The Dependent Patient A Practitioner's Guide
R O B E R T F. B O R N S T E I N
American Psychological Association • Washington, DC
Copyright © 2005 by the American Psychological Association. All rights reserved. Except as permitted under the United States Copyright Act of 1976, no part of this publication may be reproduced or distributed in any form or by any means, or stored in a database or retrieval system, without the prior written permission of the publisher. Published by American Psychological Association 750 First Street, NE Washington, DC 20002 www.apa.org To order APA Order Department P.O. Box 92984 Washington, DC 20090-2984 Tel: (800) 374-2721; Direct: (202) 336-5510 Fax: (202) 336-5502; TDD/TTY: (202) 336-6123 Online: www.apa.org/books/ E-mail:
[email protected] In the U.K., Europe, Africa, and the Middle East, copies may be ordered from American Psychological Association 3 Henrietta Street Covent Garden, London WC2E 8LU England Typeset in Goudy by Stephen McDougal, Mechanicsville, MD Printer: Sheridan Books, Ann Arbor, MI Cover Designer: Naylor Design, Washington, DC Technical/Production Editor: Emily Leonard The opinions and statements published are the responsibility of the authors, and such opinions and statements do not necessarily represent the policies of the American Psychological Association. Library of Congress Cataloging-in-Publication Data Bornstein, Robert F. The dependent patient: a practitioner's guide / Robert F. Bornstein.—1st ed. p. cm. Includes bibliographical references and index. ISBN 1-59147-203-2 (alk. paper) 1. Dependency (Psychology) I. Title. RC569.5.D47B669 2005 616.85'81—dc22 British Library Cataloguing-in-Publication Data A CIP record is available from the British Library. Printed in the United States of America First Edition
2004019485
To my students, whose insights encourage me to revisit important ideas and question longstanding assumptions. Without realizing it, they have taught me far more than I could ever teach them.
CONTENTS
Preface
xi
I.
1
Conceptual and Empirical Foundations Chapter 1. Conceptualizing Dependency Psychodynamic Models Behavioral and Social Learning Models Culture and Gender Role Cognitive Models Trait Models An Interactionist Model Integrating the Contributions of Extant Theoretical Frameworks
3 4 6 8 11 13 15
Chapter 2.
21 21 22
Chapter 3.
18
Quantifying Dependency Classifying Measures of Dependency Stand-Alone Self-Report Measures Self-Report Scales Embedded in Longer Tests Projective Tests Interviews Combining and Contrasting Test Results....
27 28 31 33
Dependency Across the Life Span Dependency as a Pervasive Life Theme The Evolution of Dependency
39 40 45
vn
Life-Span Dependency: Toward an Integrated View Chapter 4- Context-Specific Deficits and Strengths From Pervasive Passivity to Situational Variability An Interactionist Perspective on Dependency Dependency-Related Problems and Deficits.. Dependency-Related Skills and Strengths ... Clinical Implications of the Interactionist Perspective Chapter 5.
II.
Healthy and Unhealthy Dependency Distinguishing Healthy and Unhealthy Dependency Unhealthy Dependency Patterns From Laboratory to Consulting Room: Diagnosis, Assessment, and Treatment of Dependent Patients
57 58 61 63 67 70 73 74 82
86
Clinical Applications
89
Chapter 6.
91
Diagnosis
Dependent Personality Disorder in the DSM Epidemiology Differential Diagnosis and Comorbidity Effective Use of Diagnostic Information: A Framework for the Practitioner Chapter 7. Assessment Testing Versus Assessment Measuring Implicit and Self-Attributed Dependency Needs Dependency in Context: Perceptions, Defenses, and Social Support Assessing Dependency Subtypes Assessing Dependency's Impact Effective Use of Assessment Information: A Framework for the Practitioner
Vlll
54
CONTENTS
93 96 99 104 Ill 112 112 116 120 121 123
Chapter 8.
Approaches to Treatment The Psychodynamic Perspective The Behavioral Perspective The Cognitive Perspective The Humanistic-Experiential Perspective ... Effective Use of Traditional Treatment Models: A Framework for the Practitioner
Chapter 9. An Integrated Treatment Model Assimilative Integration of Dependency Treatment Models Integrating Strategies Across Domains of Patient Functioning Effective Use of Therapeutic Integration Strategies: A Framework for the Practitioner Chapter 10. Specialized Treatment Issues Contexts and Settings Alternative Treatment Modalities Special Treatment Challenges Effective Use of Alternative Treatment Strategies: A Framework for the Practitioner
131 132 135 138 141
144 151 154 156
164 173 174 176 183
185
References
193
Index
235
About the Author
243
CONTENTS
IX
PREFACE
Problematic dependency takes many forms. Some people are obviously dependent, unable to make even the smallest decision without an inordinate amount of advice and reassurance from others. But many people who do not appear dependent still bring dependency-related issues to therapy. The depressed patient who can no longer manage her life, the anxious patient who cannot leave home on his own, the borderline patient with deep-seated abandonment fears, and the elderly patient who is unable to carry out basic activities of daily living—all these people (and countless others) have difficulties involving dependency. This book is about working effectively with dependent patients. It discusses cutting-edge treatment techniques, outlines strategies for diagnosing dependency, and reviews procedures for assessing dependency-related personality dynamics that are not captured by formal diagnostic criteria. The central premise of this book is straightforward: Dependency is an important issue in clinical practice, but it is also a ubiquitous feature of human experience. We are, in the end, social creatures, bound to each other from our first days to our last. Thus, effective clinical work with dependent patients does not involve quashing dependency in all its forms, but replacing unhealthy dependency with healthy connectedness. Two events that took place nearly 20 years ago laid the groundwork for this book, and they illustrate two themes that characterize my approach to the dependent patient. The first event occurred in the mid-1980s while I was working in a psychiatric inpatient unit in upstate New York. There I met a 28-year-old bulimic woman named Marta, who—despite being highly dependent—showed considerable variability in behavior. Around her siblings Marta was clingy and insecure, but at work she was confident and assertive— completely in command. Marta's behavior did not square well with clinical writings on dependency at that time, and her unpredictable behavior was XI
difficult to comprehend. Now I understand Marta better. Her dependent behavior was unpredictable, but it was unpredictable in a very predictable way. Marta—like many dependent patients—was displaying an array of selfpresentation strategies that, while superficially different, had the common goal of pleasing those closest to her, drawing others in, and minimizing the possibility of relationship disruption. From Marta 1 learned that dependent behavior is far more diverse (and far more deliberate) than clinicians once believed. Although dependent people are motivated to seek protection and help from others, they express this need in many different ways—sometimes acting meek and passive, at other times becoming active and even quite aggressive. The second eye-opening event took place a year later, and it involved a research study that refused to come out the way we had hoped. As so often happens in the world of science, however, our "failed" investigation proved far more informative than it would have had the results turned out as expected. Joseph Masling, Frederick Poynton, and I were assessing the relationship between self-report and projective dependency scores in college students. We used well-established questionnaire and projective measures of dependency—the Depressive Experiences Questionnaire and the Rorschach Oral Dependency Scale—and we expected that scores on the two measures would show strong positive correlations. They did not. No matter how many samples we ran, the result was always the same: A very modest link between self-report and projective dependency scores. Because we had been taught to expect different measures of the same trait to correlate strongly, we assumed that something had to be wrong with one or both tests, and it took us a long time to understand what we had found. Now we know these results are but one example of a more general pattern that emerges in clinical assessment: Self-report and projective measures tap different aspects of dependency. Knowing this, clinicians must treat these assessment instruments as unique sources of data about a patient's personality and coping style. Divergences between scores on different indices of dependency can be as important and informative as convergences—sometimes more so. For the past 20 years I have studied the dynamics of dependency in clinical and research settings, and the results of these investigations form the foundation of this book. I am not alone in my interest in dependency, of course: There have been hundreds of clinical and empirical papers on this topic during the past two decades. Effective treatment of the dependent patient begins with understanding the results of these research programs, but it does not end there. Self-awareness is important as well. The ubiquitous, "experience-near" nature of dependency demands that knowledge of research and clinical evidence go hand-in-hand with self-reflection and self-scrutiny. To work effectively with dependent patients, clinicians must understand their own beliefs about dependency and their reflexive responses to dependent behavior in others. I discuss this challenge—and ways to overcome it—in the second half of the book. xii
PREFACE
To set the stage for effective clinical work with dependent patients, I have drawn upon as broad an array of sources as possible, including: • Laboratory and clinical data. Clinical studies are ideal for understanding the antecedents, correlates, and consequences of dependency, but laboratory experiments provide a unique method for manipulating dependency-related psychological processes and disentangling complex relationships among these processes. • Studies of nonclinical participants and studies of patients. Since the early 1950s, there have been more than 600 empirical studies of interpersonal dependency. At least half involved nonclinical participants, and the results of these investigations tell us much about the inter- and intrapersonal dynamics of dependency in vivo. • Studies of dependent personality traits and studies of dependent personality disorder. Some investigators operationalize dependency using dependent personality disorder symptoms; others use selfreport, projective, or behavioral indices to quantify patients' dependency levels. Because different assessment methods tap different aspects of functioning, these contrasting findings can teach us a great deal about the various ways dependency is expressed. The Dependent Patient is intended not to be a cookbook but a framework each clinician will use differently. Thus, I review a broad array of theories, assessment methods, and treatment approaches. I make recommendations in each area—useful diagnostic techniques, assessment tools, and psychotherapeutic interventions—but these recommendations are merely guidelines to be adapted to each patient and each situation. This book is structured so that research evidence informs every clinical decision confronting the practitioner. Thus, in the first half of the book I review the conceptual and empirical foundations of dependency: Conceptualizing Dependency (chap. 1), Quantifying Dependency (chap. 2), Dependency Across the Life Span (chap. 3), Context-Specific Deficits and Strengths (chap. 4), and Healthy and Unhealthy Dependency (chap. 5). In the second half of the book, I discuss clinical applications of these findings: Diagnosis (chap. 6), Assessment (chap. 7), Approaches to Treatment (chap. 8), An Integrated Treatment Model (chap. 9), and Specialized Treatment Issues (chap. 10). The Dependent Patient is based on the scientist—practitioner model but not constrained by it. It extends this model while applying the model's core principles to integrate dependency-related research and clinical data. Recent changes within the mental health professions have compelled us to reconceptualize the traditional role of the scientist-practitioner. Managed care, a changing professional landscape, prescription and hospital admitting PREFACE
Xlii
privileges, the emergence of new categories of mental health professionals— all have reshaped the life and work of the clinician. Add to this the fact that clinicians—like patients—are imperfect processors of information, subject to all sorts of perceptual and memory biases that limit our ability to apply knowledge in an objective, dispassionate way, and it is easy to see that bringing clinical science to the consulting room is fraught with pitfalls, but replete with opportunities as well. I would like to thank the many people who contributed to this book, without whom I could not have written it. I am indebted to Susan Reynolds and Mary Lynn Skutley of APA Books, whose enthusiasm for this project provided the spark that enabled me to complete it, and to Emily Welsh and Emily Leonard, whose editorial expertise improved The Dependent Patient in countless ways. I am grateful to the library staff of Gettysburg College—most especially Susan Roach and Linda Isenberger—for going beyond the call of duty to help me locate and obtain books and articles on dependency. As always, Susan and Linda saw my many requests as an opportunity—not a burden—and they tracked down every piece of information I requested, no matter how obscure. I am indebted to Carolyn Tuckey for her help in constructing readerfriendly graphs and figures. Without her creative efforts, my ideas would not have been expressed as clearly or as well. Many colleagues gave willingly of their time and energy, commenting on drafts of The Dependent Patient, providing advice and support, and improving the book tremendously. I am grateful to Kathleen Cain, Kristen Eyssell, Fritz Gaenslen, Jeffrey Johnson, Mary Languirand, Joseph Masling, Anne Sauve, and Joel Weinberger for their very helpful feedback regarding preliminary versions of various chapters. Their insightful comments and challenging questions helped clarify my thinking and enabled me to communicate these ideas more effectively.
XIV
PREFACE
I CONCEPTUAL AND EMPIRICAL FOUNDATIONS
1 CONCEPTUALIZING DEPENDENCY
Virtually every mental health professional has encountered patients who are overly dependent—patients who seem unable to make a decision on their own and alienate those around them with insatiable neediness and clinging insecurity. Studies show that rigid, inflexible dependency has myriad negative effects on a person's social, career, and romantic relationships. As every clinician knows, it can undermine treatment as well. The problem, in its most basic form, is straightforward: Many dependent patients become so comfortable in therapy's protective cocoon that they resist change to perpetuate the relationship. Impending termination brings forth an array of responses—some conscious, some unconscious—aimed at undoing progress and obviating gain. The therapist's emotional reactions to the overdependent patient's resistance run the gamut from infantilizing overprotectiveness to thinly-veiled hostility—even outright anger and resentment. There is no doubt about it: Dependent patients present unique challenges for practitioners. In this chapter, I review theoretical models of interpersonal dependency. These models conceptualize dependent traits from a variety of perspectives, and each model informs the clinician in important ways about some key features of dependency. The theoretical frameworks described in this chapter—along with the dependency assessment tools discussed in chapter 2—
serve as context for a review of empirical research in chapters 3 through 5. Ultimately, the theoretical frameworks described here help set the stage for an in-depth consideration of clinical strategies for diagnosing (chap. 6), assessing (chap. 7), and treating dependent patients (chaps. 8-10). PSYCHODYNAMIC MODELS Although the most widely studied psychoanalytic perspective on dependency is Freud's well-known "oral fixation" model, there are actually several psychodynamic frameworks relevant to this issue. These models differ with respect to the intra- and interpersonal dynamics that are presumed to underlie a dependent personality orientation, but they share a common emphasis on early relationships as the building blocks of dependency. Classical Psychoanalytic Theory In classical psychoanalytic theory, dependency is inextricably linked to events that occur during the first months of life—the period Freud termed the "oral" stage of development. In Freud's (1905/1953) model, frustration or overgratification during the infantile, oral stage was thought to result in oral fixation and an inability to resolve the developmental issues that characterize this period (i.e., conflicts regarding dependency and autonomy). As Freud (1908/1959, p. 167) noted, "one very often meets with a type of character in which certain traits are very strongly marked while at the same time one's attention is arrested by the behavior of these persons in regard to certain bodily functions." Thus, classical psychoanalytic theory postulates that the orally fixated (or "oral dependent") person will (a) remain dependent on others for nurturance, guidance, protection, and support; and (b) continue to exhibit behaviors in adulthood that mirror those of the oral stage (e.g., preoccupation with activities of the mouth, reliance on food and eating as a strategy for coping with anxiety). Empirical support for Freud's (1905/1953, 1908/1959) psychoanalytic model of dependency has been mixed. On the positive side, it is clear that early relationships with parents and other caregivers play a role in the etiology of dependent personality traits (Fisher & Greenberg, 1996; Masling, 1986). On the negative side, however, there is no evidence that oral fixation as Freud (1905/1953) described it affects dependency-related behaviors during adolescence and adulthood. Efforts to link the development of dependent traits to infantile feeding and weaning variables have produced uniformly negative results. Moreover, studies indicate that dependent children, adolescents, and adults do not show greater preoccupation than their nondependent counterparts with food- and mouth-related activities (Bornstein, 1992,1996a). Despite the theory's limitations, it would be a mistake to reject Freud's classical psychoanalytic model of dependency outright. Aside from calling 4
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
attention to the importance of early relationships in shaping dependent attitudes and behaviors, the psychoanalytic model is unique in its emphasis on unconscious determinants of dependency. As we will see in chapter 2—and again when we consider assessment and treatment issues in the second half of the book—the distinction between unconscious (or "implicit") dependency needs and conscious dependency strivings is critical in understanding the complex inter- and intrapersonal dynamics of dependency. Object Relations Theory and Self Psychology Beginning in the 1920s, the focus of psychodynamic metapsychology shifted from Freud's (1905/1953) drive-based framework to a more personcentered approach that came to be known as object relations theory (Greenberg & Mitchell, 1983). Within this framework, personality development and dynamics are conceptualized in terms of (a) self-other interactions (both real and imagined) and (b) internalized mental representations of self and significant figures (Bornstein, 2003; Galatzer-Levy & Cohler, 1993; Huprich, 2001). Although there is considerable overlap between object relations theory and self psychology—with respect to underlying assumptions as well as formal terminology—there are some noteworthy differences as well. Object relations models generally emphasize the interpersonal dynamics and enduring relationship patterns that foster and maintain dependent behavior (e.g., Fairbairn, 1952; Kernberg, 1975), whereas models derived from self psychology emphasize the role of the dependent person's self-concept and internalized self-representation in the etiology and dynamics of dependency (e.g.,Kohut, 1971, 1977).1 As Greenberg and Mitchell (1983) and others (e.g., Galatzer-Levy & Cohler, 1993) have noted, the reconceptualization of psychoanalytic concepts in relational terms introduced a fundamentally new paradigm for understanding continuity and change in personality development and dynamics. Instead of being understood solely in terms of a dynamic balance among id, ego, and superego, stability in personality was now seen as stemming from continuity in the core features of key object representations, including the self-representation (Bornstein, 1996a, 2003). Conversely, personality change was presumed to occur in part because internalized representations of self and other people evolve in response to changing life circumstances. In recent years, Blatt's (1974, 1991) theoretical framework has been the most influential object relations model of dependency. Integrating psychoanalytic principles with research on cognitive and social development, Blatt and his colleagues (e.g., Blatt & Schichman, 1983; Blatt & Zuroff, 1992) have argued that dependent personality traits result from a mental represen'Among the seminal object relations and self psychology contributions to the etiology and dynamics of dependency are those by Guntrip (1961), Jacobson (1964), Bowlby (1969, 1980), Ainsworth (1969), and Sandier and Dare (1970).
CONCEPTUALIZING DEPENDENCY
tation of the self as weak and ineffectual. Retrospective and prospective studies of parent-child interactions confirm that those parenting styles that cause children to perceive themselves as powerless and vulnerable are in fact associated with high levels of dependency later in life. As children internalize a mental representation of the self as weak, they (a) look to others to provide nurturance and support; (b) become preoccupied with fears of abandonment; (c) behave in an overtly dependent manner; and (d) show increased risk for depression and other "anaclitic" (i.e., dependency-related) psychopathologies (see Blatt & Homann, 1992, for a review of research in this area).
BEHAVIORAL AND SOCIAL LEARNING MODELS The basic premise of the behavioral perspective is straightforward: People exhibit dependent behaviors because those behaviors are rewarded, were rewarded, or—at the very least—are perceived by the individual as likely to elicit rewards. Early behavioral and social learning models of dependency were strongly influenced by the work of Hull (1943) and Mowrer (1950). Thus, dependency was initially conceptualized as an acquired drive, the impetus for which was the reduction of basic, primary drives (e.g., hunger) within the context of the infant-caregiver relationship (see Dollard & Miller, 1950, for a detailed discussion of this view). Ainsworth (1969, p. 970) provided a succinct summary of the behavioral-social learning perspective, noting that within this framework dependency is regarded as "a class of behaviors, learned in the context of the infant's dependency relationship with his mother . . . although the first dependency relationship is a specific one, dependency is viewed as generalizing to subsequent interpersonal relationships." Instrumental and Emotional Dependency As the behavioral approach gained influence during the early 1950s, researchers delineated separate categories of dependent behavior (e.g., helpseeking, reassurance-seeking, etc.). Although several frameworks were developed to address this issue, the mo'st influential early subtype model was that of Heathers (1955), who hypothesized that dependent behaviors could be usefully divided into instrumental and emotional categories. Heathers argued that in instrumental dependency, other peoples' responses serve as tools that help the individual meet some goal. Thus, the instrumentally dependent person's actions are directed primarily toward task-oriented helpseeking. In emotional dependency, other peoples' responses are reinforcing in and of themselves—merely eliciting the desired response (e.g., reassurance, support) is the dependent person's goal. Thus, the emotionally dependent person's actions tend to be focused on obtaining succorance and nurturance rather than instrumental help. Within Heathers's framework in6
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
strumental and emotional dependency were thought to have different antecedents, correlates, and interpersonal consequences. Heathers's (1955) instrumental-emotional dependency distinction was influential throughout the late 1950s and early 1960s, but it was eventually criticized by Walters and Parke (1964), Gewirtz (1972), and others. Although Heathers had conceptualized instrumental and emotional dependency as orthogonal constructs, studies showed that the behaviors associated with these two dependency subtypes were in fact strongly linked, and scores on observational measures of instrumental and emotional dependency were highly intercorrelated in children (Kagan & Mussen, 1956) and adults (Baltes, 1996). Other investigations suggested that instrumental and emotional dependency had common antecedents and parallel acquisition, maintenance, and extinction patterns (Ainsworth, 1969). Most clinicians and researchers now regard instrumental and emotional dependency as two facets of a broader dependent personality style, with most (but not all) people who show high levels of one dependency facet also showing high levels of the other.2 Social Reinforcement of Dependent Behavior A natural outgrowth of the behavioral view was the notion that dependent behaviors are shaped in social settings. Even if dependent behavior was first acquired in the child's early interactions with parents and other caregivers, this behavior must be reinforced (at least occasionally) in later relationships, or it will eventually be replaced by other social influence strategies. Because many children are rewarded for exhibiting dependent behavior in some relationships but not others, an intermittent reinforcement pattern is common— a pattern that renders dependent behavior highly resistant to extinction (Bhogle, 1978; Turkat, 1990). As children learn which behaviors are effective in eliciting the desired responses, and in which relationships these behaviors are (and are not) successful, they gradually adjust their help- and reassurance-seeking efforts to maximize rewards in different contexts. Studies confirm that intermittent reinforcement of dependent behavior plays a key role in the interpersonal dynamics of dependency—not only in children, but in adults as well. Such intermittent reinforcement patterns have been identified in a broad array of settings, including classrooms, hospitals, rehabilitation centers, and nursing homes (Baltes, 1996; Kilbourne & Kilbourne, 1983; Sroufe, Fox, & Pancake, 1983; Turkat & Carlson, 1984). As Bandura and Walters (1963) and others (e.g., Walters & Parke, 1964) pointed out, however, dependent behavior need not be reinforced directly in 2
Although Heathers's (1955) subtype model has fallen out of favor, other subtype models have proven useful in understanding the dynamics of dependency and the contrasting behaviors and relationship patterns of different dependent patients. In contrast to Heathers's behavioral framework, most contemporary dependency subtype models are derived from trait theory, using factor-analytic, clusteranalytic, and circumplex techniques (e.g., Pincus & Gurtman, 1995; Pincus & Wilson, 2001). CONCEPTUALIZING DEPENDENCY
order to be maintained. Observational learning is also important, and two processes are key in this regard: • Vicarious reinforcement. To the extent that a child notices other children (e.g., siblings, peers) being rewarded for exhibiting dependent behaviors, she is more likely to exhibit those behaviors herself. The influence of vicarious reinforcement will be particularly strong if the child (a) observes dependent behaviors being rewarded consistently and (b) admires or identifies with the person exhibiting dependent behavior (see Bandura, 1978). • Modeling. As Bandura (1977) noted, modeling—including symbolic modeling (i.e., imitation of persons pictured on film and in other media)—can play a powerful role in shaping and maintaining the developing child's help- and reassurance-seeking tendencies, even in the absence of direct or vicarious reinforcement. Symbolic modeling may be particularly important in the early acquisition of dependent traits, because many fictional characters in children's literature—especially female characters—exhibit high levels of stereotypic "helpless" dependency.3 CULTURE AND GENDER ROLE Social reinforcement of dependent behavior does not occur randomly, but is influenced by aspects of an individual's culture and his or her place within that culture. Individualistic societies typically emphasize autonomy and independence at the expense of social connectedness, and these societies tend to be particularly intolerant of dependency, especially in adults. Moreover, some cultures have rigid expectations regarding appropriate gender role-related behavior, whereas other cultures are more flexible in this domain (Cross, Bacon, & Morris, 2000; Cross & Madson, 1997). Studies indicate that gender role norms can have a powerful impact on women's and men's willingness to acknowledge underlying dependency needs (Bornstein, 1995c; Cadbury, 1991). The Impact of Cultural Context Dependent behavior is tolerated more readily in communitarian (or sociocentric) cultures than in individualistic ones (Doi, 1973; Johnson, 1993; Quantitative analyses of dependency-related behavior in children's stories were conducted by Fischer and Torney (1976) and White (1986). In both investigations female characters were portrayed as (a) more dependent and helpless than their male counterparts, (b) more likely to receive help, and (c) more receptive to receiving help. Fischer and Torney further found that exposure to these storybook characterizations had significant, measurable effects on 5-year-old children's dependencyrelated behavior in the classroom. Such results strongly support the role of symbolic modeling in shaping children's dependent behavior—at least in the short term.
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
Neki, 1976). Thus, adolescents and young adults in Japan and India have traditionally shown higher levels of self-reported dependency than those raised in North America, a pattern that continues today (Bornstein & Languirand, 2003). To a great extent, these sociocentric—individualistic cultural differences reflect prevailing norms regarding the relative importance of family ties and maintenance of group harmony. Even within a given culture, members of subgroups that value connectedness most strongly tend to score highest on self-report measures of dependency (Yamaguchi, 2004). As cultures change, attitudes regarding dependency evolve in predictable ways. Thus, as a sociocentric society becomes increasingly Westernized, population-wide dependency levels in that society tend to decrease. A second, subtler dynamic occurs as well: As individualistic values and norms are introduced into a communitarian society, people experience dependency con' flicts—discontinuities between longstanding sociocentric norms and newer individualistic values. Only recently have Japanese policymakers recognized the deleterious effects that such dependency conflicts have on their citizens' psychological adjustment. Feeling compelled to connect and compete at the same time (and with many of the same individuals), Japanese businesspeople report increased stress, frustration, and alienation (Sato, 2001). A similar process has occurred as the Indian economy has modernized and become Westernized during the past several decades (Bhogle, 1983; Singh & Ojha, 1987).4 Gender and Gender Role Gender—like culture—affects an individual's attitude regarding dependency, and societal norms are important in this context as well. Most investigations of gender differences in dependency have been carried out in Western societies (e.g., the United States, Canada, Great Britain), and in virtually every one of these investigations, women obtained significantly higher scores than men did on self-report tests of dependency (Bornstein, 1995c). Moreover, in both women and men, high self-report dependency scores were associated with high femininity and low masculinity scores on measures of gender role. Table 1.1 summarizes the results of extant studies in this area, and as this table shows, highly consistent findings have been obtained in these investigations: In almost every analysis to date, there has been a positive correlation between dependency and femininity, and a negative correlation between dependency and masculinity. 4
Although recent cultural shifts have exacerbated dependency conflicts among Japanese citizens, the tensions between individualism and collectivism actually have a long history in many communitarian societies. Moreover, even as individualistic values become firmly entrenched, underlying motivations remain stable: Strengthening social ties and maintaining group cohesion continue to shape certain behavior patterns (e.g., competitive business practices) that appear on the surface to reflect individualistic concerns.
CONCEPTUALIZING DEPENDENCY
TABLE 1.1 Dependency and Gender Role DependencyDependencymasculinity correlation femininity correlation Dependency measure
Study Anderson (1986) Bornstein, Bowers, &Bonner(1996b) Chevron, Quinlan, & Blatt(1978) Golding & Singer (1983) Klonsky, Jane, Turkheimer, & Oltmanns (2002) Sanfilipo(1994) Watson, Biderman, &Boyd(1989) Welkowitz, Lish, & Bond (1985) Zuroff, Moskowitz, Wielgus, Powers, &Franko(1983)
Women
Men
Women
Men .52*
PZ
-.34*
-.02
IDI
-.64"
-.55**
.30*
.37*
DEQ
-.48"
-.36*
.07
.42*
DEQ
-.38*
-.36*
.52*
.06
PIPD
-.08 -.36*
.07
-.28*
.14" .44**
.40*
-.29**
-.29**
.09
.09
DEQ
-.11
-.22
.20
.43*
DEQ
-.48**
-.27*
.11
.27*
DEQ
PGDS
-.01
.07
Note. PZ = Dependency subscale of the Pensacola Z Scale (Jones, 1957); IDI = Interpersonal Dependency Inventory (Hirschfeld et al., 1977); DEQ = Depressive Experiences Questionnaire (Blatt, D'Afflitti, & Quinlan, 1976); PIPD = Peer Inventory for Personality Disorders (Klonsky et al., 2002); PGDS = Peer Group Dependency Scale (Lapan & Patton, 1986). All studies except Chevron et al. (1978), Klonsky et al. (2002), and Sanfilipo (1994) used the Bern Sex Role Inventory (Bern, 1974) to assess gender role.
*p < .05. "p < .005.
A very different pattern emerges when projective measures of dependency are used. Table 1.2 summarizes the results of a large-scale metaanalysis of gender differences in dependency based on 97 published studies (Bornstein, 1995c). As the top portion of Table 1.2 shows, women scored higher than men on every questionnaire measure of dependency (and all but one of these gender difference effect sizes was statistically significant). As the bottom portion of Table 1.2 shows, however, men scored higher than women on every projective dependency test. Although these projective-test gender differences were not statistically significant for most individual measures, when the results were pooled using meta-analytic techniques, men showed a small—but statistically reliable—elevation in projective dependency test scores (d = .11, Combined Z = 1.93, p < .05).5
5 A d of. 11 suggests that, on average, men obtain projective dependency scores that are . 11 standard deviations higher than those of women. While a d of. 11 is considered a modest effect size, it is substantially larger than many well-established, widely accepted effect sizes in psychology and medicine (see Meyer, Pilkonis, Proietti, Heape, & Egan, 2001, for a summary of these effect sizes).
10
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
TABLE 1.2 Gender Differences in Dependency: Self-Report Versus Projective Measures Measure
Number of effect sizes
DEQ IDI Dy Scale MCMI DP Scale EPPS SAS LKDOS PDQ-R Other
18 16 12 8 7 5 4 3 2 20
ROD TAT HIT Other
17 3 2 4
Combined effect size (d)
Combined Z
P
8.13 6.20 10.91 5.28 5.03 3.98 4.42 3.38 0.90 7.17
<.00001 <.00001 <.00001 <.00001 <.00001 <.0001 <.00001 <.0005 ns <.00001
-2.08 -0.74 -0.54 -0.12
<.02 ns ns ns
Self-report measures .37 .42 .50 .46 .39 .48 .60 .61 .16 .33 Project measures -.17 -.09 -.17 -.07
Note. Originally published as Table 1 in "Sex Differences in Objective and Projective Dependency Tests: A Meta-Analytic Review," by R. F. Bornstein, 1995, Assessment, 2, pp. 319-331. Copyright 1995 by Psychological Assessment Resources, Inc. Reprinted with permission. A positive d reflects higher dependency scores in women than in men; a negative (Vindicates higher dependency in men. DEQ = Depressive Experiences Questionnaire (Blatt et al., 1976); IDI = Interpersonal Dependency Inventory (Hirschfeld et al., 1977); Dy Scale = Minnesota Multiphasic Personality Inventory (MMPI) Dependency Scale (Navran, 1954); MCMI = Millon Clinical Multiaxial Inventory Dependency Scale (Millon, 1987); DP Scale = Dependence-Proneness Scale (Sinha, 1968); EPPS = Edwards Personal Preference Survey Succorance Scale (Edwards, 1954); SAS = Sociotropy-Autonomy Scale (Beck, Epstein, Harrison, & Emery, 1983); LKODS = Lazare-Klerman Oral Dependency Scale (Lazare et al., 1966); TAT = Thematic Apperception Test Dependency Scale (Kagan & Mussen, 1956); HIT = Holtzman Inkblot Test Dependence Scale (Fisher, 1970).
It may be that men and women have comparable underlying dependency needs, but women are more willing than men to acknowledge these needs when asked (see Bornstein, 1995c, 1996b, for detailed discussions of this issue). Consistent with this hypothesis, projective dependency test scores (unlike the objective dependency test scores summarized in Table 1.1) are unrelated to gender role (Bornstein, Bowers, & Bonner, 1996b). The metaanalytic data in Table 1.2 not only provide important information regarding gender differences in dependency but also illustrate the impact of assessment method in studies of dependent personality traits. As discussed in chapter 2, different types of dependency tests (objective, projective, etc.) often produce very different results.
