it is not as direct as we might have envisioned. I have no cases of a parent saying, “I came to get some antibiotics” or “I would like you to prescribe antibiotics.” The design of turns that overtly lobby is quite different from this, as we will see. There is a second type of evidence that parents are reticent to overtly lobby for antibiotics. When they raise the question of antibiotics, parents often cite nonmedical circumstances as justiﬁcation for competing with the physician’s medical authority— a trip out of town, a birthday party, the inconvenience of a return visit, or an upcoming holiday during which the child’s condition could worsen. Parents also routinely appeal to other forms of life experience—other friends or family being treated in a similar way, past experience with a similar illness, and the like. Such appeals respect the physician’s medical expertise through their restriction to nonmedical accounts for their behavior. As an example, we can look at Extract 6.3, in which a mother appeals to the fact that her son is having his ﬁfth birthday party shortly and she wants him to have antibiotics so that he will be well. (6.3) 1035 (Dr. 2) 1 2 3 4 5 6 7 8 9 10 11 12 13 14

MOM: -> -> DOC: MOM: -> ->

[I’m looking for: uh uh I’m looking for a miracle from you. Okay::, heh heh Martin has: uh his very ﬁr:st major: (.) ﬁve year old birthday party tomorrow, (1.0) MOM: -> And his temperature’s been:=hh UH hundred an FI:VE, ((wail)) DOC: His birthday or is [(he) going tuh huh hah hah hah hah MOM: -> [His birthday. Tomorrow:. And he’s so: -> ^sick and I NEED UH MIRACLE! [hhha ha huh huh: DOC: [ A h : : : [: . MOM: [£I need a miracle.

OVERT FORMS OF NEOGTIATION

15 16 17 18

MOM: -> -> -> ->

135

.hh I’m h- .hh I know he probably just has thuh common cold but I’m like praying he has a horrible bacterial infection in his ears and YOU’RE GONNA C:URE IT WITH TWO DOSES OF ANTIBIOTIC [ha ha ha ha ha ha

Later, I will look at this interaction more closely. Here, I want to point out only that the parent is treating her behavior as special, and she licenses it with her son’s birthday. Although this case is quite extreme in the degree to which the mother treads into the physician’s medical territory, even here we can see that the mother orients to her behavior as problematic. In this case, the mother’s appeal is still indirect as evidenced by her use of “need” (lines 11 and 13) and “praying” (line 16) and her respect for the decision as still resting with the physician (“I’m looking for a miracle from you.” in lines 1–2). Third, parents generally do not advocate for antibiotics directly. Rather, they treat the topic as delicate either by talking about it indirectly or by mitigating their lobbying actions. Just as we observed cautiousness in the way parents offer candidate diagnoses, so, too, can we see cautiousness when parents overtly lobby for treatment. We discussed this with respect to Extract 6.3, and we can see this again in Extract 6.4 in an even more overt case of lobbying. (6.4) 161303 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

DOC:

DOC: MOM: -> DOC: MOM: -> -> DOC: MOM: -> -> -> DOC: MOM: DOC: DOC: MOM:

Uh: the virus irritates the big tubes and there’s a lot of mucus. (0.5) [So: [Because in Oct^ober he ca:me and I believe it was also for= =Something similar li[ke this or:_ [Yeah: and he got- he was on antibiotics. [Yeah:. [He- He got on antibiotics an’ .hh you know uh:m I just feel that antibiotics sort of .hh -:make him feel better: - i- in a shorter [time
In this case, the mother topicalizes antibiotics in the context of a brief narrative about when her son was on antibiotics previously. She states rather strongly that she feels they “make him feel better: - i- in a shorter time” (lines 11–12). Although it is strongly worded, the turn is hedged with “I just feel” and “that antibiotics sort of

136

PRESCRIBING UNDER PRESSURE

. . . make him. . .” as well as with markers of hesitation (the inbreaths, cutoffs, and self-repairs). So, her turn design and thus her preference for antibiotics are mitigated because she does not baldly state what she wants. Across the covert practices we examined, we saw physicians treating antibiotics as desired by parents, but this is rarely brought to the surface of the interaction (compare Extract 5.7). A rare instance is shown in Extract 6.5. In this case, two children are being treated. The boy is receiving prescription treatment but not for antibiotics. The girl is receiving antibiotics. The physician had previously stated that the girl would get antibiotics and had told the father that the son needed clear ﬂuids but said nothing about antibiotics, nor did she say what kind of antibiotics the girl would receive. Parents sometimes refer to antibiotics by indirectly referring to the “pink medicine” (a reference to the bubble gum–ﬂavored amoxicillin liquid commonly used as a ﬁrst-line antibiotic in the treatment of childhood illnesses). What is interesting is how the physician designs her treatment recommendation for the girl. (6.5) 231910 1 2 3 4 5 6 7 8

DAD: (Wha’s) he getting any=uh=that pink medicine, that [antibiotic, DOC: [I’m gonna give him thuh white one. huh? DAD: Thuh white one? DOC: Yeah. DAD: And for Leslie?, shoul’ sh- she have that pink one?, (0.5) DOC: Yeah Leslie can take thuh pink one.

In response to the father’s inquiry about antibiotic treatment for his son, the physician responds with a full form answer addressing only the color dimension but not the underlying question about antibiotics. The father advocates for the “pink one” again with his daughter, and this time the physician agrees. She says, “Yeah Leslie can take thuh pink one.” With the use of the modal “can,” the physician underscores her action as allowing the child to take the medication and thus to her role as a gatekeeper of a desirable commodity. Across the cases in this section, I have shown three things: (1) Antibiotics are desired by parents, (2) antibiotics are a negotiable commodity, but (3) parents treat lobbying for antibiotics (covertly or overtly) as treading into physician territory. The remainder of the chapter examines the speciﬁc practices through which overt negotiation is normally done.

Practices for Overtly Negotiating Antibiotics Across the data, we observe four main communication practices to overtly lobby for antibiotics: (1) direct requests for antibiotic treatment, (2) statements of desire for antibiotic treatment, (3) inquiries about antibiotic treatment, and (4) mentions of past experience with antibiotic treatment. As was mentioned earlier, even when

OVERT FORMS OF NEOGTIATION

137

table 6.1. Directness in Overt Lobbying for Antibiotics Requires Physician Response +

Requires No Physician Response –

Overtly Favors Antibiotics

Direct Requests

Statements of Desire

Does Not Overtly Favor Antibiotics

Inquiries about Antibiotics

Mentions of Past Experience with Antibiotics

parents overtly lobby for antibiotics, there are varying degrees of directness. This is exempliﬁed in table 6.1, which shows the two main indices of directness evident in these four practices. Actions that both overtly adopt a stance in favor of antibiotics and require a physician’s response are the most direct form of overt lobbying—direct requests. Actions that neither overtly favor antibiotics nor require a physician’s response are least direct though nonetheless explicit—mentions of past experience with antibiotics. Statements of desire and inquiries about antibiotics fall closer to a midpoint in their directness because statements of desire take a position in favor of antibiotics but do not require a physician’s response; conversely, inquiries about antibiotics do not overtly take a position in favor of antibiotics but do require a physician’s response. In general, the more direct forms of communication are the least frequent. In what follows, we will examine each of these types of communication behaviors in some detail. Requests for Antibiotics One way that parents initiate a negotiation of antibiotic treatment is to request it. As noted earlier, this is the most direct form of overt pressure observed in these data. It both straightforwardly indicates a parent’s preference for antibiotics and obligates the physician to respond to the request. However, it is also the least frequent practice, occurring in only two cases in the entire corpus of 882 visits. The example shown in Extract 6.6 was previously shown as an instance of treatment resistance in chapter 5 because it is positioned following a treatment recommendation. I will not go through a full analysis of this case because it was analyzed earlier, but what is most striking about this case for the present purpose is (1) that a clear request is made at all given the scarcity of such requests in the data and (2) that although the mother does make such a request, she still orients to this action as an intrusion into the physician’s domain of expertise. At this point in the visit, the doctor had already diagnosed the girl with conjunctivitis. The mother has resisted this diagnosis by inquiring about whether the treatment would be similar to the way one would treat “sinuses” (apparently indexing sinusitis—a bacterial condition typically treated with antibiotics). At the end of the physician’s response to the mother’s question, the physician states that she does not want to give the girl antibiotics (lines 1–3) but wants to treat her eyes and give her a decongestant (lines 4–5). In this environment (i.e., shortly following the physician’s

138

PRESCRIBING UNDER PRESSURE

recommendation for an alternative type of treatment) the mother explicitly raises antibiotic treatment as a possibility. (6.6) 2019 (Dr. 6) 1 2 3 4 5 6 7

DOC: DOC: DOC: DOC:

.hh So: uh:m a- at this time I don’t wanta commit ‘er to: antibiotics. Like two weeks, or three weeks, or whatever:? .h I thi:nk I’ll go ahead and treat her for the eye:s?, an’ I wanta give her some decongestant. Uh:m, .hh- Ya know i- She doesn’t look like uh:m (.) Why don’t we go ahead and try the decongestant ﬁrst.

((26 lines of discussion of treatment not shown)) 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55

MOM:

=Can I at least have thuh prescription an’ I’ll decide -> whether or not to ﬁll it in a couple day:s, DOC: For the antibiotics? MOM: Ye[ah. DOC: [Uh::m_ I really don’t like to do tha:t, because: I mean .hh She doesn’t look like she has sinusitis:. Ya know?, Uhm, if you really wanta be su:re we can go ahead and take: x rays to make su:re if it’s really opacify:, cause unnecessary treatment for sinusitis: she can get resistant to uh lot of those antibiotics? a lot of those bugs. I mean. DOC: .hh An:d it’s- it’s not really good for her:. (1.0) DOC: So:: we try to minimi:ze ya know- treatment until it’s really necessary.

)

The mother’s request is clear. The format “Can I . . . have . . .” makes conditionally relevant a granting or denial. Although the format is direct on one level, the request is for “thuh prescription” and thus neither names the drug nor the class of treatment (i.e., antibiotics) directly. By discriminating between a prescription and the drug itself, the request formulation treats the prescription as a more minimal type of treatment outcome than she would ideally have. With “at least”, she treats this as less than a request for antibiotics and goes on to indicate that she is not actually requesting antibiotics at this moment but would like a prescription that she could ﬁll when necessary. Thus, she maintains some degree of circumspection, even here.

OVERT FORMS OF NEOGTIATION

139

More transparently than in cases of covert communication behaviors, in this extract there is evidence that both the parent and the physician are oriented to the prescription of antibiotics as negotiable. First, the mother requests a prescription after the physician has recommended against antibiotics (lines 1–2). By doing this at all, she treats the prescribing decision as something that she can inﬂuence, rather than something that is decided on by the physician and that she must accept without discussion. Although the physician denies the mother’s request, there is still evidence that she, too, orients to the prescription as negotiable: The physician does not deny the request outright but states a preference against the requested course of action (“I really don’t like to do tha:t,”). This does not yet state that she will not but states her preference, thereby treating this as a discussion and a negotiation of a ﬁnal decision about the girl’s treatment. Furthermore, she provides an account for this preference that appears designed to enlist the agreement and support of the parent “because: I mean .hh She doesn’t look like she has sinusitis:. Ya know?,” (lines 46–47). Finally, she offers the parent an alternative course of action—an x-ray (line 47–49)—to determine if the girl in fact has a sinus blockage. This course of action was not previously recommended but is offered here as a compromise—a way to investigate the condition that would mandate the desired antibiotic treatment. By offering this procedure as an alternative (explicitly marked as such with “if you really wanta be su:re”), the physician treats the decision on a course of action as one that is being made jointly by the parent and physician rather than one that is handed down from the physician to the parent. By adding “if you really wanta be su:re,” the physician overtly indexes her response as acquiescing to the parent’s pressure. Although requests are rare, that they happen at all is another excellent piece of evidence for the orientation by parents and physicians alike to the negotiability of treatment and of antibiotics in particular. This case also demonstrates that even in very extreme cases like this, parents still orient to norms of behavior that give physicians primary responsibility for and authority over treatment decisions. Stating a Desire for Antibiotics Parents more frequently state a desire for antibiotics than ask for them, but statements of desire are also quite rare. They are relatively direct because they convey pressure by overtly indicating a treatment preference. Unlike requests, though, they do not make a granting or denial by the physician conditionally relevant. For this reason, statements of desire for antibiotics are somewhat less direct than requests. Despite this, physicians treat them as applying signiﬁcant pressure for antibiotics (much like requests do). We can see this illustrated in Extract 6.7, part of which was shown in Extract 6.2. Here, the mother asserts that she is hoping for antibiotics very early in the encounter. (6.7) 1035 (Dr. 2) 1 2

DOC: MOM:

Are we ready::. Hi: Doctor Sa:[nders,

140 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

PRESCRIBING UNDER PRESSURE

DOC: DOC: MOM: DOC: MOM: DOC: MOM: DOC: MOM: -> -> DOC: MOM: -> -> MOM: -> DOC: MOM: -> -> DOC: MOM: MOM: -> -> -> -> DOC: MOM: MOM: DOC: MOM: DOC: MOM: DOC: DOC:

[W- well hi:. How are you guy::s.=h .h[h [Well:We have two victims he:[:re. [Two victims. We added one when I picked [Sara up from schoo:l. [£That’s oka:y?,£ Actually: we wouldn’t be here Doctor Sanders except=h You’re ta[king the trip. huh huh huh [I’m looking for: uh uh I’m looking for a miracle from you. Okay::, heh heh Martin has: uh his very ﬁr:st major: (.) ﬁve year old birthday party tomorrow, (1.0) And his temperature’s been:=hh UH hundred an FI:VE, ((wail)) His birthday or is [(he) going tuh huh hah hah hah hah [His birthday. Tomorrow:. And he’s so: ^sick and I NEED UH MIRACLE! [hhha ha huh huh: [ A h : : : [: . [£I need a miracle. .hh I’m h- .hh I know he probably just has thuh common cold but I’m like praying he has a horrible bacterial infection in his ears and YOU’RE GONNA C:URE IT WITH TWO DOSES OF ANTIBIOTIC [ha ha ha ha ha ha ha ha. [Ha ha ha ha ha ha ha ha £.hh But I know it’s not gonna h(h)appe(h)n. .hh £#But I just said (.) you know what, .hh (.) Before I cancel this party, (1.0) .hh Tlk=[hahhh. [#Yeah:,# I- I- i- - hh. I- I’m afraid I can’tAhh[h h a h a ha h a ] hih hih hih(s). [d(h)o th(h)uh mir(h)acle.] #I mean I’d ah# ‘cause he- I think he’s gonna probably come down with the ﬂu:, ya know?,

The mother states that she is “praying he has a horrible bacterial infection in his ears and YOU’RE GONNA C:URE IT WITH TWO DOSES OF ANTIBIOTIC” (lines 27–29). That the mother is “praying” for this outcome strongly communicates her desire for antibiotics. But she differentiates between what she wants and the facts of her son’s illness. This is one way in which she maintains respect for the physician’s authority in the matter. Her desire for a bacterial diagnosis and antibiotics is couched as something that has yet to be decided (with “YOU’RE GONNA”). Additionally, by

OVERT FORMS OF NEOGTIATION

141

prefacing her preferred diagnosis (“a horrible bacterial infection”) with a more likely alternative diagnosis, “he probably just has thuh common cold” (lines 26–27), she treats her desired outcome as unlikely. This recognizes both that her son may not be able to be effectively treated with antibiotics and that she understands this. With her subsequent admission “But I know it’s not gonna h(h)appe(h)n.” (line 31), she treats her desire for a bacterial diagnosis and antibiotics as being in the doctor’s hands, which is, in turn, based in facts. But she nonetheless communicates her position as one strongly in favor of antibiotics. The parent’s animated talk also conveys her orientation to the delicacy of her action. Haakana (2001) shows that laughter is commonly used when patients report misdeeds or otherwise engage in delicate actions. Here, the misdeed appears to be the very strong pressure she is applying for antibiotic treatment, rather than allowing the physician the space to arrive at his own recommendation. Besides laughter, which inﬁltrates much of this interaction, there are also large shifts in amplitude (e.g., 29– 31) and large pitch ﬂuctuations (e.g., lines 21–22). These may, like laughter, convey the parent’s orientation to her actions as problematic. The parent also displays an understanding of the differences between viral and bacterial diagnosis and the fact that bacterial diagnoses are treatable with antibiotics. This understanding is not uncommon among parents, especially in the Seaside data from which this extract comes. In response, the doctor both denies the treatment and rejects the diagnosis being suggested. First, he says, “I’m afraid I can’t- d(h)o th(h)uh mir(h)acle.” This matches the affective tone of the parent’s talk and is ﬁtted to her earlier “I NEED UH MIRACLE!” (line 22). The declination of ability to “do the miracle” conveys the doctor’s position as unable to cure the illness. But we can also observe that the physician treats “miracle” here not so much as indexing antibiotics as a miracle drug but as a biblical miracle. By shifting his orientation to “miracle”, the physician avoids a head-on conﬂict with the parent while simultaneously conveying that the boy does not have an ear infection and that he therefore will not prescribe the desired two doses of antibiotics. That this is implicit in this turn is supported by the next turn, where the doctor suggests, “I think he’s gonna probably come down with the ﬂu:, ya know?,” (lines 39–40). Here the doctor implies that the mother’s prefatory diagnosis is correct and suggests that antibiotics would be inappropriate. This implicitly rejects the mother’s preferred diagnosis. The doctor also works to communicate this diagnosis as, so far as possible, afﬁliated with the mother. This is done both affectively and by offering this position in close proximity to the mother’s own downplaying of the likelihood of the “miracle” to occur. Although there is an effort to acknowledge the ﬁnal decision as the doctor’s, the mother nonetheless clearly states her preference, and the physician addresses that. That this appears to be working to pressure the physician is most apparent across lines 15–16/18/21–22, where the mother offers her account for wanting her son to have an ear infection and wanting antibiotic treatment—that he is having a birthday party. This case also illustrates that in the very opening of the visit, even prior to a move by the physician to establish the reason for the visit, parents can initiate a negotiation of antibiotics. As was mentioned at the outset of this chapter, overt negotiation behaviors are unlike the covert behaviors we examined in chapters 2–5 because

142

PRESCRIBING UNDER PRESSURE

they are not restricted to particular phases within the visit. Statements of a desire for antibiotics can occur at virtually any point in the visit, though they are more likely to occur, as most overt forms of negotiation do, later in the visit. We can see an example of this in Extract 6.8. The mother has already resisted the physician’s treatment recommendation and resists it again in lines 4/6. With the “and” preface, she builds continued resistance in lines 6–8 with a statement that superﬁcially opposes antibiotics but in this case appears to be building a contrast between not liking antibiotics generally but X. The physician works to afﬁliate with the dimension of the parent’s resistance that is opposed to antibiotics in building his defense, stating again that antibiotics are unnecessary and then, as we have seen is common in responding to treatment resistance, expanding on the diagnosis. This time he details the prognosis of the illness as he has seen it with other children. (6.8) 161303 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

DOC:

MOM: DOC: MOM: DOC: MOM: DOC: DOC:

MOM: DOC:

MOM: DOC: MOM: DOC:

DOC: MOM: DOC: MOM: DOC: MOM:

Uh: and ^usually they do just ﬁne with: just a good (.) decongestant cough medicine. (0.2) Well I’ve been [giving him a (.) decongestant= [(°and uh:°) =[cough medicine
OVERT FORMS OF NEOGTIATION

32 33 34 35 36 37 38 39

DOC: MOM: DOC: DOC: MOM:

143

I just feel that antibiotics sort of .hh -:make him feel better: - i- in a shorter [time
Following this, the mother again resists, using a form we will see discussed later on: mentioning past experience with antibiotics. Here, I want to focus on the mother’s comment that “I just feel that antibiotics sort of .hh -:make him feel better: - i- in a shorter time” (lines 32–33). This is a clear statement in favor of antibiotics when the parent and physician are negotiating what kind of treatment is to be given. On the surface of this interaction is also an account for why the mother wants antibiotics: because she thinks they will help her son feel better sooner. So we gain insight into why parents may pressure physicians for antibiotics in situations that are clinically inappropriate. In addition, we have another instance of a parent who crosses the implicit agreement between physicians and parents/patients to respect the physician’s cultural authority and role as medical expert. This section discussed a second practice through which parents overtly lobby physicians for antibiotics. Statements of desire for antibiotics were shown to be, on the one hand, less direct than requests for antibiotics, but on the other hand, they are nonetheless quite a strong form of pressure because they make the parent’s preference so explicit. We also observed that, once again, even in these fairly high-pressure environments, parents are oriented to their actions as treading into medical territory and show this through downgrading and mitigating their statements. Inquiring about Antibiotics One of the more common ways that parents lobby for antibiotics involves inquiring about them. Parents who ask about antibiotics—explicitly introducing them into the interaction—invite discussion of them and require a physician to respond to this treatment possibility. In all of these ways, such inquiries come to exert pressure on the physician. Inquiries occur in all phases of the visit. As an illustration of this, consider one rather unlikely place: during the history-taking phase. It is an unlikely place because, as we saw in chapter 3, this activity is generally comprised of physician questions and parent responses. But even during history taking, such inquiries occur. (6.9) 221814 1 2 3 4

DOC: MOM:

Any vomiting or diarrhea?, (0.2) Uhm no. (0.8)

144 5 6 7 8 9 10 11 12 13 14 15

PRESCRIBING UNDER PRESSURE

MOM: DOC: DOC: MOM: DOC: MOM: DOC: MOM:

She has been [like picking at her ea:r though. [(°Yeah okay_°) Okay. (2.0) Do you think she needs antibiotics?= =Let me see. (0.5) #Okay.# °Yeah:,° (0.5) Has she have any serious medical problems since birth. O^h: n^o:. nothing like that- Nothing at all.

The physician is in the middle of taking the girl’s history at line 1. The mother answers the question and adds an additional problematic symptom about the child’s ear. Following the physician’s acknowledgment of this, the mother inquires about antibiotics (line 9). The physician defers answering with “Let me see” (line 10), and then following acceptance of that (line 12), he proceeds with history taking (line 13). Inquiries such as this one are hearable as lobbying for antibiotics, even though they do not overtly take a position in favor of antibiotics, because they convey what the parent is thinking about. They are typically treated as revealing a parent’s expectation, hope, or desire, even though this is not stated. This becomes more transparent when the inquiries are positioned, as they most commonly are, following a treatment recommendation or at a point in the encounter where a nonantibiotic treatment recommendation has been implied (e.g., if the child had ear pain and the physician indicates throughout the examination that there is no evidence of any infection) or stated. Because of this positioning, these inquiries can be understood as resistant to the actual or implied treatment proposal. An example was shown in Extract 5.5, part of which is reproduced here in Extract 6.10. The physician has explained to the father that his daughter has a viral infection and has recommended “watching” it. This recommendation is essentially a no-treatment recommendation. After answering several questions related to when the child can return to school and other treatment, the physician returns to the child’s symptoms and treatment. In line 4, the father inquires as to whether the doctor is going to prescribe antibiotics. The question is designed in a way that shows that the father is oriented to this as the doctor’s decision. But following the doctor’s explanation, the father, at least retroactively, shows his question to have also been an indication of his own preference, with the account being that they would “hafta bring her ba:ck,” (line 21–22). (6.10) P201 (Dr. 7) 1 2 3 4 5 6

DOC:

She: doesn’t have anything right no:w, any symptoms of mucus or vomiti[ng that’s contagious. DAD: => [Are you gonna give her ana- antibiotics? DOC: Yeah- uh No: I don’t have anything tuh treat right now for antibiotics. Her ears look really goo:d, .hh she has

OVERT FORMS OF NEOGTIATION

7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

145

no sign of bacterial infection right no:w?, .tlkh and that’s (what) she’d get antibiotics fo:r. .hh So uh lotta times you can start out with uh virus like uh co:ld, (.) .h an:d if you- it goes on for uh while #uh:# bacteria (should) set in you can get uh secondary bacteria infection? and that’s when you need antibiotics. DOC: .hh #But- y-# otherwi:se: since she doesn’t have any source of an antibi- of uh bacterial infection?, that=uh=we just watch her. (0.3) DOC: right now. [It just means that=> ya know if she gets another fever we hafta bring her => ba:ck, DOC: .hh Well what I’ll do is she might still get uh fever: in thuh next couple uh #da:ys.# because: .h that’s th’ way viruses wor:k?, you can have- you know (how have=you) if you have uh co:ld, you can get a fever for uh few da:ys? DOC: .hh And tha:t Since she’s o:lder:, .h if something’s #uh# she (would) com[plain ( ) thuh symptom,= GIR: [( ). DOC: =then she would need tuh come back. DOC: .hh But what you ca:n do i:[sGIR: [Guess what. DOC: DOC:

At the beginning of this extract, the father has not yet accepted the physician’s proposal of over-the-counter (i.e., nonprescription) treatment with a token such as “Okay.”, “Alright.”, or “That’s ﬁne.” as is typically the case when parents agree to physician proposals of treatment. This may already be passively resisting the recommended treatment (Heritage & Seﬁ, 1992). However, in asking about antibiotics, the parent shifts to actively displaying his position in favor of antibiotic treatment. As mentioned earlier, raising antibiotics at this point in the interaction suggests resistance to the alternative nonantibiotic treatment plan that was suggested. Second, this inquiry is positioned in what appears to be midturn during the physician’s explanation of the child’s symptoms (line 4). In this sense, the father treats a no-problem explanation (lines 1–2) as problematic in itself. Finally, by inquiring about an alternative treatment, the father explicitly displays himself to be disinclined toward the previous no-treatment recommendation. After answering the parent’s question and providing an account for the negative answer, the physician reasserts that “since she doesn’t have any source of an antibi- of uh bacterial infection?, that=uh=we just watch her.”. At line 15, the father, having passed on several prior opportunities to accept the physician’s recommendation, again does not respond, and in the face of 0.3 seconds of silence, the physician

146

PRESCRIBING UNDER PRESSURE

adds an increment to her turn and again offers evidence in support of her position against treating the girl: “Her ears look really good.” In partial overlap with a similar statement in line 20, the father offers what is effectively an account for his withholding of acceptance: “It just means that- ya know if she gets another fever we hafta bring her ba:ck,” a statement that reﬂects an underlying assumption that “another fever” would be bacterial and in need of treatment. Moreover, similar to Extract 6.3, the father’s account for why this is problematic is based in nonmedical contingencies (i.e., inconvenience) rather than in biomedical issues. Returning to the doctor is an inconvenience, and this inconvenience is offered as an account for his implied preference for receiving antibiotic treatment during this encounter. Once again, this also displays the father’s orientation to antibiotic treatment as negotiable because his raising inconvenience as an account shows him to be oriented to his preferences as potentially affecting the ﬁnal prescribing outcome and thus to the decision as not wholly based on medical evidence. Following this sequence, the physician launches what appears to be a solution to this problem with “What I’ll do is. . .” and in line 32 “What you ca:n do i:s-”. But after an interruption in which the physician interacts with the child, she returns to suggest that the father could in fact call and would probably not have to bring his daughter back, concluding with: “as long as she looks this goo:d, an’ no other symptoms, .hh ya know we just- we’ll watch her.” (data not shown). The question in extract 6.10, like Extract 6.9, is designed to prefer a “Yes”. (For relevant discussions of question design, see Boyd & Heritage, 2006; Pomerantz, 1984, 1988.) If a treatment recommendation has been made or implied, as is the case in Extract 6.10, it is much more common for these inquiries to prefer a “No.” But both designs accomplish the same action in terms of raising antibiotics for discussion, requiring a physician’s response, and conveying a preference for antibiotics. The “No” preference structure may do more to orient to the physician’s medical authority and to his or her conveyed trajectory toward a no-antibiotic treatment plan. We can see an example of this turn design in Extract 6.11. Here, after the doctor has diagnosed a cold and suggested Triaminic or Robitussin (both over-the-counter cold remedies), and furthermore after the doctor has stated that the illness should “just go away by itself” (lines 1–3), the parent questions the need for antibiotics (line 42). (6.11) 2081 (Dr. 8) 1 2 3 4 5 6 7

DOC:

MOM: DOC:

#An’ thuh- (0.2) an’ thuh cough and that kind of stuff- it should hopefully just go away [by itself in the next #wee:k,# [( ) .hh And: you should- uhm_ (0.2) if you’re feeling really ba:d, (kh ta-), (.) call your (o ) cuz. . .