COGNITIVE MODELS Cognitive models of dependency focus on the ways in which a person's manner of thinking and processing information helps foster and maintain CONCEPTUALIZING DEPENDENCY
11
dependent behavior. Within this framework, Beck's (1976) concept of cognitive style is central to understanding the process through which early experiences (both family-specific and culture-wide) lead to increased dependency later in life. As Beck and Freeman (1990) and others (e.g., Overholser & Fine, 1994) noted, over time each of us develops his or her own characteristic way of thinking about the self, other people, and self-other interactions. Moreover, we construct and elaborate enduring schemas of self and others (akin to the psychodynamic concept of the internalized object representation), which shape our perceptions, thoughts, emotions, and interpersonal behaviors.
Automatic Thoughts and Negative Self-Statements Although some inconsistencies and conflicting results have been obtained in this area, developmental research suggests that two parenting styles— overprotectiveness and authoritarianism—lead to high levels of dependency in offspring (see chap. 10 for a review of these investigations). These two parenting styles involve very different interpersonal dynamics, but both send a similar message to the child, and both lead to the same end: The child comes to see herself as ineffectual and weak and perceives other people as comparatively powerful and potent (Bornstein, 1993, 1996a). Over time, such perceptions of self and others lead to a pattern of dependency-fostering automatic thoughts—reflexive self-statements that reflect the individual's perceived lack of competence. Automatic thoughts take many different forms but typically involve statements like "I can't handle this on my own" and "I'll fall apart completely unless someone helps." Studies show that automatic thoughts are difficult to control or inhibit and often occur with minimal awareness (Bargh & Chartrand, 1999). As a result, these thought patterns can be extremely resistant to change (Freeman & Leaf, 1989). Dependency-related automatic thoughts go hand-in-hand with negative self-statements—deprecating internal monologues wherein the dependent person reaffirms and elaborates various reasons (real and imagined) for his or her perceived lack of competence and skill (Overholser, 1987). Thus, within the cognitive framework, dependency is thought to be propagated through a two-stage process. First, some perceived challenge (e.g., a work-related project) triggers a dependency-related automatic thought. As the dependent person shrinks from the challenge, feelings of powerlessness and vulnerability increase, and one or more negative self-statements follow. Performance suffers and a vicious cycle ensues, as each new challenge triggers a set of cognitive responses that exacerbate the dependent person's sense of helplessness; as the dependent person's sense of helplessness increases, each new challenge seems even more insurmountable.
I2
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
Attributional Bias The cycle of automatic thoughts and negative self-statements can create a persistent attributional bias that propagates dependent behavior. In fact, many dependent individuals develop the classic self-defeating attributional style characteristic of learned helplessness: They attribute negative outcomes to internal (dispositional) factors and positive outcomes to external (situational) causes beyond their control. As a result, the dependent person may feel guilt and/or shame at each perceived failure yet be unable to take pleasure in—or feel empowered by—success. Studies have linked this type of self-defeating attributional style to increased risk for depression (Abramson, Seligman, & Teasdale, 1978; Seligman, 1991). Not surprisingly, research suggests that dependency is associated with an increased risk for depression as well (Neitzel & Harris, 1990).
TRAIT MODELS Contemporary trait models of interpersonal dependency reflect the combined influence of two very different research streams. One research stream originated with Leary's (1957) two-dimensional (love—hate and dominancesubmission) matrix for classifying personality styles. Within Leary's framework, dependency was thought to occupy the love—submission quadrant (i.e., dependent people were seen as seeking closeness with others through a pervasive pattern of submissive behavior). Leary's framework had a strong influence on Benjamin's (1974, 1996) Structural Analysis of Social Behavior (SASB) model and on various circumplex models of dependency (see Gurtman, 1992; Pincus, 2002). Within this latter category, Pincus and Gurtman's (1995) tripartite dependency framework has been most influential. Using factor- and clusteranalytic techniques to identify common elements in widely used questionnaire measures of dependency, Pincus and Gurtman (1995) identified three distinct dependency subtypes that occupy unique positions on the interpersonal circumplex: submissive dependency, exploitable dependency, and love dependency. Recent investigations using Pincus and Gurtman's framework confirm that individuals with different dependency subtypes show distinct patterns of intrapersonal functioning and interpersonal behavior (Pincus & Wilson, 2001). As discussed in chapters 7 and 9, this framework has important implications for assessment and treatment. Ironically, a second noteworthy influence on contemporary trait models came from psychodynamic theory. Investigators such as Goldman-Eisler (1950), and Lazare, Klerman, and Armor (1966) used factor-analytic methods to identify psychodynamically relevant dependency subtypes. Although
CONCEPTUALIZING DEPENDENCY
13
these subtypes are less influential now than they once were (Masling & Schwartz, 1979), scrutiny of the most widely studied trait models of dependency reveals the continuing influence of the psychodynamic and interpersonal perspectives (see Pincus, 1994, 2002, for detailed discussions of this issue). Deconstructing Dependency: The Five-Factor Model Costa and McCrae's (1985, 1992) five-factor model (FFM) is the most widely studied contemporary trait framework for conceptualizing personality traits and personality disorders. The FFM classifies personality traits along five dimensions: neuroticism, extraversion, openness, agreeableness, and conscientiousness. The FFM framework not only specifies the basic elements (or "facets") of various personality styles, but also delineates relationships among different trait dimensions and provides a context for examining change in personality structure over time (Costa & Widiger, 1994). Bornstein and Cecero (2000) conducted a meta-analysis of published FFM studies of dependency (N of studies =18) and found considerable consistency across investigations. The FFM dimensions most strongly related to questionnaire- and interview-derived dependency scores were neuroticism (r = .38), and openness (r = -.20). Other FFM dimensions showed more modest correlations with self-report dependency scores. Highly similar dependencyFFM interrelationships were obtained for clinical and nonclinical participants, supporting the generalizability of these relationships. Bornstein and Cecero's results suggest that dependency is characterized by high levels of anxiety and insecurity (neuroticism) and low levels of risk-taking and sensation-seeking (openness). Recent investigations confirmed these patterns (Mihura, Meyer, Bel-Bahar, & Gunderson, 2003). Millon's Biopsychosocial Model Millon's (1990, 1996) biopsychosocial model conceptualizes personality traits as a combination of biological predispositions, underlying psychological processes, and social-cultural influences. In Millon's framework every personality style is classified on three dimensions: fundamental life goals (enhancement vs. preservation), person-environment relationship (accommodation vs. modification), and preferred interpersonal strategy (individuation vs. nurturance). Within this perspective dependent individuals are conceptualized as being high on accommodation, high on nurturance, and midway between enhancement and preservation on the life goals dimension. Millon's (1990, 1996) biopsychosocial model of dependency represents an important integrative framework linking the trait perspective with other theoretical views (e.g., psychodynamic, behavioral, cognitive). Moreover, Millon's model continues to play an important role in bridging the gap be]4
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
tween basic theoretical work on dependency and psychotherapeutic techniques for use in the clinical setting. As discussed in chapter 6, Millon's biopsychosocial conceptualization of dependency holds up very well for some patients (e.g., those with dependent personality disorder), but less well for others (e.g., those with a mixture of dependent and borderline features), illustrating again the need to consider the impact of assessment method when evaluating dependency frameworks and findings. Only by looking beyond surface reports can the practitioner disentangle the complex elements of dependency, make meaningful distinctions among different dependent patients, and devise interventions that resonate with each patient's dependency style.
AN INTERACTIONIST MODEL Combining key elements of extant theoretical models with the results of recent studies examining situational variability in dependent behavior, Bornstein (1992, 1993, 1996a) delineated an interactionist model of interpersonal dependency. This model is summarized in Figure 1.1. As Figure 1.1 shows, a key tenet of the interactionist model is that dependency consists of four primary components: • Cognitive. A central element in a dependent personality orientation is a perception of oneself as powerless and ineffectual, along with the belief that others are comparatively powerful and potent. The interactionist model contends that dependency-related motivations, behaviors, and emotional responses all result, directly or indirectly, from the dependent person's "helpless" self-concept. • Motivational. As Figure 1.1 shows, dependent persons experience a strong desire for guidance, approval, and support from others. However, consistent with psychodynamic principles, these dependency-related urges can sometimes affect behavior with little or no awareness on the part of the dependent individual (i.e., they may be largely—even entirely—unconscious). • Affective. Dependent persons become anxious when required to function autonomously, especially when their efforts will be evaluated by a figure of authority. This anxiety can take many forms, including school refusal in children, agoraphobia in adolescents and younger adults, and pseudodementia (i.e., "false dementia") in older adults. Despite their surface differences, these dependency-related syndromes help minimize the distance between the dependent person and a protector or caregiver. • Behavioral. Dependent individuals use an array of relationshipfacilitating self-presentation strategies to strengthen ties to othCONCEPTUALIZING DEPENDENCY
15
Overprotective, Authoritarian Parenting Gender Role Socialization Cultural Attitudes Regarding Achievement/Relatedness
Cognitive Consequences: Schema of the self as powerless and ineffectual
Motivational Effects: Desire to obtain and maintain nurturant, supportive relationships
Behavior Patterns: Relationshipfacilitating self-presentation strategies (e.g., ingratiation, supplication)
Affective Responses: Performance anxiety, fear of abandonment, fear of negative evaluation Figure 1.1. An interactionist model of interpersonal dependency. As this figure shows, dependent personality traits reflect the interplay of cognitive, motivational, behavioral, and emotional factors, all of which stem from early learning and socialization experiences within and outside the family.
ers. These are summarized in Table 1.3, which lists five common dependency-related social influence styles, the selfpresentation strategy associated with each, and prototypical actions linked with each style. As Table 1.3 shows, the dependent person's relationship-facilitating self-presentation strategies include some behaviors that are passive and others that are quite active—even aggressive. 16
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
TABLE 1.3 A Taxonomy of Dependency-Related Self-Presentation Styles Characteristic support strategy
Self-presentation style Ingratiation Supplication Exemplification Self-promotion Intimidation
Prototypical actions
Ego-bolstering Performing favors Self-deprecation Submissiveness Providing help Emphasizing sacrifices Emphasize personal value/ Performance claims Exaggeration of worth accomplishments Threats/anger displays Frighten and control caregiver Breakdown displays
Create caregiver indebtedness Appear helpless and vulnerable Exploit caregiver guilt
The Etiology of Dependency As Figure 1.1 shows, three factors contribute to the initial construction of the dependent person's helpless self-schema: (a) overprotective, authoritarian parenting; (b) gender role socialization; and (c) cultural attitudes regarding achievement and relatedness. As noted earlier, overprotective, authoritarian parenting fosters the construction of a schema of the self as weak and ineffectual because these two parenting styles—alone or in combination—prevent the child from acquiring the sense of mastery and self-efficacy that follows successful learning experiences (Baker, Capron, & Azorlosa, 1996; Sroufe, Fox, & Pancake, 1983). In young girls, traditional gender role socialization experiences exacerbate this process by incorporating into the selfschema societal expectations regarding acceptable female behavior (Baumrind, 1980; Bornstein, Bowers, 6k Bonner, 1996b). Cultures that emphasize relational interdependence over individual achievement further contribute to the construction of a dependency-fostering self-schema by encouraging a view of the self as inextricably connected with others rather than as separate and autonomous (Cross et al., 2000; Gabriel & Gardner, 1999). The Dependent Self-Schema The interactionist model contends that variations in level of activation of the dependent person's helpless self-schema will influence motivations, behaviors, and emotional responses in predictable ways. Thus, events that "prime" the helpless self-schema (e.g., failure, rejection) intensify the dependent person's motivation to seek guidance, support, and protection from other people. Persons who are highly motivated to seek guidance and support will use a variety of relationship-facilitating self-presentation strategies to strengthen ties to potential caregivers and maximize the probability that they will obtain the protection and support they desire. Finally, activation of the CONCEPTUALIZING DEPENDENCY
17
helpless self-schema has important affective consequences (e.g., fear of abandonment, fear of negative evaluation by figures of authority). Although cognitive structures formed early in life mediate the motivations, behaviors, and emotional responses of the dependent person, emotional responses play a particularly important role in the dynamics of interpersonal dependency. As Figure 1.1 shows, dependency-related emotional responses (e.g., fear of negative evaluation) activate the dependent person's helpless self-schema, increasing dependency-related motivation (e.g., desire for support), and making it more likely that relationship-facilitating behavior will be exhibited. Thus, when a dependency-related emotional response occurs, a feedback loop is formed wherein this emotional response primes the dependent person's helpless self-concept. Finally, as Table 1.3 suggests, the practitioner who works with dependent patients must be prepared to encounter a broad range of behaviors that—while superficially distinct—have the common goal of strengthening ties to the therapist. Some of these behaviors (e.g., intimidation) may reflect the presence of personality traits and disorders that co-occur frequently with dependency (e.g., histrionic, borderline), but studies indicate that these behavior patterns are also found in dependent individuals without significant personality pathology (Bornstein, 1995a). As discussed in chapters 7 and 9, the contrasting selfpresentation strategies used by different dependent persons not only complicate diagnosis and assessment but also lead to very different therapist-patient dynamics and suggest that no single psychotherapeutic intervention is likely to be effective for the entire spectrum of dependent patients.
INTEGRATING THE CONTRIBUTIONS OF EXTANT THEORETICAL FRAMEWORKS Table 1.4 summarizes the key contributions of each theoretical framework to psychologists' understanding of the etiology and dynamics of dependency. As Table 1.4 shows, every framework has contributed one or more unique elements to our contemporary conceptualization of dependent personality traits. Although each framework is unique, there are some noteworthy parallels among different perspectives. In some instances similar concepts have emerged from more than one theoretical model even though these concepts are described in different terms by different theorists (e.g., the psychoanalytic concept of the dependent self-representation and the cognitive concept of the dependent self'Schema). As I review research on dependency ,in chapters 3 through 5, I will return to these frameworks to reconcile unanticipated and conflicting results. Areas wherein different theories converge are particularly helpful in this regard, although divergent aspects of different models are also useful in integrating contradictory findings. 18
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
TABLE 1.4 Key Contributions of Extant Theories of Dependency Theory Psychoanalytic theories
Key contributions Role of early relationships in the etiology of
dependency Importance of the dependent self-representation Unconscious-implicit dependency needs Behavioral models Social reinforcement of dependency Impact of modeling-observational learning Culture-based frameworks Socialization effects Dependency conflicts Underlying and expressed dependency needs Cognitive theory Importance of dependency-related schemas Automatic thoughts and negative self-statements Trait models Core components/features of dependency Subtypes of dependency Interactionist model Interrelationship of dependency's core components Links between early experiences and dependencyrelated dynamics Relationship-facilitating self-presentation strategies
Theory provides the context for data-collection and clinical work, but without valid measures of dependent traits, attitudes, and experiences, researchers could not assess the heuristic value of these theoretical frameworks, and practitioners could not use them effectively. No single measure of dependency provides a complete picture of the dependent person's intra- and interpersonal functioning, but by combining and contrasting the results obtained with different dependency scales, the clinician can gain a deeper understanding of each dependent patient and of the broader construct of dependency.
CONCEPTUALIZING DEPENDENCY
19
2 QUANTIFYING DEPENDENCY
Dependency is of interest to a wide range of practitioners and researchers, not only in psychology, but in related fields as well (e.g., sociology, political science, economics, medicine). As a result, numerous measures of dependent personality traits have been developed during the past several decades, and surveys of the social science and medical literatures indicate that there are at least 30 such measures available today (Bornstein, 1993, 1999). By understanding the strengths and limitations of different dependency scales, the clinician can make informed decisions about which tests are most useful in a given context or setting and develop strategies for combining the results of two or more dependency tests to obtain a more complete understanding of a patient's underlying and expressed dependency needs. In this chapter I review techniques for quantifying dependency. I begin by outlining the broad dimensions used by clinical researchers to classify dependency tests, then discuss key features of widely used self-report, projective, and interview measures. Finally, I present a conceptual framework for combining and contrasting the results obtained with different dependency scales. CLASSIFYING MEASURES OF DEPENDENCY There are many ways to classify and categorize measures of interpersonal dependency (see Birtchnell, 1991; Bornstein, 1999; Pincus & Gurtman, 1995). Among the most useful distinctions are 21
• Theoretical basis. Although some dependency tests are intended to be atheoretical, many are derived from—and clearly identified with—a specific theoretical framework. As a result, some scales assess dimensions of dependency that are based, at least in part, on the assumptions of a particular model (e.g., psychodynamic, cognitive). • Age of respondent. Most dependency scales are intended for use with adolescents and adults, but some are designed specifically for children. Children's dependency scales can be divided into two categories: (a) those completed by the child; and (b) those completed by a knowledgeable informant (usually a parent, but occasionally a teacher or other adult familiar with the child).1 • Clinical versus nonclinical status. While some dependency tests assess dependent traits in community and clinical participants, other scales are specifically tailored for clinical populations. In most cases clinical scales assess dependent personality disorder (DPD) symptoms rather than tapping the broader construct of trait dependency. • Test format. Dependency tests—like most measures of personality and psychopathology—can be divided into discrete categories with respect to format. Three categories are relevant in the present context: self-report scales, projective tests, and interviews. Each format has certain advantages, and certain disadvantages as well. STAND-ALONE SELF-REPORT MEASURES Because they are relatively easy to administer, self-report tests (sometimes called objective tests) are more frequently used than other types of dependency scales. The straightforward scoring of most self-report tests contributes to their widespread use in clinical and research settings, although studies suggest that the scoring of these tests is actually less objective—and more strongly affected by error and bias—than many psychologists believe (Allard, Butler, Faust, 6k Shea, 1995). The primary limitation of self-report dependency tests is their susceptibility to self-presentation effects. It is not surprising that respondents find it easier to deliberately fake their scores on self-report than on projective dependency measures (Bornstein, Rossner, Hill, & Stepanian, 1994). More'A review of the construct validity of children's dependency scales is beyond the scope of this chapter. Two of the most widely used children's dependency tests are the Children's Dependency Inventory (GDI) and the Highlands Dependency Questionnaire (HDQ). The GDI is a brief self-report instrument for use with elementary school children (Golightly, Nelson, & Johnson, 1970), while the HDQ is designed to be completed by one or both parents of a preschool/nursery school-age child based on observations of the child's behavior in the home (Berg, 1974).
22
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
over, these "fakability" effects occur regardless of whether respondents are attempting to appear overly dependent or overly autonomous. Most self-report dependency tests are "stand-alone" scales: They assess dependency and a limited number of theoretically related traits (e.g., autonomy, detachment). Other self-report dependency tests are embedded in longer assessment instruments that tap a broad range of traits. I begin with six stand-alone dependency tests that hold particular promise for use in clinical and research settings:2 Interpersonal Dependency Inventory The Interpersonal Dependency Inventory (IDI) has been one of the most widely used dependency tests during the past 25 years. The IDI consists of 48 self-statements, each of which is rated on a 4-point scale anchored with the terms Disagree (1) and Agree (4). Hirschfeld, Klerman, Gough, Barrett, Korchin, and Chodoff 's original (1977) factor analysis revealed that the 48 IDI items form three subscales: (a) Emotional Reliance on Others (ER; 18 items); (b) Lack of Social Self-Confidence (LS; 16 items); and (c) Assertion of Autonomy (AA; 14 items). Typical items from the three IDI subscales include "The idea of losing a friend is terrifying to me" (ER), "When I have a decision to make I always ask for advice" (LS), and "What people think of me doesn't affect how I feel" (AA). IDI whole-scale scores are calculated by summing the patient's scores on the ER and LS scales, then subtracting from this total their score on the AA scale (Bornstein, 1994b, 1998c). A review of evidence supporting the construct validity of the IDI as a measure of dependency is provided by Bornstein (1994b). IDI scores are positively correlated with scores on other self-report dependency measures in a variety of participant groups (Hirschfeld, Klerman, Clayton, & Keller, 1983), and predict severity of DPD symptoms in psychiatric inpatients (Loas et al., 2002). IDI scores also show adequate discriminant validity (Hirschfeld et al., 1977) and good retest reliability over 16-, 28-, 60-, and 84-week intervals (Bornstein, 1997b; Bornstein, Rossner, & Hill, 1994). Finally, IDI scores predict important dimensions of overt dependency-related behavior in laboratory, clinic, and classroom settings (Bornstein, 1995b; Bornstein & Kennedy, 1994; Hirschfeld et al., 1983).3 Depressive Experiences Questionnaire Blatt, D'Afflitti, and Quinlan's (1976) Depressive Experiences Questionnaire (DEQ) was designed to assess personality styles that place people at 2
Other, less frequently used, stand-alone dependency tests include Sinha's (1968) Dependence Proneness Scale and Lazare, Klerman, and Armor's (1966) Oral Dependency Scale. The IDI was developed by the National Institutes of Health in the 1970s and as a result, it is in the public domain, available to any clinician or researcher.
3
QUANTIFYING DEPENDENCY
23
risk for depression. Blatt et al. (1976) developed the DEQ by (a) obtaining from the published literature a representative sample of statements reflecting depressive experiences; (b) translating these statements into Likert scale format; and (c) using factor-analytic techniques to explore patterns in participants' responses to these items. This yielded a 66-item measure with three subscales: Anaclitic (dependent), Introjective (self-critical), and Efficacy (internal control). Because the original scoring system for the DEQ was complex, Welkowitz, Lish, and Bond (1985) constructed a simplified scoring method that yields comparable results; this simplified method has been used in many investigations during the past two decades. In the early 1990s an adolescent version of the DEQ (the DEQ-A) was developed for use with younger participants (Blatt, Schaffer, Bers, & Quinlan, 1992). Recent analyses of DEQ and DEQ-A items suggest that the Anaclitic scale can be divided into two narrower subscales, one tapping immature/maladaptive dependency, the other assessing mature dependency/relatedness (Blatt, Zohar, Quinlan, Zuroff, & Mongrain, 1995; Rude & Burnham, 1995). Considerable evidence has accumulated supporting the construct validity of the DEQ Anaclitic subscale as a measure of dependency. The scale shows good internal consistency, retest reliability, and convergent and discriminant validity (Blatt & Zuroff, 1992). Extensive clinical and nonclinical norms are available, facilitating use of the scale in research and treatment settings. Although much of the research involving the DEQ and DEQ-A has focused on depression risk, studies confirm that DEQ anaclitic scores predict salient aspects of functioning in a broad array of domains (Blatt, Hart, Quinlan, Leadbeater, & Auerbach, 1993; Kuperminc, Blatt, & Leadbeater, 1997). Personal Style Inventory The first formal measure developed to quantify Beck's (1976) construct of sociotropy ("social dependency") was Beck, Epstein, Harrison, and Emery's (1983) Sociotropy-Autonomy Scale (SAS). Although construct validity data for the SAS were generally supportive (Beck, Brown, Steer, & Weissman, 1991), questions regarding the independence of SAS sociotropy and autonomy subscales led researchers to develop a new measure, the Personal Style Inventory (PSI; Robins & Luten, 1991). Since the early 1990s, the PSI—and its more recent incarnation, the PSI-II (Robins, Bagby, Rector, Lynch, & Kennedy, 1997)—have been widely used in clinical research and practice. The PSI-II includes 48 self-statements, each of which is rated on a 6point scale anchored by the terms Strongly Disagree (1) and Strongly Agree (6). The 24 sociotropy items on the PSI-II compose three subscales: (a) Concerns About What Others Think (7 items); (b) Dependency (1 items); and (c) Pleasing Others (10 items). Initial validation studies showed that the PSIII Sociotropy scale has good internal consistency, adequate retest reliability 24
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
over 13 weeks, and good convergent validity with respect to other self-report dependency measures (Robins et al, 1994). Subsequent analyses demonstrated that PS1-II sociotropy scores are positively correlated with TAT-derived Need for Affiliation (nAff) scores (Kwon, Campbell, & Williams, 2001). PSI-II scores have also been shown to distinguish patients whose depressions are characterized by interpersonal sensitivity from patients whose depressions are characterized by detachment and hostility (Robins et al., 1997).4 Dependent Personality Style Scale Overholser's (1992) Dependent Personality Style Scale (DPSS) consists of 20 multiple-choice items that assess dependency-related behaviors in different situations and settings. Among the characteristics assessed by the DPSS are indecisiveness, self-doubt, passivity, submissiveness, and need for reassurance. Some DPSS items assess generalized dependent tendencies; others begin with situational anchors (e.g., "When 1 have to make a big decision . . ."; "If mistreated by someone I know well. . .") to tap context-specific dependent behaviors. Regardless of whether it is general or specific, each DPSS item is followed by four responses reflecting different degrees of dependency. Total DPSS scores are obtained by summing the 20 individual item scores. Construct validity data for the DPSS are quite strong, with the scale showing excellent internal consistency and good short-term temporal stability in clinical and nonclinical samples (Overholser, 1992,1996). DPSS scores also (a) show good convergent validity with respect to self-report and interview measures of dependency and DPD; and (b) correlate in meaningful ways with scores on measures of dependency-related traits such as loneliness, separation anxiety, interpersonal focus, and support-seeking behavior (Overholser, 1996). 3-Vector Dependency Inventory Pincus and Gurtman's (1995) 3-Vector Dependency Inventory (3VDI) occupies a unique place within the spectrum of self-report dependency tests because it is composed of items from an array of existing measures, including the IDI, SAS, and DEQ. Each 3VDI item is rated on a 6-point scale for the degree to which the statement in that item describes the respondent. The 3VDI yields three 9-item subscale scores: submissive dependence, exploitable dependence, and love dependence. While it is possible to obtain a high score on ^Because the PSI and DEQ were both designed to assess personality traits that represent diatheses for depression, it is not surprising that scores on the two scales are moderately intercorrelated (Pincus & Gurtman, 1995; Rude &. Burnham, 1995). Parallels between Blatt's (1974) anaclitic-introjective distinction and Beck's (1976) sociotropic-autonomous distinction are discussed by Bornstein (1993) and Rude and Burnham (1995). QUANTIFYING DEPENDENCY
25
all three dimensions, a primary purpose of the 3 VDI is to classify participants into different dependency groups. Recent findings confirm that individuals with different 3VD1 subtypes differ with respect to learning history, interpersonal style, relationship patterns, and other salient variables (Pincus & Wilson, 2001). Evidence supports the construct validity of the 3VDI as a measure of dependency style. Initial validation studies demonstrated that the 3VDI subscales show acceptable levels of internal consistency and predictable intercorrelations with Costa and McCrae's (1992) Five-Factor Model (FFM) domain scores (Pincus & Gurtman, 1995). Subsequent validation studies confirmed that all three 3VDI subscales show the expected patterns of gender differences and meaningful relationships with scores on self-report measures of attachment style, loneliness, and parental representations/introjects (Pincus & Wilson, 2001).
Relationship Profile Test Like the 3VDI, the Relationship Profile Test (RPT) occupies a unique niche within the spectrum of self-report dependency scales. However, in contrast to the 3VDI, which assesses contrasting dependency styles, the RPT taps three traits that occupy different points on the dependency-detachment spectrum: (a) destructive overdependence (DO); (b) healthy dependency (HD); and (c) dysfunctional detachment (DD). The RPT is the only dependency scale designed to assess both adaptive and maladaptive dependency-related traits (see Bornstein & Languirand, 2003). The RPT is a 30-item questionnaire (10 items per subscale) wherein participants respond to a series of self-statements, each of which is rated on a 5-point scale anchored by the terms Not at all true of me (1) and Very true of me (5). Representative items from the RPT subscales include "Being responsible for things makes me nervous" (DO), "Other people want too much from me" (DD), and "It is easy for me to trust people" (HD). RPT statements were derived from the theoretical and empirical literatures on dependency, detachment, and healthy dependency, and were written to tap the cognitive, emotional, motivational, and behavioral features of each personality style. Evidence supports the construct validity of the RPT as a measure of healthy and unhealthy dependency (Bornstein et al., 2002, 2003, 2004). The three RPT subscales show acceptable internal consistency, the expected patterns of interscale correlations and gender differences, and good retest reliability over 23- and 85-week intervals. In addition, the RPT DO, DD, and HD scales show good convergent and discriminant validity with respect to measures of attachment style, gender role, self-concept, self-esteem, life satisfaction, connectedness, affect regulation, and alexithymia. 26
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
SELF-REPORT SCALES EMBEDDED IN LONGER TESTS In addition to the six stand-alone dependency measures described in the previous sections, three widely used dependency scales are embedded in longer self-report instruments that assess a broad array of personality traits and psychopathologies.
Minnesota Multiphasic Personality Theory Dependency Scale The 57 items that compose Navran's (1954) Minnesota Multiphasic Personality Inventory (MMPI) Dependency (Dy) scale were selected by a panel of clinicians on the basis of their dependent content. Because these items are in the standard versions of the MMPI and MMPI-2, Dy scores can be derived from newly administered or existing MMPI protocols. Although Dy scores can also be obtained by creating a separate questionnaire containing only the relevant items, this strategy has been infrequently used by clinicians and researchers. Initial Dy scale validation studies demonstrated adequate internal reliability and good convergent and discriminant validity with respect to MMPI clinical scale scores. As expected, Dy scores were higher in psychiatric inpatients than nonclinical participants, and higher in schizophrenics than in character-disordered or neurotic patients. Subsequent studies showed that MMPI Dy scores predict overt help-seeking behavior in clinical settings (O'Neill & Bornstein, 2001), as well as self-reports of theoretically related traits and psychopathology dimensions (O'Neill & Bornstein, 1996).
Millon Clinical Multiaxial Inventory Originally developed in the 1970s, the Millon Clinical Multiaxial Inventory (MCMI; Millon, 1977) was designed to assess personality disorders (PDs) included in the third edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III, APA, 1980). The current version of the scale— the MCMI-III (Millon, Millon, & Davis, 1994)—assesses DSM-III-R (APA, 1987) and DSM-IV (APA, 1994) PD symptoms. The MCMI-III is a 175-item questionnaire that is structurally similar to the MMPI. Like the MMPI, the MCMI consists of a series of self-statements that are rated using a yes-no format. Like the MMPI, the MCMI includes an array of clinical scales, along with several validity scales (called modifying indices) that provide an interpretive context for clinical scale scores. Like MMPI scores, MCMI scores can be summarized in the form of a visual profile, facilitating pattern analysis. Finally, like the MMPI, some MCMI items are included in more than one scale, which complicates interscale comQUANTIFYING DEPENDENCY
27
parisons but makes administration more efficient by minimizing the total number of test items. The MCMI-III is one of the most widely used instruments in clinical research and practice (Groth-Marnat, 1999), and construct validity data for the scale are quite strong. The MCMI-III PD scales show good internal consistency, retest reliability, and convergent and discriminant validity (Choca, Shanley, & Van Denburg, 1992; Millon et al., 1994). Norms for the MCMIIII are based on large, heterogeneous clinical samples, facilitating use of the scale in clinical research and practice. Personality Diagnostic Questionnaire The original version of the Personality Diagnostic Questionnaire (PDQ) consisted of 163 true-false items designed to assess symptoms of DSM-III PDs (Hyler, Rieder, Williams, Spitzer, Hendler, & Lyons, 1988). Although the PDQ was intended to function as a screening instrument rather than a diagnostic tool, it is used for both purposes in research and clinical settings. The PDQ has been revised and updated several times since the late 1980s to bring scale items into line with current diagnostic criteria. The most recent version of the scale—the PDQ-IV—taps DSM-IV PD symptoms (Davison, Morven, & Taylor, 2001). Each version of the PDQ has demonstrated good convergent and discriminant validity with respect to interview and questionnaire measures of personality traits and PDs (Birtchnell & Shine, 2000; de Ruiter & Greeven, 2000). Extensive clinical norms facilitate use of the test in inpatient and outpatient settings. Along with these strengths, the PDQ-IV has one limitation that clinicians should consider: Like many self-report PD scales, the PDQ yields substantial numbers of false-positive diagnoses in nonclinical participants (Johnson & Bornstein, 1992).
PROTECTIVE TESTS Projective tests in general—and the Rorschach Inkblot Method (RIM) in particular—have been criticized vociferously in recent years (e.g., Hunsley & Bailey, 1999; Wood, Lilienfeld, Nezworski, & Garb, 2001; Wood, Nezworski, & Stejskal, 1996). To be sure, many projective methods are psychometrically flawed, with scanty norms and inadequate validity data. However, as several researchers have noted, certain projective dependency scales are actually quite strong with respect to underlying psychometrics and external (criterion-referenced) validity (Viglione, 1999; Weiner, 2000). From the practitioner's perspective, a primary disadvantage of projective dependency tests is their labor-intensiveness: Most require extensive training to administer, score, and interpret. Moreover, third-party payers rarely 28
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
reimburse practitioners for projective testing, and when they do, the reimbursement may be modest relative to the amount of effort devoted to the task.5 Despite these practical challenges, projective tests continue to be widely used—especially in clinical settings with a training component—and they have at least one important advantage over self-report measures: They are relatively immune from self-presentation effects, providing an unbiased index of an individual's underlying dependency needs (Bornstein, ZOOlb, 2002; Bornstein, Rossner, Hill, & Stepanian, 1994). Projective measures of dependency have been widely available since the late 1940s, and at least a dozen different projective dependency scales have been used at one time or another. Three projective dependency measures hold particular promise:6 Rorschach Oral Dependency Scale Masling, Rabie, and Blondheim's (1967) Rorschach Oral Dependency (ROD) scale has been the most widely used projective measure of dependency during the past 30 years, accounting for more than 80% of research involving projective dependency scores (Bornstein, 1996b, 1999). ROD scoring was derived from Schafer's (1954) writings on psychodynamic content in Rorschach responses. As a result, the ROD scale has strong psychoanalytic roots. Because standard ROD scoring involves only the free association portion of a patient's responses, ROD scores can be derived from individually administered Rorschach protocols, or from responses collected in a group setting (Masling, 1986). To control for variations in response productivity in individually administered protocols, ROD scores are expressed as percentages (i.e., the number of oral dependent responses divided by R). Regardless of whether data are collected individually or in groups, the respondent receives one point for each response that falls into any ROD category. These are summarized in Table 2.1. ROD scores show excellent interrater reliability, and adequate retest reliability over 16-, 28-, and 60-week intervals (Bornstein, Rossner, & Hill, 1994). The convergent validity of the test is well established: ROD scores predict help-seeking, conformity, compliance, suggestibility, and interpersonal yielding in laboratory and clinical settings (Bornstein, 1996b). The discriminant validity of ROD scores is supported by findings showing that 5
Because of the training and experience needed to acquire expertise in projective testing—and the labor-intensiveness of RIM scoring and interpretation—many practitioners use computer-based RIM interpretation systems. While these may be useful when scoring RIM protocols using Exner's (1993) Comprehensive System, there are no computer-based protocols for deriving ROD scores. 6 Other useful projective dependency tests include Blum's (1949) Blacky Test Oral Dependency Scale (BTODS) and Fisher's (1970) Holtzman Inkblot Test (HIT) Dependency Scale. The BTODS yields stronger results with children than adults, and the HIT's main advantage over the RIM is that the two HIT card sets facilitate retest reliability assessment in clinical and research settings. QUANTIFYING DEPENDENCY
29
TABLE 2.1 Categories of Scorable Responses on the Rorschach Oral Dependency (ROD) Scale Category
Sample responses
1. Food and drinks 2. Food sources 3. Food objects 4. Food providers 5. Passive food receivers 6. Begging and praying 7. Food organs 8. Oral instruments 9. Nurturers 10. Gifts and gift-givers 11. Good luck objects 12. Oral activity 13. Passivity and helplessness 14. Pregnancy and reproductive organs 15. "Baby-talk" 16. Negations of oral dependent percepts
Milk, whiskey, boiled lobster Restaurant, saloon, breast Kettle, silverware, drinking glass Waiter, cook, bartender Bird in nest, fat or thin man Dog begging, person saying prayers Mouth, stomach, lips, teeth Lipstick, cigarette, tuba Jesus, mother, father, doctor, God Christmas tree, cornucopia Wishbone, four-leaf clover Eating, talking, singing, kissing Confused person, lost person Placenta, womb, ovaries, embryo Patty-cake, bunny rabbit, pussy cat No mouth, woman with no breasts
Note. Originally published as Table 1 in "Construct Validity of the Rorschach Oral Dependency Scale: 1967-1995," by R. F. Bernstein, 1996, Psychological Assessment, 8, pp. 200-205. Copyright 1996 by the American Psychological Association. In Category 1, animals are scored only if they are invariably associated with eating (e.g., do not score duck or turkey unless food-descriptive phrases are used, such as roast duck or turkey leg). In Category 3, pot and cauldron are scored only if the act of cooking is implied. In Category 13, baby is scored only if there is some suggestion of passivity or frailness. In Category 14, pelvis, penis, vagina, and sex organs are not scored.
these scores are minimally related to social desirability, IQ, and locus of control (Masling, 1986; Bornstein & O'Neill, 1997). Clinical and college student norms for the ROD scale are provided by Bornstein, Hilsenroth, Padawer, and Fowler (2000) and Bornstein, Bonner, Kildow, and McCall (1997). Thematic Apperception Test Dependency Scale Originally developed by Kagan and Mussen (1956), the Thematic Apperception Test (TAT) dependency scale is based on Murray's (1938) description of need for succorance (n Succomnce), and its manifestation in TAT imagery. TAT dependency scale scoring is less standardized than that of the ROD scale, and some investigators have incorporated scoring content from Murray's other need dimensions (e.g., need for affiliation; see Masling et al., 1967; Zuckerman & Grosz, 1958). Like ROD scores, TAT dependency scores can be derived from individually administered protocols or from protocols collected in a group setting. Psychometric data for the TAT dependency scale are spotty, and the retest reliability of the scale has not been documented. However, TAT-based dependency assessment has seen a modest resurgence during the past decade, following publication of McClelland, Koestner, and Weinberger's (1989) review of TAT measures of implicit needs and motives. If current trends persist, it is likely that gaps in the psychometric literature on the TAT de-
30
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
pendency scale will be filled during the coming years, facilitating use of the scale in clinical settings. Early Memory Dependency Probe Fowler, Hilsenroth, and Handler's (1996) Early Memories Dependency Probe (EMDP) takes a unique approach to the projective assessment of dependency. Based in part on Mayman's (1968) procedures for assessing qualities of an individual's internalized object representations through an early memory interview, EMDP scoring involves assigning a single 7-point rating that captures the overall tone of a person's earliest "oral" (i.e., eatingrelated) memory, with a score of 1 indicating counter-dependence (along with conflict over dependency needs), 4 indicating mature dependence (with primarily positive associations), and 7 indicating anaclitic dependence (including painful yearning and/or fusion with the caregiver). EMDP scores can be derived from newly administered early memory interview protocols, or from archival clinical interview data (Fowler et al., 1996; Fowler, Hilsenroth, & Handler, 2000). Preliminary evidence supports the convergent and discriminant validity of the EMDP scale (Fowler et al., 1996, 2000). Although interrater reliability data are strong for various early memory rating dimensions (Karliner, Westrich, Shedler, & Mayman, 1996), the retest reliability of the EMDP has not been assessed, nor are clinical norms yet available.
INTERVIEWS All widely used interview measures of dependency assess DPD symptoms. Although interviewer observation and clinical inference contribute to these symptom ratings, every diagnostic interview requires that the respondent explicitly acknowledge an array of symptom-related behaviors to receive a diagnosis. Moreover, like questionnaires, diagnostic interviews require the respondent to generate retrospective reports of internal states (e.g., motivations, thought patterns) and past behaviors (Morrison &. Hunt, 1996). With these parallels in mind, psychometricians have argued that, in many respects, diagnostic interviews are akin to verbally administered questionnaires (see Stone et al., 2000, for detailed discussions of this issue).7 A primary advantage of interview dependency measures is that they allow the diagnostician to probe and follow up, obtaining information that
'Although EMDP scores are derived from semistructured interviews, these scores are based on a psychodynamic interpretation of the individual's early memories rather than on self-reports of symptoms and experiences. For this reason, the EMDP is considered a projective dependency test rather than an interview measure of dependency. QUANTIFYING DEPENDENCY
31
might have gone unreported on a questionnaire. In addition, the clinician may, through scrutiny of the patient's attitude, affect state, and nonverbal behavior (e.g., body posture, appearance), draw inferences that a questionnaire would not permit. Interview measures of dependency have some disadvantages as well, however. These include (a) their relative inefficiency and labor-intensiveness; (b) the potential for interviewer bias to contaminate results; (c) the need for extensive training and experience to attain acceptable levels of reliability; and (d) the paucity of nonclinical norms, limiting comparison of DPD base rates in clinical and community samples. Three structured interviews are frequently used to assess DPD in inpatient and outpatient settings: the Structured Clinical Interview for DSM Personality Disorders, the International Personality Disorder Examination, and the Structured Interview for Diagnosis of Personality. Structured Clinical Interview for DSM Personality Disorders The Structured Clinical Interview for DSM Personality Disorders (SCID-II) is a 120-item semistructured interview assessing a broad array of PD symptom criteria. It allows the diagnostician to assign PD diagnoses based on the number of symptom criteria acknowledged by the patient and also to derive a dimensional score for each PD category (Spitzer, Williams, Gibbon, & First, 1990). SCID-II scores show good interrater reliability and strong concordance with diagnoses derived from other interview and questionnaire measures (First, Spitzer, Gibbon, & Williams, 1995). SCID-II DPD scores also show acceptable levels of agreement with chart-derived DPD diagnoses (Johnson, Rabkin, Williams, Remien, & Gorman, 2000). Some studies suggest that the SCIDII may be particularly useful for assessing dependent and avoidant PDs in that it yields lower false-positive rates for these PDs than for others (Jacobsberg, Perry, & Frances, 1995). International Personality Disorder Examination The International Personality Disorder Examination (IPDE) was developed in the early 1990s to assess personality disorder symptoms and diagnoses as described in the ICD-10 and DSM-III-R (Loranger et al, 1994). It is a 157-item semistructured interview wherein the diagnostician assigns a single 3-point rating to each PD symptom (0 = absent, 1 = subdinical symptom severity, 2 - clinically significant symptom severity). Eight IPDE items are used to assess DPD, and like the SCID-II, the IPDE allows for derivation of a threshold (present—absent) diagnosis as well as a dimensional DPD score. Psychometric data support the utility and validity of the original (DSMIII—R) IPDE and the revised version, which assesses DSM—IV PD criteria 32
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
(Loranger, 1995). With respect to DPD, the IPDE shows good interrater reliability and temporal stability and strong concordance with scores on other interview measures. Structured Interview for Diagnosis of Personality The original version of the Structured Interview for Diagnosis of Personality (SIDP) was a 160-item semistructured interview designed to diagnose DSM—III PDs (Stangl, Pfohl, Zimmerman, Bowers, & Corenthal, 1985). The SIDP was subsequently revised to assess DSM—III—R PD symptoms, and while the SIDP-R retained its 160-item format, additional content domains (e.g., emotional expression, perception of threat) were added to increase the measure's ability to discriminate among similar PD categories (Pfohl, Blum, Zimmerman, & Stangl, 1989). Interviewers using the SIDP-R are explicitly instructed to supplement and corroborate interview data with informant and/ or archival information—an advantage over measures that rely solely on patient reports. The SIDP and SIDP-R show good interdiagnostician reliability, with kappa coefficients for most categories in the .60 to .80 range. The instrument also shows adequate convergent validity with respect to questionnaire and interview PD measures (Alnaes & Torgerson, 1988; Pfohl et al, 1989). Some studies suggest better validity data for dimensional than categorical PD ratings using the SIDP-R (e.g., de Ruiter & Greeven, 2000), although similar dimensional-categorical differences characterize other questionnaire and interview PD measures (see Zimmerman, 1994).8
COMBINING AND CONTRASTING TEST RESULTS Although a great deal of useful data can be obtained using individual dependency tests, additional insight regarding a patient's attitudes, motives, thoughts, and behaviors can be obtained by combining results from different measures. Comparison of questionnaire and projective test data is particularly informative in this regard, because self-report and projective measures assess different "levels" of dependency (see Bornstein, 1998a, 1998b, 2002). In chapter 7,1 discuss practical strategies for combining and contrasting the results obtained with different dependency tests. In the following sections, I review the conceptual and empirical underpinnings of these strategies. 8
A fourth assessment modality—observational—is most practical in settings wherein individuals can be observed over extended periods. Three observational methods have been used to assess dependency: (a) continuous monitoring, wherein a person is observed in a closed setting (e.g., at work), with salient aspects of behavior recorded; (b) spot sampling, wherein raters collect behavior samples at predetermined (usually random) intervals; and (c) experience sampling, wherein participants record their own behavior at predetermined intervals, usually in a diary or tape recorder. QUANTIFYING DEPENDENCY
33
Beyond the Multitrait-Multimethod Matrix For many years, researchers viewed questionnaires and projective tests as alternative ways of assessing the strength of a psychological trait or motive. Following the delineation of Campbell and Fiske's (1959) multitraitmultimethod matrix, a central goal of test development during the 1960s and 1970s was to maximize the intercorrelations of scores on objective and projective measures of any given variable. When objective and projective measures of the same construct were weakly intercorrelated, it was seen as reflecting a convergent validity problem in one or both tests. Although the "textbook" solution to this problem was to modify one or both tests until acceptable levels of test score intercorrelation were obtained, in practice modest intertest intercorrelations were typically taken as prima facie evidence against the convergent validity of the projective test, because the self-report test was presumed to be a more valid measure of the trait or need state being assessed (Watkins, Campbell, Nieberding, & Hallmark, 1995). In a landmark paper, McClelland et al. (1989) pointed out that the traditional view of objective-projective test relationships is inaccurate: Even when they are designed to measure parallel constructs, these two types of measures assess different facets of an individual's motivational state. Most self-report measures assess "explicit" (or self-attributed) needs—motives that a person acknowledges as being characteristic of his or her day-to-day functioning and experience. In contrast, projective tests assess implicit needs— motives that influence an individual's behavior automatically, often without any awareness on the individual's part that her behavior is affected by these motives. McClelland et al. (1989, pp. 698-699) argued that measures of implicit motives provide a more direct readout of motivational and emotional experiences than do self-reports that are filtered through analytic thought and various concepts of self and others, [because] implicit motives are more often built on early, prelinguistic affective experiences, whereas self-attributed motives are more often built on explicit teaching by parents and others as to what values or goals it is important for a child to pursue. A key corollary of McClelland et al.'s analysis is that implicit motive scores should be less susceptible than self-attributed motive scores to self-report and self-presentation biases, so that projective dependency tests provide a "purer" index of underlying dependency needs.
Levels of Analysis and Test Score Intercorrelations If self-report and projective dependency tests assess different aspects of a person's dependency strivings, scores on these tests should be modestly 34
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
TABLE 2.2 Intel-correlations of Self-Report and Projective Dependency Test Scores Test score intercorrelation (t)
Sample size Study Bornstein (1 998c) Bornstein (1998d) Bornstein, Poynton, & Masling(1985) Bornstein, Manning, Krukonis, Rossner, & Mastrosimone (1993) Bornstein, Rossner, & Hill (1994) Bornstein, Rossner, Hill, & Stepanian (1994) Bornstein, Bowers, & Robinson (1995) Bornstein, Bowers, & Bonner (1996a) Bornstein, Bowers, & Bonner (1996b) Bornstein & O'Neill (2000) Narduzzi & Jackson (2000) Narduzzi & Jackson (2002) Sinha(1968)
Women
Men
Women
Men
Overall
657 236
611 208
.26 .30
.21 .35
.24 .32
0
417
—
.11
.11
60
42
.67
.48
.59
54
34
—
—
.34
25
25
—
—
.46
58
25
—
—
.42
72
72
.43
.35
.39
47
40
—
—
.54
91
61
.04
.09
.06
114
0
.18
—
.18
286 0
0 40
.14 —
— .55
.14 .55
Note. All studies used the IDI and ROD scale, except for (a) Bornstein (1998c), which used the PDQ-R to assess self-attributed dependency needs; (b) Bornstein & O'Neill (2000), which used the MMPI Dy scale to assess self-attributed dependency needs; (c) Narduzzi & Jackson (2000, 2002), which used the PSI-II to assess self-attributed dependency needs; and (d) Sinha (1968), which used the Dependence Proneness Scale to assess self-attributed dependency needs and a sentence completion test to assess implicit dependency needs.
intercorrelated (Bornstein, 2002). As Table 2.2 shows, as of early 2000 there were 12 published studies assessing the correlations between scores on selfreport and projective dependency scales. Although there was some variation in the intertest correlations obtained in these investigations, the mean correlation (r) was .29. For those studies where results were calculated separately by gender, mean objective-projective test score correlations were .30 for women and .24 for men. These intertest correlation data would seem to support the hypothesis that self-report and projective tests assess different aspects of dependency, but some researchers have argued that these findings actually suggest that projective dependency tests are invalid (Wood et al., 2001). Recent data
QUANTIFYING DEPENDENCY
35
TABLE 2.3 Validity Coefficients for Self-Report and Projective Dependency Tests Test
Number of effect sizes
N
Combined effect size
Combined 2
P
Fail-safe N
Self-report tests
9
EPPS MCMI IDI DEQ MMPI DPS LKODS SAS SAS Other
7 5 5 2 2 2 6
ROD BTODS TAT HIT
21 6 4 1
9 9
485 720 424 141 320 384 67 40 40 432
.35 .17 .33 .04 .20 .34 .24 .46 .46 .10
ns
34 —
<.00001 ns <.003 <.00001 <.01 <.01 <.01 ns
108 — 17 34 2 2 2 —
<.00001 <.00001 <.001 ns
538 39 8 —
3.60 1.51 5.93 1.22 3.42 4.59 2.33 2.33 2.33 1.47
<.00001
8.49 4.51 2.84 0.31
Projective tests
1,320 323 125 40
.37 .50 .34 .12
Note. Originally published as Tables 1 and 2 in "Criterion Validity of Objective and Projective Dependency Tests: A Meta-Analytic Assessment of Behavioral Prediction," by R. F. Bornstein, 1999, Psychological Assessment, 11, pp. 48-57. Copyright 1999 by the American Psychological Association. For self-report tests, EPPS = Edwards Personal Preference Scale Succorance subscale (Edwards, 1954); MCMI = Millon Clinical Multiaxial Inventory Dependency Scale (Millon, 1987); IDI = Interpersonal Dependency Inventory (Hirschfeld et al., 1977); DEQ = Depressive Experiences Questionnaire Dependency Scale (Blatt et al., 1976); MMPI = Minnesota Multiaxial Personality Inventory Dy Scale (Navran, 1954); DPS = Dependence Proneness Scale (Sinha, 1968); LKODS = Lazare-Klerman Oral Dependency Scale (Lazare et al., 1966); SAS = Sociotropy-Autonomy Scale (Beck et al., 1983). For projective tests, ROD = Rorschach Oral Dependency Scale (Masling et al., 1967); BTODS = Blacky Test Oral Dependency Scale (Blum, 1949); TAT = Thematic Apperception Test Dependency Scale (Kagan & Mussen, 1956); HIT = Holtzman Inkblot Test Oral Dependency Scale (Fisher, 1970).
help resolve this issue and evaluate alternative interpretations of the data in Table 2.2. The results of a recent meta-analysis assessing the criterion validity of widely used self-report and projective dependency tests are summarized in Table 2.3 (Bornstein, 1999). As this table shows, as of late 1998 there were 54 published comparisons between self-report dependency test score and some index of observed dependent behavior (total N of participants = 3,013), and 32 comparisons between projective dependency test score and some index of dependent behavior (total N of participants = 1,808). When validity coefficients were calculated for each test category, the mean test score-behavior correlation (r) for self-report dependency tests was .26 (Combined Z = 9.04, p < .001), whereas the mean test score-behavior correlation for projective dependency tests was .37 (Combined Z = 9.63, p < .001). A focused comparison confirmed that the validity coefficient for projective tests did not differ from that obtained when self-report tests were used, Z = 1.28, ns. Thus, both self-report and projective dependency tests predict relevant features of dependency-related behavior. 36
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
SCORE ON OBJECTIVE DEPENDENCY TEST LOW
HIGH
Low Implicit
(_ oo W1
tg >^
LOW
u
S^W O B g
fflGH £ ^i g W
Low Implicit
Low Self-Attributed
High Self-Attributed
Low Dependency
Dependent Self-Presentation
High Implicit Low Self-Attributed
High Implicit High Self-Attributed
Unacknowledged Dependency
High Dependency
F/gure 2.7. Continuities and discontinuities between implicit and self-attributed need states: A four-cell model. Originally published as Figure 1 in "Implicit and SelfAttributed Dependency Strivings: Differential Relationships to Laboratory and Field Measures of Help Seeking," by R. F. Bornstein, 1998, Journal of Personality and Social Psychology, 75, pp. 778-787. Copyright 1998 by the American Psychological Association.
Deconstructing Objective-Projective Test Score Discontinuities The modest intercorrelations and strong criterion validity coefficients of objective and projective test scores provide an opportunity to examine naturally occurring discontinuities between implicit and self-attributed dependency needs. Figure 2.1 illustrates four outcomes that can be obtained when objective and projective dependency tests are administered to the same individual. As shown in the upper left and lower right quadrants of Figure 2.1, it is possible that a person will score high or low on both measures, which would indicate convergence between this person's self-attributed and implicit dependency scores. The other two cells in Figure 2.1 illustrate discontinuities between implicit and explicit dependency needs. In one case (i.e., high projective dependency score coupled with low objective dependency score), a person has high levels of implicit dependency needs but does not acknowledge them. These individuals may be described as having unacknowledged dependency strivings. In the other case (i.e., low projective dependency score coupled with high objective dependency score), the person has low levels of implicit dependency needs but presents him- or herself as being highly dependent. These individuals may be described as having a dependent self-presentation. Thus, by administering self-report and projective measures of dependency to the same individual, clinicians and researchers can obtain a more complete picture of that person's underlying and expressed dependency strivings. Exploration of these discontinuities may reveal important information regarding an individual's personality structure and interpersonal style. QUANTIFYING DEPENDENCY
37
3 DEPENDENCY ACROSS THE LIFE SPAN
The interactionist model in chapter 1 specified three factors that contribute to a dependent personality orientation during the first years of life: overprotective/authoritarian parenting, gender role socialization, and cultural attitudes regarding achievement and relatedness. Although early experiences lay the groundwork for the development of dependent personality traits, later events help shape these traits and influence the manner in which dependency needs are expressed. In other words, early experiences set in motion a "trajectory" of dependency, but later experiences exacerbate, inhibit, and modify this trajectory.1 From the clinician's perspective, understanding the evolution of dependency across the life span is important for two reasons: • The changing manifestations of dependency have implications for diagnosis and assessment. Because dependency-related thoughts and behaviors change over time, the symptom profiles and psychological test indicators of dependency also change as a per1
Parenting practices, socialization processes, and cultural attitudes regarding achievement-relatedness are not the only factors that foster a dependent personality orientation. Other important contributors include (a) parental permissiveness; (b) parental neglect or abuse; (c) early parental loss; and (d) premature independence pressure (see Birtchnell & Kennard, 1983; Egeland & Sroufe, 1981; Hill, Gold, & Bornstein, 2000).
39
;
son matures. Just as a clinician diagnosing depression must be cognizant of the contrasting symptom patterns that characterize this disorder during different phases of life, the clinician who seeks to assess dependency-related dynamics or diagnose dependency-related disorders must take into account the impact of patient age on the expression of underlying dependency strivings. • Etiology helps determine treatment focus. Any intervention designed to alter a dependency-related psychological process (e.g., perception of oneself as vulnerable and weak, fear of abandonment by a valued other) must be linked, conceptually and therapeutically, to a theoretical model that explains how the problematic process developed over time. Just as every pharmacological intervention seeks to correct a chemical imbalance that is presumed to underlie the disorder being treated, each psychological intervention aimed at altering some dependencyrelated process must be derived from a causal model that specifies the roots of this process. In this chapter I discuss dependency across the life span, organizing this discussion around five developmental epochs: infancy, childhood, adolescence, early and middle adulthood, and later adulthood. Within each epoch I examine factors that contribute to the consistency of dependency-related behavior over time, as well as factors that lead to age-specific variations in dependency. I begin by exploring the clinical implications of a simple—but oft-neglected—premise: Dependency-related dynamics create a context that pervades all human relationships, from infancy through old age.
DEPENDENCY AS A PERVASIVE LIFE THEME Until relatively recently, developmental researchers viewed the infant as a passive recipient of caregivers' nurturance and support. Psychologists now recognize that the typical infant is hardly helpless and rarely passive. Infants play an active role in eliciting and shaping a broad array of caregiving behaviors (Greenspan, 1989; Stern, 1985). Active though the infant may be, for most people the first months of life represent their period of maximum dependency on others. During these first months humans are utterly reliant upon caregivers for protection, nurturance, affirmation, and support. If neglected, young infants will be damaged, both psychologically and neurologically (Schore, 1994; Serbin, 1997). If abandoned, they will not survive. For many people—especially those who live into their eighties and beyond—late life is another "high dependency" period. The functional decline 40
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
that often accompanies advanced age impedes the individual's ability to carry out many activities of daily living (ADLs) they once took for granted (e.g., driving, shopping, cooking). As functional decline accelerates, older adults may be forced to rely on others for help with tasks they used to manage on their own (Bakes, 1996; Li, 2002). As will become clear, there are important differences between the expectable functional dependencies of later life and the more problematic emotional dependencies that often emerge as dependent people age (Bornstein & Languirand, 2001). Still, for many people late life is a time of increased reliance on others. Even during early and middle adulthood, when most people are as autonomous and self-sufficient as they will ever be, dependency issues pervade our relationships. Dependency-related dynamics not only affect interactions with friends, romantic partners, and colleagues at work, but also help shape intergenerational ties—those involving grandparents, parents, children, and grandchildren (Baumiester & Leary, 1995; Behrends, 2004; Fu, Hinkle, & Hannah, 1986).
The Dependency Spectrum In chapter 1,1 discussed the definition of dependency—what it means to be dependent. Ironically, clinicians and researchers have had greater difficulty agreeing upon what it means to lack dependency than what it means to have it. Several terms have been proposed to demarcate the absence of dependency, and each has a slightly different meaning: • Independence connotes a tendency to act on one's own while actively resisting the influence of others. Although the independent person presents a veneer of self-confidence and selfassurance, this person may be threatened by any attempt to sway her from her privately chosen path (Colgan, 1987). • Autonomy, like independence, connotes a tendency to act on one's own, but instead of actively resisting others' influence the autonomous person is capable of accepting outside input, evaluating it critically, and choosing whether or not to use it without feeling threatened or upset (Gurtman, 1992). More than the mere absence of dependency, autonomy (as discussed in chap. 5) is a key component of healthy, adaptive dependency. • Detachment connotes a deliberate distancing from other people, a marked absence of affiliative strivings, and avoidance of intimacy and closeness (Kantor, 1993; Millon, 1996). In its most extreme form, detachment carries with it a disconnected, almost schizoid quality—a strong aversion to all forms of social contact. DEPENDENCY ACROSS THE LIFE SPAN
41
HEALTHY DEPENDENCY
DETACHMENT
HOSTILE SUSPICIOUS CLOSED-OFF
OVERDEPENDENCE
SECURE TRUSTING FLEXIBLE
INSECURE ANXIOUS CLINGY
Figure 3.1. The dependency-detachment spectrum. As this figure shows, the converse of overdependence is detachment, which is characterized by social avoidance or withdrawal, and an absence of affiliative behaviors. At the midpoint of the dependency-detachment spectrum lies healthy dependency—autonomy and self-confidence coupled with situation-appropriate help- and support-seeking.
Figure 3.1 presents a conceptual framework for understanding the spectrum of dependency-related behaviors—the relationship between "too much" and "too little" dependency. At one end of the spectrum is detachment, characterized by social avoidance or withdrawal, and a near-total absence of affiliative ties. At the other end of the spectrum is overdependence, characterized by insecurity and excessive reliance on others for protection and support. At the midpoint of the spectrum lies healthy dependency—selfconfidence and self-directedness coupled with flexible, situation-appropriate help- and support-seeking (Bornstein & Languirand, 2003). Although the dependency-detachment spectrum summarized in Figure 3.1 is not the only way to conceptualize the link between high and low levels of expressed dependency, it is consistent with contemporary research on dependency and detachment (Bornstein, Geiselman, Eisenhart, &. Languirand, 2002) and with findings regarding the interpersonal dynamics of dependency- and detachment-related personality disorders (e.g., dependent, histrionic, antisocial, schizoid; see Bornstein, 1998b; Costa & Widiger, 1994; Millon, 1996). Moreover, the dependency-detachment spectrum provides a framework for interpreting Relationship Profile Test (RFT) responses, as well as patients' scores on the other assessment instruments described in chapter 2. Finally, the dependency-detachment spectrum provides a useful means of conceptualizing change processes in psychotherapy: As discussed in chapters 8, 9, and 10, successful treatment of the dependent patient involves replacing destructive overdependence with healthy dependency, being careful not to move the patient too far toward rigid independence or dysfunctional detachment. The Dependency-Detachment Dialectic Although the dependency-detachment spectrum captures some core elements of each person's relationship style, it would be a mistake to think of 42
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
INTERPERSONAL RELATEDNESS
1.
SELF-DEFINITION
Trust-Mistrust Autonomy-Shame
3. Initiative-Guilt 4.
Cooperation-Alienation 5. Industry-Inferiority
6. Identity-Role Diffusion 7.
Intimacy-Isolation 8. Generativity-Stagnation
9.