((30 lines discussing mother’s illness not shown)) 38 39

DOC:

Tell her: (.) tuh pinch it. ((re: bloody noses)) (0.2)

OVERT FORMS OF NEOGTIATION

40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

147

DOC:

Like h pinch it as much as she ca:n for ﬁve minutes an’ then= MOM: -> =So [he don’t even need any antibiotics then. DOC: [it’ll go awa:y, DOC: #Hm mm[:.# MOM: [Okay so I’m may need that in writing. (.) MOM: <[I mean- [I believe you and you’re thuh doctor= BOY: [( ) DOC: [-Hh .hh MOM: =[(but she-). DOC: =[.HH £for thuh:: for thuh [what. Mmhmph. MOM: [(she -) MOM: I don’t know [but she just- that’s what she said. DOC: [hm hm MOM: He needs antibiotics. [and (I’m like okay.) DOC: [Ha ha ha ha ha ha ha [ha ha ha ha ha ha ha ha ha ha ha ha ha MOM: [( ) She won’t listen to me so- and [I’m leaving ‘m with ‘er] for a week= DOC: [ha ha ha ha ha ha ha ha] MOM: [so she’ll be paranoid. DOC: [°h hu° .hh DOC: AH! ha ha huh. .h

The mother ﬁrst seeks conﬁrmation that her son does not need antibiotics (line 42). Despite the built-in structural preference for agreement, this action—like those in Extracts 6.9 and 6.10—reopens a discussion of treatment and turns the talk to antibiotics speciﬁcally. Additionally, the turn initial “So” treats this as the upshot of the doctor’s earlier explanation, which did not mention antibiotics, and thus displays the mother’s orientation to the doctor as having mentioned all of the medication likely to be suggested. After the doctor answers the mother’s question in the negative with a simple “#Hm mm:.#” (line 44), the mother indicates a desire for antibiotics even more strongly by requesting the physician’s recommendation “in writing” (line 45). This request is done on the behalf of a third party (the grandmother), which works to distance the mother from the challenging nature of the action. However, it nonetheless advocates for antibiotic treatment. Subsequently, although the physician treats this challenge in a joking way, it is notable that the mother does not afﬁliate with this stance by laughing with the physician (Jefferson, 1979, 2004). In this case, we again see the mother’s orientation to her preferences as having legitimate impact on the physician’s prescribing decision because she orients to her position as potentially altering the treatment outcome. We observed this pattern across the covert practices shown in previous chapters. In these more overt practices, it is all the more transparent: Parents view themselves as both being able and having a right to participate in their child’s treatment, even if they also display an orientation

148

PRESCRIBING UNDER PRESSURE

to this as a delicate action, but this is counterbalanced with concerns of treading into physician territory. Mentions of Past Experience with Antibiotic Treatment A ﬁnal relatively common practice parents use to explicitly communicate pressure for antibiotics is to mention a past experience with antibiotic treatment. This behavior, like inquiring about antibiotic treatment, is less direct than requests or statements of desire. In fact, mentioning a past experience with antibiotics is arguably the least direct of the methods discussed because, although it conveys a position in favor of antibiotics and overtly raises antibiotics for discussion, it neither explicitly states that preference (and therefore can be treated as simply providing information) nor makes a physician’s response conditionally relevant. Nonetheless, physicians routinely treat parents who mention past experience with antibiotics as lobbying for antibiotic treatment. For example, in Extract 6.12, the mother has already presented the child’s problem (data not shown), and in line 1 the physician is initiating the physical examination. It is during the physical examination that the mother mentions her prior experience with antibiotics. (6.12) 1113 (Dr. 1) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

DOC: MOM:

So:- Let’s take uh listen to ‘er che:st, (Alright), (.) MOM: -> Remember she- she:=uhm_ had something like this: in -> December? DOC: Uh huh, (0.5) GIR: Hhh.=.h MOM: -> (n’) She was on an antibiotic. (1.0) DOC: ^Doo doo. ((to girl)) GIR: Ksh:::::, uh. DOC: #Yeah:# Well I think that she probably got:=similar type of thi:ng, ya know [some sort of a secondary-= MOM: [Mm hm:, DOC: =.hh uh: respiratory infection in her ches:t, like #uh bronchitis an’_#

The mother’s assertion in lines 4–5/9 that her daughter had a similar problem before and was treated with an antibiotic conveys a position that the prior treatment was successful and that her daughter is in need of the same treatment again this time. Initially, such behavior might not appear to be lobbying for antibiotics. But no statement from the mother was due here. It is a self-initiated report that poses the puzzle of what the mother is doing with this. Thus, although the report might not appear to be lobbying for antibiotics, it is doing something special by virtue of its position in this

OVERT FORMS OF NEOGTIATION

149

sequential context. Also, mentions of precedents have previously been found to be a resource for pressuring interlocutors. For example, Kleinman, Boyd, and Heritage (1997) and Heritage, Boyd, and Kleinman (2001) show that in interactions between physicians and physician-insurance reviewers, tympanostomy tube surgery was more likely to be approved if the physician cited a prior use of tubes. In the present case, by reporting this previous instance of antibiotic use, the parent implies that this illness is similar to that one and thus that this treatment should also mirror that. With this action, the parent also treats this information as potentially inﬂuencing the doctor’s treatment recommendation with respect to antibiotics. The physician initially does not respond but examines the girl (lines 10–12). Following that examination, the physician agrees with the parent and suggests that this illness is in fact a “similar type of thi:ng,” making antibiotics an appropriate treatment (line 13). Subsequently, the physician goes on to prescribe antibiotics. Parent mentions of past experience also sometimes invoke their own illnesses. We saw an illustration of this in Extract 5.7 (reproduced here in 6.13). Having already fully analyzed this case in chapter 5, I will not redo that here. What I want to focus on here is that part of the parent’s resistance is to state how antibiotics were helpful in his own past experience, and because he believes the illnesses of his sons are “the same” as his own illness, the father conveys that antibiotics are his preferred treatment and thereby pressures the physician to prescribe antibiotics. (6.13) 322803 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

DOC: DAD: DOC: DAD: DOC: DAD: DOC: DAD: DOC: DAD: DOC: DAD:

->

-> -> -> ->

DOC: DAD: -> -> DOC: DAD: -> DOC: DOC: DAD: ->

I th:ink from what you’ve told me (0.2) that this is pro:bably .h uh kind of (0.2) virus infec[tion, [Uh huh, (0.4) th:at I don’t think antibiotics will ki:ll, (0.2) Well[Thee other[( ) >Go=ahead_< Yeah. .hh ( ) I had it- I had thuh symp[toms [I understand. Three weeks ago. [Right. [.hh An:d I’ve been taking thuh over the counter cough [( ) [(Good_) Uh s- ( ) coughing syrup, Nothing take away .hh Especially my sor- my [th- my throat was real= [Mm hm =sore [for (awhile- et- that) w:eek. [Uh huh °Right,° an:d (.) I start taking thuh antibiotic (0.5)

150 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38

PRESCRIBING UNDER PRESSURE

INF: eh he ((cry)) DAD: -> Yesterday. DOC: Right, DAD: -> And it (.) seemed to take care of the problem. DOC: [(Well) that’s why we’re doin’ a throat [culture. DAD: [( ) [Yeah. DOC: [is TUH SEE if they need antibiotics. DAD [( ) Yeah yeah. (0.2) DOC Cause
This practice may further emphasize the parent’s role in his sons’ health care because it treats the children’s personal history as unique and potentially distinct from the more generic illness histories that physicians may use to gauge whether a child requires antibiotic treatment. For example, a parent whose child routinely moves from viral colds to ear infections will cite this as sufﬁcient reason for the physician to prescribe the medication when the child has only a viral cold. Although both the physician and parent may say that antibiotics are generally not needed for such viruses, when the parent introduces this aspect of the child’s history, they claim that this makes this child’s needs different from those of the typical child with a cold. Another example was shown in Extract 6.8. There the mention of past experience with antibiotics was positioned after a nonantibiotic treatment recommendation and was part of resistance to this recommendation. Earlier, we focused on the parent’s statement of desire for antibiotics, which came just following her mention of prior experience with antibiotics some months previously. But, we can also observe that with this mention she suggests that antibiotics are appropriate and desired this time. (6.14) 161303 1 2 3 4 5 6 7 8 9 10 11

MOM: DOC:

DOC: MOM: DOC: MOM: DOC: MOM:

Okay, Uh: the virus irritates the big tubes and there’s a lot of mucus. (0.5) [So: [Because in Oct^ober he ca:me and I believe it was also for= =Something similar li[ke this or:_ [Yeah: and he got- he was on antibiotics. [Yeah:. [He- He got on antibiotics an’ .hh you know uh:m

OVERT FORMS OF NEOGTIATION

12 13 14 15 16 17 18 19

DOC: MOM: DOC: DOC: MOM:

151

I just feel that antibiotics sort of .hh -:make him feel better: - i- in a shorter [time
Here, we see evidence that the mother’s mention of past experience with antibiotics is lobbying for antibiotics through what she does immediately subsequently: state a desire for antibiotics by saying how she feels that antibiotics make him feel better sooner. We have now examined four ways that parents work to lobby physicians for antibiotic prescriptions. These formulations vary in their directness, but all explicitly raise antibiotics for negotiation, and all convey the parent’s position as in search of antibiotics. Through the action of lobbying for antibiotics, parents adopt a stance toward the prescribing decision as one that resides not solely with the physician but rather one that is properly shared by the parent and physician and can thus be negotiated in and through the interaction.

Physician Response to Parent Pressure Across the interactions we have examined in this chapter, we have seen through the ways that the physicians respond to parents’ overt lobbying for antibiotics that they hear parents as pressuring for, and initiating (or even escalating) a negotiation for, antibiotics, although the exact response is, of course, related to the parent’s form of overt lobbying. For instance, in Extract 6.5, after the parent advocated for “pink” medicine, the physician’s recommendation for antibiotics was designed as accommodating the parent with her use of “can have.” In Extracts 6.6 and 6.10, the physicians displayed their orientations to being in negotiation with parents through the way their denials were formulated, as well as through their defenses. In Extract 6.7, the physician formulates himself as regretful that he may not be able to perform “the miracle” of curing the child. Finally, in Extract 6.8 (shown again in 6.14), the physician rejects the parent’s account for wanting antibiotics (that they would help him feel better faster). Of course, as with other practices, perhaps the best evidence that overt negotiation inﬂuences diagnostic and treatment outcomes is cases where this is transparent. One case was analyzed in detail in Extract 5.9a–d. There, after multiple types of resistance, including very strong overt lobbying for antibiotics, the physician agrees to prescribe antibiotics, going against her earlier diagnosis of a viral infection and going against her earlier recommendation against antibiotics. In chapter 5, we were focused on treatment resistance. Here, we can observe that the change to offering antibiotics occurred after the parent shifted to overt lobbying for antibiotics.

152

PRESCRIBING UNDER PRESSURE

We can observe a similar shift in Extract 6.15. Here, the physician has done a throat culture to test for strep throat. If the culture shows bacteria, then antibiotics would be appropriate. When physicians perform a strep test, they sometimes prescribe antibiotics with the idea of stopping them if the results are negative. Alternatively, they provide a prescription only if the results are positive. So although the results are not back, there is nonetheless a prescribing decision to be made. In lines 1–2, the physician tells the mother to “call tomorrow about eleven:,” when she can get the results and the prescription if one is necessary. (6.15) 1061 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12

DOC: DOC: MOM: DOC: MOM: DOC: DOC:

-> -> => => => =>

DOC:

Mlk=Okay, so we’ll culture her, ‘n=you call tomorrow about eleven:, .hh= =Alright [so we’re not starting her= [If her culture is=[ o n a n y t h i n g, [today?, =[we’re not gonna start i- [.hh Well I tell you what I c’n do is let me give you uh sample of something just so we don’t (miss) uh day. .hh In case: ya know it is positive then we= =ya know we’ll have ( ).

In response to the physician’s directive (lines 1–2), the parent requests conﬁrmation that “we’re not starting her on anything, today?,” (lines 4/6). This works to renew a discussion of medication. In response, the physician offers what he frames as a concession with “Well I tell you what I c’n do” and then gives the mother samples of an antibiotic (line 9). As was the case in Extract 6.14, a physician’s prescribing decision is changed from no antibiotics to antibiotics in and through the interaction with the parent. In a ﬁnal example, the mother inquires about antibiotics at a point where the physician has indicated that the only treatment needed is for the child’s wheezing. (6.16) 181412 1 2 3 4 5 6 7 8 9 10

DOC:

MOM: DOC: DOC: MOM: DOC:

‘Cause he’s wheezing. So go every four hours, and then (d’) you need cough medicine or you have some at home. Uhm (0.5) I don’t know exactly. No. I haven’t had any for a while. Okay so I get you some anyway. °Okay°, (1.0) Would antibiotics be:_ Uh::: he=u=Does he have any temperature,=Yeah

OVERT FORMS OF NEOGTIATION

11 12 13 14 15 16 17

MOM: DOC: MOM:

DOC:

153

he got a little bit tiny_ Yea:[h, [Little tiny bit. So I go ahead put him on. I really want to make sure we get this out of his system before he goes back to school Monday, (0.2) Yeah_

The physician is detailing the treatment for the child’s wheezing: “so go every four hours” (with a breathing machine that the family has at home) and then recommends cough medicine. She adds, “or you have some at home.” indicating either an over-thecounter cough medicine or a prescription medication that she has prescribed for this child previously. From the physician’s subsequent offer to “get you some anyway.” it appears that it is a prescription medication. In any case, the parent answers but then asks about antibiotics as an additional type of treatment not previously offered or implied: “Would antibiotics be:_”. The physician initially indicates predisagreement with “Uh::”. The delay of an answer is hearable as predisagreement because it defers agreement and implies that agreement is contingent on that sequence. (See Schegloff, 2006, for a discussion of pre-second insert sequences.) However, after determining that a slight fever has been present, the physician does give antibiotics (line 13). Although it is not framed as “agreeing to” do this, the physician’s recommendation is directly responsive to the parent’s inquiry, and all evidence suggests that an antibiotic was not otherwise going to be recommended. Thus, as in other cases we have seen, overt lobbying can successfully be used to negotiate a change in prescription from no-antibiotics to antibiotics.

Discussion This chapter has focused on the least common form of parent pressure for antibiotics: overt pressure. As with all other types of negotiating practices, parents do not overtly pressure physicians very often. But when they do, this inﬂuences physicians and is therefore an important type of parent behavior. This chapter has demonstrated that overt pressure for antibiotics is typically done through one of four interactional practices and that all are treated by physicians as lobbying for antibiotics and all initiate (or escalate) a negotiation of antibiotics. In addition, though, this chapter showed that parents do not treat negotiation, particularly in this overt form, lightly. Rather, they treat pressuring physicians as delicate business and do so in ways that speciﬁcally maintain an orientation to the physician’s authority over the medical (and particularly the diagnostic and treatment) domain. This is an important dimension of the interaction because it helps to account for why overt pressure is relatively uncommon, whereas covert pressure is relatively more common and is therefore the primary vehicle through which parents convey their treatment preferences, as well as other concerns that affect diagnostic and treatment outcomes. This chapter also demonstrates that parents have interactional resources to pressure physicians for antibiotics during any activity within the medical visit. In this

154

PRESCRIBING UNDER PRESSURE

chapter, we saw overt negotiation used during the opening of the visit prior to the reason for the child’s visit, during the middle of history taking, during the physical examination, and, most commonly, following the treatment recommendation. Taken together, overt negotiation practices and covert negotiation practices offer a wide array of weapons with which to battle physicians who do not believe a child has a bacterial diagnosis or needs antibiotic treatment. And physicians do, as we saw quite explicitly in Extracts 6.1 and 6.13, perceive themselves to be in battle with parents at times over this issue. We might begin to wonder what physicians can do about this problem and whether our only hope is parent education. The next chapter turns to the issue of physician behavior.

7

Physician Behavior That Inﬂuences Parent Negotiation Practices

P

hysicians seeing a child for whom antibiotics are not an appropriate treatment frequently face a difﬁcult task: to secure parent acceptance of a no-problem diagnosis and a nonantibiotic treatment recommendation. The task is difﬁcult because it denies parents a treatment that they may have wanted. To make matters worse, this outcome is vulnerable to being understood as indicating that parents’ judgment was ﬂawed when they decided to visit the physician. Thus physicians also need to balance maintaining visit legitimacy with appropriate medical decision making. Because a lack of prescription treatment is, in itself, a threat to the legitimacy of the visit, each physician behavior that indicates the physician’s trajectory toward a no-antibiotics diagnosis and treatment is potentially threatening to visit legitimacy as well. But as Maynard (2003) has documented with respect to bad news, one very good resource for securing parent acceptance of bad news is to forecast the news in order to assist in the recipient’s realization of the news. It is ironic that parents treat a viral diagnosis as bad news because in many ways, physicians are delivering good news: Children who are not in need of antibiotics generally have self-limiting viral infections that will resolve in 7 to 10 days. But as we have seen in earlier chapters, telling parents that their child must simply “be sick” and that there is no cure is generally viewed as bad news. For these reasons, interactional resources that forecast that a child does not need antibiotics and help physicians maintain the legitimacy of the child’s visit are important. This chapter will discuss several types of physician resources and focus on three: online commentary, presentation of the diagnosis, and presentation of the treatment recommendation. 155

156

PRESCRIBING UNDER PRESSURE

Resources That Forecast the Diagnosis Physicians rely on a variety of resources to forecast that the child has a viral or noproblem diagnosis. Perhaps one of the earliest resources for this is to reject a parent’s bacterial candidate diagnosis. Examples of this behavior were shown in chapter 2 in the discussion of immediate responses to candidate diagnosis problem presentations. For instance, in Extract 7.1 the mother’s candidate diagnosis of “sinus infection” (line 10) is rejected with “Not necessarily:,” (line 13), despite the fact that the physician has not yet begun any verbal or physical examination of the child. (7.1) SG615 [shown earlier in Extract 2.5; 2.13] 10 11 12 13 14 15 16

MOM: DOC: MOM: DOC:

So I was thinking she had like uh sinus in[fection= [.hhh =er something.= =Not necessarily:, Thuh basic uh: this is uh virus basically:, an’=uh: .hh (.) thuh headache seems tuh be:=uh (0.5) pretty prominent: part of it at ﬁr:st uh: (0.2) .hh

The countering behavior works to shape the parent’s expectations that the child’s illness will not be diagnosed as bacterial. But it does little to legitimate the child’s visit (and thus the parent’s decision to schedule the visit). Even when physicians suggest an alternative diagnosis, this may still do little to validate the legitimacy of the visit if the proposed diagnosis is a no problem one because this implies nontreatability. When appropriate, mentions of diagnostic tests that can be done may be at least partially validating. An example of this is shown in Extract 7.2. In this case, following a bacterial candidate diagnosis but before any verbal or physical examination, the physician rejects the diagnosis of strep and counterproposes “uh viral process” (line 57). (7.2) SG316 [full presentation shown in 2.6; also shown in 2.14] 31 32 33 34 35 36

MOM: -> [(I-) I thought (0.5) maybe I better just- know if ya know strep has secondary er anything like -> that I wasn’t sure.But he hasn’t had thuh fever er thuh nausea er anything that he’[s had before. DOC: [O:kay:, DOC: .hh (Goo:d?,)

((16 lines not shown including joking about BOY having day off but not MOM)) 53 54 55 56

DOC: DOC: => => =>

££O:kay:,££ .hh Well:, (.) o:ne good thing is: that- uhm (0.5) strep infections:- respond really well tuh amoxicillin. .hh so wh:ile he may not have strep any more (.) he could still

PHYSICIAN BEHAVIOR THAT INFLUENCES NEGOTIATION

57 58 59 60 61 62 63 64 65 66 67

157

DOC: => have- uh viral process going on, he could still have just => residual sore throa:t, .h[h dry weather kind of things, .hhh MOM: [(°Okay.°) DOC: Uhm: besides having an actual infection so we can always look at those issues, .hh an then if you want we can also just retest his throat. (.) DOC: An’ make sure there’s no more strep there too. (.) MOM: Well (you it) kinda depends on what you- what you [think. DOC: [Mkay,

After countering the parent’s concern of strep (lines 54–56), including accounting for it (54–55), as well as counterproposing an alternative (lines 57–58), the physician offers to retest the boy’s throat (lines 61–64). Testing the throat may help validate the legitimacy of the visit. However, the presentation is critical. Here, it is framed as a concession to the parent with “if you want,” which may reduce its success in terms of validation. We can see that the parent, with this formulation, treats it as a sanction of her intrusion into the physician’s domain of authority with “it kinda depends on what you- what you think.” (line 66). Outright rejections of diagnostic proposals during history taking also forecast that a bacterial diagnosis is unlikely. We can see an example of this in Extract 7.3. (7.3) SG506 [shown as 3.21] 8 9 10 11 12 13 14 15 16 17 18

GMA: -> -> -> GIR: DAD: GIR: DOC: => =>

So that’s why I=uhm (1.5) we decided to bring ‘im in because (0.4) with thuh temperature. (.) (I know ) somet[hing’s gonna be wrong. [Daddy. [Stop that. [( ) Well actually uh regular cold can give you uh high fever for: three day:s. (0.4) DOC: => As long as they act okay then actually it’s- it: => (0.8) may go away by itself, (it’s) just uh virus.