Integrity-Despair
Figure 3.2. The relatedness-self-definition dialectic. Blatt (1990) argued that psyche-social development can be conceptualized as an ongoing interaction between interpersonal relatedness and self-definition. The "self-as-separate" and "self-in-relation-to-others" develop in synchrony over the years. Originally published as Figure 1 in "Interpersonal Relatedness and Self-Definition: Two Personality Configurations and Their Implications for Psychopathology and Psychotherapy," by Sidney J. Blatt. In J. L. Singer (Ed.), Repression and Dissociation: Implications for Personality Theory, Psychopathology, and Health, 1990, (pp. 299-335). Copyright 1990 by University of Chicago Press. Reprinted with permission.
dependency-related motives and behaviors as static and unchanging. Different phases of life present different interpersonal challenges and opportunities, and as these arise, dependency-related dynamics evolve in response. Building upon Erikson's (1950) psychosocial stage model, Blatt (1990, 1991; Blatt & Shichman, 1983) and others (e.g., Franz & White, 1985) conceptualized personality development in terms of a lifelong dialectic involving two contrasting motives: interpersonal relatedness (i.e., dependency, affiliativeness, connectedness) and self-definition (i.e., independence, autonomy, personal achievement). The basic elements of this dialectic are summarized in Figure 3.2. As this figure shows, certain of Erikson's stages (e.g., trust-mistrust) are characterized primarily by concerns regarding relatedness, while others (e.g., autonomy-shame) are characterized by issues of selfdefinition. As this figure also shows, each relatedness-driven stage in Erikson's model is followed by two stages devoted to issues of self-definition; this pattern repeats itself several times as the person moves through life. The "selfDEPENDENCY ACROSS THE LIFE SPAN
43
as-separate" and "self-in-relation-to-others" develop in synchrony throughout the years.2 As in Erikson's original (1950) model, Blatt (1990) contends that the issues characterizing each psychosocial stage are never fully resolved, but reemerge periodically in different contexts, including the therapeutic relationship (see Blatt & Ford, 1994). Moreover, central to Blatt's (1990) dialectical framework—like Erikson's earlier model—is the notion that intra- and interpersonal events occurring early in life help shape the events of each later stage. Thus, Blatt's dialectic makes explicit the ongoing, lifelong tension between dependency urges and autonomy strivings—between people's need to merge and connect and their need to individuate and compete (see also Galatzer-Levy & Cohler, 1993, for a discussion of this issue). The ongoing tension between merging and separating—and the intrapsychic conflict that results—has important implications for the clinician because it points to a fundamental ambivalence that pervades many patients' explicit motives and goals. Even those patients who are highly dependent will periodically grapple with issues related to self-definition and personal achievement. Even those patients who appear highly autonomous—or detached—will occasionally struggle with concerns regarding dependency. Shifting Targets and Evolving Strategies Given the opportunities and challenges of different life stages and the ongoing, inescapable tension between dependency and autonomy strivings, it is not surprising that global measures of dependency show only modest consistency from childhood through early adulthood. There have been several well-designed studies of this issue (see Bornstein, 1993, for a review), but Kagan and Moss's (1960) longitudinal findings are particularly compelling. Kagan and Moss compared scores on observational measures of dependency at ages 6 to 10 with scores on interview, observational, and self-report dependency measures obtained from the same participants at ages 20 to 29. They found childhood-adult dependency score correlations (rs) of .45 for women, and .10 for men. In other words, childhood dependency scores account for a modest portion of the variance in adult dependency scores. Does this mean that child and adult dependency levels are essentially unrelated? Probably not. A more plausible explanation of these modest correlations is that Kagan and Moss's (1960) measures—while psychometrically sound—were too global. As Ainsworth (1989) pointed out, even when the intensity of our dependency strivings remains consistent over time, the ob2
The relative importance of self-definition in Blatt's (1990) dialectic is in part a function of the individualist values that pervade many Western cultures. As noted in chapter 1, research examining societal differences in individualism—collectivism suggests that personality traits such as dependency and autonomy may be strongly influenced by these cultural norms and values (Oyserman, Coon, & Kemmelmeier, 2002; Yamaguchi, 2004). 44
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
jects—or targets—of our dependency urges change dramatically as we age. Typically, the child's dependency on the parents shifts to the peer group during adolescence, then to romantic partners, friends, and supervisors during early and middle adulthood (see Shaver & Mikulincer, 2002). The shifting expression of underlying dependency urges is both expectable and adaptive (after all, continued dependence on one's parents during adulthood is not acceptable in most Western cultures). From a purely statistical standpoint, however, these target shifts attenuate observed consistencies in dependency levels over time.3 Just as the targets of our dependency strivings change, the methods we use to obtain nurturance, guidance, and support evolve as we mature. The help- and support-eliciting self-presentation strategies that are effective at age 5 will be inappropriate at 15 or 25. In fact, one of the key distinguishing features of healthy, adaptive dependency (vs. maladaptive, inflexible dependency) is the ability to determine which strategies are most likely to obtain the desired response in the short term, while simultaneously strengthening relationships with potential caregivers over the long term (Cross, Bacon, & Morris, 2000; Lee & Robins, 1995, 1998). Moreover, because different caregivers have different needs and expectations, these strategies must be tailored to each relationship—a process that requires considerable insight and social sensitivity on the part of the dependent person (Bornstein & Languirand, 2003).
THE EVOLUTION OF DEPENDENCY With these dynamics in mind we turn our attention to the changing expression of dependency needs across the life span. In the following sections I discuss the evolution of dependency from infancy to old age, focusing on factors that shape dependency-related beliefs, motives, and behaviors. Infantile Dependency Researchers have long speculated that dependent personality traits might be influenced by genetic factors (see Buss & Plomin, 1984; Millon, 1996), and recent findings confirm these speculations (Livesley, Jang, Jackson, & Vernon, 1993). Table 3.1 summarizes the results of published studies assessing the heritability of dependency. As this table shows, five such studies have 'Although continued dependency on the parents is seen as aberrant in most Western societies, it is tolerated—sometimes even expected—in many sociocentric cultures. For example, in Japan it has been traditional for adult men to reveal to their mother personal experiences and private emotional reactions that most American men would feel awkward discussing (see Doi, 1973; Johnson, 1993). In India it is not unusual for adult men to seek parental guidance regarding personal problems (e.g., marital conflict) that most American men would be reluctant to discuss openly with their parents (Neki, 1976). DEPENDENCY ACROSS THE LIFE SPAN
45
TABLE 3.1 The Heritability of Dependency
Study Coolidge, Thede, & Jang (2001) Dworkin, Burke, Maher, & Gottesman (1976) Gottesman (1966) O'Neill &Kendler (1998)
Participants (twin pairs)
Dependency measure
dz
hf
.82
.40
.84
.56 —
.07 —
.98 .23
.29
.18
.22
.55
.37
.36
'm,
r
70 MZ/42 DZ CPNI 25MZ/17DZ MMPI 79 MZ/68 DZ CPI/MMPI 457MZ/318 IDI DZ
Torgerson et al. (2000)
Heritability estimates
92MZ/129
SCID-II
DZ
Note. MZ = monozygotic twins, DZ = dizygotic twins. For dependency measures, CPNI = Coolidge Personality and Neuropsychological Inventory for Children (Coolidge, 1998); MMPI = MMPI Dy scale (Navran, 1954); CPI = California Personality Inventory (Gough, 1956); IDI = Interpersonal Dependency Inventory (Hirschfeld et al., 1977); SCID-II = Structured Clinical Interview for DSM-III-R Axis II disorders (Spitzer et al., 1990). Heritability estimates are as follows: rm = concordance of dependency scores for monozygotic twin pairs, ra = concordance of dependency scores for dizygotic twin pairs, H2 = proportion of variance in dependency attributable to genetic factors. Gottesman (1966) reported an overall heritability estimate, but excluded monozygotic and dizygotic twin concordances.
been conducted, involving a total of 1,297 twin pairs. There was considerable variation in heritability estimates across studies (H2s ranged from .22 to .98), but when these estimates were pooled using meta-analytic techniques, an overall heritability coefficient of .32 was obtained. In other words, about 32% of the variability in self-report dependency scores is attributable to genetic factors. It is one thing to demonstrate the heritability of dependency. Determining precisely what is inherited—what physiological factors underlie observed differences in dependent behavior—is more difficult. It seems likely that the earliest precursors of dependency are certain temperament variables (e.g., withdrawal, low adaptability, negative mood) that influence subsequent dependency levels via two parallel pathways: • Attachment style. Although some temperament variables (e.g., low adaptability) may evolve directly into dependency-related traits during the first years of life, these building blocks of dependency are first reflected in an insecure attachment style— an inability to tolerate separation from the caregiver and an absence of age-appropriate autonomous behaviors (Ainsworth, 1972; Main, Kaplan, & Cassidy, 1985). Studies show that insecure attachment persists over time: High levels of insecure attachment in early childhood are associated with an insecure attachment style in adult friendships and romantic relationships (Feeney & Noller, 1990). Other investigations point to more generalized attachment effects: High levels of insecure 46
CONCEPTUAL AND EMPIRICAL
FOUNDATIONS
attachment during the first years of life are associated with a broad range of expressed dependency needs in late childhood (Sroufe, Fox, & Pancake, 1983) and early adulthood (Sperling & Herman, 1991). • Parenting. Certain infantile temperament variables (e.g., low soothability) also affect the development of dependent traits indirectly by eliciting parenting behaviors that foster dependency (e.g., overprotectiveness). While the parent whose child is calm and easily soothed gains confidence in the child's ability to cope, the parent whose child is easily upset and difficult to soothe may come to view that child as fragile and weak. Overprotective parenting is likely to follow (Bernstein, 1992, 1993). Numerous investigations have documented the complex synergy between the child's attachment behaviors and mothers' and fathers' parenting practices (see Lemery, Goldsmith, Klinnert, & Mrazek, 1999), and several studies have shown that dependent behaviors exhibited by young children elicit strong protective responses in fathers and mothers (e.g., Hunt, Browning, & Nave, 1982;Osofsky &O'Connell, 1972).4 Childhood Dependency As children grow, they internalize mental images of self and other people (what psychoanalysts refer to as introjects). They also internalize schemas of self-other interactions (what developmental psychologists call internal work' ing models). These mental images and schemas become increasingly detailed and articulated over time, a process that reflects the child's (a) ongoing interactions with parents and other significant figures; and (b) developing cognitive skills, which enable her to conceptualize the self and interpersonal world in increasingly subtle and sophisticated ways (see Blatt, 1990; Tabin, 1985). By middle childhood, the dependent child has internalized a representation of the self as weak and ineffectual, along with a belief that other people are comparatively powerful and potent (Bornstein, 1996a). These mental images and the internal working models that link them ultimately evolve into a complex array of relationship scripts—anticipated sequences of inter4
Fu, Hinkle, and Hanna (1986) used regression procedures to estimate the impact of parenting practices on the development of dependency in a sample of 150 grandmother-mother-child "units." They found that approximately 20% of the variance in late-childhood dependency levels could be accounted for via measures of parenting style. When coupled with the results summarized in Table 3.1, it appears that (a) about 30% of the variability in dependency levels is traceable to genetics (most likely temperament differences); (b) 20% is attributable to parenting attitudes and practices; and (c) 50% remains unexplained. Presumably, variables such as gender role socialization, cultural norms, and family configuration account for much of this unexplained variance (see Bornstein, 1992, 1993, 1996a).
DEPENDENCY ACROSS THE LIFE SPAN
47
personal exchange between the self and other people (Abelson, 1981; Singer & Salovey, 1991). Relationship scripts are to some degree situation-specific, but an individual's scripts typically reflect a common set of underlying goals, affect qualities, and hoped-for and feared outcomes. The dependent child's core relationship script reflects the underlying goals of helpand support-seeking, an affect tone of anxiety and insecurity, a hoped-for outcome of obtained help and support, and a feared outcome of rejection or abandonment.5 Relationship scripts are shaped by all manner of interpersonal dynamics, none more important than those that take place within the family. As Haley (1976), Minuchin (1974), and others (e.g., Brock & Barnard, 1988) have noted, our relationship styles are strongly influenced by the family system in which we are embedded. We first learn which self-presentation strategies are and are not effective by trying them out within the family. Only later do we extend these efforts to extrafamilial relationships (e.g., friends, authority figures), modifying our early tendencies based on current experiences. Within the family system, two dynamics—roles and alliances—are particularly important, both in understanding the development of dependency and in formulating treatment strategies to alter problematic dependencyrelated behaviors: • Family roles. Over time each person creates a unique role within the family, then works to preserve the niche afforded by that role. Some family roles are easy to identify (e.g., "Strong One," "Good Son," "Selfless Caregiver"); others are more subtle, and may not be consciously recognized, even by those who have assumed them. Once in place, family roles mesh into a complex, interlocking system, so changing one role disrupts all the others. This is why (as discussed in chap. 10) concurrent marital and family treatment can be an important component of psychotherapeutic work with dependent patients. • Alliances. Family members form alliances that not only meet individual needs, but also preserve the status quo and maintain each person's role. Like family roles, some alliances are obvious (e.g., Mom and Dad allied against a misbehaving son). Other alliances are hidden—sometimes so deeply that those involved 5
The notion that relationship scripts are to some degree role-specific parallels Markus and Nurius's (1986) concept of multiple possible selves. As numerous studies have shown, different roles and situations "prime" particular aspects of the self-representation, increasing the likelihood that these aspects of the self will impel social behavior (e.g., Carver, Reynolds, & Scheier, 1994; Cross & Markus, 1991). Moreover, it is not only the present-day self (or role) that helps shape social goals and actions: Our "future selves"—the selves we expect to become if things go as planned, hope to become if things go well, and fear we may become if things go badly—are also important in this context (see Hooker & Kaus, 1994; Markus & Nurius, 1986).
48
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
in the alliance are completely unaware of it. For example, an overprotective mother and her overdependent son may be unconsciously invested in maintaining their dysfunctional pattern because both members benefit, albeit in different ways: The son gets the nurturance and protection he craves, and the mother can avoid troubling personal issues by devoting all her energies to propping up her "weak" child. Thus, by late childhood two dynamics—one intrapersonal, the other interpersonal—combine to propagate and maintain dependent behavior. Intrapersonally, the child's insecure attachment style and core relationship script lead her to direct energy and attention outward, and focus on maintaining relationships with powerful others rather than cultivating the internal resources necessary to cope on her own. Interpersonally, one or more family roles and alliances may reinforce the child's dependency to maintain stability within the system. When change looms, this enmeshed system represents an additional locus of resistance that can undermine progress and obviate gain. Dependency in Adolescence Even if the intensity of an individual's underlying dependency needs does not change during adolescence, the ways in which these needs are expressed surely will. The transition to young adulthood creates intrapsychic and interpersonal conflict as the person struggles with two competing urges. On the one hand, the adolescent (no longer a child) feels increasing internal pressure to separate from the family and create his own niche in the world. On the other hand, the adolescent (not yet an adult) feels inadequately equipped to manage the challenges of independence and wants to remain connected to the family, secure in his well-practiced childhood role. As any parent can attest, when the ambivalence created by this intrapsychic conflict is coupled with the physiological changes and social demands of burgeoning adulthood, unpredictable—even stormy—behavior ensues. Adolescents have varying degrees of insight regarding the roots of their ambivalence and the factors that underlie their increasingly conflicted family relationships. In part, these variations in insight result from differences in defenses and coping styles and from characteristics of the family system in which the adolescent is embedded (e.g., communication patterns, sibling relationships; see Clark & Ladd, 2000). Insight notwithstanding, adolescents who have achieved some measure of autonomy (i.e., a secure attachment style, a healthy core relationship script) may navigate the adolescent transition with relative calm and ease. The overdependent adolescent is likely to show a less adaptive response to the stresses and challenges of this period, and exhibit one of two patterns: DEPENDENCY ACROSS THE LIFE SPAN
49
• Increased peer-group dependence. Even relatively autonomous adolescents show some degree of peer-group dependence (i.e., reliance on age-mates for affirmation and identity) as a means of effecting the transition from familial dependence to the beginnings of mature adulthood (Steinberg & Silverberg, 1986; Zirkel, 1992). Highly dependent adolescents differ from those who are more autonomous in the intensity and rigidity of their attachment to the peer group. For some dependent adolescents, peer group dependence involves submersion in the collective identity of a clique (Marcia, 1993). For others, peer group dependence is manifested via a strong attachment to a best friend, or "chum" (Hartup, 1989, 1999). Either way, a similar defensive strategy is used: The dependent adolescent denies familial dependency urges and displaces these urges onto a peer, or a peer group (Friedlander & Siegel, 1990; Lapan & Patton, 1986).6 • Regression and retreat. Some overdependent adolescents cope not by shifting their dependency urges from parents to peers, but by regressing (Galatzer-Levy & Cohler, 1993; Sullivan, 1953). These adolescents respond to each new challenge by clinging even more tightly to their familiar family role. They retreat from most peer relationships (except those which support and validate their strong family ties). The regressed adolescent remains ambivalent regarding dependency and autonomy, but copes by avoiding or repressing autonomy strivings so dependency urges dominate conscious experience. These urges may be affectively positive (if the family milieu is experienced as safe and secure) or negative (if the family milieu feels intrusive and overwhelming). For some highly dependent adolescents, vestiges of this regression can linger throughout adulthood: Even if they achieve enough separation from the family to develop a stable romantic relationship, the urge to turn to parents and other family members for guidance and support remains strong (Brock & Barnard, 1988; Alperin, 2001). Dependency in Early and Middle Adulthood The increasing variability in the expression of underlying dependency needs that marks the transition from childhood to adolescence continues as the dependent adolescent becomes a dependent young adult. To some degree, this variability in adult dependency is situation- and context-driven (as 'The dependent adolescent who uses peer ties in this way may find that, as adolescence ends, another challenge awaits: Now he must effect a transition from pseudo-intimate peer attachment to mature friendship and genuine romantic intimacy. Many dependent adolescents have considerable difficulty making this transition. 50
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
discussed in chap. 4). To some degree, this variability reflects dispositional factors—differences in self-presentation, social skills, defense style, and other trait variables (these dispositional factors are the focus of chap. 5). Studies also show that dependency-related behaviors in adulthood are in part role-related, and expressed in contrasting ways in different relationships. Discussions of relationship-based differences in dependent self-presentation are provided by Ainsworth (1989), Birtchnell (1988), and Behrens (2004). For now, it is useful to consider four domains that are particularly salient in this context: • Friendship, Two key features of overdependent friendship are possessiveness and overidentification. Possessiveness arises from the dependent person's insecurity and reflects an underlying, unexpressed fear that the friend will abandon him (Bornstein, 1995a; Pincus & Wilson, 2001). The dependent person's typical response is to perceive other people as potential threats and attempt to exclude them from the relationship. When extreme dependency is coupled with significant personality pathology, overdependent friendship can lead to overidentification, as the boundaries between two friends begin to blur. Pathological overidentification can take a variety of forms, but it often begins with imitative behaviors (e.g., dressing like the friend, adopting his manner of speech), and may escalate to magical thinking and cognitive slippage (e.g., perceiving a "unique" or "special" connection with the friend). When coupled with significant personality pathology, overidentification may culminate in threats—even violent or self-destructive behavior—if the friend attempts to terminate the relationship (see Kemberg, 1984). • Romance. In romantic relationships, the dependent person's insecure attachment is expressed through what Bornstein and Languirand (2003) termed a suffocating lovestyle—a destructive pattern of relating wherein the partner is persistently "smothered" with affection. At first, this lovestyle can be seductive and appealing (in fact, the partners of overdependent people often report that the dependent person's initial attentiveness was a major reason the relationship moved forward). Over time, however, the dependent person's sexual and emotional insecurity begin to dominate, overshadowing the earlier positive attentiveness (Bartholomew, 1997; Holmes, 1997). The dependent person's reflexive response to real or imagined distance is to cling ever more tightly, until the overwhelmed partner feels trapped and retreats. For some dependent people, this leads to a series of intense, short-lived romantic encounters. For other DEPENDENCY ACROSS THE LIFE SPAN
51
dependent people (especially those whose partner derives gratification from their neediness) it results in a long-term relationship punctuated by periods of extreme closeness and stormy withdrawal. Parenting. Studies show that many overdependent parents have difficulty setting limits on their children's behavior (Bornstein, 1993). Anxious regarding their parenting skills and fearful of alienating two important sources of support, the overdependent parent may waver between permissiveness (to strengthen ties to the child) and authoritarianism (to control the wayward child and/or please a demanding spouse). Underlying this surface variability is a fundamental need to keep the child dependent (Thompson 6k Zuroff, 1998). Some studies also show a link between extreme parental dependence and risk for perpetration of child abuse (e.g., Beasley &. Stoltenberg, 1992; Hart, Dutton, & Newlove, 1993). Several sources contribute to this risk, including the overdependent parent's (a) difficulty modulating anger; (b) inability to control the child through other means; and (c) desire to strengthen ties to the spouse by presenting herself as a flawless, "perfect" parent. Work. Dependency is associated with career indecision, especially during the transition from school to work. In the workplace, it is common for dependent people to have difficulty taking responsibility for tasks and projects and avoid leadership positions within the organization (Bornstein 6k Languirand, 2003). The dependent person's insecurity can also create conflicts with peers: Concerned with strengthening ties to potential protectors and caregivers, the dependent person tends to focus on pleasing those who seem most powerful and influential, clumsily shifting her attention from peers to superiors as opportunities emerge (Bornstein, Riggs, Hill, & Calabrese, 1996). Because the dependent person perceives each co-worker as a potential threat, she may respond by undermining others' efforts and achievements. Needless to say, this alienates colleague and supervisor alike, as both see through the dependent person's transparent efforts to succeed at others' expense.
Late-Life Dependency As noted at the outset of this chapter, it is important to distinguish the expectable functional dependencies of later adulthood from the increases in emotional dependency that occur as dependent people age. At first glance, cultural norms and expectations regarding aging would seem particularly rel52
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
evant to the development of late-life emotional dependency, but studies show that these norms and expectations affect both functional and emotional dependency in older adults (Li, 2002; Goodstein, 1985). In both domains cultural variables interact with physical and cognitive changes to produce predictable increases in dependent behavior. Not only do parallel processes occur in the functional and emotional domains, but the targets of the individual's dependency strivings overlap as well. In both areas dependency-related behaviors tend to be directed toward romantic partners, adult children, and professional caregivers (for those older adults who have experienced significant functional decline). Despite these parallels, the surface manifestations of functional and emotional dependency also differ in some important ways: • Functional dependency. Functional dependency increases as the older adult loses the ability to carry out basic (e.g., bathing, dressing) and complex (e.g., shopping, cooking) activities of daily living (ADLs). These losses usually reflect three factors, alone or in combination: (a) physical effects of aging (e.g., loss of muscle mass); (b) declines in perceptual acuity (e.g., vision and hearing loss); and (c) cognitive decline (e.g., dementia). Some older adults respond by developing compensatory dependencies—domain-specific increases in help- and support-seeking that enable the person to accommodate late-life losses (e.g., needing help with transportation to maintain financial independence; see Bornstein & Languirand, 2001). Ironically, studies show that in hospitals and long-term care facilities, older adults are often reinforced for exhibiting dependent behavior, a process that exacerbates functional decline and leads to increased helplessness, hopelessness, and depression (Bakes, 1996; Langer & Rodin, 1976). • Emotional dependency. Several biological and social processes combine to exaggerate pre-existing dependent tendencies during late adulthood. First, disinhibition resulting from neurological impairment of behavioral control mechanisms can cause moderate dependent behavior to escalate into severe dependent behavior as the person loses the ability to modulate underlying urges (Birren & Schaie, 1996). Second, late-life depression may be exhibited as pseudodementia (literally, "false dementia"), leading to increased reinforcement of dependent, helpless behavior by caregivers (and a parallel increase in dementia-like symptoms; see Goodstein, 1985). Finally, the social losses of late adulthood (e.g., death of a spouse) can bring to the surface long-buried feelings of helplessness and vulnerability, as the widowed person can no longer rely on a signifiDEPENDENCY ACROSS THE LIFE SPAN
53
TABLE 3.2 Changing Manifestations of Dependency Across the Life Span Developmental epoch Infancy
Internal dynamics
External expression
Temperament differences (e.g., low soothability)
Physical dependence; attachmentpromoting behaviors (e.g., cooing, smiling) Childhood Insecure attachment; Instrumental and dependencyemotional helpfostering seeking; dependent relationship scripts family role; dependency-based family alliances Adolescence Individuation Peer-group pressures; dependence; dependencyregressionautonomy conflicts avoidance of autonomy Early and Helpless self-concept; Role-related/situationmiddle beliefs about others; specific help-, adulthood dysfunctional support-, and nurturance-seeking schemas and introjects Later adulthood Losses of aging (ADL Increased functional decline); and emotional disinhibition; dependency internalized cultural norms
Targets of dependency strivings Parents, other caregivers
Parents, siblings, teachers, peers
Peers, older role models
Romantic partners, friends, supervisors Romantic partners, adult children, other caregivers
ADL = activities of daily living.
cant other and must learn to carry out unfamiliar tasks on her own (Bornstein & Languirand, 2001). LIFE-SPAN DEPENDENCY: TOWARD AN INTEGRATED VIEW The changing manifestations of dependency across the life span reflect a complex array of internal (psychological and biological) processes and external (familial and cultural) factors. One reason dependent psychotherapy patients present such a diverse array of styles is that each patient's pattern of internal dependency-related processes (e.g., characteristic defenses and coping mechanisms) combine with a specific set of external factors (e.g., the patient's particular interpersonal milieu) to influence personality development and behavior. The ways in which these variables interact differ from person to person, and the clinician can use the research reviewed in this chapter to develop an integrated understanding of each dependent patient's unique inter- and intra-personal dynamics.
54
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
Table 3.2 summarizes the major internal and external variables that shape dependent behavior from infancy through old age. However, a complete understanding of dependency-related motives, cognitions, and emotional responses requires that the life-span perspective summarized here be coupled with a situational analysis of dependency. In chapter 4, I begin this situational analysis by discussing context-specific deficits and strengths associated a dependent personality orientation. As this analysis proceeds, the focus of our discussion shifts from age-based variation to situational variation in dependent behavior.
DEPENDENCY ACROSS THE LIFE SPAN
55
4 CONTEXT-SPECIFIC DEFICITS AND STRENGTHS
Each of the theoretical frameworks described in chapter 1 has something important to say about how internal processes and external factors interact to influence dependency-related behavior. These models, taken together, can help the clinician gain a more complete understanding of a neglected issue in the assessment, diagnosis, and treatment of dependent patients: variability in dependency-related functioning across different situations and settings. Table 4.1 summarizes the contributions of the psychoanalytic, behavioral, culture-based, and cognitive models to a conceptualization of interpersonal dependency in terms of trait-by-situation interactions. As Table 4.1 shows, there are important similarities among these models, but noteworthy differences as well. Each model emphasizes a unique combination of internal and external variables that moderate the expression of underlying dependency needs.1 'Recent trait models of interpersonal dependency (Pincus & Gunman, 1995; Pincus & Wilson, 2001) and dependent personality disorder (Costa & Widiger, 1994; Millon, 1996) have gone further than earlier models in explaining context-driven variability in dependent behavior (compare, e.g., these recent trait views with the earlier perspectives of Cattell [1965], and Leary [1957]).
57
TABLE 4.1 Internal-External Interactions in Dependency: Four Perspectives Theoretical framework Psychoanalytic
Behavioral
Culture-based
Cognitive
Internal-external dynamic/interaction Conscious and unconscious dependency strivings are differentially influenced by situational variables; in many contexts the person's subjective perceptions of opportunity and risk override objective reality. Different relationships are associated with contrasting contingencies; these interact with acquired behavior patterns to determine context-specific help- and support-seeking strategies. Internalized cultural norms combine with external factors (e.g., societal demands, family values) to determine how underlying dependency strivings are expressed and how others' dependent behavior is interpreted. Schemas of the self, other people, and self-other interactions combine with ongoing relationship demands to influence the expression of dependencyrelated beliefs, motives, and behaviors.
Note. Detailed descriptions of these four theoretical frameworks are provided in chapter 1.
In this chapter, I explore situational variability in dependent behavior. I begin by reviewing key studies that illustrate the ways in which situational factors moderate the expression of underlying dependency needs. I then discuss the central assumptions of an integrated interactionist perspective on dependency and use this interactionist perspective to examine research on dependency-related deficits and strengths.
FROM PERVASIVE PASSIVITY TO SITUATIONAL VARIABILITY Kraeplin (1913) and Schneider (1923) were among the first theoreticians to discuss the dependency-passivity link, but the notion that high levels of dependency are associated with a passive, helpless stance in interpersonal interactions was popularized primarily by psychoanalytic theorists (e.g., Fenichel, 1945; Glover, 1925) who wrote extensively on this topic during the first decades of the 20th century. Abraham (1927, p. 400) summarized nicely the prevailing psychoanalytic view of dependency at that time when he argued that dependent persons "are dominated by the belief that there will always be some kind person—a representative of the mother, of course— to care for them and give them everything they need. This optimistic belief condemns them to inactivity . . . they make no kind of effort, and in some cases they even disdain to undertake a breadwinning occupation." Several neo-analytic theorists (e.g., Fromm, 1947; Horney, 1945; Sullivan, 1947) extended the classical psychoanalytic notion that dependency is invariably 58
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
associated with passivity and helplessness, and by mid-century this view was reified in psychology and psychiatry, having received almost universal acceptance among clinicians and researchers.2 Even today, mental health professionals and laypersons alike associate dependency with passivity, conformity, compliance, and helplessness (Birtchnell, 1988; Bornstein, 1995a; Cadbury, 1991; Tail, 1997). This view, though widespread, is only partially correct. Studies show that in certain situations the dependent person can be quite active—even assertive. Consider the results of a study by Bornstein, Riggs, Hill, and Calabrese (1996) that was among the first to demonstrate context-driven variability in dependency-related behavior. In Bornstein, Riggs, Hill, and Calabrese's (1996) investigation, samesex pairs of college students were brought to the laboratory and told they were taking part in a study of the personality-creativity link. Each pair consisted of one dependent and one nondependent student, classified using Hirschfeld et al.'s (1977) IDI. The two students were told that because they had obtained similar personality profiles in an earlier testing (which was actually the dependency prescreening), they were expected to obtain comparable creativity scores. The key manipulation came at this point: Half the participants were told that their creativity test data would be seen only by the undergraduate experimenter and the other student (the no authority condition). The remaining participants were told that their tests would be reviewed by two psychology professors who would contact them later in the semester to discuss their results (the authority condition). Participants were then given several opportunities to engage in behaviors they believed would enhance or undermine their test performance (e.g., choosing to do many or few practice items before taking the test, choosing to listen to relaxing or distracting music while being tested). The results of the experiment were clear: Dependent students "selfhandicapped" (i.e., did few practice items, chose distracting background music) in the no authority condition, because their primary goal in this situation was to be liked by the peer. However, dependent students "self-enhanced" (i.e., did many practice items, chose relaxing background music) in the authority z
Neo-analytic theorists' speculations regarding the dependency-passivity link have had far-reaching effects, even though the models from which these views were derived have become somewhat marginalized in contemporary psychology. For example, Sullivan's (1947, p. 84) notion that dependent persons "have been obedient children of a dominating parent. . . [and] go through life needing a strong person to make decisions for them" had a strong influence on the DSM-I description of "passive dependency" (the precursor of dependent personality disorder). Following Sullivan (1947), the DSM-I described the passive-dependent individual as "characterized by helplessness, indecisiveness, and a tendencyto cling to others as a dependent child to a supporting parent" (APA, 1952, p. 37). The notion that dependent personality disorder is characterized by passivity and helplessness continues to pervade the DSM-IV (APA, 1994) and DSM-IV-TR (APA, 2000), and as discussed in chapter 6, it has created significant difficulties in diagnosing dependency in clinical settings. CONTEXT-SPECIFIC DEFICITS AND STRENGTHS
59
7 -A- DEPENDENT •Q NONDEPENDENT A
Q
6 -
a
S 5o u 4-
3 -
n-
- •- Q
OH fc
0
Pi w
z2 -
CQ
S g
1I No Authority
I
I
Authority
AUTHORITY CONDITION Figure 4.1. Context-driven variability in dependency-related behavior. As this figure shows, dependent college students did many practice items when they believed that a figure of authority would be evaluating their creativity test performance, but few practice items when they believed that only a peer would have access to their results. Originally published as Figure 1 in "Activity, Passivity, Self-Denigration, and Self-Enhancement: Toward an Interactionist Model of Interpersonal Dependency," by R. F. Bornstein, J. M. Riggs, E. L. Hill, and C. Calabrese, 1996, Journal of Personality, 64, pp. 637-673. Copyright 1996 by Duke University Press. Reprinted with permission.
condition, because their primary goal had changed: Now, impressing the professors became more important than getting along with a peer. Nondependent students' behavior was unaffected by authority condition. Figure 4.1 summarizes the pattern of results obtained in this investigation. These findings illustrate the predictable variability in dependencyrelated behavior and suggest that this variability is largely a function of the 60
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
dependent person's perceptions of interpersonal risks and opportunities. With no authority figure present, being liked by a peer was paramount, but once a figure of authority entered into the equation, impressing this person became more important than getting along with a peer. Thus, these dependent students exhibited a very rational, adaptive social influence strategy: They chose to curry favor with the person who seemed best able to offer guidance, protection, and support over the long term.
AN INTERACTIONIST PERSPECTIVE ON DEPENDENCY The findings of Bornstein, Riggs, Hill, and Calabrese (1996) and others (e.g., Mongrain, Vettese, Shuster, & Kendal, 1998) confirm that dependencyrelated behavior is proactive, goal-driven, and guided by beliefs and expectations regarding the self, other people, and self—other interactions (see Mischel, 1973, 1979, 1984). Although this might seem to contradict the traditional trait view of dependency, the interactionist perspective actually complements and extends the trait approach. As Dweck and Leggett (1988, p. 270) noted, the explanatory power of dispositional variables "lies in their ability to predict what behaviors will be exhibited in various situations, not in their prediction that the same behavior will be exhibited across situations." Mischel (1984, p. 362) adopted a similar stance in explaining traitbased variability in responding, noting that: A theory of personality structure does not require a person to be characterizable by pervasive cross-situational consistency. . . . Instead of seeking high levels of consistency from situation to situation for many behaviors in a wide range of contexts, [we should] try to identify unique "bundles" or sets of temporally stable prototypic behaviors, key features that characterize the person even over long periods of time, but not necessarily across most or all possibly relevant situations.3
Thus, rather than undermining the traditional trait view, an interactionist perspective on dependency shifts the locus of stability from behavior to cognition. Even when the dependent person's interpersonal strategies vary from situation to situation, her underlying beliefs (i.e., perception of the self as weak) and goals (i.e., protection-, help-, and support-seeking) remain unchanged. Three corollaries follow from this interactionist framework: • Context is critical. Scrutinizing the milieu in which a behavior is exhibited is key in understanding dependency-related deficits and strengths. Some environments (e.g., treatment settings) 3
Recent extensions of Mischel's (1973, 1979) interactionist model are provided by Baldwin and Sinclair (1996); Cervone and Shoda (1999); and Mischel, Shoda, and Mendoza-Denton (2002).