Just prior to this, the grandmother stated that the boy commonly gets ear and throat infections. She provides support for her possible diagnosis of ear or throat infection by citing his temperature. With these behaviors, the grandmother adopts a stance toward the illness as treatable and thus, at least for the physician, as bacterial. It is this connection that the physician quarrels with in her response, which contends that “uh regular cold” (line 14) can produce the same symptoms and “(it’s) just uh virus.” (line 18). Here again, the physician is deﬁnitively forecasting a nontreatable diagnosis. On the one hand, this is useful because it conveys to the parent (or in the last case, the

158

PRESCRIBING UNDER PRESSURE

grandparent) early on that the child may have a no-problem, no-treatment outcome on the way. However, what remains as an interactional problem is that in forecasting this the physician simultaneously threatens the legitimacy of the child’s visit and the child’s treatability. Both of these issues can motivate parents to pressure physicians for antibiotics. The remainder of this chapter will look at three behaviors that deal with these issues, though the balance of forecasting, legitimating, and treatability is slightly different with each. First, we will examine physicians’ use of no-problem online comments1 (Heritage & Stivers, 1999). Second, we will examine alternative practices for delivering the viral diagnosis. And third, we will examine two practices for delivering nonantibiotic treatment recommendations. Online Commentary In earlier work, “online commentary” was deﬁned as an action that describes or evaluates the signs physicians are encountering during the physical examination (Heritage & Stivers, 1999). This behavior affords the parent some access to the physician’s diagnostic reasoning. However, although they are relevant to, and may foreshadow, the physician’s ﬁnal diagnostic evaluation, online comments are quite distinct from the ﬁnal or ofﬁcial diagnosis in two dimensions: content and positioning. In terms of content, online commentary differs from an ofﬁcial diagnosis in that it does not contain inferential reasoning in the form of conclusions about the patient’s medical condition. Rather, online commentary simply formulates the sensory evidence that is available to the physician in the course of the medical examination. In terms of positioning, online commentary occurs during the physical examination of the patient. In contrast, diagnostic evaluation is typically produced as a distinct action constituting a discrete activity within the consultation (Byrne & Long, 1976; Heath, 1992; Peräkylä, 1998). Thus, diagnostic evaluation is almost invariably spatially and temporally separated from the examination activity, and it is offered at its termination. Online commentary can be characterized not only in terms of its content and position relative to other forms of diagnostic talk but also in terms of its design. There are several key features. First, in contrast to ﬁnal diagnoses, which are, as discussed in chapter 4, treated as a central activity in the consultation, physicians and patients and/or parents treat online commentary as, at best, an intermission in and subordinate to the activity of physical examination that is under way. Second, online comments can be divided into two broad types: 1. Online commentary that describes signs that are present but mild. These comments are normally mild, downgraded, or qualiﬁed (e.g, “That’s a little bit red back there,” or “there may be a little bit of lymph node swelling on this side compared to the other side”). 2. Online commentary that describes the absence of signs. This commentary is often mitigated by the use of evidential formulations, such as “I don’t see any ﬂuid,” which make reference to the sensory evidence from which observations come (Chafe & Nichols, 1986).2

PHYSICIAN BEHAVIOR THAT INFLUENCES NEGOTIATION

159

Third, online comments addressing both present and absent signs take two primary formats: (1) reports of observations, such as “I don’t see any ﬂuid,” “°Little bit re:d°,” and “I don’t see any drainage,” or (2) assessments of what is observed, such as “Your ears look goo:d” or “This one looks perfect.” In the report format, the physician does not overtly evaluate the signiﬁcance of the observation for the patient’s health status but leaves it to parents to draw their own conclusions about it. In the assessment format, the physician provides less insight into the examination but overtly draws evaluative conclusions. As an example, we can look at Extract 7.4a, which shows a fairly complex management of a no-problem evaluation across an entire examination. In it, we see the pediatrician carefully balancing maintaining the legitimacy of the visit and ﬁrmly resisting any implied expectation for antibiotic medication. This consultation with an 11-year-old girl and her mother took place on a Monday afternoon, and the child had already missed most of her school day. The problem presentation involves a candidate diagnosis shown in chapter 2 as Extract 2.4. The pediatrician’s examination of the girl commences after a brief history taking that reveals no surprises with respect to the condition but where the mother did conﬁrm that the illness had been passed from one family member to the next, which seems to underscore her claim that this child is ill. The physical examination begins in Extract 7.4a with the primary complaint: the girl’s ears. (7.4a) 305 [problem presentation shown in Extract 2.4] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

DOC: PAT: DOC: -> MO?: DOC: -> -> PA?: DOC: PAT: DOC: DOC: PAT: DOC: DOC: MOM: MOM: DOC: -> -> MOM:

Which ear’s hurting or are both of them hurting. (0.2) Thuh left one, °Okay.° This one looks perfect, .hh (U[h:.???) [An:d thuh right one, also loo:ks, (0.2) even more perfect. ( ) Does it hurt when I move your ears like that? (0.5) No:. No?, .hh Do they hurt right now? (2.0) Not right now but they were hurting this morning. They were hurting this morning? (0.2) M[ka:y, [(You’ve had uh-) sore throat pain? (°Yes°) Let’s check your throat. <.hh=I=uhm: there’s- I don’t really see uh lotta ﬂuid build up. Some[times you’ll= [(Yeah.

160

PRESCRIBING UNDER PRESSURE

24 25 26

DOC: -> =see tha:t ya know, MOM: Uh huh, DOC: -> (But) I don’t see it right no:w,

The left ear is the ﬁrst to be examined. The physician’s online commentary at line 4, uttered during the otoscopic examination of the ear, embodies the evidential formulation that is common when online commentary is deployed to counteract patient claims. A second online comment, uttered during the examination of the right ear, is built as a cumulative addition to his ﬁrst and is deployed to similar effect. During further examination of the ears, the physician asks about current pain symptoms (lines 9, 12, and 13), ﬁnally eliciting a response (line 15) in which the girl defends herself (and her mother’s prior claim, the candidate diagnosis shown in Extract 2.4) against the possible inference that her symptoms are mild or nonexistent. At this point, the mother intervenes with a question, directed to her daughter, about “sore throat pain” (line 19), which the physician treats as raising another problem for evaluation (line 21). As he prepares for examination, he returns with further online commentary about the state of the girl’s ears (lines 21, 22, 24, 26), which embodies an evidentially formulated denial of the kind of ﬂuid buildup characteristically associated with middle ear infections. This denial includes two other mitigating elements. In characterizing ﬂuid buildup as something you “sometimes” see, the pediatrician implies that this is not a criterial feature of an ear infection or ear pain. Further, in adding that he does not “see it right now,” he leaves open the possibility that ﬂuid buildup may have been present in the past or may appear in the future. At this stage in the consultation, the effect of the physician’s online commentary is to deny the existence of the main signs associated with the girl’s chief complaint and thus what would have been support for the mother’s candidate diagnosis. Similar to denying a candidate diagnosis outright, the physician forecasts that a no-problem, nontreatable diagnosis is likely. But online commentary differs in two key ways: First, it proceeds incrementally, so the parent is gradually (rather than suddenly) brought to the realization that the diagnosis is likely to be nonproblematic and nontreatable. Second, online commentary at least claims to be supported by signs that are observable (or observably missing). Whereas an outright rejection simply asserts the physician’s authority, online commentary claims only the authority to observe what otherwise independently exists, and thus the authority is supported by facts. If we return to this case, we can see how this practice works to support, rather than undermine, the legitimacy of the visit. The examination now proceeds to the girl’s throat. (7.4b) 305 21 22 23 24 25 26

DOC: MOM: DOC: MOM: DOC:

Let’s check your throat. <.hh=I=uhm: there’s- I don’t really see uh lotta ﬂuid build up. Some[times you’ll= [(Yeah. =see tha:t ya know, Uh huh, (But) I don’t see it right no:w,

PHYSICIAN BEHAVIOR THAT INFLUENCES NEGOTIATION

27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

DOC: PAT: DOC: -> DOC: -> MOM: DOC: -> MOM: DOC: -> -> DOC: MOM: DOC: MOM: DOC:

161

(0.4) Uh: let’s see. Say ah:?, (0.5) (uh_??) (1.0) That’s uh little bit red back there, (0.2) I don’t see anything: (0.4) °Yeah.° Very good. Thank you. Huh h[uh huh (.hh) [I don’t see anything (.) that looks infected. Reall[y, °Okay.° [Uh: in thuh sense that we’re: looking at bacterial, strep throat kinda thing(s). .h[h [O[kay. [Lemme listen to ya. (1.5) Could it be [from allergie:s, [Take uh deep breath, Sit up straight?,

This examination, initiated at line 21 (and 28), eventuates in a slight change of procedure in the physician’s use of online commentary. At line 32, he observes a positive (but mild) symptom that could validate the child’s complaint of “sore throat pain” and, with it, the decision to make the medical visit. Notably, this observation, which is supportive of the mother’s decision to seek medical care for her child, is not downgraded via evidential formulation. Subsequently, having completed the examination and while preparing to listen to the girl’s lungs, he produces a more comprehensive online assessment that is also evidentially formulated: “I don’t see anything (.) that looks infected.” (line 36). Subsequent to the mother’s resistant “Really” (line 37), he qualiﬁes his previous assessment with a grammatically ﬁtted increment (lines 38–39) (Schegloff, 2001), in a way that allows that the child may still have some kind of infection while still eliminating the prospect of a bacterial infection and, by implication, the prospect of antibiotic treatment. In this exchange, then, we see both that the physician is working to forecast a no-problem diagnosis and that he is working to validate the mother’s decision to seek medical help. In this case, the physician is still facing resistance through the “Really,”, but note that the mother also moves toward acceptance with “Okay.”, particularly in line 41, where this comes following his explicit invocation of bacterial illness. The mother’s response to this outcome at line 44 is to maintain her position that her daughter has a medically treatable problem by raising the prospect of a further condition—allergies. Insofar as this inquires into a different diagnosis of the problem, it displays her acquiescence to the physician’s rejection of strep throat as a diagnosis. After an uneventful lung examination, the pediatrician moves to examine the girl’s lymph nodes. This examination is shown in Extract 7.4c.

162

PRESCRIBING UNDER PRESSURE

(7.4c) 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77

DOC:

Does it hurt when you breathe in deep like that? (1.4) ((Patient shakes head)) DOC: No:? (0.2) DOC: How ‘bout- under your chinny chin chin.<°Let’s see.° (1.5) DOC: -> No(w) there may be uh little bit of lymph no:de swelling -> on this side com[pared to the other side, MO?: [(Yeah.) (.) DOC: -> On thuh [left side, MOM: [Oh:: okay. DOC: .hh So: it would loo:k hh like she is:=uhm (.) prob’ly ﬁghting some (.) viral: upper respiratory kinda stuff, DOC: .hh More on thuh left than on thuh right, which can account for some pain maybe,

The physician’s identiﬁcation of “lymph no:de swelling” (line 68) gives implicit support to the patient’s claim that she has experienced pain primarily on the left side. Subsequently, the pediatrician makes this explicit in lines 76–77 as part of the diagnosis, which he begins with the upshot formulating “so” at line 74. The ﬁnal diagnosis is offered in two parts separated by an extended compliment about the cleanliness of the girl’s ears: (7.4d) [shown in Extract 2.21] 74 DOC: 75 76 DOC: 77 78 MOM:

.hh So: it would loo:k hh like she is:=uhm (.) prob’ly ﬁghting some (.) viral: upper respiratory kinda stuff, .hh More on thuh left than on thuh right, which c[an account for some pain maybe, [Okay. ((13 lines of ear compliment sequence removed))

92 93 94 95 96 97 98 99 100

DOC:

MOM: DOC: MOM: DOC: MOM:

Uh:- I would tell you though I don’t hhh (.) I don’t see anything that requires like antibio:tics er anythi:ng, but certainly sympto[matic treatment might be in order, [Mm. .hh O[kay. [Uhm: anything from vaporizers tuh maybe some chloraseptic kinda stuff for thuh [throat, lozenges might be better, [Oh:. Okay.

The ﬁrst part of the problem discussion (the diagnostic evaluation in lines 74–77) is supportive of the mother and daughter’s decision to seek medical care and draws on

PHYSICIAN BEHAVIOR THAT INFLUENCES NEGOTIATION

163

earlier online comments (e.g., Extract 7.4c, lines 68–69) that were also supportive of that decision. The second half (lines 92–93) builds from the adverse online commentary—especially that in (7.4b)—and then clearly rejects antibiotic treatment (line 93) in favor of symptomatic over-the-counter remedies (line 94/98–99). And as the physician moves from diagnosis to treatment, he begins with the phrase “I would tell you though” (line 92). In this way, he builds the recommendation as contrastive with the notion that viral conditions require antibiotic treatment, and hence in contrast to any position the mother might hold in favor of antibiotic treatment, without contradicting her earlier claim made both by her presence at the clinic and the candidate diagnosis (Extract 2.4) that the child is sick. His use of the evidential formulation (“I don’t see anything. . .”) revives the relevance of the observations reported in his earlier online comments and reinvokes their signiﬁcance as evidence for the position he is currently taking. Across this sequence, the mother responds to both the supportive and adverse aspects of the diagnostic evaluation with an acknowledgment token “okay”, which accepts the physician’s evaluation. At line 100, this acceptance becomes more marked with the addition of “Oh” (Heritage, 1984a; Heritage & Seﬁ, 1992). Subsequent to this, the mother discusses the merits of several commercial remedies in a cordial way and without contesting any aspect of the physician’s conclusions. This interaction proceeds from a situation in which both the patient and her mother initially viewed the patient’s complaint as signiﬁcant and were defensive (extract 7.4a, lines 15, 19; extract 7.4b, line 44) when that stance seemed to be threatened, to a situation in which they acquiesced to the physician’s nonantibiotic treatment without active resistance or signs of disappointment. I would argue that the primary reason for this is that the legitimacy of the visit has been maintained, at least partially. Online commentary’s role in the progressive construction of powerful support for the physician’s ﬁnal diagnostic evaluation in this case seems clear. Parents are in a position where contesting the online comments is very difﬁcult because they represent a physician’s authoritative way of seeing, hearing, feeling, or assessing. And when online commentary is used, inferential resources for the diagnostic evaluation and the lack of antibiotic treatment are built up incrementally. The result is a more persuasively formulated case for the ﬁnal evaluation and recommendation than would be obtained without the use of online commentary. Moreover, when online comments have been used, it is more difﬁcult for parents to contest ﬁnal diagnoses without exhuming the online observations that they have already let pass. But as Heritage (2006) discusses, this use of online commentary not only ramps up the physician’s claim to authority by offering unchallengeable observations but also is “evidence formulating” (Peräkylä, 1998), and in this way it makes the physician’s diagnosis a bit more accountable. In sum, then, the use of online commentary is an interactional resource with which physicians can build a case for a no-problem diagnosis and thus a case for no treatment, while still reassuring patients of the rightness of their decision to seek medical assistance. Because parents are more likely to be satisﬁed when physicians make use of this resource, it appears to be an excellent means for diminishing resistance to no-problem, no-antibiotic treatment outcomes.

164

PRESCRIBING UNDER PRESSURE

Presentation of No-Problem, No-Treatment Conditions A second interactional domain I will focus on is how physicians present their diagnoses and treatment recommendations. Physicians most commonly formulate their noproblem diagnoses and no-antibiotic treatment recommendations as announcements, but there are still quite different ways that they can be, and are, formulated. When we examine cases where physicians make use of these alternative ways of formulating diagnoses and treatment recommendations, we can observe that parents respond differently depending on the format and that this is consequential for whether and how they resist no-problem diagnoses and nonantibiotic treatment recommendations. The remainder of this chapter will focus on a parallel discussion of one dimension of diagnosis delivery formats and treatment recommendation formats: whether they frame the announcement afﬁrmatively or negatively. Background The diagnosis delivery and treatment recommendation activities are, as discussed earlier, rather different. In chapters 4 and 5, we observed that the normative structure of the treatment recommendation activity was such that parent acceptance was due upon completion of the physician’s initial treatment recommendation, whereas the diagnosis delivery made no such response conditionally relevant. And although physicians did not pursue parent acceptance of a diagnosis, if acceptance was not forthcoming following a treatment recommendation, physicians routinely pursued, or did interactional “work” to secure, this acceptance. Given the problematic nature of parent resistance (especially to the treatment but also to the diagnosis), the question to be addressed here is whether the way that physicians deliver their diagnoses and treatment recommendations promotes or inhibits parent resistance. This section will ﬁrst describe the main formats that are observed in each activity. From this point, we will see that the implications and explanations of these formats vary and then discuss each in turn. Afﬁrmative Announcements When the physician delivers a diagnosis or a treatment recommendation, the most common turn design involves an afﬁrmative announcement of (in the case of the diagnosis) what the condition is or (in the case of treatment) what should be done for the patient’s problem. As we will discuss more later, physicians sometimes offer multiple diagnostic statements, but in the Metro data, 55% of ﬁrst diagnoses and 54% of all diagnoses (whether ﬁrst or subsequent) are afﬁrmative announcements. So across the board, this is the dominant type of diagnosis. An illustration of an afﬁrmative diagnosis is shown in Extract 7.5. As the physician completes his physical examination (lines 1–5), he moves into a summary of his examination ﬁndings (5–6) and then the diagnosis “that: kind of viral stuff.” (lines 10–11). The physician does not specify a condition name, but as noted brieﬂy earlier, such diagnoses are very rare in the context of nonbacterial URI diagnoses, so this is not unusual. What I want to focus on is that the condition is an announcement, and it is delivered afﬁrmatively.

PHYSICIAN BEHAVIOR THAT INFLUENCES NEGOTIATION

165

The stress on “is” appears designed to deliver the diagnosis as conﬁrmatory of something the mother had said (Stivers, 2005a), though the mother had not previously asserted a claim that this was a viral infection. This delivery may do additional work to secure parent acceptance because if it is built as conﬁrmatory, however contrived, it may be more likely to be accepted. (7.5) 323 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

DOC:

I see that (tooth coming in but not) (0.2) DOC: any other ones, (0.4) DOC: ( ) (.) An’ she looks uh little congested_ But I don’t see anything wrong with her ears or thro^a:t. MOM: (Oh good.) (0.7) MOM: Oh: my g[oodn(h)e:ss. DOC: -> [S:o I would say: that this is that: kind -> of viral stuff. MOM: Okay.= DOC: =An’ I would keep doing thuh clear ﬂuids: (.) for twelve hours,= MOM: =(Mm [hm,) DOC: [An’ the:n (.) start thuh banana:s, ri:ce, applesau:[ce, toast, (kind of diet_) MOM: [Okay.

Positive announcements are not necessarily delivered deﬁnitively. For instance, in Extract 7.6 the physician downgrades the epistemic certainty of the diagnosis with “prob’ly” (a reduced version of “probably”). As the extract begins, the physician is reaching the conclusion of his physical examination as he transitions into a diagnosis of the primary reason for the visit: the cold symptoms (see lines 5–6). The diagnosis is offered in line 7: “That stuff is prob’ly caused by uh virus.” and restates this on its way to treatment in lines 8–9. (7.6) 324 1 2 3 4 5 6 7 8 9 10

DOC:

.h But=(uhm) uh lotta kids do:_ when they= especially when they cut their molars they- pull at their ear, MOM: Yeah:. [(No,) DOC: [Uhm, (0.2) Now as far as=uh (0.5) fe:ver, hh (0.4) co:ld, (0.4) congestio:n, runny nose, -> That stuff is prob’ly caused by uh virus. DOC: .hh So=uhm (0.2) .tlkh he- he’s- ya know he’s prob’ly got uh virus that he’s gonna get over:=an’ he’s prob’ly (getting it) over already.

166 11 12 13 14 15

PRESCRIBING UNDER PRESSURE

MOM: DOC: DOC:

#Mm hm,# Ya know getting over it already_ On his ow:n. (0.2) .h An’ you guys (could) of course give him something for congestio:n like uh little- .h [( -)

The ﬁnal example in Extract 7.7 shows a diagnosis that is offered without any epistemic downgrading or mitigation but nonetheless gives insight into the physician’s stance toward the diagnosis as not what she assumes the parent wants to hear with her use of the turn initial “Unfortunately”. (7.7) 215 1 2 3 4 5 6 7

DOC: -> -> MOM: MOM: DOC: MOM:

Unfortunately he caught uh bad co:ld. He caught thuh [one I think that [Da:d had. [(ah:-) [(Yes.) And he gave it to me too: [an’- (thuh whole family.) [
The cases shown in Extract 7.5–7.7 illustrate the most common diagnosis delivery format. This format is even more commonly used for treatment recommendations. In the Metro data, when treatment recommendations are given, 94% of all ﬁrst recommendations are done with afﬁrmative announcements, and 93% of all treatment recommendations (no matter how many) are afﬁrmative announcements. We can see examples of announcements of treatment recommendations if we return to extracts 7.5 and 7.6. In both cases, following the afﬁrmative diagnosis, the physician moves to afﬁrmatively recommend treatment as well. In 7.5, this is in lines 13–14/15–16, where the physician suggests keeping the child on clear ﬂuids and then moving to mild solid foods. In 7.6, after suggesting that the child will get better on his own (and thus indicating that no treatment is necessary), he goes on to offer an afﬁrmative treatment recommendation of something for the congestion (lines 14–15). Although the last word of this recommendation is not clear, and the physician goes on to later further specify a recommendation, here I want to note only that in both of these cases a treatment recommendation for a particular course of action is offered. Afﬁrmative announcements of treatment do not necessarily involve medication. This is illustrated in Extract 7.8. (7.8) 100101 1 2 3

DOC: MOM:

So I think he’s (just has) a co:ld_ an’ it’s just making him feel kinda of crummy. (°Yeah:.°)

((10 lines not shown - expansion of diagnosis))

PHYSICIAN BEHAVIOR THAT INFLUENCES NEGOTIATION

14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

DOC: MOM: DOC: -> -> -> DOC: -> -> -> -> DOC: -> -> DOC: DOC: -> MOM:

167

A:lright. and I’ll stand away from you so you don’t (ha[fta me anymore.) [(Hah hah hah .hh heh heh) Uhm_ (0.2) but:=uh: (.) I think thuh best thing tuh do at this point (would)/(will) jus’ be try elevating his head at night. give him plenty of liquids to drink. .hh you c’n- .h if you c’n put something other- under one end of the ma:ttress, so thuh whole ma:ttress is on uh sla:nt or under one one end of thuh bed, so thuh whole bed is tilted, .hh that way his >head’s up a little bit<. That helps thuh mucus tuh drain better. (.) for him. Right.

Here, the physician offers her diagnosis in lines 1–2. After providing some evidence for this in terms of the fever (data not shown), she outlines the treatment she recommends for the boy in lines 17–25. The physician’s recommendation is framed as a recommendation of what should be done: elevating the boy’s head at night and giving him plenty of liquids. Although there is no medication being recommended, the treatment recommendation is nonetheless formatted afﬁrmatively. This section has shown that one way that both announcements of ﬁnal diagnoses and announcements of treatment recommendations are delivered is through an afﬁrmative turn design. This turn format is not the only turn format used by physicians. It is certainly the most common format when there is a bacterial diagnosis and treatment to be prescribed. In no-problem and no-treatment scenarios, it is not uncommon for the physician to offer no comment on the diagnosis at all, particularly if online comments were delivered during the examination. However, many physicians do go on to formulate ﬁnal diagnoses. Because, as we saw in chapter 4, this is an area where parent resistance can be consequential, when physicians deliver diagnoses of no-problem conditions, it is important to understand whether and how turn design affects parent behavior. We now turn to a second primary turn format for diagnosis deliveries and treatment recommendations. Negative Announcements Both diagnoses and treatment recommendations can be formulated to announce the action negatively: what the condition is not or what is not recommended. With respect to the diagnosis, this may take the form of ruling out either a single diagnosis or a class of diagnoses. In Extract 7.9, a “bacterial infection,” as a class of infections, is ruled out (lines 14–15). Although the ruling out is part of a response to the naming of a problem following the physical examination, it nonetheless offers a ruling out diagnosis that in turn the parent can deal with as such.

168

PRESCRIBING UNDER PRESSURE

(7.9) 383507 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

MOM:

Joseph’s so wa^:rm. I mean: (.) you know that’s not usually how he i^s. MOM: .hh Ay^! DOC: Uh:m (.) MOM: Eh! (.) DOC: But you haven’t measured uh f- you’ve checked him:, and he hasn’t had a fever, MOM: No:, n[o (fever but) I gave him= INF: [((screaming throughout)) MOM: =Pedi[acare today:_ DOC: [Why don’t- He may be- His temperature may be a little bit more than usual?, (0.5) but -> (0.4) uhm (0.5) >y’know< I don’t see any -> bacterial infection. MOM: Oh my:_

Most commonly, single diagnoses are ruled out, as we can see in Extract 7.10 in lines 1–2. Here, the physician rules out any infection and then goes on to speciﬁcally rule out ear and throat infections by stating that they are “ﬁne”. There is no afﬁrmative statement here about what the child does have, although she does go on to account for the sore throat with the cough. That the physician is done (at least for the moment) is evidenced by her just subsequent move into treatment recommendation at line 8. (7.10) 302501 1 2 3 4 5 6 7 8

DOC: MOM: DOC: MOM: DOC: DOC:

So with Clarissa right now she doesn’t have any infection. Ears and throat’s: ﬁne. Okay. So=uhm:: she just has- you know this- the throat hur:ts because of the coughing. [Uh huh [(an’)/(it) usually irritates the throat. I’ll give her another coughing formula the same- an:’=

As we saw before, the announcements can vary in terms of epistemic certainty and other dimensions of stance. For instance, whereas the diagnosis in Extract 7.10 is delivered without mitigation or downgrading as simple fact, the diagnosis in Extract 7.11 includes “I’m not really worried about” as opposed to the simple factual statement “She doesn’t have”. Here, the physician reports on her physical examination ﬁndings regarding the lung examination and then rules out a diagnosis of bronchitis before moving on to a treatment recommendation. (7.11) 160706 1

DOC:

I hear- I don’t hear the mucus sitting in there uh lot,

PHYSICIAN BEHAVIOR THAT INFLUENCES NEGOTIATION

2 3 4 5 6 7 8 9 10 11

MOM: DOC: MOM:

MOM: DOC: MOM:

169

I mean I get a sense (th’) it’s rattling a little bit but Not a whole bunch, Mm hm, So I’m not really worried about a bronchitis; °Mkay_° .hh Uhm (1.0) But she is uh little- taking a little to what we say exhale, to get the air ou:t, ((Nods))/(.) So I’m gonna help her out with a different cough syrup. °Mkay.°

In the Metro data, ruled out diagnoses comprised 18% of ﬁrst diagnoses and 21% of all diagnoses. Like negative diagnoses, treatment recommendations are sometimes delivered against a class of treatment, though most commonly they rule out a particular treatment. As an example of the former, we can look at Extract 7.12. (7.12) 100114 1 2 3 4 5 6 7 8 9 10

DOC:

Yeah her lungs sound great. (0.5) DOC: .h So- I’m pretty sure this is all just viral. (.) DOC: What I will d[o is uh culture of her throat just to be= MOM: [Okay. DOC: -> =uh hundr’d percent su:re_ [but I don’t think we need to= MOM: [(Okay_) DOC: -> =put her on any medica[tion_ MOM: [Okay.

Here the physician moves from an afﬁrmative diagnosis in line 3 to a test recommendation (lines 5/7) and to her treatment recommendation, “I don’t think we need to put her on any medication_” (lines 7/9). In contrast to the afﬁrmatively designed recommendation, in this example the physician announces what she recommends against (“any medication”) rather than asserting what should be done. As mentioned in previous chapters, when physicians use terms like “medicine,” “medication,” or “treatment,” parents routinely understand them to be indicating antibiotics, and physicians generally appear to be intending that reference. Conversely, when physicians talk about symptomatic treatment, they are indicating that they will not be prescribing. It does though here suggest not just “Not X” but no treatment at all. It is much more common, though, that physicians recommend against particular medications, with antibiotics being the most common treatment that is recommended against.3 Of course, this is most common when the diagnosis is not bacterial. In the Metro data, among cases with an initial nonbacterial diagnosis, this type of treatment

170

PRESCRIBING UNDER PRESSURE

recommendation format occurred 16% of the time (Mangione-Smith, Elliott, Stivers, McDonald, & Heritage, 2006). We can see an example of this in Extract 7.13. (7.13) 150607 1 2 3

DOC: -> But in the meanti::me no:: antibiotics or anything yet. DOC: Okay?, MOM: Yeah.

In line 1, the physician names antibiotics as an instantiation of medication that is unnecessary at this time (with “or anything”). Also note that when acceptance of the treatment recommendation is not immediately forthcoming after line 1, it is pursued by the physician in line 2. This displays, at minimum, the physician’s orientation to the treatment having been completed at line 1. This section has shown that both diagnostic announcements and treatment recommendations can be designed negatively by ruling out a diagnosis or treatment possibility. If one goal of pediatrician-parent interactions in these contexts is to minimize parent resistance to no-problem diagnoses and nonantibiotic treatment recommendations, then the next question becomes, Do these different turn designs make a difference for how parents receive diagnoses and treatment recommendations and whether they resist? The next sections will examine these issues. Parent Responses to Diagnoses In chapter 4, we observed that oftentimes diagnoses are not responded to. So unlike following treatment recommendations, following diagnoses, only active resistance is notable. Silence following a diagnosis does not have the same import as it does following a treatment recommendation.4 These data show no statistical association between a negative diagnosis and parent diagnosis resistance. One account for this is that negative diagnoses in isolation are relatively rare. Thus, parents may view them as generally on their way to a positive diagnosis: a feature that is different from treatment recommendations in no small part because there is no normative orientation to response as relevant following an initial diagnosis. Despite this lack of distributional evidence, diagnosis resistance is nonetheless arguably engendered by a ruled out diagnosis. We can see an example in Extract 7.14. Following an initial negative diagnosis, the parent is resistant. She initiates a sequence that delays the physician’s progress to the treatment recommendation. Initially, this appears to resist the implied nontreatment, but as the mother elaborates, she appears to be resisting the diagnosis in that she goes on to say that her son is not himself. The physician treats the mother as quarreling with the diagnosis and, on the one hand, works to legitimize the mother’s claim that the child is ill (line 10) while, on the other hand, also defending the diagnosis with more negative diagnoses. (7.14) 383507 1 2

MOM:

Joseph’s so wa^:rm. I mean: (.) you know that’s not usually how he i^s.

PHYSICIAN BEHAVIOR THAT INFLUENCES NEGOTIATION

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

MOM: DOC: MOM: DOC: MOM: INF: MOM: DOC:

MOM: DOC: DOC: MOM: DOC: MOM: DOC: MOM: DOC:

171

.hh Ay^! Uh:m (.) Eh! (.) But you haven’t measured uh f- you’ve checked him:, and he hasn’t had a fever, No:, n[o (fever but) I gave him= [((screaming throughout)) =Pedi[acare today:_ [Why don’t- He may be- His temperature may be a little bit more than usual?, (0.5) but (0.4) uhm (0.5) >y’know< I don’t see any bacterial infection. Oh my:_ [so what do you think I should do to uhm_ [So_ (°Don’t kick tha::t?,°)((to older child)) keep him_ hh (cuz) I was so worried about him cause he’s just not him. he’s kinda .h He’s sick. He- he is si:ck. I- do not dispute that. [(but that-) [hah=hah=hah=hah=hah_ Is it an ear infection or a pneumonia or [(something sinus,) I don’t see that. [(lookit) there’s no ear infection. I don’t see any of that. ((Knock on door; DOC answers it))

Earlier, it was claimed that although negative diagnoses are not necessarily clearly resisted more frequently than afﬁrmative diagnoses, they do, by design, make resistance easier. This is probably because, in stating what the child does not have, the physician implies that the child is not ill and thus risks delegitimizing the visit. The same problem occurs in the treatment recommendation. However, there, because parent acceptance is due immediately upon completion of the recommendation, we can observe that negative treatment recommendations are more likely to be resisted. Parent Resistance to Recommendations against Particular Treatment Although diagnosis resistance is nearly as common as active treatment resistance, treatment resistance is much more common when passive treatment resistance is taken into account. In addition, it arguably represents a bigger problem in that parents are more likely to persist with resistance and escalate in the context of treatment discussions because physicians will not, as previously shown, move to the next activity without parent acceptance.

172

PRESCRIBING UNDER PRESSURE

Parents are more likely to actively resist a treatment recommendation if its ﬁrst component involves a ruling out rather than an afﬁrmative recommendation. This is supported both quantitatively and qualitatively. Quantitatively, in the Metro data, parents were 24% more likely (39% vs. 15%) to resist a physician’s treatment recommendation when physicians ruled out the need for antibiotics than when they formulated the nonantibiotic treatment recommendation afﬁrmatively (p <.001) (Mangione-Smith et al., 2006). Qualitatively, this pattern is very visible, as is the same pattern with passive resistance. As a ﬁrst example, we can look at Extract 7.15, which we previously analyzed as an example of treatment resistance in chapter 5. (7.15) 322803 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

DOC: DAD: DOC: DAD: DOC: DOC: DAD: DOC: DAD: DOC: DAD:

->

-> -> -> ->

DOC: DAD: -> -> DOC: DAD: -> DOC: DOC: DAD: -> INF: DAD: -> DOC: DAD: -> DOC: DAD: DOC:

I th:ink from what you’ve told me (0.2) that this is pro:bably .h uh kind of (0.2) virus infec[tion, [Uh huh, (0.4) th:at I don’t think antibiotics will ki:ll, (0.2) Well- [( ) [Thee other>Go=ahead_< Yeah. .hh ( ) I had it- I had thuh symp[toms [I understand. Three weeks ago. [Right. [.hh An:d I’ve been taking thuh over the counter cough [( ) [(Good_) Uh s- ( ) coughing syrup, Nothing take away .hh Especially my sor- my [th- my throat was real= [Mm hm =sore [for (awhile- et- that) w:eek. [Uh huh °Right,° an:d (.) I start taking thuh antibiotic (0.5) eh he ((cry)) Yesterday. Right, And it (.) seemed to take care of the problem. [(Well) that’s why we’re doin’ a throat [culture. [( ) [Yeah. [is TUH SEE if they need antibiotics.

Following the diagnosis (lines 1–2), the physician offers a recommendation against antibiotics (line 4). Passive treatment resistance begins immediately subsequently (line 5), and then the father escalates to active resistance. “Well” projects disagreement (Pomerantz, 1984). Following the physician’s “>Go=ahead_<” (line 8), the

PHYSICIAN BEHAVIOR THAT INFLUENCES NEGOTIATION

173

father continues with active resistance in lines 9/11/13–14/16–17/19/22/24/26. This constitutes resistance because across this stretch of talk, the father builds a case for why antibiotics—the treatment the physician recommended against—may be effective: They were effective in treating a similar illness that the father recently had. A second case is shown in Extract 7.16. Here, the physician offers a recommendation against antibiotics in line 1. She pursues agreement to this proposal in line 2 and receives this in line 3. But immediately following the agreement, the mother inquires about further testing of the child, which the physician agrees with. Although the sequence is possibly closed at this point, and the physician initiates a move to closing in line 8 with an inquiry to the child about the visit not being “so scary,”, in overlap, the mother initiates active resistance (line 9). (7.16) 150607 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

DOC: DOC: MOM: MOM: DOC: MOM: MOM: DOC: MOM: -> -> -> -> DOC:

But in the meanti::me no:: antibiotics or anything yet. Okay?, Yeah. Did you want her tuh get that ultra sound? Yes I want her [to get thuh ultra-sound too. [Okay. Okay. ( ) ((to girl)) [Alright see: [( ) not so scary, [So [Should we- bring her? Eh- Should we bring her i:n? See e- my husband gets just rea:l insistent that(.) “there’s some’in’ wrong with her” because she keeps getting sick. So: She’s been sick for (how is-) well last time was thuh urwas thuh kidney. (Or well-) .hh both time was urin- urinary problem ri:ght?

In line 9, the mother begins her turn with another inquiry about future action: “Should we bring her i:n?”. This is then immediately accounted for with an upgrade to resistance by invoking her husband (lines 10–12). As resistance, the parent claims that what the physician has offered them up until this point cannot be defended to a third party. The physician addresses the mother’s action as resistance by restarting an investigation of the child’s illness. As mentioned earlier, it is not only how but also whether a parent responds that shapes whether and how physicians continue with treatment recommendations. We can see this exempliﬁed in Extract 7.17. (7.17) 170812 1 2 3 4 5

DOC: 1-> As you know they’re viral infections, so there’s 1-> no point in any a- any ant- antibiotics. (0.5) DOC: 2-> Simply control thuh cou:gh with .hh whatever 2-> your favorite cough medicine is,

174

PRESCRIBING UNDER PRESSURE

The physician rules out antibiotics in lines 1–2. It is only following a lack of uptake of this recommendation that a recommendation for other treatment is offered (lines 4–5). This pattern comprised the majority of the cases where a recommendation against treatment was followed by a recommendation for particular treatment. We have seen evidence that ruling out antibiotics is more likely to engender parent resistance than afﬁrmatively recommending nonantibiotic treatment. But the question remains as to why that would be the case and what this reveals about parents’ concerns in the medical visit. One analysis is that, like negative diagnoses, ruling out antibiotic treatment undermines the legitimacy of the visit. Additionally, by suggesting no treatment, the physician implies that the condition is untreatable. After all, these children do have illness symptoms for which the parents are looking for a solution. I will argue that a sufﬁcient treatment recommendation necessarily involves advice about a solution for these symptoms. By deﬁnition, treatment recommendations that rule out treatment do not involve this advice. In the next section, I discuss what a sufﬁcient treatment recommendation looks like and provide evidence that this is what underlies the pattern discussed earlier that shows more parent resistance to ruling out treatment recommendations. Sufﬁcient Treatment Recommendations As we have discussed earlier, when parents arrive at a decision to bring their child to the pediatrician, this generally means that they no longer feel able to self-manage the child’s problem. They turn to pediatricians for expertise and a solution to the problem. The treatment recommendation represents the latter. Here, we consider what constitutes “a solution” or a “sufﬁcient” treatment recommendation. We will see that parents generally treat a treatment recommendation as insufﬁcient (in the sense of not providing a solution) if it (1) fails to provide an afﬁrmative action step, (2) is nonspeciﬁc, or (3) minimizes the signiﬁcance of the problem. By contrast, a sufﬁcient treatment recommendation asserts a speciﬁc next action step afﬁrmatively and treats it as a wholly legitimate recommendation, thereby treating the patient’s problem as legitimate. Adult patients in the acute care context display an orientation to receiving a treatment recommendation as a relevant next activity following a diagnosis delivery (Robinson, 2003). One type of evidence for this is that when treatment is not immediately forthcoming, patients pursue a treatment recommendation. In these data, this pattern is also present: Parents ask about and thus pursue a treatment recommendation if no treatment recommendation is offered. But what is more prominent is that some treatment recommendations proposed by physicians are responded to by parents as insufﬁcient. So the absence of a treatment recommendation may be part of a more signiﬁcant normative orientation by parents to the relevance of having a sufﬁcient treatment recommendation offered by physicians. This section will be concerned with providing evidence in support of this claim. By deﬁnition, all cases of recommendations against antibiotics fail to provide an afﬁrmative next action step, and I argue that it is for this reason that they are more likely to be resisted. Evidence that it is the lack of an afﬁrmative action step that is at issue comes ﬁrst from the observation that parents pursue this dimension of treatment following recommendations against particular treatment. We can see this in Extract

PHYSICIAN BEHAVIOR THAT INFLUENCES NEGOTIATION

175

7.18. Following the physician’s recommendation against an antibiotic (line 5), the mother inquires about a medication that she can provide (line 9). (7.18) 322708 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

DOC:

.hh So: I think it’s just (.) one uh thuh (.) thi:ngs: kids get one thing after another sometimes, MOM: M[kay. DOC: [Nothing serious here, DOC: .mh Nothing that I can see that an antibiotic would help, MOM: Okay; (.) DOC: [Uh:m MOM: -> [So uh:m (.) should I continue with thuh Tyleno:l? er_ DOC: Tylenol if he’s uncomfortable. (.) DOC: [With fever ‘n (0.2) headache. MOM: [(‘kay) DOC: or anything [like that. MOM: [(Okay.)

Although in some cases parents actively advocate for antibiotics (as shown in chapter 6), in other cases, resistance and other such behaviors may be rooted in a concern for receiving speciﬁc treatment. In this case, we can observe that the parent does not resist the lack of an antibiotic per se but displays her concern, and thus the root of this resistance, to be the lack of an afﬁrmative next action step. We can see that the mother does not resist the lack of a prescription because Tylenol is an over-the-counter recommendation. Rather, she pursues a concrete afﬁrmative recommendation. That it is not always “a prescription” that parents are looking for is further supported by Extract 7.19. After the physician recommends against one over-the-counter cold medication (lines 1–2), the mother inquires about another form of nonprescription treatment that she could offer her daughter (line 8). (7.19) 170802 1 2 3 4 5 6 7 8 9 10 11

DOC: -> Uh::m o- nl- unfortunately we probably can’t give her -> stuff .hh like Sudafed. (.) DOC: Because that’d crank her blood pressure up_ an’ we don’t need tha:t. MOM: Right. (1.0) MOM: -> Okay: so give her Tylenol?,= DOC: =Yeah. (0.2) DOC: for discomfort.

176

PRESCRIBING UNDER PRESSURE

The parent displays her concern to be not with the denial of Sudafed or with the failure of the physician to recommend antibiotics but rather with what the mother can do for her child. Note, too, that in all cases, the parents’ questions are designed to prefer a “yes” answer and so to prefer the proposal of an afﬁrmative course of action. Parents appear to be oriented not primarily to the relevance of receiving a treatment recommendation but rather to receiving an afﬁrmative next action step. These cases show that parents are oriented to a minimally sufﬁcient treatment recommendation as necessarily including such an action step, and without an afﬁrmative next action step, parents treat the recommendation as insufﬁcient. Parent Responses to Recommendations for Particular Treatments Generally, when physicians recommend nonantibiotic treatment, if they do it afﬁrmatively, it is not resisted. This pattern is illustrated in Extract 7.20. (7.20) 150614 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

DOC: DOC:

DOC: 1-> DOC: 1-> DOC: DAD: => DOC: 2-> 2-> BOY: DAD: => DOC:

Looks like he has a co:ld,=h It’s just uh virus, not uh bacteria;=his lungs sound really good,=it’s just .h all irritation up here;= =(and)/(that) he’s coughing thuh- .h throat looks uh little red_ but there’s no puss or anything; .hh ear is just uh little (.) slightly pi:nk and .h it’s uh combination for with thuh stuffy no:se_ .hh so=w:e have=to .h clear thuh nose. Ya know like ((exhaling noise))/(0.2) reduce thuh congestions that will help him uh lot. [.hh [>Okay.<= =An’ I’m gonna give you some cough medicine that has some decongestant in it. ((whispering))/ ((DAD nods)) M[kay. [That will help him out.

Following a diagnosis of a cold (line 1) and the explication of the evidence for that diagnosis (lines 2–7), the physician afﬁrmatively recommends nonantibiotic treatment: cough medicine. Besides being formatted afﬁrmatively, the physician also formats her recommendation speciﬁcally—she recommends a type of cough medicine (lines 13–14). Earlier, I mentioned that speciﬁcity was the second criteria of a sufﬁcient treatment recommendation. Although the cough medicine may or may not turn out to be prescription, what appears to be important for whether resistance is likely to be engendered is that the recommendation is both speciﬁc and for a concrete next action step. Here, the cough medicine is not named, but the physician states that she is “gonna give you some” and speciﬁes that it has “some decongestant in it.”. Both of these aspects of the turn indicate that she has in mind a particular medication, and in

PHYSICIAN BEHAVIOR THAT INFLUENCES NEGOTIATION

177

this way she is being speciﬁc in her recommendation. This is subsequently accepted both visibly (line 15) and verbally (line 16). A similar example was shown in Extract 7.8. Earlier, we saw that the physician diagnosed a cold and recommended elevation and ﬂuids in an afﬁrmative formatted treatment recommendation. The physician’s treatment recommendation offers the parent concrete afﬁrmative action steps. Although she slightly minimizes her recommendation with “just” in line 18, in general, her recommendation is formulated as an unmitigated positive announcement of what needs to be done, in spite of the fact that the recommendation does not involve antibiotics. There is a moment of delay in acceptance (line 26) that leads to the physician’s use of an increment to recomplete the recommendation (line 27) (Schegloff, 2001), but there is both acknowledgment at line 28 and acceptance (“Oh okay”) following the (possibly) responsive explanation of the illness and justiﬁcation for the treatment (data not shown). A similar case is shown here in Extract 7.21. (7.21) 100115 1 2 3 4 5 6 7 8

DOC:

MOM: DOC: MOM: DOC: MOM:

If her uhm if her cough: does really become croupy tonight, (0.4) just be sure you use the coo- in fact you can use cool mist hu[midiﬁer anyway in the room. [°Okay.°/((nodding)) That will help her feel more comfortab[le. [((nods)) Keep thuh nose open. Okay.

Here again, the physician delivers her treatment recommendation beginning in line 1. It is initially projected as a conditional with “if”, but this is revised in line 2 with “in fact” and then in line 3 with “anyway”. The recommendation is afﬁrmative and speciﬁc—a cool mist humidiﬁer. These cases are not only afﬁrmative. In them, physicians satisfy all of the conditions outlined earlier for sufﬁcient treatment recommendations: They are afﬁrmative, speciﬁc, and nonminimized treatment recommendations. Because recommendations for treatment by deﬁnition satisfy the criteria of being afﬁrmative, this may explain why they are less likely to be resisted. But the other requirements for sufﬁciency are also important. Evidence for this is that when recommendations for treatment are resisted, they typically fail on one of the latter two dimensions: Typically, either they involve a vague or nonspeciﬁc treatment recommendation, or the physician minimizes the treatment recommendation. We can see an example of this in Extract 7.22. The physician recommends against antibiotics and for medicine for the girl’s stuffy nose, cough, and fever, but he fails to offer any speciﬁc treatment recommendation. At line 19, the mother resists the recommendation with “just over the counter?,”; in response, the physician offers to “check off my favorites for you,” and goes on to provide a speciﬁc treatment plan. This revision is accepted by the parent (line 25), providing evidence that a physician’s speciﬁcity is important to parents in their orientation to what constitutes a sufﬁcient treatment recommendation.

178

PRESCRIBING UNDER PRESSURE

(7.22) 383412 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

DOC:

.hh Which is goo:d that means that she doesn’t need any antibiotics. (.) DOC: because this is probably, (.) caused by uh virus, DOC: .hh [an:d=eh as you may kno:w antibiotics don’t kill= MOM: [Mm hm, DOC: =viruses. [soMOM: [Mm hm, DOC: .hh uh: and this is- (p)/(.) uh lotta kids this, -> [(i’s) pretty common;=so .hh treatment will be: MOM: [Mm hm, DOC: =you know medicine- that’re gonna make her comfortable and treat her symptoms. so .hh you c’d get her medicine that’s gonna make her nose less stuffy an’ °make it° less runny, an’ uh medicine for thuh cou:gh?, DOC: .hh An:d=uh you know something for thuh fever like (you’ve)/(we’ve) been doing, DOC: .hh [uhMOM: [Just over thuh counter?, (.) MOM: jisDOC: Over thuh counter_ yeah, DOC: I have some here_ I’ll- I’ll check off my favorites for you, MOM: £Okay,£

Similarly, recommendations for treatment are also more likely to be resisted if they fail on the third dimension of recommendation sufﬁciency: minimization. We can see an example of this in Extract 7.23. The physician does not initially receive acceptance when he recommends against antibiotics. He proceeds to recommend Robitussin afﬁrmatively, but he presents his overall recommendation in a downgraded manner with “just”. In contrast with Extract 7.8 where one component of the treatment was downgraded, here the “just” downgrades and minimizes the entire recommendation. The recommendation here implies a contrast between stronger treatment and Robitussin, thereby treating his recommendation of Robitussin as a minimal sort of treatment. The passive resistance in line 4 is addressed by the physician with one account for the lack of uptake—that the parent has tried this medicine already and found it to be inadequate. (7.23) 151213 1 2 3 4

DOC: DOC: DOC: -> =>

She probably doesn’t need antibiotics. I mean uh: most of these are viral. .h uhm hh .h I’d- just give her Robitussin, (0.7)

PHYSICIAN BEHAVIOR THAT INFLUENCES NEGOTIATION

5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

179

DOC: -> MOM: => => DOC: MOM: => DOC:

Have you- (.) given her any of tha[t, [Uh:m I’ve given her the Tylenol: an’ stuff but .h she just seems so uncomfortable when she cou:ghs; that [ya know_ [Uh huh_ I feel like I need to give her something uh little bit Yeah. .h Right. well I- (s-) she deﬁnitely needs cough medicine tuh thin out thuh mucus. DOC: so that when she does cough (that) it’ll come up; (0.5) DOC: .h [an’ thenMOM: [And then her congestion? DOC: Pard’n me, MOM: => Also her congestion. (.) DOC: Yeah. (.) DOC: -> Well, .hh they have uh Robitussin PE: that (could-) MOM: Sh sh sh sh sh [sh sh sh ((to child)) DOC: [take care of tha:t. MOM: => Oh okay.

In response, the mother resists the treatment further by asserting that her daughter is uncomfortable and needs “something a little bit”, which idiomatically suggests that the likely next term was “stronger”. The physician responds to the parent’s resistance by agreeing that cough medicine is necessary, but he does not provide a speciﬁc responsive recommendation that indicates an upgrade on his prior recommendation. The mother then again resists the treatment by inquiring about treatment for her daughter’s other symptom: congestion. With this inquiry, similar to other examples shown earlier, the mother conveys her perception that the physician’s prior suggestion was insufﬁcient. In response, the physician offers a speciﬁc recommendation “Robitussin PE” (line 22), and this is accepted with “Oh okay.” in line 25. In the treatment recommendation phase of the visit, offering prescription treatment such as antibiotics is one way that physicians show parents that they were correct in having sought medical attention. Physicians legitimize the child’s visit through the prescription of medication. As mentioned earlier, offering no treatment is, conversely, understandable as delegitimizing the visit. When a treatment is afﬁrmatively and speciﬁcally recommended, this may do work to counter this problem. But when physicians generalize to “whatever your favorite cough medicine is” or orient to the treatment as minimal or arbitrary, this creates or perhaps intensiﬁes the problem parents face when the physician offers no treatment recommendation or recommends against a particular treatment because with a minimized or vague treatment plan (at least in the URI context), physicians both delegitimize the patient’s visit and fail to provide a solution, thereby underscoring that the condition is not treatable. This section has argued, then, that treatment resistance is best avoided through the use of sufﬁcient treatment recommendations: recommendations that are afﬁrma-

180

PRESCRIBING UNDER PRESSURE

tive in design, speciﬁc, and not minimizing. Thus, this provides a second resource physicians can make use of in securing parent acceptance of a nonantibiotic treatment recommendation. Diagnosis Delivery Although diagnosis resistance was not clearly associated with negative diagnoses, what can be observed is that afﬁrmative diagnoses, like afﬁrmative treatments, can nonetheless fail to be optimally designed. Like treatments, parents are more likely to resist diagnoses that are minimized. This suggests that whereas treatability is primarily at issue with respect to treatment recommendations, the maintaining of legitimacy may be more of an issue with respect to diagnoses, though both are clearly issues throughout the visit. This section examines cases involving diagnoses that are minimized to provide evidence for the claim that minimized diagnoses are more likely to engender resistance than non-minimized but still afﬁrmative diagnoses. As an initial example, we can look at Extract 7.24. The physician afﬁrmatively identiﬁes blisters and cold sores at the end of a physical examination and then moves into an afﬁrmative but minimized diagnosis in lines 7–8. (7.24) 1126 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

DOC: MOM: DOC: MOM: DOC: -> -> MOM: => DOC: DOC: DOC: MOM: => => DOC:

MOM: MOM:

°Yeah:.° You know actually what those a:re °pr=h° .hh are primarily blisters back there. Yea:h? It’s almost like she’s got cold sores in thuh back of ‘er throa:t. (Oh:[::.)/(Aw:::.) [And u:sually that’ll go along with this just being viral. (.) [Really.= [#er-# =Y:eah. .hh One ‘v thuh teachers told me it might be stre:p so:[:_ [.mlk Yeah we are starting to see some strep so I’m gonna culture just in case .hh she’s got both going on at the same ti:me but- .hh when you see: (you know)/(any uh) those #uh:# (thuh)/(that) white stuff you see back there is- is really not: like pus pus but it’[s ya know like she’s got blisters n’ [Oh yeah:_ Oh:::.