CONTEXT-SPECIFIC DEFICITS AND STRENGTHS
61
accommodate dependent behavior more easily than other environments (e.g., work settings) do. As a result, the same behavioral tendency (e.g., help-seeking) may be problematic in some contexts, but adaptive in others. • Perception is paramount. The objective risks and opportunities associated with a particular situation are less important than the dependent person's subjective perceptions of opportunity and risk (see Table 4.1). As psychodynamic (Horowitz, 1991; Bucci, 1997) and cognitive theorists (Kelly, 1955; Neimeyer, 1992) have noted, it is the individual's personal construction of situational contingencies that ultimately guides behavior. • People, seek contexts that further their goals. To some degree we all must adapt to the situations in which we find ourselves. However, as Bandura (1978, 1991) pointed out, whenever possible we seek contexts that accommodate our needs. Once immersed in a setting, we find ways to modify the milieu so that it meshes even more smoothly with our goals. When a setting proves unaccommodating and we find we cannot alter it, we may choose to move on (though as we will see, dependent persons have greater difficulty leaving unaccommodating situations than nondependent persons do). Not only does an interactionist perspective on dependency have some noteworthy theoretical implications, it also has important implications for diagnosis, assessment, and treatment. I discuss these in detail in chapters 6 throughlO. Briefly, they are as follows: • Diagnosis. Diagnostically, findings such as those reported by Bornstein, Riggs, Hill, and Calabrese (1996) and others (e.g., Pincus & Wilson, 2001) point to the importance of looking beyond the DSM-IV dependent personality disorder criteria, which emphasize dependency-related passivity and ignore situation-driven behavioral variability (Bornstein, 1997b). To obtain a more complete diagnostic picture, the clinician must explore a broad range of Axis I and Axis II diagnoses (including subsyndromal symptom patterns) that capture the passive and active components of dependency. • Assessment. These findings further suggest that a thorough understanding of dependency-related dynamics requires a multimodal assessment strategy. The clinician must not only explore the internal processes that underlie dependent behavior (i.e., motives, thoughts, emotional responses), but should also assess external factors (e.g., family dynamics, career demands, social milieu) that cause the dependent patient to respond differently in different contexts and settings. 62
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
• Treatment. Findings regarding situational variability in dependent behavior can enhance treatment effectiveness. As discussed in chapter 1, dependent patients exhibit a broad array of selfpresentation styles that vary from setting to setting. By understanding the context-specific deficits and strengths of each dependent patient, the clinician can structure treatment to minimize a patient's dependency-related problems and enhance dependency-related skills. In the following sections I use the interactionist framework to explore the context-specific deficits and strengths associated with high levels of interpersonal dependency.
DEPENDENCY-RELATED PROBLEMS AND DEFICITS Dependency-related deficits arise in many arenas, but most consistently (and most prominently) in three related domains. Insecurity, Jealousy, and Commitment Studies by Markus (1977) and Bargh (1982) were among the first to suggest that dependent persons are preoccupied with concerns regarding vulnerability and helplessness, and predisposed to interpret ambiguous social information in ways that confirm these concerns. In a later investigation using very different measures and procedures, Bornstein, Leone, and Galley (1988) obtained results consistent with those of Markus (1977) and Bargh (1982): When college students who had been prescreened for level of dependency using Masling, Rabie, and Blondheim's (1967) ROD scale were asked to provide open-ended self-descriptions, they consistently described themselves as timid, fearful, vulnerable, and weak. Given the dependent person's "helpless" self-concept, it is not surprising that dependency is associated with an insecure attachment style (Sperling & Berman, 1991; Zuroff & Fitzpatrick, 1998) and with a tendency to experience high levels of anxiety and jealousy in friendships and romantic relationships (Bringle & Buunk, 1985; Buunk, 1982). Mongrain's (1998) finding that dependent college students frequently sought help and reassurance from friends and romantic partners offers further support for the central role of insecurity in dependency-related social behavior (see also Mongrain, Lubbers, & Struthers, 2004). Even when they are involved in a stable, long-term relationship, dependent people's insecurity can cause them to ruminate about how they are perceived by their partner and spend a great deal of time pondering strategies that might strengthen the relationship and minimize the possibility of future rejection (Caspi, Bern, & Elder, 1989; Overholser, 1992). CONTEXT-SPECIFIC DEFICITS AND STRENGTHS
63
In a particularly ambitious and innovative investigation of this dynamic, Simpson and Gangestad (1991) examined the link between dependency and commitment in a sample of 241 unmarried monogamous couples. Not only did Simpson and Gangestad find positive correlations between self-reported dependency levels and level of romantic commitment in women (r = .49) and men (r = .59), they also found positive correlations between dependency levels and perceptions of the partner's level of commitment (rs were .29 in women and .28 in men). There are at least two plausible explanations for these latter results. The more optimistic interpretation is that—consistent with their underlying needs for support and reassurance—dependent persons cultivate ties with romantic partners who are strongly committed to the relationship. Less optimistically, it may be that dependent individuals overestimate their partner's degree of commitment, displaying a kind of self-serving bias that helps quell their anxiety regarding the possibility of relationship disruption (though at the cost of misperceiving the partner's feelings, intentions, and goals). Compliance, Exploitation, and Victimization A plethora of studies confirm that dependent persons are highly suggestible (Jakubczak & Walters, 1959; Ojha, 1972), compliant with external demands (Bornstein & Masling, 1985; Masling, O'Neill, & Jayne, 1981; Melley, Oltmanns, &. Turkheimer, 2002), and unusually susceptible to group pressure (Kagan & Mussen, 1956; Masling, Weiss, & Rothschild, 1968). These behaviors are a product of at least two factors: (a) lack of self-confidence (Alam, 1986; Ojha, 1978) and (b) fear of negative evaluation by others, especially figures of authority (Bornstein, Masling, & Poynton, 1987). Given the underlying motives and emotional patterns of the dependent person, it is not surprising that dependent individuals are particularly upset by relationship conflict (Allen, Home, & Trinder, 1996). Not only does actual or anticipated relationship disruption lead to increased anxiety and depression in dependent individuals, it also results in increased illness risk because chronic interpersonal stress diminishes immunocompetence and increases susceptibility to pathogens (Bornstein, 1993, 1995b). As Table 4.2 shows, a meta-analysis of research in this area demonstrated that the magnitude of the dependency-illness link is substantially larger than the magnitude of the personality-illness link obtained for other trait variables (Bornstein, 1998e), including those traits (like anger and hostility) that have traditionally been seen as key risk factors for disease (Friedman & BoothKewlev, 1987).4 4
The dependency-illness effect size (r) of .30 suggests that dependency levels by themselves account for about 9% of the variance in illness rates (other traits in Table 4.2 account for between 0% and 3% of illness risk). Follow-up analyses (Bornstein, 1998e) confirmed that the dependency-illness effect size is consistent across study design (retrospective vs. prospective), type of participant (child vs. adult), and outcome measure (patient report vs. objective illness index).
64
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
TABLE 4.2
Personality-Illness Effect Sizes: Dependency Versus Other Trait Variables Trait variable Dependency Trait anxiety Trait depression Anger-hostility Extraversion
Number of effect sizes
Effect size estimate (r)
P
24 39 39 32 34
.30 .18 .16 .11 -.03
<.00001 <.00001 <.00001 <.00001 ns
Note. Dependency effect size data are from Bornstein (1998e); other trait data are from Friedman and Booth-Kewley (1987). Number of effect sizes represents the number of published trait-illness comparisons for each variable.
Dependent people's concern with pleasing others and maintaining interpersonal ties not only places them at increased illness risk, but also renders them vulnerable to exploitation and victimization. Two sets of findings are germane in this context. First, studies show that dependent women have a significantly greater number of unprotected sexual contacts than do nondependent women with similar demographic profiles, in part because they have difficulty setting limits and resisting the partner's persuasive efforts (Lavan & Johnson, 2002). Second, studies indicate that dependent women are at increased risk for being victimized by their partners, both physically and sexually (Kalmuss & Straus, 1982; Rusbult & Martz, 1995; Watson et al., 1997). Apparently, a desire to maintain close ties to others at all costs can sometimes lead the dependent person to tolerate mistreatment, even when this mistreatment escalates into violence. Immaturity, Infantilization, and Abuse Given the dependent person's strong desire to maintain interpersonal ties—even in the face of mistreatment—one might expect that dependency would invariably be associated with accommodating, other-centered behavior. In fact, this is not always the case. When dependent persons are required to function autonomously they sometimes regress and exhibit an array of immature responses that alienate those around them. Thus, Sroufe, Fox, and Pancake (1983) found that, when challenged, dependent schoolchildren displayed a pattern of theatrical helplessness that angered both teachers and peers. Parallel findings were obtained by Fichman, Koestner, and Zuroff (1996, 1997), who examined the links among helplessness, homesickness, and peer rejection in dependent children at summer camp. Echoing Fichman et al.'s (1996, 1997) results, Holmbeck and Wandrie (1993) found that high levels of dependency were associated with significant adjustment difficulties in firstyear college students. CONTEXT-SPECIFIC DEFICITS AND STRENGTHS
65
TABLE 4.3 Studies of Dependency and Sociometric Status
Study
Sample
Dunnington (1957) Marshall & McCandless (1957)
15 nursery school children 36 preschool children
McCandless, Bilous, & Bennett (1961) Miller & Stine (1951) Moore & Updegraff (1964)
26 preschool children
Wiggins & Winder (1961)
7104th-6th grade boys
1662nd-4th graders 62 nursery school children
Dependency measure Behavior during structured play Behavior in open classroom situation Behavior in open classroom situation Projective test imagery Behavior in open classroom situation Peer and teacher ratings
Dependencysociometric Sociometric status status measure correlation Peer ratings
-.64
Peer ratings Teacher ratings Observer ratings Teacher ratings Observer ratings
-.34 -.31
Peer ratings
-.22
Peer ratings
-.24
Peer ratings
-.33
-.32 -.27 -.33
Note. All studies except Wiggins and Winder (1961) used mixed-sex samples. In all studies higher dependency scores were associated with lower Sociometric status ratings.
With these findings in mind, it is not surprising that dependent children tend to be less well liked than their nondependent counterparts. Remarkably consistent results have emerged in this area: As Table 43 shows, every study of the dependency-sociometric status relationship in schoolchildren has reported an inverse correlation between dependency levels and popularity ratings, regardless of whether ratings were made by age mates or adults (the overall dependency-sociometric status correlation in these investigations was -.33).5 Although no studies of the dependency-sociometric status relationship have been conducted with adults, Speed and Gangestad (1997) found that women and men alike view dependency as an undesirable trait in a potential romantic partner. Little wonder that dependent adolescents and adults report pervasive feelings of isolation and loneliness (Mahon, 1982; Wiseman, 1997). The theatrical helplessness of dependent schoolchildren is mirrored in the attention- and succorance-seeking behavior of regressed psychiatric inpatients (Hollender, Luborsky, & Harvey, 1970). Parallel findings emerge outside the clinical setting, where Keinan and Hobfoll (1989) found that 'Interestingly, Gordon and Tegtemeyer (1983) found that dependent children are generally unaware of their diminished Sociometric status: They found only a small (r = -.13) and statistically nonsignificant relationship between projective dependency scores and children's estimates of their own popularity in a mixed-sex sample of 6-year-olds.
66
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
dependent pregnant women showed debilitating anxiety before and during childbirth, but when the partner was present in the delivery room this anxiety subsided to tolerable levels. Later studies showed that, even after the child is born, dependent mothers continue to experience pervasive anxiety and insecurity unless they receive frequent reassurance from the partner (Besser, Priel, & Wiznitzer, 2000; Priel & Besser, 2000). In addition to (or possibly in lieu of) seeking support from their partner, some dependent mothers cope with their insecurity by fostering dependency in their children. In a compelling demonstration of this dynamic, Thompson and Zuroff (1998) found that dependent mothers systematically undermined their adolescent daughters' efforts to succeed in a skill-based computer game, presumably because they felt threatened by the daughters' success and the increased autonomy that might ensue. When the daughters in Thompson and Zuroff s (1998) study performed well, they were criticized by their mothers; when they performed badly, their efforts were met with praise. Parallel findings were not obtained when this study was replicated using mother-son pairs (Thompson & Zuroff, 1999), suggesting that dependent mothers' dependency-fostering behaviors may be specific to daughters. Finally (and somewhat ironically), several investigations indicate that high levels of dependency are associated with increased likelihood of perpetrating abuse. Initial community-based studies found that dependent mothers showed higher-than-expected rates of child abuse (Kertzman, 1980; Melnick & Hurley, 1969), in part because these mothers had difficulty modulating negative affect through internal means. Studies also indicate that highly dependent men who are sexually immature may be at risk for engaging in child sexual abuse (Chantry & Craig, 1994; Fisher, 1969). Along somewhat different lines, community-based and forensic studies suggest that dependent men may be at increased risk for engaging in spouse abuse when they perceive the relationship to be threatened by increased autonomy in the partner. Conflicting results have been obtained in these investigations, however, and it appears that observed dependency—spouse abuse links are in part a function of the way in which dependency is assessed. Significant dependency-spouse abuse relationships are obtained in studies that use trait dependency measures (Holtzworth-Monroe, Stuart, & Hutchinson, 1997; Murphy, Meyer, & O'Leary, 1994), but when dependency is operationalized in terms of DPD symptoms and diagnoses, significant dependency-spouse abuse links are not found (Beasley & Stoltenberg, 1992; Hart, Dutton, & Newlove, 1993). DEPENDENCY-RELATED SKILLS AND STRENGTHS Three dependency-related skills and strengths have been examined in detail: interpersonal sensitivity, situation-appropriate help-seeking, and successful patienthood. CONTEXT-SPECIFIC DEFICITS AND STRENGTHS
67
Interpersonal Sensitivity The dependent person's desire to obtain and maintain nurturant, supportive relationships should, over time, lead to increased sensitivity to interpersonal cues. After all, to the degree that a person can accurately infer the feelings, beliefs, and desires of others, that person will be better able to cultivate social ties, obtain help and support, and maintain close relationships with protectors and caregivers. Masling, Johnson, and Saturansky (1974) were the first researchers to explore the dependency-interpersonal sensitivity link. After dividing a mixedsex sample of undergraduates into dependent and nondependent groups, the authors created dyads consisting of one dependent and one nondependent participant and asked the two members of each dyad to spend 15 minutes getting acquainted with each other. Following the conversations, participants individually completed a questionnaire that assessed an array of beliefs and personal characteristics. Then each participant completed a second copy of the questionnaire as they thought their partner would. Masling et al. found that dependent men were more accurate than nondependent men in predicting their partner's attitudes and beliefs. However, there were no differences between dependent and nondependent women on this dimension. When Masling et al. (1974, Experiment 2) replicated their first experiment using Peace Corps trainee roommates in place of college students, they obtained virtually identical results: Dependent men were more accurate predictors of their roommates' responses than were nondependent men, but a parallel relationship was not found in women. Masling, Shiffner, and Shenfeld (1980) extended Masling et al.'s (1974) findings to the clinical setting, comparing dependent and nondependent psychotherapy patients' ability to infer their therapists' personal beliefs. A methodology similar to that of the earlier study was used, except that dependent and nondependent patients completed attitude questionnaires as they thought their therapists would, on two occasions: (a) after three therapy sessions and (b) after termination. In this study dependent patients of both genders provided more accurate answers than did matched nondependent patients—a finding that is even more noteworthy because the issues assessed (e.g., abortion, gun control) are not topics about which therapists typically disclose.6 6 In an innovative extension of Masling et al.'s (1974, 1980) interpersonal sensitivity results, Masling, O'Neill, and Katkin (1982) found that dependent male college students showed increased autonomic arousal following cold, impersonal treatment (but not warm, friendly treatment) by a male confederate. Nondependent students showed no differential responding in the warm versus cold conditions. Apparently, dependent persons are not only accurate social perceivers, but they are physiologically reactive as well. These findings dovetail with data indicating that dependency is a risk factor for illness and disease. It may be that repeated episodes of physiological arousal in response to actual or anticipated relationship conflict diminish immunocompetence (and increase illness risk) in dependent persons.
68
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
Situation-Appropriate Help-Seeking Although meta-analyses indicate that dependent persons are at increased risk for illness, once symptoms appear, the dependent person's active approach to managing these symptoms becomes a strength. Several studies have found that dependent persons delay less long than nondependent persons in seeking treatment following symptom onset (Brown & Rawlinson, 1975; Greenberg & Fisher, 1977; Stamler & Palmer, 1971). These findings are consistent across different illness types (e.g., heart disease, cancer), symptom patterns (e.g., physical, psychological), measures of dependency (e.g., objective, projective), and participant groups (e.g., children, adults). This decreased latency in medical help-seeking is clearly an adaptive feature of dependency in that it sets the stage for earlier diagnosis, more rapid intervention, and more positive treatment outcome (Bornstein, 1993).7 At least two processes contribute to the dependent person's willingness to seek treatment quickly following symptom onset: • Reflexive help-seeking. Rapid treatment-seeking is one manifestation of the generalized, almost reflexive help-seeking stance typical of dependent persons. As noted earlier, this tendency is exhibited in a wide range of contexts (see Bornstein, 1993; Bornstein & Languirand, 2003), and while it can be a problem in social and work settings, in treatment settings it is an asset. • Positive attitudes regarding physicians and therapists. The dependent person's decreased treatment latency also reflects a positive perception of physicians and therapists (Greenberg & Fisher, 1977; Juni & LoCascio, 1985; Tyrer, Mitchard, Methuen, & Ranger, 2003). Studies show that dependent individuals tend to view treatment professionals as benign pseudo-parental figures, a perception that develops early in life (Parker & Lipscombe, 1980). Interestingly, the same help-seeking tendency that prompts dependent persons to seek help quickly following symptom onset also influences their behavior in academic settings (Bornstein & Kennedy, 1994). Because they are predisposed to seek guidance from professors and academic advisors when challenged, dependent college students deal with minor academic difficulties before these difficulties escalate into serious problems. As a result, dependent college students obtain significantly higher grade point averages than do nondependent students, even when scholastic aptitude is controlled for statistically. The dependent person's help-seeking tendencies lead to more rapid intervention following symptom onset, but they also lead to overuse of medical services, both in inpatient (O'Neill &. Bornstein, 2001) and outpatient settings (Bornstein, Krukonis, Manning, Mastrosimone, & Rossner, 1993). CONTEXT-SPECIFIC DEFICITS AND STRENGTHS
69
Successful Patienthood Early studies of dependency and patienthood showed that dependent medical and psychiatric patients are viewed by physicians and therapists as being more cooperative and compliant than nondependent patients (e.g., Davis & Eichorn, 1963; Lorr & McNair, 1964b). Later studies confirmed that these perceptions are at least partially correct: Dependent patients show greater treatment compliance and follow-through than do nondependent patients with similar problems. Thus, Nacev (1980) found that dependent outpatients missed fewer psychotherapy sessions than did matched nondependent patients. Poldrugo and Forti (1988) obtained even stronger results, finding that patients with diagnoses of dependent personality disorder (DPD) showed more consistent attendance in outpatient treatment for alcoholism than did patients with other personality disorders (PDs). Moreover, 75% of DPD patients (vs. 33% of patients with other PDs) completed the 1-year course of treatment and remained abstinent for the duration of the study. Similar results were subsequently obtained by Fisher, Winne, and Ley (1993), who compared group therapy completion rates in outpatient sexual abuse survivors with various PD diagnoses. Consistent with the findings of Poldrugo and Forti (1988), a significantly greater proportion of DPD than other-PD patients in Fisher et al.'s investigation completed the full 6-month course of treatment. The findings of Poldrugo and Forti (1988) and Fisher et al. (1993) are compelling, but not all investigations in this area have yielded positive results. Although no studies reported an inverse relationship between dependency and treatment compliance or completion, several investigations obtained nonsignificant relationships between these variables (e.g., Fals-Stewart, 1992; McMahon, Kelley, & Kouzekanani, 1993). It may be that the dependency-compliance link is stronger for certain forms of treatment than others, but additional data are needed to address this question directly.8
CLINICAL IMPLICATIONS OF THE INTERACTIONIST PERSPECTIVE The studies reviewed in this chapter point to the complexity of dependent behavior, the importance of looking beyond the traditional "passivity" view of dependency, and the need to conceptualize dependency in terms of trait-by-situation interactions. Although these studies used a diverse array of 8
Although dependency is associated with some noteworthy treatment strengths, it is linked with some important deficits as well. As discussed in chapter 10, dependent patients often have difficulty terminating treatment (Greenberg & Bornstein, 1989) and have a greater number of "pseudoemergencies" (i.e., false alarms that require therapist intervention) than nondependent patients (Emery & Lesher, 1982).
70
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
methodologies to explore a broad range of topics, their findings converge to confirm a single, simple principle with important clinical implications: The same core features of dependency that lead to maladaptive behavior in some situations lead to adaptive behavior in others. Thus, the insecurity and lack of self-confidence that create discomforting anxiety and jealousy in close relationships also help foster sensitivity to subtle verbal and nonverbal cues. The dependent person's need to please other people leads to problematic suggestibility in some situations, but it also promotes compliance with medical and psychotherapeutic regimens. The feelings of vulnerability and helplessness that combine with interpersonal stress to increase illness risk also cause the dependent person to seek treatment quickly once symptoms appear. Among the many findings described in chapter 4, one stands out as particularly noteworthy (and more than a little ironic) from the perspective of the clinician: Whereas dependency can lead some individuals to tolerate exploitation and abuse, it causes others to become perpetrators of abuse themselves. One might wonder how the same underlying trait can lead to victimization in some situations and abuse perpetration in others, but this "paradox" illustrates nicely the importance of focusing on trait-by-situation interactions to deconstruct dependent behavior. When the dependent person's strong desire to maintain close ties is expressed in the context of a dysfunctional relationship with few exit options, the likelihood that the dependent person may be exploited or abused increases. When this same desire to maintain close ties is expressed within the parental or spousal role—and directed toward a child or partner who shows signs of increased independence—the dependent person may attempt to stifle this burgeoning autonomy and maintain the status quo. In conceptualizing dependency from an interactionist perspective, it is important to keep in mind that assessment and treatment settings have great symbolic meaning to the dependent patient insofar as psychologists, physicians, and other treatment professionals represent figures of tremendous authority, insight, and power. Some patients may be overly dependent in treatment but not elsewhere, and the clinician must be sensitive to this possibility when formulating diagnoses or drawing inferences from psychological tests. These issues—and the clinical strategies they suggest—are discussed in detail in chapters 6 and 7, but in their most basic form they can be summarized as follows: Given the powerful impact of treatment context on behavior, the practitioner must be careful not to make what social psychologists call the fundamental attribution error, attributing everything the dependent patient does to internal (dispositional) factors and ignoring the impact of contextual cues on dependency-related responding. Finally, the findings reviewed thus far have implications for the therapist-patient dynamic. Given dependent persons' strong desire to maintain close ties, the practitioner working with a dependent patient must anCONTEXT-SPECIFIC DEFICITS AND STRENGTHS
71
ticipate considerable resistance when change seems imminent. As Caspi et al.( 1989, p. 395) noted: Dependency as an individual interactional style may well be even more self-perpetuating than [other personality styles], because dependent individuals are positively motivated to select and construct environments that sustain their dependency.... Dependent persons recruit and attach themselves to others who will continue to provide the nurturance and support they seek, [and] become increasingly skilled at evoking from others those nurturing responses that reinforce their dependency.
As discussed in chapters 8, 9, and 10, Caspi et al.'s (1989) insight has been confirmed in the clinical setting. Not only does resistance to therapeutic growth originate within the dependent patient, but also in the patient's interpersonal milieu (e.g., parents, siblings, spouse, children). For this reason adjunctive marital and family therapy can be an important—sometimes critical—mode of intervention for effecting long-term behavior change in dependent patients.
72
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
5 HEALTHY AND UNHEALTHY DEPENDENCY
Chapter 4 focused on situational variability in dependency and the ways in which internal processes and external factors combine to determine the outward expression of underlying dependency needs. In this chapter I examine individual differences in dependency-related behavior. Although many mental health professionals associate dependency with a pervasive pattern of help- and reassurance-seeking, studies show that people actually express underlying dependency needs in many different ways. Some manifestations of dependency are reasonably adaptive; others are not. Moreover, maladaptive expressions of dependency take a variety of forms—passive and active, obvious and subtle, consistent and variable, direct and indirect. As discussed in chapter 3, concerns regarding dependency and relatedness affect us from infancy through old age, and the mere presence of dependency strivings—even relatively intense dependency strivings—is not, by itself, problematic. This chapter is divided into two parts. First, I discuss differences between unhealthy dependency and healthy dependency, providing a framework the clinician can use to distinguish patients whose difficulties reflect maladaptive dependency from patients who appear overly dependent, but whose difficulties actually lie in other areas. Second, I describe the most com-
73
mon unhealthy dependency styles encountered in clinical settings, using two recently developed theoretical frameworks to conceptualize and classify these dependency styles.
DISTINGUISHING HEALTHY AND UNHEALTHY DEPENDENCY Numerous factors—some situational, others stemming from practitioner expectations and experience—interfere with accurate interpretation of patient behavior. Studies show that these distortions can bias the clinician's perceptions of a broad range of personality and psychopathology variables, including—but not limited to—dependency (see Cantor, Mischel, & Schwartz, 1982; Cantor, Smith, French, & Mezzich, 1980). These information-processing distortions have different biasing effects on different personality traits, but in most cases they lead the clinician to overperceive dependency in patients, especially female patients. Two factors are responsible for this widespread "dependency overestimation bias": • Context and role effects. Because mental health professionals associate dependency with pathology, many practitioners interpret the mere act of treatment-seeking as evidence of dependency (Cadbury, 1991; Games, 1984). As a result, they attribute dependency-related characteristics to anyone who enters voluntarily into the patient role. This process—which occurs automatically and unconsciously—is essentially a clinical manifestation of what social psychologists call the fundamental attribution error: peoples' tendency to disregard contextual information and attribute any salient behavior exhibited by another person to dispositional (i.e., trait) factors (Gilbert & Malone, 1995). Given societal stereotypes regarding gender and gender role, this attribution error is particularly likely to occur when a woman enters into the patient role (Banaji, Hardin, & Rothman, 1993; Gilbert, 1987). • Presenting complaints. Many of the concerns that patients discuss early in treatment are associated with dependency in the minds of clinicians. These include depression, anxiety, low selfesteem, lack of assertiveness, jealousy and insecurity, ambivalence regarding termination of a longstanding relationship, and exploitation or victimization by a trusted other. Some of these issues are empirically linked with dependency (e.g., depression, exploitation), but others (e.g., termination ambivalence) are not (Blatt & Zuroff, 1992; Nietzel & Harris, 1990). Nonetheless, on hearing these presenting complaints clinicians may instinctively develop a working hypothesis of underlying depen74
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
dency in the patient. Once in place, such a hypothesis can be difficult to revise or reject.1 Misperceptions notwithstanding, not all dependency-related behaviors are dysfunctional. As discussed in chapter 4, some of these behaviors are actually quite adaptive and promote healthy functioning in social, medical, and work settings. Thus, it is useful, both conceptually and clinically, to distinguish healthy dependency from unhealthy dependency. Two brief definitions capture the essential features of these contrasting relationship styles: " Unhealthy dependency is characterized by intense, unmodulated dependency strivings that are exhibited indiscriminately and reflexively across a broad range of situations. • Healthy dependency is characterized by dependency strivings that—even when strong—are exhibited selectively (i.e., in some contexts but not others) and flexibly (i.e., in situationappropriate ways). In the following sections I place the concept of healthy dependency in an appropriate theoretical context and explore the contrasting interpersonal and intrapsychic dynamics of healthy dependent and unhealthy dependent persons. I then review assessment tools the practitioner can use to supplement clinical observation in determining whether a particular patient expresses underlying dependency needs in a healthy or unhealthy way.
Conceptual Issues Although unhealthy dependency has a long history in psychology and psychiatry, the concept of healthy dependency is relatively new. Speculation regarding the possibility that dependent urges could be expressed adaptively first appeared during the 1950s, in the writings of psychoanalytic (e.g., Goldman-Eisler, 1951) and trait theorists (e.g., Leary, 1957). These theorists used different terminologies and conceptual frameworks, but they agreed that certain forms of dependency play a critical role in early social development and ultimately form the basis of healthy adolescent and adult relationships. Following the delineation of these psychodynamic and trait models, behavioral researchers explored the value of dependent behaviors in facilitating learning and skill acquisition (e.g., Hartup,
'As Mahoney (1977) noted, this tendency to cling to an initial hypothesis even in the face of contradictory data reflects confirmatory bias on the part of the clinician—selective attention to supporting evidence, coupled with inattention to (or discounting of) evidence that contradicts one's favored position. Confirmatory bias is problematic in the clinical setting, but it is not limited to this context. Studies show that confirmatory bias also distorts the processing of hypothesis-relevant information by experienced scientists (Mahoney, 1976, 1987).
HEALTHY AND UNHEALTHY DEPENDENCY
75
1963; Sinha & Pandey, 1972). These behavioral analyses helped refine and reify the instrumental-emotional dependency distinction initially described by Heathers (1955) and others.2 Despite periodic references to adaptive features of dependency during the 1950s and 1960s, the notion that dependent urges could actually enhance adjustment and functioning was not taken seriously until two developments—one theoretical, one empirical—laid the groundwork for a paradigm shift in this area. In the theoretical realm, dependency was reconceptualized within the context of attachment theory by Ainsworth (1972, 1989), Bowlby (1969, 1973), and others (e.g., Alperin, 2001; Hetherington, 1999). Using these frameworks, researchers discovered a surprising degree of continuity in dependency-related attachment behaviors from childhood through adolescence and beyond (Feeney &Noller, 1990; Gjerde, 2001; Main, Kaplan, & Cassidy, 1985). The other major development took place in the empirical arena, where Masling and his colleagues' pioneering studies demonstrated the adaptive features of dependent personality traits in psychotherapy patients, college students, and community participants (see Masling, 1986, for a review). These studies showed that when high levels of dependency are coupled with effective defenses and good coping skills, they can actually enhance functioning in a variety of social and occupational contexts. Later investigations (e.g., Bornstein & Kennedy, 1994; Pincus & Wilson, 2001) confirmed and extended Masling's (1986) results. In its contemporary usage, the term healthy dependency overlaps with several other constructs in psychology, sociology, and medicine. These include adaptive dependency (Bornstein, 1998d), compensatory dependency (Bakes, 1996), interdependence (Cross & Madson, 1997), connectedness (Clark & Ladd, 2000), and mature dependency (Baumeister & Leary, 1995). The essential features of these five constructs are summarized in Table 5.1. As the definitions in Table 5.1 show, these constructs overlap to some degree, but they are not isomorphic. Moreover, healthy dependency differs from each of these constructs, both conceptually (Bornstein, 1995a) and empirically (Bornstein &Languirand, 2003).3
2
Early humanistic theorists also made reference to the adaptive features of dependency (e.g., Maslow, 1958; Rogers, 1963), but these frameworks have had less influence than other models on contemporary approaches to diagnosis, assessment, and treatment of the dependent patient (cf. Florian, Mikulincer, & Hirschberger, 2002; Harrison, 1987). 3 Healthy dependency also overlaps with the construct of insecure attachment (Collins & Read, 1990; Sundin, Armelius, & Nilsson, 1994), but there are noteworthy differences as well. Studies show that insecure attachment is associated with substantially greater behavioral consistency than healthy dependency, which is expressed in very different (even diametrically opposing) ways in different relationships (Bornstein, Riggs, Hill, & Calabrese, 1996; Heiss, Herman, & Sperling, 1996). Beyond these behavioral differences, studies confirm that healthy dependency scores are only modestly related to scores on measures of insecure attachment, with correlations typically in the r = 30-.40 range (Bornstein et al., 2003; Pincus & Wilson, 2001).