The minimizer “just” treats the illness as less signiﬁcant than it might be if it were not viral. And in response, the parent ﬁrst resists with a newsmark “Really.” Following

PHYSICIAN BEHAVIOR THAT INFLUENCES NEGOTIATION

181

reconﬁrmation, the parent goes on to challenge the physician’s diagnosis by counterproposing a bacterial diagnosis: strep throat. Another way in which this diagnosis is presented as a minimal or less signiﬁcant one is that whereas bacterial illnesses are typically named “strep throat” or “sinusitis” or are labeled as “infections” as in “sinus infection” or “ear infection”, viral illnesses, are often not called infections. This further delegitimizes the visit because there is greater risk that the parent will hear the physician to be implying that the child is not actually ill. Another example of a minimized diagnosis is shown in Extract 7.25. Here, the physician announces the diagnosis afﬁrmatively, but similar to 7.24, it is minimized in “uh little viral col:d,”; this minimization is accentuated with the contrastively prefaced “but it doesn’t look like anything much,” which is resisted with the newsmark “Really,” in partial overlap. (7.25) 104 [shown earlier in 4.4] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

DOC: -> -> DOC: -> MOM: => DOC: DOC: MOM: DOC: MOM: => => DOC: MOM: => DOC: MOM: => DOC:

.hh Uh:m, I think probably he’s- ya know has uh little viral col:d, His nose is uh little stuffy, .hh Uh:m (.) tl=His throat does look uh little bit re:d [but it doesn’t look like any[thing ] much,=^Yeah, [Right. [Really,] An’ I think thuh redness is really -again like I say from dripping down thuh ba:ck, .hh Uh:m his chest is perfectly clear. There’s nothing in his lungs at all[:. [Okay, .hh An’ I- ya know (you see) thuh fevers have gone dow:n, .hh uh:m_ I was just concerned cuz he’s been so cranky an’ I thought well [there must be something botherin’ ‘im= [^Well:=[that I can’t: =[Well (that) could be. I mean-= =see_ What will happen.
After the physician details justiﬁcation for her claim that there is no problem, the mother offers more substantive resistance with “I was just concerned cuz he’s been so cranky. . .”. Even when physicians do not actively minimize the presentation of a viral illness, it is vulnerable to being heard as minimized because of the absence of a true diagnostic label. This is shown in Extract 7.26. (7.26) 161303 1 2

DOC:

It’s- It’s=uh: .h It’s something that’s - b- been going through the schoo:ls the last two tuh three wee:ks. like

182 3 4 5 6 7 8 9 10 11 12 13

PRESCRIBING UNDER PRESSURE

MOM: MOM: DOC: DOC: MOM:

crazy:, .h ^I don’t know if she had the same thing or not. .hh It’s possible.=Some kids it’s very mi:ld an’ then others get it an’ uh ( ) they’re really hard like this [but- .h It’s uh c- it’s uh virus, It’s uh kind of [Mm: hm, ﬂu:, It- h- he doesn’t seem to have an infection? (.) .hh Uh yi=y: it’s uh viral infection. [<See (what) the problem is .hh you know you think . . . [A viral infection.

The critical point here is that the diagnosis is presented as “something that’s - b- been going through the schoo:ls” and then “it’s uh virus” and “it’s uh kind of ﬂu:,” which is as close to a diagnostic label as the physician gets here. And the physician treats this as indicating that the physician found no infection (line 9): something that she treats as problematic. It is interesting that the cutoff of “c-” in line 6 may represent the initiation of repair on what would have been “uh cold”, in which case “uh virus” upgrades the diagnosis to a medical category. But even so, it is problematic. These cases offer evidence that diagnoses, like treatment recommendations, are less likely to be resisted if they are delivered in such a way that they do not make light of the child’s illness or the treatment recommendation but rather underscore the legitimacy of the parent’s decision to bring the child in for medical evaluation. Thus, just as the last section asserted the importance of formulating the treatment recommendation afﬁrmatively, speciﬁcally, and without minimization, this section argues that diagnoses will be less likely to be resisted if they are delivered without minimization. The principles of supporting visit legitimacy, addressing treatability, and forecasting a nonantibiotic outcome can be handled successfully in ways that are quite different than have been outlined here. Extract 7.27 provides an illustration of this. What appears to be more important than the exact format of the counseling is that the basic principles are oriented to by the physician. In this case, the physician packages his diagnosis and treatment recommendation together, projected as a multiunit explanation from the outset with “There’s some good news, and there’s some bad news.” (7.27) 211701 1 2 3 4 5 6 7 8 9

DOC:

MOM: MOM: DOC: MOM: DOC:

Okay. .h There’s some good news, and there’s some bad news. Okay, uhm the good news is she doesn’t look like she doesn’t look like she has any kind of a bacteria i[nfection.=okay,=So she doesn’t need any antibiotics. [Okay Okay, Her ears look °M[m hm,° [Okay so no ear infection, .h doesn’t look like uh strep

PHYSICIAN BEHAVIOR THAT INFLUENCES NEGOTIATION

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 42

MOM: DOC:

MOM: DOC: MOM: DOC: MOM: DOC: MOM: DOC:

MOM: DOC: MOM: DOC: DOC:

MOM: DOC:

MOM: DOC: DOC: MOM: DOC: MOM:

183

throat at a:ll, Okay. (.)] [(( gesturing like nose picking ))] [as much as possible. Mkay, cuz only time will= [Mm hm, =actually=uh get rid of the viru[s also. [Okay.

Relative to the concern of minimization, what can be seen here is that the no-problem diagnosis and nonantibiotic treatment recommendation are explicitly identiﬁed as good news. The illness the child has is labeled “bad news,” and this validates the visit and the parent’s decision to come. The illness not only is not minimized but also has its signiﬁcance heightened by calling it an infection. Moreover, no antibiotics is also explicitly identiﬁed as good news, whereas the offering of ﬂuids is not identiﬁed as bad news but provided as an afﬁrmative, speciﬁc, and nonminimized solution to the infection. This case suggests that a range of formal variations can be used by physicians in successfully dealing with parents in the delivery of no-problem diagnoses and nonantibiotic treatment. What this section has asserted is that in the counseling

184

PRESCRIBING UNDER PRESSURE

phase of the visit, diagnoses and treatment recommendations will be less likely to engender resistance if they validate the child’s illness and provide a solution to the child’s problem.

Discussion In the context of routine viral upper respiratory infections, if physicians are to work to avoid inappropriate prescribing, they must attempt to also avoid parent interactional behaviors that pressure them for antibiotics. This chapter has argued that physicians who forecast a no-problem diagnosis and a nonantibiotic treatment to parents may help to get parents to accept this outcome, especially if this outcome is managed in such a way that the legitimacy of their medical visit is maintained and the treatability of the child is acknowledged. This chapter discussed several resources for forecasting a no-problem diagnosis and focused on three resources that both forecast this news and work to maintain the legitimacy of the visit and/or address the treatability of the child. First, online commentary delivered during the physical examination forecasts a no-problem diagnosis and a no-treatment recommendation. This practice not only forecasts the news but also, by providing parents with insight into how a noproblem diagnosis was reached, assists parents in accepting this outcome. Second, when physicians deliver a viral diagnosis with mitigation, parents were shown to be more likely to resist the diagnosis. Diagnoses that are delivered afﬁrmatively and do not minimize the child’s illness were argued to be optimal in securing parent acceptance of the nonantibiotic treatment outcome. Finally, treatment recommendations, like diagnoses, can be formulated positively or negatively. This chapter demonstrated that parents are less likely to resist a treatment recommendation that is formulated afﬁrmatively, provides a speciﬁc next action step or solution to the child’s problem, and does not minimize this solution. In this way, this resource both addresses the legitimacy of the visit and provides a solution to the problem the parent has sought help for—treatability. The prior chapters in this book demonstrated the important role that parents play in these pediatric encounters in terms of shaping the visit outcome. Although physicians “hold the pen” and thus, in the end, are the party truly accountable for inappropriate prescribing, this outcome has been shown to be very much an interactional product. Chapters 2–6 showed a number of ways in which parents mobilize interactional resources that are likely—for a variety of different reasons—to have an impact on the prescribing decision. This chapter showed several interactional resources that some physicians make use of to successfully retard parent resistance and thus steer clear of an interactional environment that is more likely to lead to inappropriate antibiotic prescribing.

8

Conclusion

W

hen physicians in developed countries like the United States prescribe antibiotics to children who they believe have a viral upper respiratory tract infection, they generally know that they are committing a medical error. But what we have seen throughout this book is that the diagnosis and treatment of upper respiratory tract infections is not simply the result of applying a clinical algorithm. Rather, the diagnosis and treatment are arrived at in and through a moment-by-moment interaction with the parents and children. Recall Mangione-Smith and colleagues’ ﬁnding that where physicians perceive that parents expect an antibiotic, they are signiﬁcantly more likely to prescribe it for a viral condition (Mangione-Smith, Elliott, Stivers, McDonald, & Heritage, 2006; Mangione-Smith, McGlynn, Elliott, Krogstad, & Brook, 1999). Thus, although it may sound rather straightforward to say “no antibiotics” for a viral upper respiratory tract infection, a close look at pediatric interactions like those shown throughout this book suggests that it is actually much more difﬁcult to deny a sick child and the parents who simply want their child to feel better. So, at the root of a large-scale global health problem, as well as a classic social dilemma, is a micro-level problem in social interaction.

Summary of Findings In chapters 2–5, we examined four phases of the acute pediatric visit and observed that in each both parents and physicians have interactional resources for negotiating the diagnostic and treatment outcome of the visit. These are summarized in table 8.1.

185

186

PRESCRIBING UNDER PRESSURE

table 8.1. Parent Interactional Practices That Pressure Physicians for Antibiotics Interactional Practice Candidate diagnosis presentation

Primary Phase of Use

Frequency of Use

Reason for the visit

26%

Mention of additional symptoms History taking Mention of alternative diagnoses History taking

9% 12%

Diagnosis resistance Active treatment resistance Overt pressure

17% 19% 8%

Diagnosis delivery Treatment recommendation Treatment recommendation

We ﬁrst observed that during the problem presentation, parents can take a stance toward the child’s illness as bacterial through their use of a candidate diagnosis. We saw that when parents introduce their children’s problems with a candidate diagnosis, as opposed to simply presenting their children’s symptoms, physicians treat them as seeking conﬁrmation of their view of the child’s illness but, more important, as looking for antibiotic treatment for it. Parents may use a candidate diagnosis for reasons other than communicating a desire for antibiotics: They may wish to underscore why they thought the problem required a medical visit or be genuinely concerned that the illness is serious. But physicians tend to focus on the desirability of antibiotics, as evidenced by their responses to candidate diagnoses either immediately or later in the visit. This shows that candidate diagnoses are one interactional resource through which parents can participate in and shape the diagnosis and treatment outcome. During the history-taking phase of the visit, we observed that with each question physicians ask, they display whether they are taking a stance toward that symptom as problematic or not. Thus, with each question, parents gain insight into what sort of trajectory the physician is on: toward a problem diagnosis and treatment or toward a no-problem diagnosis and nonantibiotic treatment. In this phase, we further observed that parents, particularly on hearing a question or series of questions that convey the physician to be on a no-problem trajectory, mention additional symptoms to push physicians away from their current no-problem trajectory, or they mention alternative possible diagnoses to encourage physicians toward an alternative trajectory. Again, these interactional practices may be motivated by a desire for antibiotics, by a concern that the child’s illness is more serious than the physician appears to be treating it, or by a concern that their child needs some form of help to get well and they do not know what to do. Physicians typically treat these behaviors, like candidate diagnoses, as primarily indexing a desire for antibiotics, so by mentioning additional problematic symptoms or alternative possible diagnoses during history taking, parents can also inﬂuence the visit outcome.

CONCLUSION

187

Next, when physicians deliver their diagnoses, we observed that this afﬁrmatively takes a position toward the illness as problematic or not. If the physician takes a position that the illness is nonproblematic, this action potentially threatens both the legitimacy of the child’s medical visit and the treatability of the child. Thus, the parent may, for either or both reasons, be concerned that they are not going to receive any solution to their child’s problem, though this concern may be motivated by various issues: fears of having wasted the physician’s time, because they desire antibiotics, or because they feel that the physician has not understood the child’s illness in that the child is clearly sick and in need of treatment. Their primary resource for negotiating the outcome of the visit in this phase is to resist the diagnosis. We saw that diagnosis resistance could be accomplished quite minimally because virtually any sequence-initiating action delays progress of the ongoing physical examination (in the case of online diagnosis) or to treatment (as in the case of ofﬁcial diagnoses). But parents must actively resist if they are to take issue with this phase of the visit because response is not normatively required. This phase is, in some ways, particularly interesting; it is where the physician’s expertise and authority are, at least in theory, at their highest. Empirically, the lack of a normative response to a diagnosis delivery is in line with this theoretical position. However, parents do nonetheless intrude into this domain, though they do so most commonly in ways that downplay this intrusion and rarely directly question the diagnosis. Regardless of the directness, diagnosis resistance is yet another interactional resource for shaping the diagnosis and treatment decision. Resistance during the diagnosis phase is rather different from what we observed in the treatment phase, where physicians and parents are far more oriented to a shared responsibility and domain of authority. This was evidenced by the normative organization that makes parent acceptance of a suggested treatment conditionally relevant before a physician will initiate a move to the next activity or to closing the encounter. In line with this, but in contrast with the diagnosis context, here parents who fail to respond at all are treated as resisting. Therefore, a broader range of resistance to treatment recommendations, including both passive and active types of resistance, is available to parents. This phase is also the ﬁnal phase where there is a structurally provided opportunity for visit outcome negotiation, and thus it is perhaps not surprising that the negotiation in this phase is typically the most overt, including questions about whether antibiotics are necessary. But even here, parents’ behaviors may be driven by different factors. Although a desire for antibiotics is at times plain, at other times, the same collocation of issues that can motivate behaviors at other points in the visit are analyzable here: Parents may be pushing for treatment because they feel that the legitimacy of their visit has been threatened; they may feel that their child is seriously ill and will get still sicker if not properly treated; they may feel that without medicine their child will be sick longer, that they will be out of work longer, that the child will be out of day care longer, and so on. But again, physicians tend to primarily address treatment resistance as a behavior driven by a desire for antibiotics. In chapter 6, we examined situations when negotiation is most transparent: when parents overtly lobby for antibiotics. We saw that overt lobbying took the form of

188

PRESCRIBING UNDER PRESSURE

one of four main practices that varied in their directness but all pushed overtly for antibiotics and could occur at any point in the visit. We saw examples occurring at the very opening of the medical encounter, during history taking, amid the physical examination, and through to the treatment recommendation. We also saw that diagnosis resistance or treatment resistance could at times involve overt lobbying for antibiotics, though most of the time it remained covert. Examining the more overt cases of physician-parent negotiation over antibiotics provided yet further support for the claims that true negotiation is going on, despite the fact that a vast majority of the negotiation goes on covertly. Observing that negotiation is occurring and how it is being done is only part of the story. If we want to attempt to reduce inappropriate prescribing, we must also understand what is driving these behaviors so we understand how best to reduce them. We dealt with the complexity behind parent behaviors most in chapter 7, which showed that physicians could minimize parent resistance to no-problem visit outcomes with behaviors that worked to maintain the legitimacy of the visit, forecast to the parent early in the visit that their child might not have a condition that is treatable with antibiotics, and addressed the general treatability of the child’s illness. Forecasting the “bad news scenario” that a condition is not problematic and that there will be no antibiotics helps ease parents into this outcome and may therefore help them accept it (Maynard, 2003). This may then be why behaviors such as rejecting a candidate diagnosis or addressing additional symptoms or diagnostic possibilities are useful. However, these behaviors do nothing to maintain the legitimacy of the visit. By contrast, online commentary, when used as discussed here, can work to forecast a no-problem diagnosis and a no-antibiotics treatment early in the visit— before an ofﬁcial diagnosis is reached. It also provides parents with insight into why the diagnosis is what it is and reassures parents that each possibly concerning symptom was investigated. Moreover, online commentary also give parents insight into what minor problems the child has that may have caused the child discomfort or the symptoms that led the parent to schedule the visit. Thus, online commentary can also work to maintain the legitimacy of the visit at exactly the point where there was a possible threat to its legitimacy. Similarly, when physicians formulate diagnosis announcements straightforwardly and afﬁrmatively, they avoid threatening the legitimacy of the visit. By contrast, diagnoses that minimize the illness, no matter how well intentioned, do threaten visit legitimacy. Chapter 7 took the position that physicians delivering no-problem diagnoses would do best to present the diagnosis as legitimate without minimization because this, at the very least, does not actively threaten the legitimacy of the visit any more than null ﬁndings do by deﬁnition. Finally, we observed that, when physicians deliver treatment recommendations, both negative announcements and positive announcements that are vague or nonspeciﬁc or that minimize the treatment are problematic and more likely to meet with parent resistance. This is perhaps the strongest evidence that parents are not necessarily primarily driven by a desire for antibiotics but that antibiotics embody a cure for their child’s illness, and a cure is what they are in search of. If a physician can offer them an alternative solution, they may be more willing to try it and not resist than if the physician denies them this.

CONCLUSION

189

Solutions Parent pressure and physician responses to it are reﬂections of a societal-level problem, and it is partially the society as a whole that is responsible for the current state of affairs. In chapter 1, we analyzed the use of antibiotics in the face of the bacterial resistance problem as a commons dilemma. Jared Diamond’s Collapse: How Societies Choose to Fail or Succeed (2005) offers numerous examples of how some societies destroy themselves through poor decision making, often with respect to commons dilemmas. One of the latter chapters asks the question “Why do some societies make disastrous decisions?” The problem we are discussing here is a disaster in the making: We are quickly reaching a time when a great many illnesses will no longer be curable with antibiotics, and we will probably once again risk losing the lives of especially vulnerable members of our society—children, the elderly, and the immunocompromised—to such illnesses. According to Diamond, four levels of deﬁcit cause societies to fail in solving their problems: failure to anticipate a problem; failure to perceive a present problem; failure to try to solve a perceived problem; failure to solve a problem despite attempts. In our present case, many (but by no means all) parents do not perceive the problem at any level and thus are in no position to contribute to a solution. This, then, is one level of our problem. As long as parent pressure remains a part of the equation, a problem remains at both the individual and societal levels. Thus, parents must become aware of the problem so that they can then become part of the solution. The primary means thus far for heightening parent awareness of the bacterial resistance and inappropriate antibiotic-prescribing problem has been public health campaigns (e.g., Finkelstein et al., 2001; Madle, Kostkova, Mani-Saada, Weinberg, & Williams, 2004; Perz et al., 2002; Wheeler et al., 2001). Various international and national associations have web resources for educating consumers, and research about the impact of this is going on (Madle et al., 2004). But most efforts focus on the difference between bacterial and viral conditions and what antibiotics can and cannot treat. Although this is clearly relevant, it does not entirely handle the problem for multiple reasons. One reason that was observable in these data, as well as in my own conversations with parents, is that even parents who understand that antibiotics do not cure viral infections sometimes pressure physicians for antibiotics. This, then, reﬂects a critical disconnect between medical thinking and parent thinking. The disconnect appears to be rooted in an understanding that viral infections are less serious than bacterial infections. As I mentioned before, physicians have, to a big extent, exacerbated this problem in the way they talk about viral infections. What parents have learned is that if their children have “minor symptoms,” then they have viral infections, whereas if their children are “very sick,” then they probably have bacterial infections and need antibiotics. So parents who believe their child is “very sick” will pressure for antibiotics and will be unlikely to accept that their child has a viral illness because the child is sicker than they have been taught is associated with such illnesses. Even if public health campaigns address this issue, as long as physicians reinforce this model through their own way of talking about viral infections, the problem will persist.

190

PRESCRIBING UNDER PRESSURE

Related to this issue is a disconnect between illness models that parents and the medical community may have. Parents across these data view fevers and pain as symptoms of infection that require treatment, most often antibiotic treatment. These symptoms are, for them, not associated with minor colds but with “major” bacterial infections or at least “infections” (by which they mean illnesses treatable with antibiotics). Once again, many physicians and public health campaigns miss opportunities to teach parents that these symptoms exist in both types of infections. Thus, both physicians and public health campaigns must work to address parent models of illness. Both need to make the connection clearer between prescribing for a cold and the risks to other community members. And both need to incorporate a way of discussing viral infections that no longer minimizes either the illness itself or the ability we have to help children who suffer from these infections. Returning to Diamond’s four levels of deﬁcit, physicians are, at least intellectually, aware of the problem. They anticipated it as early as 1945, and the literature reﬂects a broad awareness of the scale of the problem of bacterial resistance, as well as of the link with inappropriate antibiotic prescribing. Still, a range of issues here contribute to a continuing problem: First, although physicians “know” about the problem, they cannot actually see it. The slow trend toward less effective antibiotics conceals the problem in the same way that Diamond argues global climate change was concealed. This makes it easier to do the wrong thing under pressure. Probably more important, though, is the failure of physicians to attempt to solve the problem. Much research attempts to modify physician behavior in prescribing antibiotics (Avorn & Soumerai, 1983; Bauchner & Philipp, 1998; Belongia et al., 2001; Coley et al., 2000; Davis, Thomson, Oxman, & Haynes, 1995; Doyne et al., 2004; Finkelstein et al., 2001; Perz et al., 2002). And the education on both the parent and physician sides appears to be contributing to a reduction in antibiotic use (Finkelstein et al., 2003). But it is still clear that physicians continue to prescribe inappropriately. Our discussion of why physicians commit this medical error has shown that they are responding to perceived pressure by parents in a very understandable prosocial way: At least sometimes, they succumb to the perceived pressure. But this book also points to a range of interactional lessons. First, although parent behaviors such as using a candidate diagnosis or mentioning additional problematic symptoms may “feel” like pressure for antibiotics, they are not always motivated by this goal. Physicians may need to explicitly register when parents use these behaviors but then attempt to sort out whether the behavior is driven more by parent anxiety about the illness, a desire for a solution or something to do for their child, or a desire for antibiotics. In all cases, the use of online commentary has been shown to help prepare parents for a no-problem, no-antibiotics treatment outcome while maintaining the legitimacy of the visit. Moreover, nonminimized, afﬁrmative, and speciﬁc diagnoses and treatment recommendations may help physicians manage parents who use these behaviors. Second, across-the-board use of afﬁrmative, nonminimized, and speciﬁc diagnoses and treatment recommendations reduces parent resistance yet also educates parents to the idea that viral illnesses are real illnesses with real treatments. Based

CONCLUSION

191

on how physicians talked about viral illnesses in these data, parents were receiving consistent reinforcement of the idea that viral illnesses were minor relative to bacterial illnesses and were not treatable. Physicians can implicitly educate parents that viruses are every bit as serious as bacterial illnesses and are treatable (i.e., something can be done even if they cannot be cured), and this may, in time, reduce parent pressure for antibiotics. Third, although physicians do need to educate parents, the place to educate them about the role of antibiotics in upper respiratory tract infections is after parents have received an afﬁrmative, speciﬁc, and nonminimized diagnosis and treatment recommendation. Only then are parents in a position to learn. Prior to receiving this, parents are far more likely to hear it as an insufﬁcient treatment recommendation. Thus, ruling out the need for antibiotics and explicating the connection between viral infections, bacterial infections, and antibiotics should be done following the initial treatment recommendation. Returning once again to Diamond’s four levels, we can recognize that for the reasons described previously, parents and physicians have not been trying to solve the problem—whether due to lack of perception of a problem or because of pressure. But as has been mentioned, at the societal level, there are efforts being made to curb antibiotic prescribing for viral infections. We can ask the question, Why have these efforts not yet worked? This returns us to our commons dilemma. For societal-level efforts to work, both physicians and parents must not only perceive the problem but also perceive themselves to be part of the solution. Physicians may suffer from perceiving the prescribing of antibiotics in individual cases to be either (1) worth the damage done to society or (2) not truly contributing sufﬁciently to the problem to actually constitute wrongdoing. With respect to the former, physicians who perceive themselves to be at risk of alienating a parent by not prescribing, or who cannot get the parent to leave without prescribing, may well feel inclined to prescribe, even though they know that they should not. As one physician put it to me, “If I don’t give them the antibiotic they want, they’ll just go next door.” With respect to the latter, the perceived problem is spread out so diffusely that individuals may not sufﬁciently perceive their actions to matter. These dimensions are critical to solutions to commons dilemmas. If people feel that their actions will not have any discernible effect on the situation, then they will not cooperate with what is better for the group (Kollock, 1998). We can also speculate on another dimension of this problem that is commonly at issue with commons dilemmas: group identity. Physicians and parents alike, even if they perceive and understand the problem of inappropriate antibiotics prescribing, may nonetheless prioritize their own needs over society’s if they fail to perceive themselves as intimately connected with the society. Kramer and Brewer (1986a, 1986b) show that in simulated commons dilemmas, subjects are more willing to exhibit personal restraint if they identify as members of a group. Diamond highlights this as well (2005). Kollock notes that one reason individuals may be more willing to cooperate if they feel they are part of a group is that a collective social identity may increase member altruism (Kollock, 1998). Alternatively, it may be more strategic (e.g., see Karp, Jin, Yamagishi, & Shinotsuka, 1993). As Kollock summarizes, the

192

PRESCRIBING UNDER PRESSURE

effects of group identity may stem from a “belief in the interdependencies of group members and expectations of reciprocity among the members” (1998: 198). Even when reciprocity is not logically possible, the expectation is sometimes manifest (e.g., Karp et al., 1993). We can only speculate that whether parents and physicians feel themselves to have a stronger collective identity with their community is an important factor here. Parents who recognize the impact of their actions on society and who feel a strong sense of collective identity may be less likely to push for antibiotics. Similarly, physicians who believe that a child is suffering from a viral illness and therefore should not be prescribed antibiotics may be less likely to do so if they feel their actions will adversely affect the society. If physicians do not believe that their actions matter, then, as with similar types of social dilemmas, they may be more likely to prioritize their individual patient. If they are receiving pressure, then succumbing to that pressure, for interactional reasons, may be more attractive. It is certainly possible that a sense of collective identity is playing a role in this dilemma. With all of these factors, and especially this last factor, we would expect to see quite different rates of prescribing internationally, nationally, and regionally. And this is certainly the case, as we saw in chapter 1. To date, there has been no study that attempts to account for why some countries have lower rates of inappropriate antibiotic prescribing, though these rates have, as discussed earlier, been viewed as a primary contributor to the rate of bacterial resistance in the country. Among the many possible contributing factors outside what we discussed in the last section are health policy guidelines, economic factors such as drug costs, cultural attitudes toward illnesses, the role of physicians, and medication, as well as the role that various health care providers such as pharmacists play in the health care system. But in addition to these might be, as discussed in the last section, the degree to which individuals feel that they are part of a local community. Whether or not this is present in a given culture, this attitude should be present in medical schools, and physicians who begin to bring this way of thinking into their practices may make it possible for parents to see social connections that they might not otherwise see. And these issues do not take away from the legitimacy of their visit at all, a further beneﬁt to their inclusion in any discussion of antibiotic prescribing. Just as the problem of antimicrobial resistance is multifaceted, any solution to the problem will be as well. One alteration that might make a difference in countries such as the United States and Australia would be a push at the level of the medical school to encourage physicians in visits such as those examined in this book to maintain a focus on what their actions do to the larger society and to the vulnerable members of the population in particular. Instead of telling parents about the differences between viral infections and bacterial infections, perhaps parents need to hear stories like the one told by Alexander Fleming in his Nobel Prize acceptance speech, quoted in chapter 1. Finally, the problems facing developing countries with respect to bacterial resistance are as much the problems of the Americans as our problems are to developing countries like Bangladesh and Vietnam because, unlike humans, bacteria do not know nor do they respect national or regional boundaries. International travel makes

CONCLUSION

193

it possible for a country like The Netherlands to be quite effective in changing antibiotic prescribing health policies within their borders and yet still face problems with bacterial resistance. Currently, we are on track for a disastrous outcome: a return to the era of no cure for many bacterial infections. Our challenge is to change both our policies and our way of thinking about antibiotic prescribing.