76
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
TABLE 5.1 Healthy Dependency in Context: Related Theoretical Constructs Construct Adaptive dependency Compensatory dependency Interdependence
Connectedness
Mature dependency
Definition Situation-specific help- and support-seeking that facilitates task performance and strengthens interpersonal ties Goal-directed dependent behavior in one domain that enhances autonomous functioning on one or more other domains Mutual (usually constructive) dependency between two individuals, groups, or larger entities (e.g., corporations, nations) The experience of closeness/intimacy coupled with a firm sense of autonomy/individuality ("self-inrelation-to-others") Similar to connectedness, with the additional implication that the individual is willing to lean on others and be influenced by external sources
Note. Detailed discussions of these constructs are provided by Tail (1997), Baumeister and Leary (1995), Clark and Ladd (2000), Cross and Madson (1997), and Hetherington (1999).
Contrasting Self-Presentations Although healthy dependent and unhealthy dependent patients often present with similar problems, they differ in the manner in which they describe these problems. Four aspects of a patient's initial self-presentation (summarized in the top portion of Table 5.2) can help the clinician determine whether that patient's dependency is healthy or unhealthy: • Insight. At the outset of therapy, patients with unhealthy dependency typically show little insight regarding the role that underlying dependency needs play in their current difficulties. Healthy dependent patients not only show greater initial insight, but as therapy progresses they also show greater capacity to use this insight to effect positive change (Blatt & Ford, 1994; Bomstein, 1994c, 1998a). Healthy dependent patients display what Weissmark and Giacomo (1998) referred to as relationship reactivity: They are able to integrate therapist feedback into their working self-concept and use this information to modify dysfunctional behavior patterns. • Social skills. Although research confirms that dependency is associated with sensitivity to interpersonal cues (Masling, Johnson, & Saturansky, 1974; Masling, Schiffner, & Shenfeld, 1980), unhealthy dependent patients are unable to apply this information effectively in real-world situations. Even when patients with unhealthy dependency accurately decode subtle verbal and nonverbal signals, their ability to use this knowlHEALTHY AND UNHEALTHY DEPENDENCY
77
TABLE 5.2
Unhealthy and Healthy Dependency: Constrasting Self-Presentations and Intrapsychic Dynamics Dimension
Unhealthy dependency
Insight
Poor at outset; modest gains over time
Social skills
Compromised by chronic anxiety To be "fixed" (or "cured") by an omniscient therapist Few close relationships; fixed roles and limited range of dynamics Compulsive ("mindless") help- and supportseeking Relatively immature (e.g., projection, denial); ineffective in managing anxiety Poor; regression under stress
Therapeutic goals Relationship quality Urgency of experienced dependency needs Defense effectiveness
Impulse control Cognitive complexity
Simple, conceptually unsophisticated selfrepresentation
Healthy dependency Good at outset; potential for greater gain over time Able to apply social information effectively Enhanced self-efficacy, intimacy, and parenting-career skills Dense social network; role flexibility Selective, situationappropriate help- and support-seeking Relatively mature (e.g., sublimation, humor); effective in managing anxiety Good; can moderate impulses through internal means Complex, conceptually sophisticated selfrepresentation
Note. The first four dimensions in this table reflect self-presentation differences; the second four dimensions reflect contrasting intrapsychic dynamics.
edge is impaired by chronic anxiety that drains cognitive resources and leads to "mindless" (i.e., reflexive) support- and reassurance-seeking (Alam, 1986; Overholser, 1996). Because healthy dependent patients have a lower baseline anxiety level, they apply social information more effectively and display better social skills than do patients with unhealthy dependency.4 Therapeutic goals. Even when healthy and unhealthy dependent patients report similar presenting complaints, they almost invariably have different therapeutic goals. Unhealthy dependent patients typically enter treatment with the explicit or implicit agenda of being "fixed" or "cured" by an omniscient, om4
The "mindlessness" of unhealthy dependency is a key factor in its maladaptive impact on social and occupational functioning. As Langer (1989) pointed out, longstanding behavior patterns tend to be exhibited automatically and reflexively whenever relevant situational cues are encountered (see also Bargh & Chartrand, 1999). In the case of dependency, these cues typically take the form of protectors and caregivers (and even here, the dependent person is likely to misperceive others as potential caregivers when in fact they are not). As Bornstein and Languirand (2003) noted, by replacing "mindless" with "mindful" responding, the dependent individual can begin to express underlying dependency needs in more adaptive, situation-appropriate ways.
78
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
TABLE 5.3 Unhealthy and Healthy Dependency: Contrasting Relationship Patterns Relationship domain Friendship Romance Parenting Work
Unhealthy dependency pattern
Healthy dependency pattern
Possessiveness, jealousy, Openness, relationship overidentification, identity flexibility, clear boundaries merging Self-validating/narcissistic Mature intimacy/relatedness, intimacy, sexual insecurity open communication Autonomy-squelching parenting, Authoritative parenting, flexible variable/inconsistent limitlimit-setting setting Difficulty taking responsibility, Interdependence, cooperation, self-promotion, territoriality collegiality, group focus
Note. Detailed discussions of these contrasting relationship patterns are found in Bornstein and Languirand (2003).
nipotent therapist (Coen, 1992; Emery & Lesher, 1982; Overholser, 1987). The therapeutic goals of the healthy dependent patient, in contrast, tend to center on enhancing selfefficacy, improving parenting and career skills, and developing greater intimacy in close relationships (Bornstein & Languirand, 2003). • Relationship quality. Unhealthy dependent patients differ from healthy dependent patients in the quality of their interpersonal ties (Nelson, Hammen, Daley, Burge, & Davila, 2001). The unhealthy dependent patient's relationships invariably share a common core dynamic, with the patient assuming the role of care receiver, and those closest to him functioning primarily as caregivers (Rathus & O'Leary, 1997; Whiffen & Aube, 1999). Healthy dependent patients have a greater number of ongoing relationships than do unhealthy dependent patients (i.e., a denser network of social ties), and these relationships are more diverse, characterized by a broader array of roles (Cadbury, 1991; Miller, 1979). Table 5.3 summarizes the contrasting relationship patterns of healthy dependent and unhealthy dependent patients in four key domains—friendship, romance, parenting, and work—and illustrates these role flexibility differences. Divergent Intrapsychic Dynamics Healthy dependent and unhealthy dependent patients not only differ with respect to outward self-presentation, but private experience as well. Four domains of intrapersonal functioning (summarized in the bottom portion of Table 5.2) distinguish healthy from unhealthy dependency: HEALTHY AND UNHEALTHY DEPENDENCY
79
Urgency of experienced dependency needs. Even when their underlying dependency needs are equally intense (as reflected in comparable scores on projective measures; see Bornstein, 1999), healthy dependent and unhealthy dependent patients expertence their dependency urges differently. Unhealthy dependency is characterized by intense cravings for support that have a compulsive, pressured quality. Healthy dependency is characterized by urges that are more intermittent and less intense. This difference is in part a product of the individual's self-concept: The patient with unhealthy dependency has a pervasive sense of vulnerability that cuts across a broad range of contexts (Blatt, 1974,1991;Bomstein, 1996a). The healthy dependent patient's self-representation is more balanced and less extreme; her underlying vulnerability is attenuated by a sense of self-efficacy in some situations and settings. Defense effectiveness. Dependency is associated with what Ihilevich and Gleser (1986) termed a "turning against self defense configuration, wherein anger and resentment are bottled up and channeled inward (Berman & McCann, 1995). Within the context of this internalizing style, however, patients with unhealthy dependency tend to rely on less mature—and less effective—defenses (e.g., projection, denial; see Lingiardi et al., 1999; Vaillant, 1994). Healthy dependent patients utilize more mature defenses (e.g., sublimation, humor), which enable them to moderate anxiety more effectively.5 Impulse control. The more mature defenses of healthy dependent patients enhance their ability to manage urges through internal means. Thus, healthy dependent patients are better able to control dependency-related impulses (e.g., helpseeking) that—when expressed indiscriminately—lead to problems in social and occupational functioning (Mongrain, 1998; Nelson, Hammen, Daley, Burge, & Davila, 2001). Patients with unhealthy dependency have greater difficulty controlling these impulses and may regress under stress, displaying immature, childlike behavior that alienates those close to them (e.g., physical clinging, theatrical helplessness; see Haaga, Fine, Terrill, Stewart, & Beck, 1995; Hollender, Luborsky, & Harvey, 1970; Overholser, 1996). Cognitive complexity. The self-representation of the healthy dependent patient is more complex and nuanced than that of the unhealthy dependent patient: It has been integrated at a 'Detailed discussions of the links between dependency and defense style are provided by Cramer, Blatt, and Ford (1988); Devens and Erickson (1998); and Levit (1991). 80
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
higher conceptual level (Blatt, 1991) and contains a broader array of self-relevant traits (Cross et al., 2000; Kuperminc, Blatt, & Leadbeater, 1997; Little & Garber, 2000). As a result, the healthy dependent patient is capable of productive introspection, flexible interpersonal responding, and effective coping with challenge and loss (Neuberg & Fiske, 1987; Solomon & Haaga, 1993; Tjosvold & Fabrey, 1980; Yasunaga, 1985).6 Assessment Tools As Table 5.3 illustrates, a variety of indicators can help the clinician distinguish healthy dependent from unhealthy dependent patients during the initial intake and therapy sessions. Some of these indicators (e.g., therapeutic goals) are revealed through patients' reports of current mental states, whereas others (e.g., insight) are reflected in patients' descriptions of past and present relationships, and still others (e.g., defense effectiveness, impulse control) are best assessed via careful observation of patient functioning within the clinical setting. Beyond these clinical data, two psychological tests may be useful in distinguishing healthy from unhealthy dependency, and in refining the clinician's initial impressions. Both instruments were introduced briefly in chapter 2. Here I focus on practical use of these instruments in the clinical setting. The Depressive Experiences Questionnaire for Adolescents (DEQ-A) Although the DEQ-A—like the original version of the DEQ—was designed to assess anaclitic and introjective personality traits. Subsequent cluster and facet analyses of Anaclitic scale items revealed that these items form two subscales (Blatt, Zohar, Quinlan, Luthar, & Hart, 1996; Blatt, Zohar, Quinlan, Zuroff, & Mongrain, 1995). Blatt and his colleagues labeled these subscales Dependence and Relatedness, and they demonstrated that scores on the Dependence subscale predict maladaptive behavioral tendencies akin to those of insecure attachment, whereas scores on the Relatedness subscale predict a higher level of functioning reminiscent of Baumeister and Leary's (1995) mature dependency. Ten DEQ-A items (2, 19, 22, 23, 26, 28, 38, 42, 46, and 52) compose the Dependence subscale, while eight items (9, 20, 32, 34, 45, 50, 55, and 65) compose the Relatedness subscale. Norms and construct validity data for these subscales are provided by Blatt, Zohar, Quinlan, Luthar, and Hart (1996) and Blatt, Zohar, Quinlan, Zuroff, and Mongrain 'Cognitive complexity is also adaptive in the cateer domain, and several investigations have shown that relative to those people who are less cognitively complex, cognitively complex persons (a) are able to make vocational choices that fit well with their skills and interests and (b) show more rapid career progress and higher levels of career satisfaction (Bodden, 1970; Feist, 1994; Stein, 1994). HEALTHY AND UNHEALTHY DEPENDENCY
81
(1995). The DEQ-A can help distinguish unhealthy from healthy dependency in patients in their mid- to late teens. The Relationship Profile Test (RPT) The RPT yields three scores, two of which—dysfunctional overdependence (DO) and healthy dependency (HD)—are useful in identifying healthy dependent and unhealthy dependent patients (Bomstein & Languirand, 2003). DO and HD scores can also serve as a baseline for tracking therapeutic progress, because studies indicate that successful treatment of dependent patients is associated with increases in HD scores and decreases in DO scores. Norms and construct validity data for the RPT are provided by Bornstein, Geiselman, Eisenhart, and Languirand (2002); Bornstein, Languirand, et al. (2003); and Bornstein, Geiselman, et al. (2004). These data focus on the differential relationships of DO and HD scores with scores on measures of attachment style, self-concept, self-esteem, identity, relatedness, gender role, life satisfaction, and affect regulation. The RPT is useful in distinguishing unhealthy from healthy dependency in patients age 18 and older.
UNHEALTHY DEPENDENCY PATTERNS By late adolescence, people develop relatively stable ways of expressing underlying dependency needs. Two theoretical frameworks are useful in conceptualizing these contrasting expressions of dependency: Pincus and Gurtman's (1995) Three-Vector Model and Bornstein and Languirand's (2003) Four-Pattern Model. Although these frameworks use different categories and classification schemes, there is considerable convergence between the two perspectives, both conceptually and empirically. The surface differences between these models are due largely to their different levels of analysis: Pincus and Gurtman's (1995) framework was derived from sophisticated circumplex analyses of participants' responses to widely used self-report dependency tests, whereas Bornstein and Languirand's (2003) framework was based on synthesis of the empirical literature examining convergences and divergences in the results obtained with different types of dependency measures (i.e., selfreport, projective, interview, behavioral).
Pincus and Gurtman's Three-Vector Model Beginning with the assumption that dependency reflects a core underlying motivation to obtain and maintain nurturant, supportive relationships, Pincus and Gurtman (1995) explored the latent structure of dependency by analyzing links between participants' scores on widely used self-report dependency measures and their scores on circumplex and five-factor model 82
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
TABLE 5.4 Exploitable, Submissive, and Love Dependency Core features Dimension
Exploitable dependency
Submissive dependency
Love dependency
Predominant interpersonal pattern
Tolerance of exploitation/ mistreatment
Underlying traits (five-factor domain scores) Adult attachment style Pathological attachment behaviors
High on neuroticism
Difficulty resisting Strong need for external influence succorance and close ties with others High on neuroticism; High on neuroticism low on openness
Secure/fearful
Fearful
Compulsive careseeking
Parental introjects
High paternal affiliation
Compulsive careNone (marked seeking angry absence of and angry withdrawal) withdrawal High maternal and High maternal and paternal affiliation; paternal affiliation high maternal control
Secure
Note. Detailed discussions of these findings are provided by Pincus and Gurtman (1995) and Pincus and Wilson (2001).
trait measures (Gurtman, 1992; Pincus, 1994). They found that clusters of dependency test items were distributed in meaningful ways across three separate regions (or vectors) of the interpersonal circumplex and represented distinct, unhealthy dependency subtypes. These test item clusters ultimately evolved into the three 3VDI subscales (see Pincus & Gurtman, 1995). Table 5.4 summarizes some key features of the three dependency subtypes described by Pincus and Gurtman (1995) and Pincus and Wilson (2001). The intra- and interpersonal dynamics of these three dependency subtypes differ as well, and dovetail with the trait patterns summarized in Table 5.4: • Submissive dependency. Submissive dependents are anxious and insecure, and they attempt to strengthen interpersonal ties by exaggerating their vulnerability. They are particularly sensitive to interpersonal disruption and go to great lengths to prevent conflict situations from occurring. Submissive dependent people assume a passive role in many relationships, display the selfpresentation style of supplication (see Table 1.3), and devote considerable energy to accommodating other people's needs to minimize the possibility that they will have to function autonomously. • Exploitable dependency. Exploitable dependents—like submissive dependents—tend to acquiesce to others' needs and demands. However, in contrast to the submissive dependent, the HEALTHY AND UNHEALTHY DEPENDENCY
83
exploitable dependent experiences a powerful, compelling urge to seek protection from a valued other. This can sometimes lead to victimization, as the exploitable dependent person has difficulty terminating dysfunctional relationships or setting limits on others' behavior. Supplication and exemplification are the exploitable dependent's favored self-presentation styles. • Love dependency. Love dependents experience strong affiliative strivings, and accompanying these strivings is an underlying confidence that they will be successful in maintaining close ties and obtaining the help and support they desire. Love dependents take a more active approach than other types of dependent people to cultivating ties with valued others, use a broader array of self-presentation styles, and are more socially skilled than submissive and exploitable dependents.7 Bornstein and Languirand's Four-Pattern Model Following an extensive review of the empirical literature on the etiology and dynamics of dependency, Bornstein and Languirand (2003) identified four unhealthy dependency patterns that emerge by late childhood, coalesce during adolescence, and evolve throughout adulthood in response to changing circumstances. Preliminary findings support the validity of this fourpattern model, although additional data are needed to confirm the utility of the model in clinical settings and assess the covariation of these dependency patterns with other personality traits and PDs. Table 5.5 shows the prototypic social influence strategies of individuals with different unhealthy dependency patterns, and the characteristic reactions of others to these social influence strategies. As Table 5.5 shows, these social influence strategies elicit a broad range of reactions, ranging from affection and nurturance to hostility and anger. Although some persons show features of a single unhealthy dependency pattern, many show features of more than one pattern (Bornstein, Geiselman, Eisenhart, & Languirand, 2002). Moreover, these patterns may shift over time, as longstanding behavior tendencies are challenged within and outside therapy (Steele, van der Hart, & Nijenhuis, 2001). The core features of Bornstein and Languirand's (2003) four unhealthy dependency patterns are as follows: • Helpless dependency. People with a helpless dependency pattern maintain ties to others by exaggerating their vulnerability. They 7 As might be expected, Pincus and Wilson (2001) found small to moderate positive correlations among scores on the three 3VDI subscales. For example, in their initial validation sample of 921 nonclinical participants, subscale intercorrelations were as follows: Love-Exploitable, r = .52; LoveSubmissive, r = .21; Exploitable-Submissive, r = .49.
84
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
TABLE 5.5
Unhealthy and Healthy Dependency: Prototypic Social Influence Strategies and Characteristic Reactions of Others Dependency pattern Helpless Hostile Hidden Conflicted
Protypic social influence strategy
Characteristic reactions of others
Help- and support-seeking; clinging; crying; theatrical helplessness Stated or implied threats; selfdestructive gestures; intimidation Weakness; vulnerability displays; illness/symptom exaggeration Inconsistency; emotional lability; rapid shifts between dependency and autonomy
Supportive-protective urges, alternating with periods of frustration and distancing Fear, guilt, hidden resentment; eventual withdrawal/ relationship termination Desire to protect and nurture, coupled with infantilization (and possible exploitation) Anxiety, confusion, anger, ambivalence; periodic efforts to restore relationship stability
present themselves as weak and emotionally needy, and they often appear childlike and immature—easily frustrated and quick to cry. It is tempting to think of helpless dependent people as passive and "fragile," but in fact they are not. Their helplessness is a tool through which they draw others in and trap people into gratifying their dependency needs. Hostile dependency. Hostile dependent people maintain ties to others by intimidating them. On the surface, the hostile dependent person may appear to be in turmoil, barely functioning, and on the verge of breaking down. When one looks closely, however, it becomes clear that this surface appearance is deceiving: In reality, the hostile dependent person is very much in control. This control usually stems from some implied or stated threat (e.g., a suicide gesture), which is designed to exploit people's guilt and prevent them from ending the relationship. Hidden dependency. Individuals who display hidden dependency behave in an overdependent manner, but oftentimes their dependent behavior is so subtle and indirect that its true nature is not recognized by the person, or by those around her. Hidden dependency can take many forms, from feigned illnesses to imaginary allergies, but whatever form it takes, hidden dependency functions much like the helpless and hostile dependency patterns: It traps people into remaining involved in an unsatisfying relationship they might otherwise end. Conflicted dependency. Conflicted dependent people show markedly inconsistent behavior, wavering between periods of exHEALTHYAND UNHEALTHY DEPENDENCY
85
treme overdependence and superficial, short-lived episodes of autonomy. Such people can be especially difficult to interact with over extended periods because they are so unpredictable. Oftentimes the conflicted dependent person will show features of hostile dependence during her "overdependent" periods and hidden overdependence during her "autonomous" periods.8
FROM LABORATORY TO CONSULTING ROOM: DIAGNOSIS, ASSESSMENT, AND TREATMENT OF DEPENDENT PATIENTS The developmental antecedents, interpersonal dynamics, and clinical correlates of healthy dependency are discussed in detail elsewhere (Baumeister & Leary, 1995; Bornstein & Languirand, 2003; Clark & Ladd, 2000; Cross, Bacon, &Norris, 2000; Gabriel & Gardner, 1999; Hetherington, 1999; Sato, 2001). Although these researchers approach the topic from a variety of perspectives, their analyses converge to confirm that healthy dependency helps promote optimal functioning in many different areas of life. In the context of clinical work with dependent patients, healthy dependency is important because it helps define the goal—the ideal endpoint—of treatment. Consistent with the conclusions that emerged from an analysis of dependency across the life span (chap. 3), research confirms that effective therapeutic work with dependent patients must focus on replacing unhealthy dependency with healthy dependency, being careful not to move the patient too far toward independence/detachment. The remainder of this book focuses primarily on unhealthy dependency, with healthy dependency providing background and context for this discussion. Research on healthy dependency becomes increasingly important as I move from diagnosis to assessment to treatment, and it becomes central when I delineate an integrated treatment model for therapeutic work with dependent patients. As I move from laboratory to consulting room, I continue to use research evidence to guide the discussion. The studies reviewed in chapters 1 through 5 are relevant to clinical work with dependent patients and to these studies will be added investigations of diagnosis and comorbidity, testing and assessment, and treatment process and outcome. By combining laboratory evidence with clinical data, the clinician gains a deeper understanding of the interpersonal and intrapsychic dynamics of dependency in clinical settings.
'Patients with hostile and conflicted dependency also show features of borderline PD, so care must be taken to distinguish these patterns in the clinical setting. Because there is considerable symptom overlap in hostile dependent, conflicted dependent, and borderline patients, assessment (rather than diagnostic) data are particularly useful in distinguishing these syndromes (see chap. 7 for a discussion of this issue).
86
CONCEPTUAL AND EMPIRICAL FOUNDATIONS
Our discussion of clinical applications begins with a review of diagnostic issues (chap. 6) and an analysis of dependency-related assessment techniques (chap. 7). I then outline the major psychotherapeutic approaches that have been used with dependent patients (chap. 8) and combine the most effective elements of these approaches into an integrated treatment model (chap. 9). Finally (chap. 10), I discuss special treatment issues and adjunct treatment modalities that enhance the effectiveness of traditional psychotherapeutic work with dependent patients.
HEALTHY AND UNHEALTHY DEPENDENCY
87
6 DIAGNOSIS
Dependent personality disorder (DPD) occupies a unique place in the history of dependency research. The DSM-IV (APA, 1994) definition of DPD overlaps with contemporary conceptualizations of interpersonal dependency, but there are important differences as well. Patients diagnosed with DPD tend to score high on questionnaire measures of dependency, but these relationships—even when they are statistically significant—are usually modest in magnitude. As is true of self-report (but not projective) dependency tests, there are gender differences in DPD, with women acknowledging a greater number of dependency-related symptoms than men do. However, as is true of projective (but not self-report) dependency tests, DPD symptoms covary with mood, with negative mood states associated with increases in DPD symptoms.1 From the practitioner's perspective, DPD diagnosis represents a unique source of information about dependency, similar in certain respects to the information gleaned from other sources, but also contributing data that no 'In evaluating the impact of mood on dependency test scores, it is important to distinguish changes in depressive symptoms from variations in everyday mood states. Studies indicate that changes in depressive symptoms affect both self-report and projective dependency scores (Masling, 1986; Hirschfeld, Klerman, Clayton, & Keller, 1983), whereas more minor variations in mood states affect projective—but not self-report—dependency scores (Bornstein, Bowers, & Bonner, 1996a).
91
other measure can provide. For the most part, DPD diagnoses and symptom ratings tap the unhealthy, problematic aspects of dependency, but do not capture dependency's more adaptive features (see chap. 5). The clinician's task is to determine how DPD symptom information complements and extends other information about each patient's underlying and expressed dependency needs. Three principles are useful in this regard: • DPD diagnosis provides a shorthand symptom picture. As Widiger (1993) noted, there is much more heterogeneity within diagnostic categories than many clinicians realize, and this is particularly true for personality disorder (PD) categories. Nonetheless, a DPD diagnosis can be a convenient shorthand summary of a patient's dependency-related difficulties, especially when Axis II data are accompanied by a comprehensive multiaxial assessment. • Diagnosis helps locate DPD in relation to other psychological disorders. DPD shows predictable patterns of comorbidity with other Axis I and Axis II diagnoses. The comorbidity patterns reported by clinical researchers do not always reflect the assertions of the DSM-IV and DSM-IV-TR, however, which is why an awareness of empirical findings in this area (not just the DSMIV and DSM-IV-TR guidelines) is critical in predicting which other syndromes a DPD patient is likely to show, both at present and in the future. • Diagnosis sets the stage for treatment planning. As discussed in chapters 8, 9, and 10, a variety of interventions, alone and in combination, have been used to treat excessive dependency. These include psychodynamic, behavioral, cognitive, humanistic, experiential, and pharmacological treatment regimens, many of which have been tested empirically in inpatient and outpatient settings. Accurate information regarding DPD symptoms can help the clinician implement a treatment plan tailored to the patient's current functioning. In this chapter, I discuss DPD diagnosis. I begin by reviewing the history of DPD in the DSM series to place the current symptom criteria into an appropriate context. I then discuss research on the validity of the DSM-IV DPD symptoms, offer an alternative set of criteria that are aligned more closely with current findings in this area, and discuss issues related to DPD diagnosis (e.g., gender differences, prevalence rates, differential diagnosis, comorbidity). Finally, I outline a framework for maximizing the utility of dependencyrelated diagnostic information and provide guidelines for implementing this framework.
92
CLINICAL APPLICATIONS
DEPENDENT PERSONALITY DISORDER IN THE DSM The diagnostic criteria for DPD have changed considerably during the past 50 years. In the DSM-I (APA, 1952), DPD was a subtype of the passiveaggressive personality, identified as the "passive-aggressive personality, passive-dependent type." These passive-dependent persons were characterized by "helplessness, indecisiveness, and a tendency to cling to others as a dependent child to a supporting parent" (APA, 1952, p. 37). As noted in chapter 4, the DSM-I description of passive-dependent PD was strongly influenced by mid-century psychodynamic conceptualizations of dependency (e.g., Sullivan, 1947). Interest in dependency and DPD declined during the early 1960s, as the influence of the psychodynamic perspective waned and other theoretical perspectives on dependency had not yet taken hold (Ainsworth, 1969; Bornstein, 1993). Thus, in the DSM-II (APA, 1968), DPD was relegated to a catch-all category of "other personality disorders of specified types," a grouping that also included the "immature" personality. The DSM-II provided no description of the symptoms underlying passive-dependent PD and no hypotheses regarding the etiology of the disorder. Finally, a full-fledged category of DPD was included in the DSM-III (APA, 1980), with the disorder defined in terms of three broad symptoms: (a) passivity in interpersonal relationships; (b) a tendency to subordinate one's needs to those of others; and (c) lack of self-confidence. The DSM-III conceptualization of DPD emphasized submissiveness, timidity, insecurity, and immaturity, and focused on dependent people's willingness to put others' needs before their own (Kaplan, 1983; Millon, 1981). The DSM-IH-R (APA, 1987), DSM-IV (APA, 1994), and DSM-IVTR (APA, 2000) DPD criteria represented a substantial improvement over earlier versions in several respects. By the time the DSM-III-R was published, the psychodynamic roots of DPD emphasized in the DSM-I had given way to a more eclectic, integrative perspective. In addition, the DPD symptoms described in the DSM-III-R and DSM-IV captured a wide range of dependency-related behaviors and affective responses—a great improvement over the three broad (but vague) symptoms in the DSM-III. Moreover, during the construction of the DSM—IV, care was taken to reduce the overlap between DPD and other Axis II disorders (Hirschfeld, Shea, & Weise, 1991), and one DSM—III—R DPD symptom (being "easily hurt by criticism or disapproval") was dropped from the DSM-IV to improve the discriminant validity of the DPD symptom criteria.2 2
Because the symptom criteria themselves were not revised in the DSM-IV-TR, I simply refer to DSM-IV DPD symptoms for the remainder of the chapter. Where substantive changes were made in the background or textual material for DPD in the DSM-IV-TR, I distinguish between earlier and later versions of the manual.
DIAGNOSIS
93
Limitations of DSM-IV DPD Symptoms Even though the DSM-IV DPD symptom criteria are stronger than those in earlier editions of the manual, two problems remain. The first problem has to do with an implied link between dependency and passivity; the second concerns external validity. • The passivity problem. Although the nomenclature has been changed and "passive-dependent personality disorder" no longer exists, there is still a strong implied association between dependency and passivity throughout the DPD symptom criteria— an association that is clearly contradicted by recent research on dependency (Bornstein, 1997a; Pincus, 2002). As discussed in chapters 4 and 5, submissive passivity is but one of many self-presentation strategies used by dependent individuals to strengthen ties to others. • The external validity problem. Flaws in the DSM—IV DPD criteria are also a product of the methods used to validate these criteria. As is the case for most PDs, researchers have focused primarily on evaluating the internal reliability of the DPD symptom criteria and on assessing the relationship of DPD symptoms to other Axis I and Axis II diagnoses (Widiger, Frances, Pincus, Davis, & First, 1991). For the most part, issues related to external (i.e., criterion-referenced) validity have not been addressed.3 Empirical Evidence The left column of Table 6.1 lists the essential features and eight symptom criteria used to diagnose DPD in the DSM-IV. The right column of this table summarizes empirical evidence related to the essential features and each symptom criterion. As Table 6.1 shows, the essential feature of DPD is generally consistent with research on dependency, with one caveat: Studies indicate that dependency-related motivations and behaviors stem from a view of the self as powerless and ineffectual. Without question, a "helpless" selfconcept is an essential feature of DPD (Blatt, 1990; Bornstein, 1996a; Overholser, 1992). Four of the eight DSM—IV DPD symptoms (1,5,6, and 8) are supported by the results of laboratory and field studies of dependency. However, researchers have never examined empirically the assertion that the DPD pa3
The problems here were twofold. First, members of the DSM-IV Personality Disorders Task Force did not conduct field trials to assess the utility of the proposed DPD symptom criteria (Widiger, personal communication, September 19, 1995). Second, they did not review empirical studies assessing the behavior of dependent individuals in laboratory and field settings (First, personal communication, September 20, 1995; Reich, personal communication, September 24, 1995).