This page intentionally left blank

APPENDIX: TRANSCRIPT SYMBOLS

1. Temporal and Sequential Relationships

[ [ ] ]

=

(0.5)

A. Overlapping or simultaneous talk is indicated in a variety of ways. Separate left square brackets, one above the other on two successive lines with utterances by different speakers, indicates a point of overlap onset, whether at the start of an utterance or later. Separate right square brackets, one above the other on two successive lines with utterances by different speakers, indicates a point at which two overlapping utterances both end, where one ends while the other continues, or simultaneous moments in overlaps that continue. B. Equal signs ordinarily come in pairs: one at the end of a line and another at the start of the next line or one shortly thereafter. They are used to indicate two things: 1. If the two lines connected by the equal signs are by the same speaker, then there was a single, continuous utterance with no break or pause, which was broken up to accommodate the placement of overlapping talk. 2. If the lines connected by two equal signs are by different speakers, then the second followed the ﬁrst with no discernible silence between them or was “latched” to it. C. Numbers in parentheses indicate silence, represented in tenths of a second; what is given here in the left margin indicates 0.5 seconds of silence. Silences may be marked either within an utterance or between utterances. 195

196

APPENDIX

(.)

D. A dot in parentheses indicates a “micropause,” hearable but not measured; ordinarily less than 0.2 seconds.

2. Aspects of Speech Delivery (Including Intonation)

. ? , ?, _ ; :: word

° °two° ^ >< <> < hhh

A. The punctuation marks are not used grammatically, but to indicate intonation. The period indicates a falling, or ﬁnal, intonation contour, not necessarily the end of a sentence. Similarly, a question mark indicates rising intonation, not necessarily a question, and a comma indicates “continuing” intonation, not necessarily a clause boundary. A combined question mark and comma indicates a rise stronger than a comma but weaker than a question mark. An underscore following a unit of talk indicates level intonation. The semicolon indicates that the intonation is equivocal between ﬁnal and “continuing.” B. Colons are used to indicate the prolongation or stretching of the sound just preceding them. The more colons, the longer the stretching. C. A hyphen after a word or part of a word indicates a cutoff or self-interruption. D. Underlining is used to indicate some form of stress or emphasis, either by increased loudness or higher pitch. The more underlining, the greater the emphasis. E. The degree sign indicates that the talk following it was markedly quiet or soft. When there are two degree signs, the talk between them is markedly softer than the talk around it. G. The caret indicates high pitch. Within a word, it indicates a pitch rise and fall. H. The combination of “more than” and “less than” symbols indicates that the talk between them is compressed or rushed. Used in the reverse order, they indicate that a stretch of talk is markedly slowed or drawn out. The “less than” symbol by itself indicates that the immediately following talk is “jump-started,” that is, sounds like it starts with a rush. I. Hearable aspiration is shown where it occurs in the talk by the letter “h”— the more h’s, the more aspiration with each “h” representing approximately 0.1 seconds. The aspiration may represent breathing, laughter, or the like. If

APPENDIX

197

it occurs inside the boundaries of a word, in may be inclosed in parentheses to set it apart from the sounds of the word. If the aspiration is an inhalation, it is shown with a dot before it.

3. Other Markings ((

))

A. Double parentheses are used to mark transcriber’s descriptions of events, rather than representations of them. Thus ((cough)), ((sniff)), ((telephone rings)), and the like. (word) B. When all or part of an utterance is in parentheses, or the speaker identiﬁcation is, this indicates uncertainty on the transcriber’s part but represents a likely possibility. ( ) Empty parentheses indicate that something that could not be understood is (or is possibly) being said.

This page intentionally left blank

NOTES

Chapter 1 1. Although many countries have public health campaigns to educate patients about antibiotics, those who do not understand that there are social consequences to their use certainly will not recognize a decision about antibiotic usage as a social dilemma at all. For those patients who do, and certainly for most physicians, antibiotic usage does represent such a dilemma. 2. Although patients and parents are acting on a sort of commons dilemma, in actuality antibiotics are not rational at the individual level if they do not help. But the point is that if they are perceived as helpful at the individual level but costly at the collective level, then we do, in fact, have a commons dilemma. 3. See Foster (1974) for an interesting discussion of key differences between a sociology of medicine and an anthropology of medicine perspective. Chapter 2 1. Parents can, even in the very opening of the visit, overtly pressure physicians for antibiotics (see chapter 6, Extract 6.7), but to do so they must buck the normative phase structure of the visit and initiate a course of action that counters the one normally followed in acute care visits. This is vanishingly rare in these data. 2. The percentage actually rises to 26% when “implied candidate diagnoses” are taken into account. This will be discussed later in the chapter. Here, I begin with a discussion of the explicit candidate diagnoses. 3. In linguistics, the term “marked” is used in many different ways. Here, when I refer to “unmarked,” I mean the normal, usual, or “default” way of doing some action. When I use the term “marked,” I mean that the form departs from the usual. This use is discussed by Levinson (1983, 1987, 2000).

199

200

NOTES

4. For a discussion of ﬁrst versus second position and epistemic rights, see Heritage (1998), Heritage and Raymond (2005), and Stivers (2005a). Chapter 3 1. Also see Heinemann’s discussion of preference and polarity in Danish (2005) and Koshik’s discussion in another institutional context (2002). 2. Also closely related is Sacks’s original idea of preference—that events can be characterized in such a way as to rank an activity such as dinner as including talk (a “ﬁrst preference” invitation), as opposed to a lower activity such as talk (which does not include dinner) (Sacks, 1992, pp. 367–369). And Drew’s “maximal properties of description” (1992) partially overlaps with a concept called “minimization” (Levinson, 1987). All of these ideas circle around the fundamental observation that speakers and hearers do not state everything possible—all events are partially described—but which absences are understood in what ways is contingent on a number of factors, including presumptions interactants make about the relevant circumstances surrounding the activity. Chapter 5 1. The placement of the treatment recommendation prior to the diagnosis may help to avoid further sequences of parent resistance. 2. Although this claim requires more investigation, it appears that the period-intoned “Okay” and “Alright” are treated as doing acceptance, whereas these objects said with comma or level intonation may be offering only acknowledgment. Beach (2001) has explored prosodic variation in the token “okay.” The period-intoned “Okay.” or “Alright.” appears to be a more minimal form of acceptance than the “That’s ﬁne” type of assessment. It may be that fuller forms of acceptance are treated as optimal in this environment and therefore lead most directly to activity closure. 3. Here and in some of the later examples, I do not have access to video to determine if there is visible behavior during these silences. However, as will be shown particularly in the next section, if there were any visible behavior such as nodding during these spaces, the physicians do not treat this as sufﬁcient to accept the recommendation but rather treat it as indicating that the activity is (or should be) continuing (Schegloff, 1982). 4. If the physician had elicited agreement from the child—which is explicitly sought— this might have helped in gaining a somewhat coerced acceptance by the mother. It is in this sense that I see this as a practice for pursuing parent acceptance. 5. This ordering parallels the larger principle of interaction that conﬂict-producing turns are typically delayed. Preference organization in second pair part deliveries displays this principle. There, delays of various sorts precede dispreferred responses, thus allowing the just prior speaker an opportunity to reformulate the ﬁrst pair part and thereby remove the relevance of the dispreferred second pair part (Heritage, 1984b; Pomerantz, 1984). A similar pattern is observable with respect to shifts from passive to active treatment resistance. 6. A similar instance of this shift from a recommendation to do nothing to a recommendation to do something was seen in Extract 5.4a, where the physician ﬁrst suggested waiting and then suggested lots of ﬂuids. Although only an incremental shift in position, this change may nonetheless be responsive to parent resistance of a “do nothing” approach to managing the child’s illness. 7. This was a concern implied early in the visit. The mother offered only the symptom of a runny nose as her problem presentation. However, apparently drawing on the patient’s chart, the doctor offered for conﬁrmation a description of the color of the discharge. Built into this

NOTES

201

turn is the doctor’s understanding that the discharge is currently “greenish,” thus conveying the understanding that the parent is worried about sinusitis. 1 2 3 4 5 6 7

MOM: DOC: DOC:

She’s [had uh r- runny nose off an’ on for about two weeks. [Uh huh ( ) Okay:_ (0.8) ((DOC hearably erasing something)) DOC: -> An:d initially: was it kinda clea:r? an’ then it -> started [g e t t i n g t h i s: greenish co[lor, MOM: -> [Initially clear but it got (.) green. [Right.

8. Here, the physician offers a straightforward plan for what should be done in the future. In previous cases where doctors introduced plans similar to this, it was offered by way of suggesting when they would reconsider treatment or by way of addressing when the parent would need to return to the ofﬁce. In Extract 5.8a, by contrast, the plan is a move toward closing the encounter by providing a straightforward future action plan. 9. This difference is probably informed by two factors: Coding was slightly more conservative in the Metro data set, so some behaviors that were counted as initiating a negotiation in the Seaside data set were not counted in the Metro data. In addition, though, the latter data were much more heterogeneous in terms of ethnicity and socioeconomic factors. This is important because, for instance, African Americans did not resist the treatment in any case in the Metro data, whereas other ethnic groups did (Mangione-Smith, Elliott, Stivers, McDonald, & Heritage, 2006). Chapter 7 1. “Problem” online commentary is also used (Mangione-Smith, Stivers, Elliott, McDonald, & Heritage, 2003). This chapter will not address this practice because although it forecasts a diagnosis, it generally forecasts a bacterial diagnosis and thus is outside the realm of this chapter. All references to “online commentary” here will refer to no-problem online commentary. 2. Note that this is a way of downgrading the claim as well (Chafe & Nichols, 1986). The claim “I don’t see any ﬂuid” is not as strong as “There isn’t any ﬂuid” because it leaves open the possibility of there being ﬂuid that is unseen. 3. Sudafed, an over-the-counter medication, and prescription allergy medications were other examples of medication that were ruled out. This was not at all frequent. 4. This is the basic argument of noticeable absences observed by Schegloff (1968). At any given point, any number of actions may be absent, but most of them were not relevant for the sequence in progress, so they are not “noticeably absent.” Here, the point is that because nothing is relevant following a diagnosis, silence does not mark a noticeable absence in this context.

This page intentionally left blank

REFERENCES

Adam, D. (2002). Global antibiotic resistance in Streptococcus pneumoniae. Journal of Antimicrobial Chemotherapy, 50, 1–5. Albrich, W. C., Monnet, D. L., & Harbarth, S. (2004). Antibiotic selection pressure and resistance in Streptococcus pneumoniae and Streptococcus pyogenes. Emerging Infectious Diseases, 10, 514–517. Aronoff, S. C. (1996). Antimicrobials in children and the problem of drug resistance. American Family Physician, 54, 44. Athreya, B. H., & Silverman, B. K. (1985). Pediatric Physical Diagnosis. Norwalk, CT: Appleton-Century-Crofts. Atkinson, J. M. (1984). Public speaking and audience responses: Some techniques for inviting applause. In J. M. Atkinson & J. Heritage (Eds.), Structures of Social Action: Studies in Conversation Analysis (pp. 370–409). Cambridge: Cambridge University Press. Avorn, J., & Solomon, D. H. (2000). Cultural and economic factors that (mis)shape antibiotic use: The nonpharmacologic basis of therapeutics. Annals of Internal Medicine, 133, 128–135. Avorn, J., & Soumerai, S. B. (1983). Improving drug-therapy decisions through educational outreach. A randomized controlled trial of academically based “detailing.” New England Journal of Medicine, 308, 1457–1463. Baquero, F., Baquero-Artigao, G., Canton, R., & Garcia-Rey, C. (2002). Antibiotic consumption and resistance selection in Streptococcus pneumoniae. Journal of Antimicrobial Chemotherapy, 50, 27–38. Barden, L. S., Dowell, S. F., Schwartz, B., & Lackey, C. (1998). Current attitudes regarding use of antimicrobial agents: Results from physicians’ and parents’ focus group discussions. Clinical Pediatrics, 37, 665–672. Bartoloni, A., Cutts, F., Leoni, S., Austin, C. C., Mantella, A., Guglielmetti, P., et al. (1998). Patterns of antimicrobial use and antimicrobial resistance among healthy children in Bolivia. Tropical Medicine and International Health, 3, 116–123.

203

204

REFERENCES

Bates, B., Bickley, L. S., & Hoekelman, R. A. (1995). Physical Examination and History Taking (6th ed.). Philadelphia: Lippincott. Bauchner, H., & Philipp, B. (1998). Reducing inappropriate oral antibiotic use: A prescription for change. Pediatrics, 102, 142–145. Beach, W. (2001). Some prosodic displays of incongruity in interaction. Paper presented at the National Communication Association, Atlanta, GA. Beckman, H. ,& Frankel, R. (1984). The effect of physician behavior on the collection of data. Annals of Internal Medicine, 101, 692–696. Beckman, H. B., Markakis, K. M., Suchman, A. L., & Frankel, R. M. (1994). The doctorpatient relationship and malpractice: Lessons from plaintiff depositions. Archives of Internal Medicine, 154, 1365–1370. Bell, N. (2002). Antibiotic resistance: The Iowa experience. American Journal of Managed Care, 8, 988–994. Bell, R. A., Wilkes, M. S., & Kravitz, R. L. (1999). Advertisement-induced prescription drug requests: Patients anticipated reactions to a physician who refuses. Journal of Family Practice, 48, 446–452. Belongia, E. A., Knobloch, M. J., & Kieke Jr., B. A. (2005). Impact of statewide program to promote appropriate antimicrobial drug use. Emerging Infectious Diseases, 11, 912–920. Belongia, E. A., Sullivan, B. J., Chyou, P. H., Madagame, E., Reed, K. D., & Schwartz, B. (2001). A community intervention trial to promote judicious antibiotic use and reduce penicillin-resistant Streptococcus pneumoniae carriage in children. Pediatrics, 108, 575–583. Blanchard, C. G., Labrecque, M. S., Ruckdeschel, J. C., & Blanchard, E. B. (1988). Information and decision-making preferences of hospitalized adult cancer patients. Social Science and Medicine, 27, 1139–1145. Borzo, G. (1997). Consumer drug ads booming: FDA reviews restrictions. American Medical News, 40, 1, 37. Bosu, W. K., & Ofori-Adjei, D. (1997). Survey of antibiotic prescribing pattern in government health facilities of the Wassa west district of Ghana. East African Medical Journal, 74, 138–142. Boyd, E. A. (1997). Constructing histories and negotiating care: Professional discourse during medical peer review. Unpublished dissertation, University of California, Los Angeles. Boyd, E. A. (1998). Bureaucratic authority in the “company of equals”: The interactional management of medical peer review. American Sociological Review, 63, 200–224. Boyd, E. A., & Heritage, J. (2006). Taking the patient’s medical history: Questioning during comprehensive history-taking. In J. Heritage & D. Maynard (Eds.), Communication in Medical Care: Interactions between Primary Care Physicians and Patients (pp. 151– 184). Cambridge: Cambridge University Press. Braddock, C. H., Edwards, K. A., Hasenberg, N. M., Laidley, T. L., & Levinson, W. (1999). Informed decision making in outpatient practice: Time to get back to basics. Journal of the American Medical Association, 282, 2313–2320. Braithwaite, A., & Pechere, J. C. (1996). Pan-European survey of patients’ attitudes to antibiotics and antibiotic use. Journal of International Medical Research, 24, 229–238. Britten, N. (1994). Patient demand for prescriptions: A view from the other side. Family Practice, 11, 62–66. Britten, N., Jenkins, L., Barber, N., Bradley, C. P., & Stevenson, F. A. (2003). Developing a measure for the appropriateness of prescribing in general practice. Quality and Safety in Health Care, 12, 246–250. Britten, N., Stevenson, F., Gafaranga, J., Barry, C., & Bradley, C. (2004). The expression of aversion to medicines in general practice consultations. Social Science and Medicine, 59, 1495–1503.

REFERENCES

205

Britten, N., Stevenson, F. A., Barry, C. A., Barber, N., & Bradley, C. P. (2000). Misunderstandings in prescribing decisions in general practice: Qualitative study. British Medical Journal, 320, 484–488. Britten, N., & Ukoumunne, O. (1997). The inﬂuence of patients’ hopes of receiving a prescription on doctors’ perceptions and the decision to prescribe: A questionnaire study. British Medical Journal, 315, 1506–1510. Brody, D. S. (1980). The patient’s role in clinical decision making. Annals of Internal Medicine, 93, 718–722. Brody, D. S., Miller, S. M., Lerman, C., Smith, M. D., & Caputo, C. (1989). Patient perception of involvement in medical care: Relationship to illness attitudes and outcomes. Journal of General Internal Medicine, 4, 506–511. Brody, D. S., Miller, S. M., Lerman, C. E., Smith, D. G., Lazaro, C. G., & Blum, M. J. (1989). The relationship between patients’ satisfaction with their physicians and perceptions about interventions they desired and received. Medical Care, 27, 1027–1035. Brown, P., & Levinson, S. C. (1987). Politeness: Some universals in language usage. Cambridge: Cambridge University Press. Butler, C. C., Kinnersley, P., Prout, H., Rollnick, S., Edwards, A., & Elwyn, G. (2001). Antibiotics and shared decision making in primary care. Journal of Antimicrobial Chemotherapy, 48, 435–440. Bybee, J., Perkins, R., & Pagliuca, W. (1994). The evolution of grammar. Chicago: University of Chicago Press. Byrne, P. S., & Long, B. E. L. (1976). Doctors talking to patients: A study of the verbal behaviours of doctors in the consultation. Exeter: Royal College of General Practitioners. Cars, O., Mölstad, S., & Melander, A. (2001). Variation in antibiotic use in the European Union. Lancet, 357, 1851–1853. Cassileth, B. R., Zupkis, R. V., Sutton-Smith, K., & March, V. (1980). Information and participation preferences among cancer patients. Annals of Internal Medicine, 92, 832–836. Centers for Disease Control, National Center for Health Statistics. (2004). National Ambulatory Medical Care Survey, 2004 [Data ﬁle]. Available from CDC Web site, http://www. cdc.gov/nchs/about/major/ahcd/ahcd1.htm Chafe, W., & Nichols, J. (1986). Evidentiality: The linguistic coding of epistemology. Norwood, NJ: Ablex. Clayman, S., & Heritage, J. (2002). The news interview: Journalists and public ﬁgures on the air. Cambridge: Cambridge University Press. Clayman, S. E., Elliott, M., Heritage, J., & McDonald, L. (2006). Historical trends in questioning presidents 1953–2000. Presidential Studies Quarterly, 36, 561–583. Clayman, S. E., Heritage, J., Elliott, M., & McDonald, L. (in press). When does the watchdog bark? Conditions of aggressive questioning in presidential news conferences. American Sociological Review. Clift, R. (2001). Meaning in interaction: The case of “‘actually’.” Language, 77, 245–291. Cockburn, J., & Pit, S. (1997). Prescribing behavior in clinical practice: Patients’ expectations and doctors’ perceptions of patients’ expectations—A questionnaire study. British Medical Journal, 315, 520–523. Cohen-Cole, S. A. (1991). The medical interview: The three-function approach. St. Louis: Mosby. Coley, K. C., Skledar, S. J., Fine, M. J., Yealy, D. M., Gleason, P. P., Ryan, M. L., et al. (2000). Changing physician prescribing behavior: The community-acquired pneumonia intervention trial. American Journal of Health System Pharmacy, 57, 1506–1510. Coulter, A. (1997). Partnerships with patients: The pros and cons of shared clinical decisionmaking. Journal of Health Services Research Policy, 2, 112–121.

206

REFERENCES

Cristino, J. M. (1999). Correlation between consumption of antimicrobials in humans and development of resistance in bacteria. International Journal of Antimicrobial Agents, 12, 199–202. Dagan, R. (2000). Clinical signiﬁcance of resistant organisms in otitis media. Pediatric Infectious Disease Journal, 19, 378–382. Davis, D. A., Thomson, M. A., Oxman, A. D., & Haynes, R. B. (1995). Changing physician performance: A systematic review of the effect of continuing medical education strategies. Journal of the American Medical Association, 274, 700–705. Deber, R. B. (1994). Physicians in health care management. The patient-physician partnership: Decision making, problem solving and the desire to participate. Canadian Medical Association Journal, 151, 423–427. Deeks, S. L., Palacio, R., Ruvinsky, R., Kertesz, D. A., Hortal, M., Rossi, A., et al. (1999). Risk factors and course of illness among children with invasive penicillin-resistant Streptococcus pneumoniae. The Streptococcus pneumoniae Working Group. Pediatrics, 103, 409–413. Diamond, J. (2005). Collapse: How Societies Choose to Fail or Succeed. New York: Viking Penguin. Doern, G. V., & Brown, S. D. (2004). Antimicrobial susceptibility among community-acquired respiratory tract pathogens in the USA: Data from PROTEKT US 2000–01. Journal of Infection, 48, 56–65. Doyne, E. O., Alfaro, M. P., Siegel, R. M., Atherton, H. D., Schoettker, P. J., Bernier, J., et al. (2004). A randomized controlled trial to change antibiotic prescribing patterns in a community. Archives of Pediatric and Adolescent Medicine, 158, 577–583. Drew, P. (1992). Contested evidence in a courtroom cross-examination: The case of a trial for rape. In P. Drew & J. Heritage (Eds.), Talk at Work: Social Interaction in Institutional Settings (pp. 470–520). Cambridge: Cambridge University Press. Drew, P. (1997). “Open” class repair initiators in response to sequential sources of trouble in conversation. Journal of Pragmatics, 28, 69–101. Drew, P. (2006). Misalignments in “after-hours” calls to a British GP’s practice: A study in telephone medicine. In J. Heritage & D. Maynard (Eds.), Communication in Medical Care: Interactions between Primary Care Physicians and Patients (pp. 416–444). Cambridge, England: Cambridge University Press. Drew, P., & Heritage, J. (1992a). Analyzing talk at work: An introduction. In P. Drew & J. Heritage (Eds.), Talk at Work (pp. 3–65). Cambridge: Cambridge University Press. Drew, P., & Heritage, J. (Eds.). (1992b). Talk at Work. Cambridge: Cambridge University Press. du Pré, A. (2000). Communicating about Health: Current Issues and Perspectives. Mountain View, CA: Mayﬁeld. Dunnell, K., & Cartwright, A. (1972). Medicine Takers, Prescribers, and Hoarders. London: Routledge and Kegan Paul. Elwyn, G., Edwards, A., & Kinnersley, P. (1999). Shared decision-making in primary care: The neglected second half of the consultation. British Journal of General Practice, 49, 477–482. Emanuel, E. J., & Emanuel, L. L. (1992). Four models of the physician-patient relationship. Journal of the American Medical Association, 267, 2221. Emerson, C. (1983). Bakhtin and Vygotsky on internalization in language. Quarterly Newsletter of the Laboratory of Comparative Human Cognition, 5(1), 9–13. Ende, J., Kazis, L., Ash, A., & Moskowitz, M. A. (1989). Measuring patients’ desire for autonomy: Decision making and information-seeking preferences among medical patients. Journal of General Internal Medicine, 4, 23–30.