94
CLINICAL APPLICATIONS
TABLE 6.1
DSM-IV and DSM-IV-TR Dependent Personality Disorder Criteria: Empirical Evidence Criterion
Relevant findings
A pervasive and excessive need to be Studies indicate that the "pervasive and taken care of that leads to submissive excessive need to be taken care of" and clinging behavior and fears of results from the dependent person's separation, beginning by early view of the self as powerless and adulthood and present in a variety of ineffectual, not from attachment-based contexts, as indicated by five (or more) fears of separation (Bornstein, Riggs, of the following: Hill, & Calabrese, 1996; Livesley et al., 1990). Studies support the assertion that DPD is associated with "submissive and clinging behavior and fears of separation" (Birtchnell & Kennard, 1983; Keinan & Hobfall, 1989). 1. Has difficulty making everyday Supported by laboratory and field studies decisions without an excessive amount (Masling et al., 1968; Tribich & Messer, of advice and reassurance from others 1974). 2. Needs others to assume responsibility Never tested directly for most major areas of his or her life 3. Has difficulty expressing disagreement Contradicted by findings from laboratory with others because of fear of loss of studies. Under certain conditions the support or approval dependent person expresses disagreement vociferously (Bornstein et al., 1987; Bornstein, Riggs, Hill, & Calabrese, 1996). 4. Has difficulty initiating projects or doing Contradicted by findings from laboratory things on his or her own (because of a and field studies. Dependent persons lack of self-confidence in judgment take the initiative in medical rather than a lack of motivation or (Greenberg & Fisher, 1977) and energy) academic settings (Bornstein & Kennedy, 1994; Masling et al., 1981). 5. Goes to excessive lengths to obtain Supported by laboratory and field studies nurturance and support from others, to (Agrawal & Rai, 1988; Weiss, 1969) the point of volunteering to do things that are unpleasant 6. Feels uncomfortable and helpless Supported by laboratory and field studies when alone because of exaggerated (Masling et al., 1981; Simpson & fears of being unable to care for him- or Gangestad, 1991). herself 7. Urgently seeks another relationship as Never tested directly, a source of care and support when a close relationship ends 8. Is unrealistically preoccupied with fears Supported by laboratory and field studies of being left to take care of him- or (Berg, 1974; Birtchnell, 1988) herself Note. Originally published as Table 1 in "Dependent Personality Disorder in the DSM-IV and Beyond," by R. F. Bornstein, 1997, Clinical Psychology: Science and Practice, 4, pp. 175-187. Copyright 1997 by Oxford University Press. Reprinted with permission.
tient needs others to assume responsibility for most major areas of life (Symptom 2), nor have they tested the proposition that the DPD patient urgently DIAGNOSIS
95
seeks another relationship as a source of care and support when a close relationship ends (Symptom 7). These symptoms might well be valid predictors of pathological dependency, but as yet there is no evidence supporting their inclusion in the diagnostic manual.4 Two DSM-IV DPD symptoms are contradicted by the results of relevant empirical studies. Contrary to Symptom 3, studies show that under certain conditions (e.g., when competing for the approval of an authority figure), dependent persons express disagreement quite vociferously (Bomstein, Masling, & Poynton, 1987; Bornstein, Riggs, Hill, & Calabrese, 1996). Contrary to Symptom 4, studies indicate that dependent people initiate certain kinds of projects and activities with no prompting whatsoever from others (e.g., Greenberg & Fisher, 1977; Masling, O'Neill, & Jayne, 1981). DPD in the DSM-V To bring the DPD symptom criteria into line with empirical research on dependency, several changes must be made in future versions of the DSM. In particular, the DPD symptoms should be modified to (a) reflect the centrality of the dependent person's view of the self as weak and ineffectual and (b) make explicit the fact that dependency is associated with a variety of active and passive behaviors that are linked by an underlying motive to cultivate relationships with those people best able to offer protection and support over the long term. Table 6.2 presents a tentative set of empirically validated DPD symptom criteria, along with research evidence supporting the inclusion of each proposed criterion. These criteria include a modified version of the DSM-IV essential feature, which now emphasizes a view of the self as weak and ineffectual. Four DSM-IV symptoms (Symptoms 2, 5, 6, and 8) have been retained in their present form, but four symptoms are new and reflect (a) the dependent person's "helpless" self-concept (Symptom 1); (b) the relationship-facilitating features of dependency-related behavior (Symptoms 3 and 4); and (c) the dependent person's performance anxiety and fear of negative evaluation, which stem from a desire to maintain nurturant, supportive relationships with potential caregivers (Symptom 7). EPIDEMIOLOGY Beyond findings related to the symptom criteria themselves, considerable evidence has accumulated bearing on the distribution of DPD in different segments of the population. These epidemiological data provide impor4
The absence of data supporting DPD Symptom 7 is particularly problematic because this pattern of behavior is characteristic of insecure attachment (Shaver & Mikulincer, 2002; Sperling & Herman, 1991) and may represent a confound in the diagnostic criteria for DPD (Livesley, Schroeder, & Jackson, 1990).
96
CLINICAL APPLICATIONS
TABLE 6.2
Empirically Validated Diagnostic Criteria for Dependent Personality Disorder Criterion A view of the self as weak and helpless that motivates the individual to seek nurturant, protective relationships and engage in a variety of active and passive behaviors to maintain those relationships. This pattern begins by early adulthood and is present in a variety of contexts, as indicated by five (or more) of the following: 1. Perceives him- or herself as powerless and ineffectual and believes that others are comparatively powerful and potent 2. Has difficulty making everyday decisions without an excessive amount of advice and reassurance from others 3. Uses a variety of self-presentation strategies (e.g., ingratiation, supplication, exemplification, selfpromotion) to obtain and maintain nurturant, supportive relationships 4. Focuses his or her efforts on strengthening a relationship with the person most likely to be able to offer help and support over the long term 5. Goes to excessive lengths to please others, to the point of volunteering to do things that are unpleasant 6. Feels uncomfortable or helpless when alone because of exaggerated fears of being unable to care for him- or herself 7. Shows high levels of performance anxiety and fears of negative evaluation, especially by figures of authority 8. Is unrealistically preoccupied with fears of being left to take care of him- or herself
Supporting evidence Birtchnell & Kennard (1983); Bornstein et al. (1986); Bornstein et al. (1996); Simpson & Gangestad (1991); Sroufe et al. (1983)
Bornstein et al. (1996); Parker & Lipscombe (1980) Masling et al. (1968); Tribich & Messer (1974) Bornstein et al. (1996); Caspi et al. (1989); Emery & Lesher (1982)
Bornstein et al. (1987); Keinan & Hobfoll (1989); Overholser (1992); Rossman (1984) Agrawal & Rai (1988); Weiss (1969) Masling et al. (1981); Simpson & Gangestad (1991) Goldberg et al. (1989); Schlenker & Weigold (1990); Singh (1981) Berg (1974); Birtchnell (1988)
Note. Originally published as Table 2 in "Dependent Personality Disorder in the DSM-IVand Beyond," by R. F. Bornstein, 1997, Clinical Psychology: Science and Practice, 4, pp. 175-187. Copyright 1997 by Oxford University Press. Reprinted with permission.
tant information regarding DPD prevalence rates and subgroup (i.e., age, culture, and gender) differences. Prevalence Rates Studies offer mixed support for the DSM-IV and DSM-IV-TR contention that DPD is "among the most frequently reported Personality Disorders DIAGNOSIS
97
encountered in mental health settings" (APA, 2000, p. 723). On the one hand, it is clear that DPD is diagnosed quite frequently in psychiatric inpatients: Studies typically report DPD prevalence rates of between 15% and 25% in hospital and rehabilitation settings (Jackson et al., 1991; Oldham et al., 1995). On the other hand, the base rate of DPD in outpatients is not particularly high. In most studies it ranges from 0 to 10% (Klein, 2003; Mezzich, Fabrega, & Coffman, 1987; Poldrugo & Forti, 1988), considerably lower than the prevalence rates of several other PDs (e.g., borderline, histrionic, narcissistic, avoidant, obsessive-compulsive). The contrasting DPD prevalence rates in inpatients and outpatients suggest that dependent patients might present themselves in such a way as to receive inpatient hospitalization at unusually high rates. This would certainly be consistent with the dependent person's desire to be nurtured and cared for by figures of authority. Alternatively, increases in dependent behavior might follow—not precede—inpatient treatment, leading to false-positive DPD diagnoses in inpatients. As Booth (1986, p. 418) noted, the dependency that research has shown to be so widespread among [hospital] residents has been linked to the nature of institutional regimes. ... As residents grow more inured to residential life, so they become more dependent on the routine imposed on their lives.
Until studies test directly the "motivated hospitalization" and "institutional dependency" explanations for the high prevalence rates of DPD in psychiatric inpatients, it will be impossible to know which (if either) of these hypotheses accounts for extant findings in this area. Impact of Culture Findings regarding culture and age effects support the DSM-IV assertion that both variables influence the overt expression of dependency strivings and help determine the likelihood that a patient will receive a DPD diagnosis. Two cultural influences in particular are now well-established: The unusually high rates of overt dependent behavior displayed by people raised in Japan (Doi, 1973; Johnson, 1993) and India (Kaul, Mathur, & Murlidharan, 1982; Neki, 1976), both of which have been linked to traditional socialization practices in these countries. Although broad-based epidemiological base rate data are scanty, preliminary evidence suggests that dependency-related personality pathology is in fact diagnosed at higher rates in Japan than in North America and Western Europe (no studies have addressed this issue in India). Effects of Age Studies also show that the ways in which dependent behaviors are expressed vary with age. As noted in chapter 3, high levels of dependency in 98
CLINICAL APPLICATIONS
young children are associated with a propensity to seek physical contact and comfort from others, and with insecure attachment behaviors (e.g., clinging, hovering). In young adults dependency needs are commonly expressed via overt or implied requests for help, guidance, protection, and support (see Shaver & Kazan, 1993; Shaver & Mikulincer, 2002). In older persons dependency needs are often expressed indirectly (Bornstein & Languirand, 2001), sometimes displayed as somatic symptoms and physical complaints, and at other times manifested as cognitive confusion ("pseudodementia").
Gender Differences Women consistently receive DPD diagnoses at higher rates than men do (e.g., Alnaes & Torgerson, 1988; Jackson et al., 1991; Loranger, 1995; Oldham et al., 1995). Prior to 1994 there were eight published studies comparing DPD prevalence rates in men and women (see Bornstein, 1993, for a listing of these studies). In every one of these investigations, women received higher rates of DPD diagnoses than men. When data from these eight studies were combined using meta-analytic techniques, the overall base rate of DPD diagnoses (collapsing across type of setting in which data were collected) was 8% in men and 11% in women (Bornstein, 1993, 1997b). This difference is statistically significant (X2 [1, N = 5,965] = 13.53, p = .0005) and suggests that the base rate of DPD is about 40% higher in women than in men. Virtually identical findings have been obtained in studies using DSM-IV criteria to diagnose DPD, regardless of whether these data were collected via questionnaire or diagnostic interview (Baker, Capron, &. Azorlosa, 1996; Corbitt & Widiger, 1995; Melley, Oltmanns, & Turkheimer, 2002). It is not surprising that given these strong and consistent results, the passage on DPD gender differences was revised in the DSM—IV—TR. The DSM—IV asserted that "the sex ratio of this disorder is not significantly different than the sex ratio of females within the respective clinical setting" (APA, 1994, p. 667). It now reads: "In clinical settings, this disorder has been diagnosed more frequently in females, although some studies report similar prevalence rates among males and females" (APA, 2000, p. 723).
DIFFERENTIAL DIAGNOSIS AND COMORBIDITY Differential diagnosis information in the DSM-IV and DSM-IV-TR is included to make the clinician aware of alternative syndromes that may produce similar symptom patterns so that misdiagnosis will be minimized. In contrast, comorbidity information alerts the clinician to syndromes that are expected to "cluster" within patients at higher-than-expected rates, so that these co-occurring disorders will not be underdiagnosed. Comorbidity inforDIAGNOSIS
99
TABLE 6.3 DPD in the DSM-IV: Differential Diagnosis and Comorbidity Information DSMaxis
Differential diagnoses
1
Mood disorders Panic disorders Agoraphobia Borderline PD Histrionic PD Avoidant PD
II
Comorbid diagnoses Mood disorders Anxiety disorders Adjustment disorder Borderline PD Histrionic PD Avoidant PD
Note. DPD differential diagnosis and comorbidity information is identical in the DS/W-/l/(APA, 1994) and DSM-/V-7R (APA, 2000).
mation can also play an important role in treatment planning: For many disorders, including DPD, different interventions are needed depending upon which (if any) comorbid syndromes are present. The DSM-IV-TR DPD differential diagnosis and comorbidity information is summarized in Table 6.3. No extant studies have assessed the validity of this differential diagnosis information (which would involve tracking common clinician errors made in diagnosing dependent patients). However, numerous investigations have explored comorbidity patterns associated with DPD. Axis I Comorbidity As Table 6.3 shows, the DSM-IV lists three Axis I comorbid diagnoses for DPD. Evidence supports the continued inclusion of these three categories in future versions of the DSM and suggests that several other Axis I disorders also co-occur with DPD at higher-than-expected rates. Mood Disorders
Nietzel and Harris's (1990) meta-analysis of the dependency-depression relationship indicated that in a variety of participant groups, depression level is positively correlated with the intensity of expressed dependency needs. Dependency-depression correlations in Nietzel and Harris's meta-analysis ranged from .19 to .33, suggesting that dependency levels can account for a modest portion of the variance in depression. Since the early 1990s, there have been more than a dozen additional studies of the dependencydepression link, and these have yielded results consistent with those of Nietzel and Harris (1990). Some longitudinal studies indicate that dependency and DPD levels remain elevated even after depression remits (e.g., Frank, Kupfer, Jacob, & Jarrett, 1987; Overholser, 1990), although others report that lessening of depression is accompanied by a decrease in self-reported dependency (e.g., Joffe & Regan, 1988; Klein, Harding, Taylor, & Dickstein, 1988). Both laboratory (Smith, O'Keefe, & Jenkins, 1988) and field investigations (Hammen, Ellicott, & Gitlin, 1989) confirm that dependency represents a 100
CLINICAL APPLICATIONS
diathesis for depression, with interpersonal stress or relationship conflict leading to increased depression in dependent (but not nondependent) persons. Anxiety Disorders Dependency and DPD have been linked with a broad array of anxiety disorders, including agoraphobia (Chambless, Renneberg, Gracely, Goldstein, & Fydrich, 2000), social phobia (Sans & Avia, 1994), generalized anxiety disorder (Jansen, Arntz, Merckelbach, & Mersch, 1994), panic disorder (Stewart, Knize, & Pihl, 1992), and obsessive-compulsive disorder (Sciuto et al., 1991). In most studies observed dependency-anxiety links are moderate in magnitude. Unfortunately, the design of most of these investigations does not allow strong conclusions to be drawn regarding causal relationships between dependency and anxiety, and the only study to assess this issue directly obtained mixed results. Mavissakalian and Hamann (1987) found a significant decrease in DPD symptoms following successful agoraphobia treatment, though DPD levels remained elevated even in patients who responded most completely.5 Adjustment Disorder Although no studies have assessed the link between DPD and adjustment disorder symptoms, Piper, Ogrodniczuk, Joyce, McCallum, Weideman, and Azim (2001) found a modest—but statistically significant—link between dependency levels and intensity of grief following death of a loved one in psychiatric outpatients (r = .21). At the very least, dependency appears to be associated with a comparatively strong response to interpersonal loss (see also Allen, Home, & Trinder, 1996), a finding that dovetails with research showing that dependency can be a diathesis for depression. Substance Use Disorders DPD is associated with elevated rates of a broad array of substance use disorders, including alcohol (Poldrugo & Forti, 1988), opiates (Calsyn & Saxon, 1990), cocaine (Weiss, Mirin, Michael, & Sollogub, 1986), and polydrug abuse (Nace, Davis, & Gaspari, 1991). However, the dependencysubstance use relationship is neither strong nor specific. Several other PDs are at least as strongly linked with substance use as is dependent PD (e.g., antisocial, borderline, narcissistic), and these relationships hold regardless of whether PD levels are assessed via interview or questionnaire (O'Boyle, 1993). Some studies suggest that dependency levels increase as substance abuse continues (Blatt, Rounsaville, Eyre, & Wilber, 1984), a pattern that is particu!
These DPD-anxiety disorder links echo findings that show that high levels of dependency in children and adolescents are associated with school phobia (Berg, 1974; Berg & McGuire, 1974) and test anxiety (Chadha, 1983; Devito & Kubis, 1983). DIAGNOSIS
101
larly strong for alcohol abuse (Vaillant, 1980). Thus, dependency may be both a predictor and consequence of substance abuse.6 Eating Disorders
Several dozen studies have examined the relationships between dependency/DPD levels and risk for anorexia and bulimia (see Bornstein, 2001a, for a review). The results of these investigations are remarkably consistent. Findings indicate that (a) there is a positive relationship between dependency/DPD levels and intensity of eating disorder symptoms (r = .25); (b) this relationship is comparable in anorexic and bulimic participants; (c) although DPD symptom levels are elevated in eating-disordered participants, so are the symptoms of several other PDs (e.g., borderline, avoidant, obsessive-compulsive); and (d) when eating disorder symptoms remit, dependency levels decrease. Thus, there is a statistically significant link between DPD and eating disorders, but this relationship is modest in magnitude, nonspecific, and varies with eating disorder symptom levels.7 Somatization, Conversion, and Dissociative Disorders
Several investigations have reported significant links between DPD and symptoms of conversion disorder (Mersky & Trimble, 1979) and somatization disorder (Hayward & King, 1990). In the best-designed study of this issue, Hayward and King obtained DPD-somatization symptom level correlations (rs) of .47 in men and .12 in women from a nonclinical (community) sample. Findings regarding the DPD—dissociative disorder link have been less compelling: Although researchers report positive correlations between DPD levels and dissociative symptoms (including dissociative identity disorder symptoms) in a variety of participant groups, other PDs (e.g., avoidant, schizotypal, borderline) are more strongly associated with dissociative symptoms than is DPD (Dell, 1998; Fink, 1991; Waldo & Merritt, 2000). Axis II Comorbidity The DSM-IV and DSM-IV-TR list three Axis II PDs—borderline, histrionic, and avoidant—as comorbid with DPD (see Table 6.3), and studies confirm that these three PDs do in fact co-occur with DPD at higher-thanexpected rates (Becker, Grilo, Edell, & McGlashan, 2001; Blais, Hilsenroth, Castlebury, Fowler, & Baity, 2001; Grilo, Anez, & McGlashan, 2002). How6
Interestingly, dependency also predicts tobacco use (Fisher & Fisher, 1975; Kline & Storey, 1980), although in this domain it appears that susceptibility to peer pressure is key (Bornstein, 1993). Dependent adolescents have greater difficulty than do nondependent adolescents in resisting negative peer influences, and they are therefore more likely to experiment with—and become habitual users of—tobacco products. 'Similar results are obtained in studies of pathological overeating: Overweight individuals obtain significantly higher projective and self-report dependency scores than do matched normal-weight controls (Elfhag, Barkeling, Carlsson, & Rossner, 2003; Weiss & Masling, 1970), but dependency levels decrease when formerly obese persons return to average weight (Marshall & Neill, 1977).
102
CLINICAL APPLICATIONS
TABLE 6.4 Relationship of Dependent Personality Disorder (DPD) to other Personality Disorders (PDs): Implications for Comorbidity Correlation with DPD diagnosis-symptom rating
PD Category
Barber & Morse (1994)
Ekselius et al. (1994)
Meyer et al. (2001)
Sinha & Watson (2001)
Wise (1996)
Cluster A Paranoid Schizoid Schizotypal
.38 .20 .48
.57 .24 .50
Antisocial Borderline Histrionic Narcissistic
.40 .57 .47 .48
.24 .66 .37 .48
Avoidant Compulsive
.25 .35
.66 .59
.05 -.29 -.25
.46 .12 .42
.34 .39 .46
.13 .42
.35 .37
Cluster B -.09
.22 .29 .07
-.22 -.28
-.26 -.31
Cluster C .20
-.06
.62 .32
.77 .32
Note. Barber and Morse (1994) used interview and questionnaire PD measures in a mixed-sex sample of 116 outpatients. Ekselius et al. (1994) used an interview PD measure in 388 inpatients and outpatients and 176 community controls. Meyer et al. (2001) used consensus PD diagnoses derived from chart records in a mixed-sex sample of 149 inpatients and outpatients. Sinha and Watson (2001) used questionnaire PD measures in a mixed-sex sample of 293 college students. Wise (1996) used questionnaire PD measures in a mixed-sex sample of psychiatric inpatients.
ever, a detailed survey of the literature in this area indicates that DPD is actually associated with a much broader array of Axis II PDs than the DSMIV and DSM-IV-TR suggest. Table 6.4 summarizes DPD Axis II comorbidity information for five representative investigations conducted since the early 1990s (Barber & Morse, 1994; Ekselius, Lindstrom, vonKnorring, Bodlund, &Kullgren, 1994; Meyer, Pilkonis, Proietti, Heape, & Egan, 2001; Sinha Si Watson, 2001; Wise, 1996). These investigations used DSM-III-R or DSM-IV PD criteria, assessed PDs using a variety of techniques (e.g., interviews, consensus diagnoses, questionnaire responses), and collected data from inpatient, outpatient, community, and college student participants. Some studies operationalized comorbidity as diagnostic overlap or co-occurrence; others calculated correlations between DPD and other PD symptom ratings. Despite differences in measures, procedures, and participant samples, the studies in Table 6.4 produced reasonably consistent results. These may be summarized as follows: • DPD symptoms are positively related to the majority of DSM-IV PDs. Although some negative DPD-other PD correlations have been reported, the vast majority of correlations in Table 6.4 are positive and statistically significant. Most are in the moderate range, though there is considerable variability in this regard. DIAGNOSIS
103
These "generalized PD intercorrelation" findings dovetail with recent reviews of the PD literature, which suggest that the current PD diagnostic categories lack discriminant validity (Bornstein, 1998f) and cannot be reliably distinguished in realworld settings (Westen & Shedler, 1999). DPD symptoms are as strongly related to PDs from Clusters A and B as those from Cluster C. As the data in Table 6.4 show, DPD is as strongly linked with ostensibly unrelated PDs (e.g., paranoid, antisocial) as with PDs that are conceptually and dynamically similar (e.g., avoidant, histrionic). For example, Ekselius et al. (1994) found a DPD-paranoid PD correlation (r) of .57. Barber and Morse (1994) found a DPD-antisocial PD correlation of .40. Again, these data speak to the generalized, nonspecific overlap among DSM-IV PDs. DPD Axis II comorbidity relationships are consistent across assessment modalities and participant groups. Although Meyer, Pilkonis, Proietti, Heape, and Egan (2001) obtained somewhat weaker comorbidity links than the other investigations in Table 6.4, studies using different procedures and participant samples have yielded consistent results. Moreover, DSM—III—R and DSM—IV diagnostic criteria yield highly similar DPD comorbidity patterns, suggesting that revisions in these criteria have had minimal effects on PD comorbidity.
EFFECTIVE USE OF DIAGNOSTIC INFORMATION: A FRAMEWORK FOR THE PRACTITIONER The research reviewed in this chapter represents a framework for using diagnostic information to structure the assessment and treatment of dependent patients. Eight guidelines help shape this process: Go Beyond DPD Symptoms Given the limitations of existing DPD diagnostic criteria (Table 6.1) and the possibility that alternative criteria might capture more accurately the core elements of problematic dependency (Table 6.2), it is important that the clinician go beyond extant DPD symptoms when making diagnoses. Effective diagnosis of the dependent patient requires a two-pronged approach: While continuing to render formal DPD diagnoses using established symptom criteria, the clinician should also extend and supplement these diagnoses using additional dimensions that are more consistent with empirical studies in this area. It is best to think of DSM—IV DPD symptoms as the
104
CLINICAL APPLICATIONS
starting point—not the end point—of a complete understanding of dependency-related personality pathology.8 In supplementing the DSM-IV DPD criteria, two neglected features of dependency are key: the dependent person's "helpless" self-concept and her biased view of other people as exceedingly powerful and potent. As discussed in chapter 7, these are areas wherein psychological assessment data can complement and enrich diagnostic information by making explicit the patient's internalized representations of self and others. Until the DPD symptoms are revised to reflect the centrality of these core issues, it will be necessary to assess these domains using other (nondiagnostic) instruments. Take a Broad Perspective In addition to going beyond the existing DPD criteria, effective use of diagnostic information requires that the clinician recognize and challenge four assumptions that may—if unacknowledged—undermine diagnosis and treatment planning. To varying degrees, we have touched upon these assumptions in this and earlier chapters. Here they are, in more formal (and memorable) terms: • Do not equate dependency with passivity. As discussed in chapter 4, dependent persons can be quite active—even aggressive—in strengthening ties to potential caregivers. Simply because a patient is dependent, do not assume he will be passive. Simply because a patient is passive, do not assume he is dependent. • Do not equate dependency with pathology. Excessive dependency is often associated with difficulties in social and occupational functioning, but as the research reviewed in chapters 4 and 5 indicated, dependency is associated with certain strengths as well. These strengths form the core of effective treatment planning. • Do not assume that the absence of a DPD diagnosis means dependency is irrelevant to treatment. Two issues are germane here. First, the DSM threshold model imposes an arbitrary cutoff distinguishing pathological from "normal" functioning, but patients who experience four (rather than the required five) symptoms do not differ appreciably from patients who receive a DPD diagnosis (see Livesley & Jang, 2000, for a discussion of this issue). Second, as discussed in chapter 2, underlying dependency 8
The DSM-IV DPD symptom criteria, though flawed, map nicely onto certain unhealthy dependency subtypes. In particular, patients diagnosed with DPD using current criteria show prominent features of Pincus and Gunman's (1995) exploitable dependents and Bornstein and Languirand's (2003) helpless dependents.
DIAGNOSIS
105
needs are often expressed indirectly and may not be captured by the DPD criteria, which tap self-attributed (but not implicit) dependency strivings. • Do not assume that a dependent selj''presentation always reflects excessive dependency. Studies show that some people choose to present themselves as being dependent even when they are not (see Figure 2.1). Just as the absence of self-reported dependency does not always indicate a lack of underlying dependency urges, the presence of self-reported dependency does not necessarily reflect strong dependency needs. Consider the Impact of Diagnostic Method In Chapter 2, I reviewed construct validity data for widely used DPD questionnaires and interviews. No measure is perfect, and each has its limitations. Ideally, diagnostic information should be collected using multiple measures, with convergences and divergences among measures scrutinized by the diagnostician. Practically, this is rarely possible. At the very least, the clinician should interpret diagnostic data—including archival diagnostic data— in the context of the instrument used to collect them. At present, evidence supporting the validity and reliability of the SCIDII (Spitzer et al., 1990) and IPDE (Loranger, 1995) is quite strong, and either interview can yield useful information regarding DPD, including dimensional DPD scores. Construct validity data for the SIDP-R (Pfohl, Blum, Zimmerman, & Stangl, 1989) are also strong, but less plentiful than those available for the SCID-II and IPDE. As data continue to accumulate, the usefulness of the SIDP-R in clinical settings should increase. Both widely-used questionnaire measures of DPD are also well validated, and either could yield useful diagnostic information to supplement or replace interview-derived data. The MCMI-III (Millon, Millon, & Davis, 1994) taps a broader range of pathologies than does the PDQ-IV (Davison, Morven, & Taylor, 2001) and has a more comprehensive and refined set of moderator/validity indices. Be Aware of Culture, Gender, and Age Effects Diagnostic criteria for DPD—and other PDs as well—are limited in the degree to which they reflect cutting-edge findings regarding gender, culture, and age, although recent revisions of the DSM have brought improvements in each of these domains. Additional changes are needed, and until these are made, the clinician should take into account • The effects of cultural background on experienced and reported dependency. As noted in chapter 1, individuals raised in sociocentric cultures have very different attitudes regarding depen106
CLINICAL APPLICATIONS
dency than do individuals raised in individualistic cultures (see also Castillo, 1997, for a detailed discussion of this issue). • Gender differences in willingness to acknowledge dependency urges and strivings. As discussed in chapter 2, men are less willing than women to discuss dependency-related thoughts and feelings, even in the context of a therapeutic relationship (in fact, many men have difficulty acknowledging such thoughts and feelings privately as well). • Age-related changes in dependence-related symptoms and behaviors. As the research reviewed in chapter 3 indicated, dependent strivings are expressed very differently during different phases of life. Because the DSM-IV DPD criteria emphasize dependency-related behaviors characteristic of early and middle adulthood, making accurate DPD diagnoses in adolescents and older adults can be difficult (see Table 3.2 for a summary of findings in this area). Use Comorbidity Data to Disentangle Dependency-Related Dynamics DPD shows high rates of comorbidity with other psychological disorders, but these comorbidity patterns have very different meanings on Axes I and II. On Axis I, DPD is associated with higher-than-expected rates of depression, anxiety disorders, adjustment disorder, substance use disorders, eating disorders, and somatization—dissociative—conversion disorders. These comorbidity patterns can help the clinician understand how a particular patient experiences and expresses problematic dependency (e.g., through selfmedication or pathological eating). Comorbidity information, when scrutinized carefully and supplemented with psychological test data, may provide useful clues regarding a patient's defense and coping style. As the studies summarized in Table 6.4 demonstrated, DPD shows substantial overlap with the majority of Axis II disorders. Without question, these DPD—other PD links are in part a consequence of differential diagnosis limitations in DSM-IV Axis II, but they also reflect the fact that people who have longstanding personality difficulties in one area tend to have difficulties in other areas as well. Personality pathology is less categorical than the DSM—IV implies, and Axis II comorbidity information can provide considerable insight regarding a patient's personality dynamics (e.g., the dependent patient with borderline PD has a very different personality structure than the dependent patient with avoidant PD). Distinguish Primary From Secondary Dependency Some clinical researchers have conceptualized dependency as a risk factor for Axis I pathology (e.g., Blatt & Zuroff, 1992); others have developed frameDIAGNOSIS
107
works to explain how dependency may result from, rather than lead to, illness (Saviola, 1981). Research suggests that both perspectives are correct. For the majority of Axis I disorders that are comorbid with DPD (e.g., substance abuse, depression, anxiety disorders, eating disorders), dependency represents both diathesis and consequence. These causal links differ for different patients, however: Some patients show longstanding problematic dependency that precedes Axis I onset, others show increased dependency as Axis I symptoms persist, and still others show a combination of both processes. At the level of the individual patient, disentangling causal relationships between DPD and Axis I pathology requires careful documentation of the onset and course of different symptoms and syndromes. For each Axis I disorder that a dependent patient presents, it is important to determine whether dependency was primary (i.e., preceded Axis I symptom onset) or secondary (i.e., followed Axis I onset). In making these distinctions, the clinician should keep in mind that even within a particular patient, dependency may be primary to some Axis I syndromes and secondary to others. The same is true for Axis III. The DSM-IV-TR notes that DPD is often associated with general medical conditions, although it does not specify which conditions are most strongly linked with dependency. Paralleling the patterns obtained on Axis I, research in this area suggests that dependency represents a risk factor for physical illness (Bornstein, 1998e), as well as a consequence of illness (Hoare, 1984) and its treatment (Booth, 1986). Just as the clinician must distinguish primary from secondary dependency in the context of Axis I, she must distinguish primary from secondary dependency in patients who qualify for one or more diagnoses on Axis III. Pay Particular Attention to Axis IV Because dependency-related stressors (e.g., interpersonal conflict, relationship disruption) play a role in the onset of depression (Blatt & Zuroff, 1992), eating disorders (Halmi, 1997), and physical disorders (Bornstein, 1995b), careful diagnosis on Axis IV is crucial. Axis IV information not only helps the clinician determine whether a diathesis-stress process may have been involved in the onset of one or more existing syndromes, but can also help the clinician predict whether stress-related disorders are likely to occur in the future. Any DSM-IV-TR Axis IV category could potentially be relevant in this context, but given the dependent person's concern with obtaining and maintaining nurturant, supportive relationships, several categories will be particularly important. These are • Problems with primary support group • Problems related to the social environment
• Educational problems (if they involve discord with teachers or classmates) 108
CLINICAL APPLICATIONS
• Occupational problems (if they involve difficulties with supervisor or co-workers) • Housing problems (if they involve conflicts with neighbors or landlord) • Other psychosocial and environmental problems (if they involve conflicts with nonfamily caregivers such as a counselor, social worker, or physician) Integrate Diagnostic and Assessment Information Because diagnoses are derived primarily from patient self-reports, it is important to recognize the limitations of this information. Patients bring to treatment a broad array of agendas and self-presentation styles, and they are not always forthright in disclosing symptom-related information. Obtaining accurate Axis II data can be particularly challenging, because PDs by definition involve limited insight and systematic distortions in the patient's perception of self, others, and self—other interactions. Given these limitations, the clinician must take care to place diagnostic information in context and appreciate the unique contributions of different data sources to a complete understanding of the patient's inter- and intrapersonal functioning (e.g., self-reports, reports from knowledgeable informants, behavioral observations, project!ve-test responses). Depending upon the particular circumstances of each patient, various combinations of these data sources may be particularly useful. The clinician's task is to integrate this information mindfully, within the context of an appropriate conceptual framework, and with a working knowledge of the strengths and limitations of each data source. These are the issues taken up in chapter 7.