REFERENCES

207

Evans, B. J., Kiellerup, F. D., Stanley, R. O., Burrows, G. D., & Sweet, B. (1987). A communications skills programme for increasing patients’ satisfaction with general practice consultations. British Journal of Medical Psychology, 60, 373–378. Faden, R. R., Becker, C., Lewis, C., Freeman, J., & Faden, A. I. (1981). Disclosure of information to patients in medical care. Medical Care, 19, 718–733. Fallowﬁeld, L., Hall, A., Maguire, G. P., & Baum, M. (1990). Psychological outcomes of different treatment policies in women with early breast cancer outside a clinical trial. British Medical Journal, 301, 575–580. Feikin, D. R., Schuchat, A., Kolczak, M., Barrett, N. L., Harrison, L. H., Lefkowitz, L., et al. (2000). Mortality from invasive pneumococcal pneumonia in the era of antibiotic resistance, 1995–1997. American Journal of Public Health, 90, 223–229. Fidler, D. P. (1998). Legal issues associated with antimicrobial drug resistance. Emerging Infectious Diseases, 4, 169–177. Finkelstein, J. A., Davis, D. A., Dowell, S. F., Metlay, J. P., Soumerai, S. B., Rifas-Shiman, S. L., et al. (2001). Reducing antibiotic use in children: A randomized trial in 12 practices. Pediatrics, 108, 1–7. Finkelstein, J. A., Metlay, J., Davis, R. I., Rifas, S., Dowell, S. F., & Platt, R. (2000). Antimicrobial use in deﬁned populations of infants and young children. Archives of Pediatrics and Adolescent Medicine, 154, 395–400. Finkelstein, J. A., Stille, C., Nordin, J., Davis, R. I., Raebel, M. A., Roblin, D., et al. (2003). Reduction in antibiotic use among US children, 1996–2000. Pediatrics, 112, 620– 627. Fischer, T., Fischer, S., Kochen, M. M., & Hummers-Pradier, E. (2005). Inﬂuence of patient symptoms and physical ﬁndings on general practitioners’ treatment of respiratory tract infections: A direct observation study. Retrieved from http://www.biomedcentral. com/1471-2296/6/6. Fisher, S. (1991). A discourse of the social: Medical talk/power talk/oppositional talk? Discourse and Society, 2, 157–182. Fisher, S., & Todd, A. D. (Eds.). (1983). The Social Organization of Doctor-Patient Communication. Washington, DC: Center for Applied Linguistics. Fleming, A. (1929). On the antibacterial action of cultures of a penicillium, with special reference to their use in the isolation of B. inﬂuenzae. British Journal of Experimental Pathology, 10, 226–236. Fleming, A. (1945). Penicillin: Nobel Lecture. Retrieved from http://nobelprize.org/medicine/ laureates/1945/ﬂeming-lecture.pdf. Foster, G. M. (1974). Medical anthropology: Some contrasts with medical sociology. Medical Anthropology Newsletter, 6, 1–6. Frankel, R. (1984). From sentence to sequence: Understanding the medical encounter through microinteractional analysis. Discourse Processes, 7, 135–170. Freidland, I. R. (1995). Comparison of the response to antimicrobial therapy of penicillinresistant and penicillin susceptible pneumococcal disease. Pediatric Infectious Diseases Journal, 14, 885–890. Freidson, E. (1970a). Profession of Medicine: A Study of the Sociology of Applied Knowledge. Chicago: University of Chicago Press. Friedson, E. (1970b). Professional Dominance. Chicago: Aldine. Froom, J., Culpepper, L. J. M., Green, L. A., de Melker, R. A., Grob, P., Heeren, T., et al. (2001). A cross-national study of acute otitis media: Risk factors, severity, and treatment at initial visit. Report from the International Primary Care Network (IPCN) and the Ambulatory Sentinel Practice Network (ASPN). Journal of the American Board of Family Parctice, 14, 406–417.

208

REFERENCES

Frosch, D. L., & Kaplan, R. M. (1999). Shared decision making in clinical medicine: Past research and future directions. American Journal of Preventive Medicine, 27, 1139– 1145. Gani, L., Arif, H., Widjaja, S. K., Adi, R., Prasadja, H., Tampubolon, L. H., et al. (1991). Physicians’ prescribing practice for treatment of acute diarrhoea in young children in Jakarta. Journal of Diarrhoeal Disease Research, 9, 194–199. Gardner, R. (1997). The conversation object Mm: A weak and variable acknowledging token. Research on Language and Social Interaction, 30, 131–156. Gill, V. T. (1998). Doing attributions in medical interaction: Patients’ explanations for illness and doctors’ responses. Social Psychology Quarterly, 61, 342–360. Gomez, J., Banos, V., Ruiz Gomez, J., Herrero, F., Nunez, M. L., Canteras, M., et al. (1995). Clinical signiﬁcance of pneumococcal bacteraemias in a general hospital: A prospective study 1989–1993. Journal of Antimicrobial Chemotherapy, 36, 1021–1030. Gonzales, R., Malone, D. C., Maselli, J. H., & Sande, M. A. (2001). Excessive antibiotic use for acute respiratory infections in the United States. Clinical Infectious Diseases, 33, 757–762. Gonzalez, R., Steiner, J. F., & Sande, M. A. (1997). Antibiotic prescribing for adults with colds, upper respiratory tract infections and bronchitis by ambulatory care physicians. Journal of the American Medical Association, 278, 901–904. Greenberger, N. J., & Hinthorn, D. R. (1993). History Taking and Physical Examination: Essentials and Clinical Correlates. St. Louis: Mosby. Greenﬁeld, S., Kaplan, S. H., Ware, J. E., Yano, E. M., & Frank, J. L. H. (1988). Patients’ participation in medical care: Effects on blood sugar control and quality of life in diabetes. Journal of General Internal Medicine, 3, 448–457. Gums, J. G. (2002). Assessing the impact of antimicrobial resistance. American Journal of Health System Pharmacy, 59, S4–S6. Guyon, A. B., Barman, A., Ahmen, J. U., Ahmen, A. U., & Alam, M. S. (1994). A baseline survey on use of drugs at the primary health care level in Bangladesh. Bulletin of the World Health Organization, 72, 265–271. Haak, H. (1988). Pharmaceuticals in two Brazilian villages: Lay practices and perceptions. Social Science and Medicine, 27, 1415–1427. Haakana, M. (2001). Laughter as a patient’s resource: Dealing with delicate aspects of medical interaction. Text, 21, 187–219. Halkowski, T. (2006). Realizing the illness: Patients’ narratives of symptom discovery. In J. Heritage & D. Maynard (Eds.), Communication in Medical Care: Interactions between Primary Care Physicians and Patients (pp. 86–114). Cambridge: Cambridge University Press. Hamm, R., Hicks, R., & Bemben, D. (1996). Antibiotics and respiratory infections: Are patients more satisﬁed when expectations are met? Journal of Family Practice, 43, 56–62. Harbarth, S., Albrich, W., & Brun-Buisson, C. (2002). Outpatient antibiotic use and prevalence of antibiotic resistant pneumoncocci in France and Germany: A sociocultural perspective, CDC 8(12), Retrieved from: http://www.cdc.gov/ncidod/EID/vol8no12/010533.htm. Harbarth, S., & Samore, M. H. (2005). Antimicrobial resistance determinants and future control. Emerging Infectious Diseases, 11, 794–801. Hardin, G. (1968). The tragedy of the commons. Science, 162, 1243–1248. Haug, M., & Lavin, B. (1983). Consumerism in Medicine: Challenging Physician Authority. Beverly Hills, CA: Sage. Heath, C. (1981). The opening sequence in doctor-patient interaction. In P. Atkinson & C. Heath (Eds.), Medical Work: Realities and Routines (pp. 71–90). Aldershot, England: Gower.

REFERENCES

209

Heath, C. (1992). The delivery and reception of diagnosis and assessment in the general practice consultation. In P. Drew & J. Heritage (Eds.), Talk at Work (pp. 235–267). Cambridge: Cambridge University Press. Heinemann, T. (2005). Where grammar and interaction meet: The preference for matched polarity in responsive turns in Danish. In A. Hakulinen & M. Selting (Eds.), Syntax and Lexis in Conversation: Studies on the Use of Linguistic Resources in Talk-in-Interaction (pp. 375–402). Amsterdam: John Benjamins. Helman, C. G. (1978). “Feed a cold, starve a fever”: Folk models of infection in an English suburban community, and their relation to medical treatment. Culture, Medicine and Psychiatry, 2, 107–137. Heritage, J. (1984a). A change-of-state token and aspects of its sequential placement. In J. M. Atkinson & J. Heritage (Eds.), Structures of Social Action (pp. 299–345). Cambridge: Cambridge University Press. Heritage, J. (1984b). Garﬁnkel and Ethnomethodology. Cambridge: Polity Press. Heritage, J. (1998). Oh-prefaced responses to inquiry. Language in Society, 27, 291–334. Heritage, J. (2002). The limits of questioning: Negative interrogatives and hostile question content. Journal of Pragmatics, 34, 1427–1446. Heritage, J. (2006). Revisiting authority in physician-patient interaction. In M. Maxwell, D. Kovarsky, & J. Duchan (Eds.), Diagnosis as Cultural Practice (pp. 83–102). New York: Mouton de Gruyter. Heritage, J., Boyd, E., & Kleinman, L. (2001). Subverting criteria: The role of precedent in decisions to ﬁnance surgery. Sociology of Health and Illness, 23, 701–728. Heritage, J., & Greatbatch, D. (1986). Generating applause: A Study of rhetoric and response at party political conferences. American Journal of Sociology, 92(1), 110–157. Heritage, J., & Lindström, A. (1998). Motherhood, medicine and morality: Scenes from a medical encounter. Research on Language and Social Interaction, 31, 397-438. Heritage, J., & Maynard, D. (2006). Communication in Medical Care: Interactions between Primary Care Physicians and Patients. Cambridge: Cambridge University Press. Heritage, J., & Raymond, G. (2005). The terms of agreement: Indexing epistemic authority and subordination in assessment sequences. Social Psychology Quarterly, 68, 15–38. Heritage, J. & Robinson, J. (2006a). Accounting for the visit: Giving reasons for seeking medical care. In J. Heritage & D. Maynard (Eds.), Communication in Medical Care: Interactions between Primary Care Physicians and Patients (pp. 48–85). Cambridge: Cambridge University Press. Heritage, J. & Robinson, J. D. (2006b). The structure of patients’ presenting concerns: Physicians opening questions. Health Communication 19(2), 89–102. Heritage, J., & Seﬁ, S. (1992). Dilemmas of advice: Aspects of the delivery and reception of advice in interactions between health visitors and ﬁrst time mothers. In P. Drew & J. Heritage (Eds.), Talk at Work (pp. 359–417). Cambridge: Cambridge University Press. Heritage, J., & Sorjonen, M.-L. (1994). Constituting and maintaining activities across sequences: And-prefacing as a feature of question design. Language in Society, 23, 1–29. Heritage, J., & Stivers, T. (1999). Online commentary in acute medical visits: A method of shaping patient expectations. Social Science and Medicine, 49, 1501–1517. Hibbard, J. H., & Weeks, E. C. (1985). Consumer use of physician fee information. Journal of Health and Human Resources Administration, 7, 321–335. Himmel, W., Lippert-Urbanke, E., & Kochen, M. M. (1997). Are patients more satisﬁed when they receive a prescription? The effect of patient expectations in general practice. Scandinavian Journal of Primary Health Care, 15, 118–122. Holmberg, S. D., Solomon, S. L., & Blake, P. A. (1987). Health and economic impacts of antimicrobial resistance. Reviews of Infectious Diseases, 9, 1065–1078.

210

REFERENCES

Hogerzeil, H. V., Bimo, Ross-Degnan, D., Laning, R. O., Ofori-Adjei, D., Santoso, B., et al. (1993). Field tests for rational drug use in twelve developing countries. Lancet, 342, 1408–1410. Horn, L. (1989). A Natural History of Negation. Chicago: University of Chicago Press. Hossain, M. M., Glass, R. I., & Khan, M. R. (1982). Antibiotic use in a rural community in Bangladesh. International Journal of Epidemiology, 11, 402–405. Hui, L., Li, X. S., Zeng, X. J., Dai, Y. H., & Foy, H. M. (1997). Patterns and determinants of use of antibiotics for acute respiratory tract infections in children in China. Pediatric Infectious Disease Journal, 16, 560–564. Humphrey, N. K. (1988). The social function of intellect. In R. Byrne & A. Whiten (Eds.), Machiavellian Intelligence: Social Expertise and the Evolution of Intellect in Monkeys, Apes, and Humans (pp. 13–26). Oxford: Clarendon Press. Igun, U. A. (1994). Reported and actual prescription of oral rehydration therapy for childhood diarrhoeas by retail pharmacists in Nigeria. Social Science and Medicine, 39, 797–806. Indalo, A. A. (1997). Antibiotic sale behaviour in Nairobi: A contributing factor to antimicrobial drug resistance. East African Medical Journal, 74, 171–173. Jacobs, M. R., Felmingham, D., Appelbaum, P. C., Grüneberg, R. N., & Group, T. A. P. (2003). The Alexander Project 1998–2000: Susceptibility of pathogens isolated from community-acquired respiratory tract infection to commonly used antimicrobial agents. Journal of Antimicrobial Chemotherapy, 52, 229–246. Jefferson, G. (1979). A technique for inviting laughter and its subsequent acceptance/declination. In G. Psathas (Ed.), Everyday Language: Studies in Ethnomethodology (pp. 79–96). New York: Irvington Publishers. Jefferson, G. (1981). The Abominable “Ne?”: A Working Paper Exploring the Phenomenon of Post-Response Pursuit of Response. Occasional Paper No. 6, Department of Sociology, University of Manchester, Manchester, England. Jefferson, G. (1983). Notes on a systematic deployment of the acknowledgement tokens ‘yeah’ and ‘mm hm.’ In G. Jefferson (Ed.), Two Papers on Transitory Recipientship (pp. 1–18). Tilburg University: Tilburg Papers in Language and Literature 30. Jefferson, G. (1988). On the sequential organization of troubles-talk in ordinary conversation. Social Problems, 35(4), 418–441. Jefferson, G. (1990). List construction as a task and interactional resource. In G. Psathas (Ed.), Interaction Competence (pp. 63–92). Washington, DC: International Institute for Ethnomethodology and Conversation Analysis; University Press of America. Jefferson, G. (2004). A note on laughter in “male-female” interaction. Discourse Studies, 6, 117–133. Kaiser, L., Lew, D., Hirschel, B., Auckenthaler, R., Morabia, A., Heald, A., et al. (1996). Effects of antibiotic treatment in the subset of common-cold patients who have bacteria in nasopharyngeal secretions. Lancet, 347, 1507–1510. Kaplan, S. H., Greenﬁeld, S., & Ware, J. (1989). Assessing the effects of physician-patient interactions on the outcomes of chronic disease. Medical Care, 27, S110–126. Karp, D., Jin, N., Yamagishi, T., & Shinotsuka, H. (1993). Raising the minimum in the minimal group paradigm. Japanese Journal of Experimental Social Psychology, 32, 231–240. Kassirer, J. P. (1994). Incorporating patients’ preferences into medical decisions. New England Journal of Medicine, 330, 1895–1896. Kleinman, L., Boyd, E., & Heritage, J. (1997). Adherence to prescribed explicit criteria during utilization review: An analysis of communications between attending and reviewing physicians. Journal of the American Medical Association, 278, 497–501. Kollock, P. (1998). Social dilemmas: The anatomy of cooperation. Annual Review of Sociology, 24, 183–214.

REFERENCES

211

Korsch, B., Gozzi, E. K., & Francis, V. (1968). Gaps in doctor-patient communication. Pediatrics, 42, 855–871. Koshik, I. (2002). A conversation analytic study of yes/no questions which convey reversed polarity assertions. Journal of Pragmatics, 34, 1851–1877. Kozyrskyj, A. L., Dahl, M. E., Chateau, D. G., Mazowita, G. B., Klassen, T. P., & Law, B. J. (2004). Evidence-based prescribing of antibiotics for children: Role of socioeconomic status and physician characteristics. Canadian Medical Association Journal, 171, 139– 145. Kramer, R. M., & Brewer, M. B. (1986a). Effects of group identity on resource use in a simulated commons dilemma. Journal of Personality and Social Psychology, 38, 504–516. Kramer, R. M., & Brewer, M. B. (1986b). Social group identity and the emergence of cooperation in resource conservation dilemmas. In H. A. M. Wilke, D. M. Messick, & C. Rutte (Eds.), Experimental Social Dilemmas (pp. 205–234). Frankfurt: Verlag Peter Lang. Kravitz, R. L., Bell, R. A., & Franz, C. E. (1999). A taxonomy of requests by patients (TORP): A new system for understanding clinical negotiation in ofﬁce practice. Journal of Family Practice, 48, 872–878. Kravitz, R. L., Epstein, R. M., Feldman, M. D., Franz, C. E., Azari, R., Wilkes, M. S., et al. (2005). Inﬂuence of patients’ requests for direct-to-consumer advertised antidepressants: A randomized controlled trial. Journal of the American Medical Association, 293, 1995–2002. Kunin, C. M., Lipton, H. L., Tupasi, T., Sacks, T., Scheckler, W. E., Jivani, A., et al. (1987). Social, behavioral, and practical factors affecting antibiotic use worldwide: Report of task force 4. Review of Infectious Diseases, 9, S270–S284. Lerner, G. (1996). On the “semi-permeable” character of grammatical units in conversation: Conditional entry into the turn-space of another speaker. In E. Ochs, E. A. Schegloff, & S. Thompson (Eds.), Interaction and Grammar (pp. 238–276). Cambridge: Cambridge University Press. Levine, M. N., Gafni, A., Markham, B., & MacFarlane, D. (1992). A bedside decision instrument to elicit a patient’s preference concerning adjuvant chemotherapy for breast cancer. Annals of Internal Medicine, 117, 53–58. Levinson, S. C. (1983). Pragmatics. Cambridge: Cambridge University Press. Levinson, S. C. (1987). Minimization and conversational inference. In M. Papi & J. Verschuren (Eds.), The Pragmatics Perspective (pp. 61–129). Amsterdam: J. Benjamins. Levinson, S. C. (2000). Presumptive Meanings: The Theory of Generalized Conversational Implicature. Cambridge, MA: MIT Press. Levy, S. B. (1992). The Antibiotics Paradox: How Miracle Drugs Are Destroying the Miracle. New York: Plenum. Lindström, A. (1995). ‘Or’ -constructed inquiries as a resource for probing the relevance of prior talk in Swedish conversation. Paper presented at The Georgetown Linguistics Society meeting, Washington, DC. Little, P., Dorward, M., Warner, G., Stephens, K., Senior, J., & Moore, M. (2004). Importance of patient pressure and perceived pressure and perceived medical need for investigations, referral, and prescribing in primary care: Nested observational study. Retrieved from http:// bmj.bmjjournals.com/cgi/content/full/328/7437/444?maxtoshow=&HITS=10&hits= 10&RESULTFORMAT=&author1=Little&andorexactfulltext=and&searchid=1& FIRSTINDEX=0&sortspec=relevance&resourcetype=HWCIT Madle, G., Kostkova, P., Mani-Saada, J., Weinberg, J., & Williams, P. (2004). Changing public attitudes to antibiotic prescribing: Can the Internet help? Informatics in Primary Care, 12, 19–26.

212

REFERENCES

Mainous, A. G., Cheng, A. Y., Garr, R. C., Tilley, B. C., Everett, C. J., & McKee, M. D. (2005). Nonprescribed antimicrobial drugs in Latino community, South Carolina. Emerging Infectious Diseases, 11, 883–888. Mainous, A. G., Hueston, W. J., & Clark, J. R. (1996). Antibiotics and upper respiratory infection: Do some folks think there is a cure for the common cold? Journal of Family Practice, 42, 357–361. Mamun, K. Z. (1991). Prevalence and genetics of resistance to commonly used antimicrobial agents in faecal enterbacteriaceae from children in Bangladesh. Unpublished Ph.D. Thesis, University of Liverpool, Liverpool, England. Mangione-Smith, R., Elliott, M. N., McDonald, L., & McGlynn, E. A. (2002). An observational study of antibiotic prescribing behavior and the Hawthorne effect. Health Services Research, 37, 1603–1623. Mangione-Smith, R., Elliott, M. N., Stivers, T., McDonald, L., Heritage, J., & McGlynn, E. A. (2004). Racial/ethnic variation in parent expectations for antibiotics: Implications for public health campaigns. Pediatrics, 113, e385–e394. Mangione-Smith, R., Elliott, M. N., Stivers, T., McDonald, L. L., & Heritage, J. (2006). Ruling out the need for antibiotics: Are we sending the right message? Archives of Pediatric and Adolescent Medicine 160, 956–952. Mangione-Smith, R., McGlynn, E. A., Elliott, M. N., Krogstad, P., & Brook, R. H. (1999). The relationship between perceived parental expectations and pediatrician antimicrobial prescribing behavior. Pediatrics, 103, 711–718. Mangione-Smith, R., McGlynn, E. A., Elliott, M. N., McDonald, L., Franz, C. E., & Kravitz, R. L. (2001). Parent expectations for antibiotics, physician-parent communication, and satisfaction. Archives of Pediatric and Adolescent Medicine, 155, 800–806. Mangione-Smith, R., Stivers, T., Elliott, M. N., McDonald, L., & Heritage, J. (2003). The relationship between online commentary use and prevention of inappropriate antibiotic prescribing by pediatricians. Social Science and Medicine, 56, 313–320. Maynard, D. W. (2003). Bad News, Good News: Conversation Order in Everyday Talk and Clinical Settings. Chicago: University of Chicago. McCaig, L., Besser, R. E., & Hughes, J. M. (2002). Trends in antimicrobial prescribing rates for children and adolescents. Journal of the American Medical Association, 287, 3096– 3102. McCaig, L. F., Besser, R. E., & Hughes, J. M. (2003). Antimicrobial drug prescriptions in ambulatory care settings, United States, 1992–2000. Emerging Infectious Diseases, 9, 432–437. McCaig, L. F., & Hughes, J. M. (1995). Trends in antimicrobial drug prescribing among ofﬁce-based physicians in the United States. Journal of the American Medical Association, 273, 214–219. McCormick, A. W., Whitney, C. G., Farley, M. M., Lynﬁeld, R., Harrison, L. H., Bennett, N. M., et al. (2003). Geographic diversity and temporal trends of antimicrobial resistance in Streptococcus pneumoniae in the United States. Nature Medicine, 9, 424–430. Melker, R. A., & Kuyvenhoven, M. M. (1994). Management of upper respiratory tract infections in Dutch family practice. Journal of Family Practice, 38, 353–357. Memish, Z. A., Venkatesh, S., & Shibl, A. M. (2003). Impact of travel on international spread of antimicrobial resistance. International Journal of Antimicrobial Agents, 21, 135–142. Mendonca, P. J., & Brehm, S. S. (1983). Effects of choice on behavioral treatment of overweight children. Journal of Social Clinical Psychology, 1, 343–358. Metlay, J. P., Shea, J. A., Crossette, L. B., & Asch, D. A. (2002). Tensions in antibiotic prescribing. Journal of General Internal Medicine, 17, 87–94.

REFERENCES

213

Mishler, E. (1984). The Discourse of Medicine: Dialectics of Medical Interviews. Norwood, NJ: Ablex. Nava, J. M., Bella, F., Garau, J., Lite, J., Morera, M. A., Marti, C., et al. (1994). Predictive factors for invasive disease due to penicillin-resistant Streptococcus pneumoniae: A population-based study. Clinical Infectious Diseases, 19, 884–890. Neu, H. C. (1992). The crisis in antibiotic resistance. Science, 257, 1064–1073. Okeke, I. N., Laxminarayan, R., Bhutta, Z. A., Duse, A. G., Jenkins, P., O’Brien, T. F., et al. (2005). Antimicrobial resistance in developing countries: Part I: Recent trends and current status. Lancet, 5, 481–493. Ong, L., de-Haes, J. C., Hoos, A., & Lammes, F. (1995). Doctor-patient communication: A review of the literature. Social Science and Medicine, 40, 903–918. World Health Organization (1999). Containing Antimicrobial Resistance: Review of the Literature and Report of a WHO Workshop on the Development of a Global Strategy for the Containment of Antimicrobial Resistance. Geneva: World Health Organization. Orr, P. H., Scherer, K., MacDonald, A., & Moffatt, M. (1993). Randomized placebo controlled trials of antibiotics for acute bronchitis: A critical review of the literature. Journal of Family Practice, 36, 507–512. Otters, H. B. M., van der Wouden, J. C., Schellevis, F. G., van Suijlekom-Smit, L. W. A., & Koes, B. W. (2004). Trends in prescribing antibiotics for children in Dutch general practice. Journal of Antimicrobial Chemotherapy, 53, 361–366. Pachter, L. M. (1994). Culture and clinical care: Folk illness beliefs and behaviors and their implications for health care delivery. Journal of the American Medical Association, 271, 690–694. Palmer, D. A., & Bauchner, H. (1997). Parents’ and physicians’ views on antibiotics. Pediatrics, 99, 862–863. Pantell, R. H., Steward, T. J., Dias, J. K., Wells, P., & Ross, A. W. (1982). Physician communication with children and parents. Pediatrics, 70, 396–402. Papper, S. (1970). The undesirable patient. Journal of Chronic Diseases, 22, 777–779. Parsons, T. (1951). The Social System. New York: Free Press. Pendleton, D. (1983). Doctor-patient communication: A review. In D. Pendleton & J. Hasler (Eds.), Doctor-Patient Communication (pp. 5–53). New York: Academic. Pennie, R. A. (1998). Prospective study of antibiotic prescribing for children. Canadian Family Physician, 44, 1850–1856. Peräkylä, A. (1998). Authority and accountability: The delivery of diagnosis in primary health care. Social Psychology Quarterly, 61, 301–320. Perz, J. F., Craig, A. S., Coffey, C. S., Jorgensen, D. M., Mitchel, E., Hall, S., et al. (2002). Changes in antibiotic prescribing for children after a community-wide campaign. Journal of the American Medical Association, 287, 3103–3109. Peyrot, M., Alperstein, N. M., van Doren, D., & Poli, L. G. (1998). Direct-to-consumer ads can inﬂuence behavior: Advertising increases consumer knowledge and prescription drug requests. Marketing Health Services, Summer, 27–32. Pinto, M. B., Pinto, J. K., & Barber, J. C. (1998). The impact of pharmaceutical direct advertising: Opportunities and obstructions. Health Marketing Quarterly, 15, 89–101. Pomerantz, A. (1984). Agreeing and disagreeing with assessments: Some features of preferred/dispreferred turn shapes. In J. M. Atkinson & J. Heritage (Eds.), Structures of Social Action: Studies in Conversation Analysis (pp. 57–101). Cambridge: Cambridge University Press. Pomerantz, A. (1988). Offering a candidate answer: An information seeking strategy. Communication Monographs, 55, 360–373.

214

REFERENCES

Quagliarello, A. B., Parry, C. M., Hien, T. T., & Farrar, J. J. (2003). Factors associated with carriage of penicillin-resistant Streptococcus pneumoniae among Vietnamese children: A rural-urban divide. Journal of Health Population Nutrition, 21, 316–324. Quirk, R., Greenbaum, S., Leech, G., & Svartvik, J. (1972). A Grammar of Contemporary English. London: Longman. Radyowijati, A., & Haak, H. (2003). Improving antibiotic use in low-income countries: An overview of evidence on determinants. Social Science and Medicine, 57, 733–744. Raymond, G. (2003). Grammar and social organization: Yes/no interrogatives and the structure of responding. American Sociological Review, 68, 939–967. Raymond, G. (2004). Prompting action: The stand-alone ‘so’ in sequences of talk-in-interaction. Research on Language and Social Interaction 37, 185–218. Reeder, L. G. (1972). The patient-client as a consumer: Some observations on the changing professional-client relationship. Journal of Health and Social Behavior, 13, 406–412. Reichler, M. R., Allphin, A. A., Breiman, R. F., Schreiber, J. R., Arnold, J. E., McDougal, L. K., et al. (1992). The spread of multiply resistant Streptococcus pneumoniae at a day care center in Ohio. Journal of Infectious Diseases, 166, 1346–1353. Robinson, J. D. (1998). Getting down to business: Talk, gaze, and body orientation during openings of doctor-patient consultations. Human Communication Research, 25, 97–123. Robinson, J. D. (2003). An interactional structure of medical activities during acute visits and its implications for patients’ participation. Health Communication, 15, 27–57. Roter, D., & Hall, J. (1992). Doctors Talking with Patients/Patients Talking with Doctors: Improving Communication in Medical Visits. Westport CT: Auburn House. Roter, D. L. (2002). The Roter Method of Interaction Process Analysis. Retrieved from www. rias.org. Ruusuvuori, J. (2000). Control in the medical consultation: Practices of giving and receiving the reason for the visit in primary health care. Unpublished doctoral dissertation: University of Tampere. Sacks, H. (1992). Lectures on Conversation, Vol. 2 (Fall 1968–Spring 1972). Oxford: Blackwell. Sacks, H., & Schegloff, E. A. (1979). Two preferences in the organization of reference to persons and their interaction. In G. Psathas (Ed.), Everyday Language: Studies in Ethnomethodology (pp. 15–21). New York: Irvington. Sacks, H., Schegloff, E. A., & Jefferson, G. (1974). A simplest systematics for the organization of turn-taking for conversation. Language, 50, 696–735. Sanchez-Menegay, C., & Stalder, H. (1994). Do physicians take into account patients’ expectations? Journal of General Internal Medicine, 9, 404–406. Sargent, J., & Welch, H. G. (2001). Reducing “unnecessary” antibiotic use in primary care: Hard rules, soft calls. Effective Clinical Practice, 4, 136–138. Schegloff, E. A. (1968). Sequencing in conversational openings. American Anthropologist, 70, 1075–1095. Schegloff, E. A. (1972a). Notes on a conversational practice: Formulating place. In D. Sudnow (Ed.), Studies in Social Interaction (pp. 75–119). New York: Free Press. Schegloff, E. A. (1972b). Sequencing in conversational openings. In J. J. Gumperz & D. Hymes (Eds.), Directions in Sociolinguistics: The Ethnography of Communication (pp. 346–380). New York: Holt, Rinehart and Winston. Schegloff, E. A. (1977). Identiﬁcation and recognition in interactional openings. In I. de Sola Pool (Ed.), The Social Impact of the Telephone (pp. 415–450). Cambridge, MA: MIT Press.

REFERENCES

215

Schegloff, E. A. (1982). Discourse as an interactional achievement: Some uses of “uh huh” and other things that come between sentences. In D. Tannen (Ed.), Analyzing Discourse (Georgetown University Roundtable on Languages and Linguistics 1981) (pp. 71–93). Washington, DC: Georgetown University Press. Schegloff, E. A. (1996). Turn organization: One intersection of grammar and interaction. In E. Ochs, S. Thompson, & E. Schegloff (Eds.), Interaction and Grammar (pp. 52–133). Cambridge: Cambridge University Press. Schegloff, E. A. (2000). On granularity. Annual Review of Sociology, 26, 715–720. Schegloff, E. A. (2001). Increments: Where they are and what they do. Paper presented at the Linguistic Institute, Santa Barbara, CA. Schegloff, E. A. (2006). A primer for Conversation Analysis: Sequence Organization. Cambridge: Cambridge University Press. Schegloff, E. A., Jefferson, G., & Sacks, H. (1977). The preference for self-correction in the organization of repair in conversation. Language, 53, 361–382. Schulman, B. A. (1979). Active patient orientation and outcomes in hypertensive treatment. Medical Care, 17, 267–280. Schwartz, B. (1999). Preventing the spread of antimicrobial resistance among bacterial respiratory pathogens in industrialized countries: The case for judicious antimicrobial use. Clinical Infectious Diseases, 28, 211–213. Schwartz, R. H., Freij, B. J., Ziai, M., & Sheridan, M. J. (1997). Antimicrobial prescribing for acute purulent rhinitis in children: A survey of pediatricians and family practitioners. Pediatric Infectious Disease Journal, 16, 185–190. Sheridan, M. (1994). Parents’ reporting of symptoms in their children: Physicians’ perceptions. Hawaii Medical Journal, 53, 216–217. Sleath, B., Svarstad, B., & Roter, D. (1997). Physician vs patient initiation of psychotropic prescribing in primary care settings: A content analysis of audiotapes. Social Science and Medicine, 44, 541–548. Smolinski, M. S., Hamburg, M. A., & Lederberg, J. (Eds.). (2003). Microbial Threats to Health: Emergence, Detection, and Response. Washington, DC: Institute of Medicine. Starr, P. (1982). The Social Transformation of American Medicine. New York: Basic Books. Steffen, R., & Lobel, H. O. (1996). Travel medicine. In G. C. Cook (Ed.), Manson’s Tropical Diseases (Vol. 20, pp. 407–420). London: W. B. Saunders. Stevenson, F. A., Greenﬁeld, S. M., Jones, M., Nayak, A., & Bradley, C. P. (1999). GPs’ perceptions of patient inﬂuence on prescribing. Family Practice, 16, 255–261. Stivers, T. (1998). Pre-diagnostic commentary in veterinarian-client interaction. Research on Language and Social Interaction, 31, 241–277. Stivers, T. (2000). Negotiating antibiotic treatment in pediatric care: The communication of preferences in physician-parent interaction. Doctoral Dissertation. University of California, Los Angeles. Stivers, T. (2001). Negotiating who presents the problem: Next speaker selection in pediatric encounters. Journal of Communication, 51, 1–31. Stivers, T. (2002a). Participating in decisions about treatment: Overt parent pressure for antibiotic medication in pediatric encounters. Social Science and Medicine, 54, 1111–1130. Stivers, T. (2002b). “Symptoms only” versus “candidate diagnoses”: Presenting the problem in pediatric encounters. Health Communication, 14, 299–338. Stivers, T. (2005a). Modiﬁed repeats: One method for asserting primary rights from second position. Research on Language and Social Interaction, 38, 131–158. Stivers, T. (2005b). Non-antibiotic treatment recommendations: Delivery formats and implications for parent resistance. Social Science and Medicine, 60, 949–964.

216

REFERENCES

Stivers, T. (2005c). Parent resistance to physicians’ treatment recommendations: One resource for initiating a negotiation of the treatment decision. Health Communication, 18, 41–74. Stivers, T. (Submitted). Stance, alignment and afﬁliation during story telling: Nodding as a token of preliminary afﬁliation. Stivers, T., & Heritage, J. (2001). Breaking the sequential mould: Answering “more than the question” during comprehensive history taking. Text, 21, 151–185. Stivers, T., Mangione-Smith, R., Elliott, M. N., McDonald, L., & Heritage, J. (2003). Why do physicians think parents expect antibiotics? What parents report vs what physicians perceive. Journal of Family Practice, 52, 140–148. Strong, P. (1979). The Ceremonial Order of the Clinic. London: Routledge. Swenson, S. L., Buell, S., Zettler, P., White, M., Ruston, D. C., & Lo, B. (2004). Patient centered communication. Journal of General Internal Medicine, 19, 1069–1079. Szasz, P. S., & Hollender, M. H. (1956). A contribution to the philosophy of medicine: The basic model of the doctor-patient relationship. Archives of Internal Medicine, 97, 585– 592. Terasaki, A. (2004). Pre-announcement sequences in conversation. In G. Lerner (Ed.), Conversation Analysis: Studies from the First Generation (pp. 171–223). Washington, DC: University Press of America. Thomas, K. (1999). Prospective multicentre hospital surveillance of Streptococcus pneumoniae disease in India. Lancet, 353, 1216–1221. Thompson, S. C., Pitts, J. S., & Schwankovsky, L. (1993). Preference for involvement in medical decision-making situational and demographic inﬂuences. Patient Education and Counseling, 22, 133–140. Todd, A. D., & Fisher, S. (Eds.). (1993). The social organization of doctor-patient communication. Norwood, NJ: Ablex. Todd, J. K., Todd, N., Damato, J., & Todd, W. A. (1984). Bacteriology and treatment of purulent nasopharyngitis: A double blind, placebo-controlled evaluation. Pediatric Infectious Disease, 3, 226–232. Tuckett, D., Boulton, M., Olson, C., & Williams, A. (1985). Meetings between Experts: An Approach to Sharing Ideas in Medical Consultations. London: Tavistock. U.S. Department of Health and Human Services, Public Health Service (2000). Healthy People 2010: Understanding and Improving Health (2nd ed.). Washington, DC: U.S. Government Printing Ofﬁce. Uppal, R., Sarkar, U., Giriyappanavar, C. S., & Kacker, V. (1993). Antimicrobial drug use in primary health care. Journal of Clinical Epidemiology, 46, 671–673. Vinson, D. C., & Lutz, L. J. (1993). The effect of parental expectations on treatment of children with a cough: A report from the ASPN. Journal of Family Practice, 37, 23–27. Vuckovic, N., & Nichter, M. (1997). Changing patterns of pharmaceutical practice in the United States. Social Science and Medicine, 44, 1285–1302. Waitzkin, H. (1991). The Politics of Medical Encounters. New Haven, CT: Yale University Press. Wald, E. R., Guerra, N., & Byers, C. (1991). Upper respiratory tract infections in young children: Duration of and frequency of complications. Pediatrics, 87, 129–133. Warren, J. (1998, December 28). Too much of a good thing. San Diego Union Tribune, pp. E1, E3. Watanabe, H., Sato, S., Kawakami, K., Watanabe, K., Oishi, K., Rikitomi, N., et al. (2000). A comparative clinical study of pneumonia by penicillin-resistant and -sensitive Streptococcus pneumoniae in a community hospital. Respirology, 5, 59–64.

REFERENCES

217

Watson, R. L., Dowell, S. F., Jayaraman, M., Keyserling, H., Koczak, M., & Schwartz, B. (1999). Antimicrobial use for pediatric upper respiratory infections: Reported practice, actual practice and parent beliefs. Pediatrics, 104, 1251–1157. Weber, J. T., & Courvalin, P. (2005). An emptying quiver: Antimicrobial drugs and resistance. Emerging Infectious Diseases, 11, 791–793. West, C. (1984). Routine Complications: Troubles with Talk between Doctors and Patients. Bloomington: Indiana University Press. Whalen, J., Zimmerman, D. H., & Whalen, M. R. (1988). When words fail: A single case analysis. Social Problems, 35(4), 335–362. Whalen, M., & Zimmerman, D. H. (1987). Sequential and institutional contexts in calls for help. Social Psychology Quarterly, 50, 172–185. Wheeler, J. G., Fair, M., Simpson, P. M., Rowlands, L. A., Aitken, M. E., & Jacobs, R. F. (2001). Impact of a waiting room videotape message on parent attitudes toward pediatric antibiotic use. Pediatrics, 108, 591–596. Whitney, C. G., Farley, M. M., Hadler, J., Harrison, L. H., Lexau, C., Reingold, A., et al. (2000). Increasing prevalence of multidrug-resistant Streptococcus pneumoniae in the United States. New England Journal of Medicine, 343, 1917–1924. World Health Organization. (2000). Overcoming antimicrobial resistance. World Health Organization Report on Infectious Diseases 2000. Retrieved August 18, 2006, from http:// www.who.int/infectious-disease-report/2000/intro.htm Wise, R., Hart, T., Cars, O., Streulens, Helmuth, R., Huovinen, P., et al. (1998). Antimicrobial resistance is a major threat to public health. British Medical Journal, 317, 609–610.

This page intentionally left blank

INDEX

Additional symptoms, mention of, 63–67, 71–76, 186 Alternative diagnosis, mention of, 67–70, 186 Afﬁliation, 41, 53, 81, 141–142, 147 Afﬁrmative announcements, 164–167, 169, 188. See also Diagnosis delivery; Treatment recommendation Antibiotics history of, 4–5 inquires about, 143–148 mentions of past experience with, 148–151 preference for, 139, 144 rate of use, 3, 7, 9, 12 regulation of, 6 relevance of, 39 requests for, 10, 126, 137–139 statement of desire for, 139–143, 150 use in livestock, 5 Antimicrobial resistance, 4, 5 determinants of, 5–8 rates of, 6, 8, 9 Australia, 192 Avorn, J., 7 Bacterial resistance. See Antimicrobial resistance

Bangladesh, 192 Barber, J. C., 11 Beach, W., 200n Belgium, 8 Bell, R. A., 11, 12 Boyd, E. A., 25, 53, 55, 149 Brewer, M. B., 191 Britten, N., 21 Byrne, P. S., 23, 70 Candidate Diagnosis. See Problem presentation, candidate diagnosis Centers for Disease Control (CDC), 6, 13 Children in the pediatric visit, role of, 16 Clayman, S. E., 52 Commons dilemma, 7–8, 191, 199n. See also Social dilemma Concession, 75, 125, 146 Conditional relevance, 79, 87, 106–114, 148, 164, 187 Conﬁrmation, 46, 47, 48, 49, 82, 165 Consumerist movement, 11 Contingency plan, 75 Conversation analysis, 13–15 Counseling phase, 43–49 Course of action, 52 Cultural authority, 78, 79 Culture, 7

219

220

INDEX

Danish, 200n Developing countries, 5–6 Diagnosis delivery, 78, 164–171, 180–184, 187 online, 77, 83 Diagnosis resistance, 77–104, 180–184, 187 Diamond, J., 189 Differential diagnosis, 52, 58 Direct to consumer advertising, 4, 11–12 Discourse analysis, 13–14 Doctorability, 17, 18, 53 Domain of expertise, 10, 28, 88, 106, 129, 132, 137, 153, 187–188 Drew, P., 52, 84, 200n Epistemic territory, 14, 29, 48, 78, 89, 92, 106, 129, 133–136. See also Domain of expertise Escalation, 153, 171, 172 Evidential formulation, 160, 163, 201n Fisher, S., 14 Fleming, A., 3, 5, 7, 192 Foster, G. M., 199n France, 8, 9 Franz, C. E., 11 Germany, 8 Gill, V., 28 Guidance model, 13 Haakana, M., 74, 141 Halkowski, T., 18 Hall, J., 11, 19 Hardin, G., 7–8 Haug, M., 11 Heath, C., 78 Heinemann, T., 200n Helman, C. G., 78 Heritage, J., 14, 28, 52, 53, 55, 70, 78, 79, 81–82, 113, 114, 149, 163, 200n History taking, 51–76, 186 Hong Kong, 9 Humphrey, N. K., 8

Jefferson, G., 16, 84 Kleinman, L., 149 Kollock, P., 191–192 Korsch, B., 13, 16 Koshik, I., 200n Kramer, R. M., 191 Kravitz, R. L., 11, 12 Lavin, B., 11 Legitimacy, 17–21, 39, 50, 54, 62, 69, 97, 124–125, 155–158, 160, 163, 179–184, 187–188, 192 Levinson, S. C., 55–56, 199n Lifeworld agenda, 14, 53. See also Mishler, E. Long, B. E. L., 23, 70 Malpractice, 7 Mangione-Smith, R., 185 Markedness marked, 53, 78, 199n unmarked, 27, 38, 49, 164–167 Maynard, D., 14, 155 Medical agenda, 14 Medical authority, 133, 153. See also Cultural authority Medical error, 7, 185 Mexico, 9 Mishler, E., 14 Misuse of antibiotics, 5 in developed countries, 6–8, 9 in developing countries, 5–6 Negative announcements, 167–170. See also Diagnosis delivery; Treatment recommendation Negotiation, 23, 187 of diagnosis, 40–50, 72–76, 96 of treatment, 72–76, 84, 102, 120, 124– 129, 131–154 Netherlands, The, 8, 193 Newsmarks, 81–88, 92, 99, 180–181

Illness model, 50, 190 India, 6 Israel, 9

“Oh”-preface, 53, 163 Online commentary, 91, 158–163, 188, 190, 201n Overt negotiation, 103, 131–154, 187

Japan, 9

Parent expectations, 9

INDEX

Parent pressure motivations for, 98 physician citation of, 9, 10 physician response to, 117, 151–154 Parsons, T., 11, 18 Patient satisfaction, 11 Patient/parent participation, 10–13, 50, 53, 78 Peräkylä, A., 78, 92 Physician perceptions disconnect with parent expectations, 10 of parents, 10, 190 See also Parent expectations Pinto, J. K., 11 Pinto, M. B., 11 Portugal, 8 Power, 14 Prescribing rates. See Antibiotics, rate of use Presuppositions, 52, 53, 70 Problem presentation, 23–50 Problem presentation candidate diagnosis, 24, 28–36, 39–50, 67, 133, 156, 159, 186, 199n symptoms-only, 24, 25–27, 36–39 Progressivity, 71, 79 Prosody, 57, 92, 94, 113, 122, 165 Public health campaigns, 189–190, 199n Question design negative polarity item, 54, 55 optimization, 53, 54, 55, 56 preference, 53, 54, 55, 63, 79, 89, 200n problem attentiveness, 55, 56 recipient design, 53 Questioning the diagnosis, 92–97 an examination ﬁnding, 88–92 Raymond, G., 53, 200n Roter, D., 11, 12, 13, 19 Roter Interaction Analysis System (RIAS), 13 Ruling out treatment. See Negative announcements Ruusuvuori, J., 24, 28 Sacks, H., 84, 200n

221

Schegloff, E. A., 84, 201n Seﬁ, S., 81–82, 113, 114 Sequence expansion, 63, 71, 75, 81–97 Shared decision making, 11 Sheridan, M., 19 Sleath, B., 12 Social action, 14 Social dilemma, 7, 185, 191, 199n Solomon, D. H., 7 Sorjonen, M., 70 Spain, 8, 9 Stance, 53, 76 Stance adjustment of, 75 towards diagnosis, 55, 60, 85, 93, 166 towards problem, 39, 48, 52, 61, 68, 78, 163, 186 towards treatment, 57, 59, 119, 147, 151 Streptococcus pneumoniae, 3, 6 Strong, P., 18 Structural organization of the visit, 23, 79, 82 Svarstad, B., 12 Switzerland, 8 Todd, A. D., 14 Treatability, 17, 20–21, 25–27, 29, 39, 50, 53, 54, 97–104, 124–125, 179, 187 Treatment change, 124–129, 152–153 recommendation, 106–114, 164–184, 174–176 Treatment resistance, 105–130, 114–115, 163, 171–180, 187 active, 114–130 passive, 109, 113, 124, 145 Troubles resistance, 17, 18, 19, 59, 61, 125, 149 United Kingdom, 8 Vietnam, 192 West, C., 14 Wilkes, M. S., 12 World Health Organization (WHO), 5, 6

Language without Rights (Oxford Studies in Sociolinguistics)

Read more

under pressure

Read more

Under Pressure

Read more

Queerly Phrased: Language, Gender, and Sexuality (Oxford Studies in Sociolinguistics)

Read more

Gender and Conversational Interaction (Oxford Studies in Sociolinguistics)

Read more

Queerly Phrased: Language, Gender, and Sexuality (Oxford Studies in Sociolinguistics)

Read more

Claiming Power in Doctor-Patient Talk (Oxford Studies in Sociolinguistics)

Read more

Coherence in Psychotic Discourse (Oxford Studies in Sociolinguistics)

Read more

Girls Under Pressure

Read more

Caution- Contents under Pressure

Read more

Hot Under Pressure

Read more

Welfare States under Pressure

Read more

Hot Under Pressure

Read more

Hot Under Pressure

Read more

Grace Under Pressure

Read more

Caution - Contents under Pressure

Read more

Therapeutic Ways with Words (Oxford Studies in Sociolinguistics)

Read more

Girls Under Pressure

Read more

Teachers Under Pressure

Read more

Hot Under Pressure

Read more

Hot Under Pressure

Read more

Planning Under Pressure

Read more

Interpreting As a Discourse Process (Oxford Studies in Sociolinguistics)

Read more

Sociolinguistic Variation: Critical Reflections (Oxford Studies in Sociolinguistics)

Read more

Sociolinguistic Perspectives on Register (Oxford Studies in Sociolinguistics)

Read more

Discourse and Identity (Studies in Interactional Sociolinguistics)

Read more

Discourse and Identity (Studies in Interactional Sociolinguistics)

Read more

Gender and Politeness (Studies in Interactional Sociolinguistics)

Read more

Antibiotics

Read more

Imagination under pressure 1789-1832

Read more

Recommend Documents

Language without Rights (Oxford Studies in Sociolinguistics)

LANGUAGE WITHOUT RIGHTS OXFORD STUDIES IN SOCIOLINGUISTICS General Editors: Nikolas Coupland Adam Jaworski Cardiff Un...

under pressure

Under Pressure - 1 This is a work of fiction. Names, characters, places, and incidents either are the product of the a...

Under Pressure

Under Pressure (current title) e-version 5.0 aka The Dragon in the Sea (original title) aka 21st Century Sub (second t...

Queerly Phrased: Language, Gender, and Sexuality (Oxford Studies in Sociolinguistics)

QuEERLY P H R A S E D OXFORD STUDIES IN SOCIOLINGUISTICS Edward Finegan, General Editor Editorial Advisory Board Doug...

Gender and Conversational Interaction (Oxford Studies in Sociolinguistics)

Gender and Conversational Interaction OXFORD STUDIES IN SOCIOLINGUISTICS Edward Finegan, General Editor EDITORIAL ADV...

Queerly Phrased: Language, Gender, and Sexuality (Oxford Studies in Sociolinguistics)

QuEERLY P H R A S E D OXFORD STUDIES IN SOCIOLINGUISTICS Edward Finegan, General Editor Editorial Advisory Board Doug...

Claiming Power in Doctor-Patient Talk (Oxford Studies in Sociolinguistics)

C L A I M I N G POWER IN D O C T O R - P A T I E N T TALK OXFORD STUDIES IN SOCIOLINGUISTICS Edward Finegan, General...

Coherence in Psychotic Discourse (Oxford Studies in Sociolinguistics)

COHERENCE IN PSYCHOTIC DISCOURSE Oxford Studies in Sociolinguistics Edward Finegan, General Editor Editorial Advisory...

Girls Under Pressure

Caution- Contents under Pressure