DIAGNOSIS
109
7 ASSESSMENT
Diagnosis provides a shorthand summary of a patient's symptoms, along with contextual information (e.g., comorbid pathology, salient stressors) that may have implications for clinical prediction and treatment planning. Assessment goes deeper and helps disentangle the complex array of dispositional and situational factors that interact to determine a patient's subjective experiences, core beliefs, coping strategies, and behavior patterns. Put another way, diagnosis is key to understanding pathology; assessment is key to understanding the person with the pathology. In this chapter, I discuss psychological assessment of the dependent patient. I begin by exploring differences between testing and assessment, and how these two tasks relate to each other in the clinical setting. I then review assessment instruments the practitioner can use to gain a deeper understanding of the dependent patient's intrapsychic dynamics and interpersonal style. Next, I discuss strategies for classifying patients into the dependency subtypes introduced in chapter 5 and procedures for assessing the impact of patients' dependency on their environment and social milieu. Finally, I outline an integrated framework for effective use of assessment information with dependent patients.
Ill
TESTING VERSUS ASSESSMENT Psychologists often use the terms testing and assessment interchangeably, but in fact they mean very different things. Handler and Meyer (1998, pp. 4-5) provided an excellent summary of the conceptual and practical differences between psychological testing and psychological assessment. They wrote: Testing is a relatively straightforward process wherein a particular test is administered to obtain a particular score or two. Subsequently, a descriptive meaning can be applied to the score based on normative, nomothetic findings.... Psychological assessment, however, is a quite different enterprise. The focus here is not on obtaining a single score, or even a series of test scores. Rather, the focus is on taking a variety of testderived pieces of information, obtained from multiple methods of assessment, and placing these data in the context of historical information, referral information, and behavioral observations in order to generate a cohesive and comprehensive understanding of the person being evaluated.1
Handler and Meyer's (1998) insightful analysis has been echoed by numerous clinicians and clinical researchers (e.g., Groth-Marnat, 1999; Grove, Zald, Lebow, Snitz, & Nelson, 2000). Gates (1999, p. 637) put it well when he noted that in the realm of psychological assessment, "art rests on science." Psychological testing requires precision, objectivity, and the kind of scientific detachment that facilitates accurate data-gathering. Psychological assessment involves integration, synthesis, and clarification of ambiguous—even conflicting—evidence. These are skills that cannot be taught directly, but must be built up over time, through experience. MEASURING IMPLICIT AND SELF-ATTRIBUTED DEPENDENCY NEEDS Assessment of the dependent patient begins with measurement of that patient's implicit and self-attributed dependency needs. As noted in chapter 2, implicit dependency needs are typically assessed via projective tests; selfattributed dependency needs are assessed via self-report (questionnaire) measures. Normative data from clinical samples are required for the practitioner to determine the overall level (i.e., intensity) of each patient's implicit and 'Elsewhere, Handler and Meyer (1998, p. 5) provided a useful comparison of the testing-assessment relationship in psychology and medicine, noting that The medical counterpart to psychological testing is found when technicians or medical personnel obtain scores on such instruments as a blood pressure gauge or a thermometer. . . . However, the medical counterpart to psychological assessment is when a physician takes the information from these various tests and places them in the context of a patient's symptomatic presentation and history in order to accurately understand the full scope of his or her condition.
112
CLINICAL APPLICATIONS
TABLE 7.1 Advantages and Disadvantages of Self-Report and Projective Dependency Tests Type of test Self-report
Projective
Advantages
Disadvantages
Easy to administer and score
Easily faked; susceptible to selfreport and self-presentation effects Often have subscales tapping Some self-report tests have subtle gender, age, and different subtypes of facets of dependency cultural biases Can help stimulate therapeutic May increase defensiveness, dialogue regarding avoidance, and resistance in dependency-related thoughts some patients and experiences Low face validity; resistant to Time-consuming and laborself-report and selfintensive; require training and presentation effects experience to administer, score, and interpret properly Can assess implicitMost projective tests do not unconscious-latent aspects of yield information regarding dependency the patient is subtypes or facets of unable or unwilling to dependency acknowledge directly
Note. Additional information regarding these measures, along with a review of widely used self-report and projeotive dependency tests, is provided in chapter 2.
self-attributed dependency strivings. Moreover, in evaluating self-attributed dependency needs, it is important to use gender-specific norms, because virtually every self-report dependency test yields significant gender differences, with women scoring higher than men (see Table 1.2 for a summary of findings in this area). Table 7.1 summarizes the advantages and limitations of self-report and projective measures of dependency (see also Bornstein, 2002, for a detailed discussion of this issue). Although the clinician should be aware of the contrasting strengths and weaknesses of self-report and projective tests, scrutiny of Table 7.1 confirms that neither measure is inherently superior. Each type of test contributes unique information regarding a patient's functioning and, when data from these tests are combined, the clinician can learn a great deal about the patient's private experience and public self-presentation. The key to effective assessment of the dependent patient lies in integrating the results obtained with these different measurement tools. The first step in this process involves choosing which tests to use. Selecting Appropriate Dependency Tests Masling et al.'s (1967) ROD scale is the most well-validated projective dependency test available, and in most situations this will be the measure of ASSESSMENT
113
TABLE 7.2 Key Features of Widely Used Self-Report Dependency Tests
Test
Conceptual/ theoretical basis
Trait/ DPD
#of items
Dy
Rationally derived
Trait
57
IDI
Rational/ factor Trait analysis
48
DEQ
Psychodynamic Trait
66
PSI
Cognitive
Trait
48
MCMI DS/W-based
DPD
175
DSM-based
DPD
163
PDQ
DPSS Rational/DS/W- Trait based
20
3VDI Circumplex/ Trait dependency subtypes RPT Rationally Trait derived
27
30
Item format
Distinguishing Features
Can be used in archival analyses and derived from existing MMPI protocols Extensive clinical norms; Likert subscales tap different dependency-related traits Likert Psychodynamic basis; clinical and community norms; diathesis-stress findings Likert Cognitive basis; diathesisstress findings T/F Extensive clinical norms; multiple pathology scales T/F Extensive clinical norms; multiple pathology scales Multiple Taps context-specific dependent behavior and choice generalized dependency Multiple scores tapping Likert different dependency subtypes Likert Separate scores for unhealthy dependency and healthy dependency T/F
Note. In column 3, tests labeled Trait assess trait dependency; tests labeled DPD assess dependent personality disorder symptoms. Complete scale names and construct validity findings are in chapter 2.
choice for clinicians who seek to assess a patient's underlying dependency needs (see Bornstein, 1996b, 1999, 2002). The choice is more complicated for self-report dependency tests. The practitioner's decision in this domain should be based in part on the available criterion-referenced validity data (discussed in detail in chap. 2 and summarized in Table 2.3), but these data are not yet available for all self-report dependency scales. Other considerations, such as test length, item format, ease of administration, and theoretical basis of a test will also be relevant in making this decision. To assist the clinician in choosing among self-report dependency tests, Table 7.2 summarizes some key features of the most widely available measures. Certain test characteristics (e.g., a cognitive orientation) might represent a positive attribute for certain clinicians but a negative attribute for others. In addition, certain test features (e.g., assessment of trait dependency versus DPD) can help determine which measure is optimal in a given situation or setting. As clinical norms for various scales are expanded and refined, the clinician may choose to supplement or replace a previously used self114
CLINICAL APPLICATIONS
report measure with one that allows for more precise distinctions and accurate predictions. The test battery should evolve over time in response to accumulating evidence and the emergence of psychometrically stronger instruments. Integrating Implicit and Self-Attributed Dependency Scores Figure 2.1 summarized a four-cell model for analyzing continuities and discontinuities between a patient's implicit and self-attributed need states. By contrasting a given patient's self-report and projective dependency scores, the practitioner can determine whether that patient is best described as having low dependency, high dependency, unacknowledged dependency, or a dependent self'presentation. Beyond this initial classification, self-report and projective test data can be combined to draw tentative conclusions in four domains: • Describing personality. Different patterns of implicit and selfattributed dependency scores are associated with different personality styles. For example, while high dependency patients tend to be overtly dependent, patients with unacknowledged dependency are more likely to show histrionic, borderline, or narcissistic features. Detailed discussions of personality styles associated with different implicit-self-attributed test score patterns are provided by Bornstein (1998b, 1998d). • Delineating pathology. Whereas high dependency patients are at increased risk for depression and anxiety disorders (Overholser, 1990; Stewart, Knize, & Pihl, 1992), patients with unacknowledged dependency are more likely to develop eating disorders or somatization symptoms (Almgren, Nordgren, & Skantze, 1978; Bornstein & Greenberg, 1991). Patients with a dependent selfpresentation are at risk for a broad array of psychopathologies, but they also tend to exaggerate physical and psychological symptoms and receive a higher-than-expected number of falsepositive diagnoses (O'Neill & Bornstein, 1990). • Documenting strength. As discussed in chapters 4 and 5, dependent personality traits can be adaptive in certain contexts and situations. A key ingredient in adaptive dependency is insight (Blatt, Zohar, Quinlan, Zuroff, & Mongrain, 1995; Gabriel & Gardner, 1999), and in this respect the patient who shows high dependency typically has stronger psychological resources (and better psychotherapy potential) than the patient with unacknowledged dependency needs. • Deconstructing defense. Different patterns of implicit and selfattributed dependency scores are associated with different defense styles. Patients with unacknowledged dependency needs score ASSESSMENT
115
high on measures of repression and denial, whereas patients with a dependent self-presentation displace anger and other negative emotions onto the self (Herman & McCann, 1995; Bornstein, Greenberg, Leone, & Galley, 1990; Vaillant, 1994). These patterns vary considerably from person to person, however, and a complete picture of a patient's defense and coping style requires that these constructs be assessed directly using established defense style measures. I discuss this issue in the following section.
DEPENDENCY IN CONTEXT: PERCEPTIONS, DEFENSES, AND SOCIAL SUPPORT Assessment of the dependent patient begins with measurement of implicit and self-attributed dependency needs, but it must not end there. The next step involves exploring three domains of functioning that provide critical information regarding the patient's perceptions of self and others, coping pattern, and social milieu. Perceptions of Self and Others As discussed in chapter 1, studies confirm that a key element in a dependent personality orientation is a perception of oneself as vulnerable and weak and of others as comparatively powerful and potent (Bornstein, 1996a; Huprich, 2001; Pincus & Wilson, 2001). No extant measure of dependency includes subscales that quantify these dimensions, so additional testing is needed to assess the patient's perceptions of self and other people. Blatt, Chevron, Quinlan, and Wein's (1981) index of Qualitative and Structural Dimensions of Object Representations (QSDOR) is the best measure in this area. The QSDOR asks the patient to provide open-ended descriptions of significant figures; the figures to be described are determined by the therapist. For each target figure, the patient is given a blank sheet of paper, at the top of which is a simple instruction: DESCRIBE . Any type of person, real or imagined, can be described in this way (see Blatt et al., 1981; Blatt, Chevron, Quinlan, Schaffer, & Wein, 1988), but for the dependent patient, three figures are particularly useful: self, mother, and father. These descriptions vary considerably, from terse lists of physical characteristics to lengthy narratives with rich descriptions of the individual's internal life and personal history.2 2 The QSDOR has been updated substantially since it first appeared in 1981 (see, e.g., Blatt, Bers, & Schaffer, 1993), as new research findings and improved scoring procedures have been integrated into the scale. The practitioner who intends to use this measure should become familiar with the basic elements of the QSDOR and with recent research on the construct validity of the measure (e.g., Auerbach & Blatt, 1996; Blatt et al., 1996).
J 16
CLINICAL APPLICATIONS
QSDOR-derived descriptions of significant figures are scored in three areas: • Personal qualities. Twelve trait dimensions are included in this content domain, and each is scored on a 7-point scale anchored with the terms None (1) and Very (7). Representative QSDOR trait dimensions include affectionate, ambitious, benevolent, judgmental, and punitive. • Ambivalence. The degree of ambivalence (i.e., internal contradiction and inconsistency) in the description is rated on a 3point scale anchored by the terms None (1) and Very (3). • Conceptual kvel. The conceptual level of each description is rated on a 9-point scale, with primitive, concrete descriptions receiving low ratings, and sophisticated descriptions that reflect an awareness of the person's needs, emotions, and changing internal states receiving the highest ratings. Based on the work of Piaget (1954), and Werner and Kaplan (1963), conceptual-level scores in the QSDOR range from SensorimotorPreoperational (1), through External Iconic (5), to Conceptual (9). Information regarding QSDOR scoring and interpretation—along with an array of scored practice descriptions—is provided by Blatt, Bers, and Schaffer (1993) and Blatt et al. (1981, 1988). The construct validity of the measure is well established, and studies show that both content (i.e., trait) and structural dimensions of descriptions (i.e., ambivalence, conceptual level) predict important features of behavior and adjustment in a variety of participant groups (Auerbach & Blatt, 1996; Blatt & Ford, 1994; Blatt, Stayner, Auerbach, & Behrends, 1996). In addition to providing unique information regarding the patient's mental representations of self and significant others, the QSDOR can be a useful bridge between assessment and treatment. These open-ended descriptions often act as cues (or "triggers") that help patients gain insight into past experiences, current relationships, unacknowledged feelings, and implicit attitudes and beliefs. These insights can stimulate constructive dialogue and shape the initial course of therapy.3 Coping and Defense Style In chapter 5, I noted that dependency is often associated with a "turning against self" defense style wherein anger and aggression are bottled up and channeled inward. Within this generalized introjective stance, however, 3
Although perceptions of self and parents are particularly informative, the clinician should not feel constrained with respect to the range of figures evaluated. The QSDOR is designed to quantify structural and content features of all types of mental representations, including actual persons (e.g., siblings, spouse, supervisor, therapist) and imagined/fantasy figures (e.g., the perfect therapist, an ideal romantic partner).
ASSESSMENT
] 17
there is considerable variation in the defenses used by different dependent patients. Whereas lower-functioning dependent patients tend to rely on primitive defenses like projection and denial, higher-functioning dependent patients use more mature defenses and coping strategies like sublimation and humor (see Berman & McCann, 1995; Lingiardi et al., 1999). Two measures are particularly useful in examining the dependent patient's defense style:4 the Defense Style Questionnaire and the Defense Mechanisms Inventory. The Defense Style Questionnaire (DSQ; Bond, Gardner, Christian, &Sigal, 1983) The DSQ includes 88 self-statements describing characteristic thoughts, emotions, and behavior patterns. The respondent is asked to rate each statement on a 9-point scale anchored with the terms Strongly Disagree (1) and Strongly Agree (9). Factor analyses revealed that DSQ items form four scales, each tapping a different defense configuration: (a) imrnature-maladaptive (e.g., repression, projection); (b) image-distorting (e.g., splitting, primitive idealization); (c) self-sacrificing (e.g., reaction formation, denial); and (d) mature (e.g., sublimation, humor). Construct validity data for the DSQ are strong, and studies confirm that DSQ defense scores predict psychological adjustment, life satisfaction, physical health, and psychotherapy potential (Bond et al., 1983; Bullitt & Farber, 2002; Johnson, Bornstein, & Krukonis, 1992). A primary advantage of the DSQ is its user-friendly format and ease of administration; a primary disadvantage of the test is its high face validity and susceptibility to self-presentation effects (although the DSQ does include a Lie Scale to assess these biases). The Defense Mechanisms Inventory (DMI; Gleser & Ihilevich, 1969) The DMI consists of 10 brief vignettes describing interpersonal conflict situations. Four questions follow each vignette, with five responses provided for each question. Respondents select their most likely and least likely response; these are then tallied across vignettes to yield five defense style scores: turning against object (TAO), projection (PRO), f>rinci£>afeation (PRN), turning agaimt self (TAS), and reversal (REV). Evidence supporting the construct validity of the DMI is good, and DMI defense scores predict a broad array of psychological variables. The DMI is also useful for tracking changes in defense style over the course of therapy (Ihilevich & Gleser, 1986,1991). However, the test has three limitations: (a) its format can be confusing for patients who are not psychologically minded; (b) a relatively narrow range of situations is assessed (e.g., no situations involving sexuality are included); 4
Beyond these self-report tests, an array of Rorschach-derived defense scales can be useful in assessing dependent patients (see Cramer, 2002; Lerner, 1991; and Levit, 1991, for reviews).
J 18
CLINICAL APPLICATIONS
and (c) like the DSQ, the DMI is susceptible to self-presentation effects (although, unlike the DSQ, the DMI does not include a validity scale to assess self-report confounds). Social Skills and Social Support As discussed in chapter 6, high levels of interpersonal conflict are associated with increased risk for depression in dependent persons (Hammen, Ellicott, & Gitlin, 1989; Zuroff & Mongrain, 1987). Other studies (described in chap. 4 and summarized in Table 4.2) show that persistent interpersonal stress places the dependent person at increased risk for physical illness (Allen, Home, &Trinder, 1996; Bornstein, 1995b). The social—interpersonal aspect of experienced stress appears to be central in both pathology domains: Other types of stressors (e.g., achievement-related, financial) do not increase the dependent person's risk for depression and physical illness to nearly the same degree. Two variables have been shown to buffer the negative effects of interpersonal stress in dependent individuals: social skills and social support. Good social skills help the dependent person minimize interpersonal conflict and avoid relationship disruption (Bornstein, 1993; Masling, 1986). Good social support helps the dependent person cope more effectively with interpersonal conflict when it cannot be avoided (Priel & Besser, 2000; Priel & Shahar, 2000). Thus, assessment of these variables is useful in placing a patient's dependency into an appropriate social context and documenting some key dependency-related strengths. Although there are psychometrically sound measures of social skills for use with psychiatric inpatients and outpatients (e.g., Riggio's, 1986, Social Skills Inventory), studies suggest that valid and reliable information in this domain can be obtained informally by experienced practitioners during the initial treatment sessions (Huffcut, Conway, Roth, & Stone, 2001). Accurate measurement of social support requires more formal, standardized testing. Two measures are useful: the Perceived Social Support Scale and the Social Network Scale. The Perceived Social Support Scale (PSS; Procidano & Heller, 1983)
The PSS includes 40 self-statements tapping various aspects of social support and perceptions of close relationships. Respondents circle Yes, No, or Don't Know for each PSS item, and responses are tallied to yield two 20-item scores: (a) PSS-Fr (Perceived Social Support from Friends) and (b) PSS-Fa (Perceived Social Support from Family). Construct validity data confirm that PSS scores predict overall levels of psychopathology, anxiety, and adjustment, with lower levels of support associated with less adequate functioning in each domain (Procidano, 1992; Procidano & Heller, 1983). Like many questionnaires, the PSS has high face validity, but its modest length and ASSESSMENT
119
user-friendly format make it very useful in clinical settings. An additional advantage of the PSS is that it allows the clinician to determine the individual's primary locus of support by yielding separate friend and family scores. Studies show that PSS-Fr and PSS-Fa scores are positively—but modestly—correlated (rs ranged from .21 to .35 in initial validation samples), suggesting that perceived social support may be best assessed in domainspecific rather than global terms (Procidano, 1992). The Social Network Scale (SNS; Kaplan, 1975) The SNS assesses the density and quality of the respondent's social network using a two-stage process. The first section of the scale includes 10 questions focusing on breadth of the social network. Respondents are then asked to list four persons with whom they have discussed personal problems during the past 6 months and rate each person on seven dimensions: (a) frequency of contact; (b) accessibility; (c) time spent discussing the respondent's problems; (d) time spent discussing the other person's problems; (e) range of topics discussed; (f) importance of the relationship; and (g) helpfulness of the other person. Ratings of the person's social network as a whole can be derived by summing the individual scores, but the SNS also allows for derivation of separate person-specific support ratings. The construct validity of the SNS is well established in an array of participant groups (Lin, Dean, & Ensel, 1986), and studies suggest that it may be particularly useful for assessing social support in dependent persons (Overholser, 1996). The primary advantage of the SNS is the level of specificity provided by the range of ratings obtained; the primary disadvantages of the scale are its complexity and length.
ASSESSING DEPENDENCY SUBTYPES Thus far I have focused on quantifying implicit and self-attributed dependency needs and supplementing these data with information regarding the patient's perceptions of self and others, coping and defense style, and social support system. Assessment can also be used to determine whether a patient fits one or more of the dependency subtypes discussed in chapter 5. Different assessment strategies are needed, depending upon whether the clinician seeks to classify the patient within the context of Pincus and Gurtman's (1995) three-vector model or Bornstein and Languirand's (2003) fourpattern model. Exploitable, Submissive, and Love Dependency Pincus and Gurtman's (1995) 3VDI is the measure of choice for determining the degree to which a patient shows features of exploitable, subtnis120
CLINICAL APPLICATIONS
sive, or love dependency. The 3VDI consists of 27 self-statements (9 per category), reflecting a variety of dependency-related feelings and experiences. The respondent rates each statement on a 6-point scale anchored with the terms Not at all like me (1), and Very much like me (6). Representative items from the three 3VDI subscales include "I find it difficult to say 'no' to the requests of friends" (exploitable), "I am certainly lacking in self-confidence" (submissive), and "I find it difficult to be separated from the people I love" (love). Construct validity data for the 3VDI (described in chaps. 2 and 5 and summarized in Table 5.4) are strong, and the scale's modest length and ease of use enhance its effectiveness in clinical settings. Helpless, Hostile, Hidden, and Conflicted Dependency Unlike Pincus and Gurtman's (1995) three-vector model, no single measure is available to classify patients into the categories described in Bornstein and Languirand's (2003) four-pattern model. However, the clinician can use a combination of life history information, behavioral observation, and defense style data to make classification decisions in this domain. Table 7.3 summarizes the prototypic patterns of life history, interview, and defense style data associated with helpless, hostile, hidden, and conflicted dependency. Few patients will show complete consistency across all three of these areas, because few patients represent "pure" dependency subtypes. However, to the degree that converging results are obtained across these three domains, strong conclusions may be drawn regarding a patient's characteristic dependency pattern. To the degree that inconsistent or conflicting results are obtained across these areas, this suggests that a patient may represent a "mixed" dependency subtype, with features of more than one pattern (see Bornstein & Languirand, 2003).
ASSESSING DEPENDENCY'S IMPACT The final domain of assessment involves exploring dependency's impact on the patient's social, family, and work relationships. As Lanyon and Goodstein (1997) noted, this information—like all biographical and lifehistory information—should come from multiple sources whenever possible. Thus, it is best when the patient's self-reports can be supplemented by archival records and the reports of knowledgeable informants. • Self'reports. When asking patients to provide information regarding the impact of their dependency on others, it is important to move beyond general impressions and global descriptions, using behavioral anchoring to increase the validity of these self-reports (Arvey, 1992; Stokes & Cooper, 2001). This reASSESSMENT
121
TABLE 7.3
Classifying Patients Using Bornstein and Languirand's Four-Pattern Model Dependency pattern Helpless
Hostile
Hidden
Conflicted
Defense style Life history data
Interview behavior
DSQ
Domination by others; Overtly dependent; SS-ID exploitation/abuse; pattern of help- and passivity and reassurance-seeking; compliance in most or self-deprecation all relationships Acting out to preclude Incipient breakdown IM abandonment; threats/gestures; manipulative/selfescalation when destructive behavior challenged Indirect expressions of Lack of insight; SS-IM dependency; lack of defensive regarding insight regarding dependency when relationship challenged; difficulties; somatizing avoidance of difficult issues/topics tendencies Inconsistent behavior; ID Pattern of intense, stormy relationships; disparity between fluctuation between content of speech idealization and and emotional devaluation; little response; inability to use insight change in selfdefeating behavior productively over time
DMI
TAS
PRO
PRNREV
TASTAO
Note. DSQ = Defense Style Questionnaire (Bond et al., 1983); DMI = Defense Mechanicsms Inventory (Gleser & Ihilevich, 1969). For DSQ, SS = self-sacrificing, ID = image-distorting, IM = immaturemaladaptive. For DMI, TAS = turning against self; PRO = projection; PRN = principalization; REV = reversal; TAO = turning against object. Complete assessment requires use of only one defense style measure; information for both measures is provided so the practitioner can use either scale.
quires that the clinician break down dependency into its narrower, more easily identified components. Rather than asking a patient whether his or her "dependency" or "dependent behavior" has caused problems at work, the clinician should inquire about specific dependency-related responses—help-seeking, reassurance-seeking, anxiety regarding evaluation, and so on. Questions should be phrased in domain- and context-specific language, to obtain information about the impact of the patient's dependency in different areas of life (e.g., work, family, friendship). When a patient describes dependency-related behavior in generalized terms, the clinician can use follow-up questions to focus the patient's description and increase its precision. The clinician should ask for concrete examples of dependency-related events and encourage the patient to describe specific behaviors exhibited by all the key figures. It is some-
122
CLINICAL APPLICATIONS
times helpful to use a clipboard or flip-chart to list these behaviors in sequence, so the interaction can be examined in detail (and preserved for future reference). Archival data. Archival data (sometimes called biodata) take many forms, including past therapy records, school or summer camp reports, and personal documents (e.g., diary entries, letters, email correspondence). The key to effective use of archival data is to explore its significance with the patient, seeking input and feedback at each step (Lerner & Tubman, 1989; Stokes, Mumford, & Owens, 1994). Thus, the clinician should ask the patient to describe the circumstances surrounding a written or email exchange—the antecedents and consequences of the interaction. Similarly, school and camp records are most useful when used as springboards for discussion of dependencyrelated issues and life events (especially those involving interactions with authority figures and caregivers). Knowledgeable informants. If key persons in the dependent patient's life (e.g., spouse, parents, children) are willing to attend part of an assessment session, their feedback can provide an invaluable perspective. Sometimes reports from knowledgeable informants dovetail surprisingly well with patients' selfreports, confirming the accuracy of the patient's perceptions and inferences. In other cases the reports of knowledgeable informants diverge dramatically from patients' descriptions (see Harkness, Tellegen, & Waller, 1995). These discontinuities not only inform the practitioner about the dynamics of the patient's interpersonal milieu (Ready & Clark, 2002), but can also provide useful material for the initial stages of therapy.
EFFECTIVE USE OF ASSESSMENT INFORMATION: A FRAMEWORK FOR THE PRACTITIONER To maximize the usefulness of psychological test data, an overarching framework is needed to guide the assessment process. Thus far I have reviewed the major categories of information that should be collected when a dependent patient is assessed. In the following sections, six guidelines are described for using this information in clinical prediction and treatment planning. Tailor Assessment to the Individual Patient A truly comprehensive assessment of the dependent patient would involve administering the traditional "standard" test battery, then supplementASSESSMENT
123
TABLE 7.4 A Psychological Test Battery for Use With Dependent Patients Dimension-domain Implicit dependency needs Self-attributed dependency needs Perceptions of self and others Coping and defense style Social support
Measure(s)
Scoring/interpretation/construct validity data
ROD scale Bornstein (1996b, 1999) IDI, DEQ, PSI, See Table 7.2 for information etc. regarding individual scales, and chapter 2 for construct validity data SQDOR Blatt, Chevron, Quinlan, Schaffer, & Wein(1988) DSQ, DMI Bond et al. (1983); Ihilevich & Gleser (1986, 1991) PSS scale; Procidano & Heller (1983); Lin, SNS Dean, & Ensel (1986)
Note. Full names of scales, along with detailed descriptions, are in chapters 2 and 7.
ing this battery with an array of dependency-specific measures. Practical considerations do not always allow for such extensive testing, however. Administration of one test from each category listed in Table 7.4 requires at least 90 minutes, excluding the RIM. If the RIM is administered along with these other measures, an additional 60 minutes or more of testing time may be needed.5 The practitioner must therefore be selective, tailoring each assessment battery to the individual patient. This may mean omitting certain measures during the initial testing sessions (though these measures can always be administered later, if warranted). Three principles are useful here: • Assessment of implicit and self-attributed dependency strivings is usually advisable. Unless both self-report and projective measures are administered, an incomplete picture of the patient's underlying and expressed dependency needs will be obtained. • Completion of the QSDOR is helpful in most cases. This is true because the QSDOR: (a) provides valuable data regarding the patient's internal object world and (b) is a useful tool for stimulating therapeutic dialogue. • For patients who show significant character pathology, administration of a defense styk measure will be particularly informative. For higher-functioning patients (especially those with symptoms 'Traditionally, the standard psychological assessment battery for adolescents and adults included the MMP1, RIM, intelligence test (e.g., Wechsler Intelligence Scale for Children [WISC] or Wechsler Adult Intelligence Scale [WAIS]), and a brief neurological screen such as the Bender (1938) VisualMotor Oestalt Test. This standard battery has evolved in recent years, partly in response to managed care pressures which compel clinicians to minimize the number of tests used, and partly in response to studies which suggest that tailoring test batteries to individual patients makes assessment more efficient and effective (see Gates, 1999; Groth-Marnat, 1999; McGrath, 2001).
124
CLINICAL APPLICATIONS
of depression), a social support scale may replace the defense style measure. Efficiency in testing can be maximized through judicious use of existing data. If a patient completed the MMPI or RIM within the past 12-18 months and no major changes have taken place in the patient's functioning that call into question the validity of these data, either test protocol can be rescored for dependency. ROD scores derived from the RIM have excellent long-term retest stability (Bornstein, Ressner, & Hill, 1994). Although MMPI Dy scores are not the most extensively validated index of self-attributed dependency needs, they are psychometrically adequate (Bornstein, 1999), stable over time (Russo et al., 2001), and can be used in place of other measures if circumstances dictate (Bornstein & O'Neill, 2000). Document Strengths as Well as Deficits Information in two areas is useful here. First, the clinician should document the patient's environmental assets and resources, focusing on (a) perceived social support and (b) the impact of dependency on the patient's social milieu (even if dependency is problematic in some areas, it may be adaptive in other domains). Second, the clinician should document the patient's per' sonal strengths and resources. Among the topics addressed in this latter section are • Insight. Insight—or the potential for increased insight—is a predictor of success in many forms of therapy (Blatt & Ford, 1994; Weissmark & Giacomo, 1998). Patients who show consistency in their implicit and self-attributed dependency scores tend to have greater insight than patients who show discontinuities in these areas (Bornstein, 1998a). Patients who show features of helpless dependency have greater insight than patients with hostile, hidden, or conflicted dependency (Bornstein & Languirand, 2003). • Motivation. As discussed in chapters 8 and 9, most dependent patients are highly motivated to please the therapist, especially during the initial phase of treatment (Heller &. Goldstein, 1960; Poldrugo & Forti, 1988). This may change as therapy progresses and the patient becomes increasingly conflicted about autonomy (Greenberg & Bornstein, 1989), but during the early stages dependency helps build the therapist-patient bond and facilitates the transference (Juni & LoCascio, 1985; Lorr & McNair, 1964b). • Frustration tolerance. Because many dependent patients reflexively seek help when difficulties arise, frustration tolerance can be an important psychological resource. The dependent patient ASSESSMENT
125
with adequate frustration tolerance will have relatively few "pseudo-emergencies"; the dependent patient with inadequate frustration tolerance may make frequent attempts to contact the therapist after hours and between scheduled appointments (Emery & Lesher 1982). • Coping. Two issues are relevant here: social skills and defense style. Dependent patients with a mature defense style cope more effectively with interpersonal stress and conflict than do patients with a less mature defense style (Maffei et al., 1995; Vaillant, 1994). Patients with strong social skills are better able to cultivate and maintain nurturant, supportive relationships than are patients with weaker social skills (Bornstein, 1993; Masling, 1986). View Assessment as a Snapshot of a Changing Scene Traditionally clinicians have conceptualized assessment results as a stable picture of a patient's personality and functioning, but as Gates (1999) pointed out, assessment results only capture one moment in time: They reflect longterm functioning, but they do not describe it completely. Thus, the clinician should expect assessment results to lose their predictive power over time. Moreover, as Masling (1960) and others (e.g., Rosenthal, 1966) have shown, situational context influences assessment results, and subtle variations in clinician-patient interaction can have powerful effects on assessment process and outcome. It is also important to remember that psychological assessments do not take place at random times in patients' lives, but typically occur during particularly stressful or troubled periods. They are likely to represent a "worst-case" picture of the patient's emotional state, stress level, and coping resources. This does not mean that assessment results are invalid, but they do have limited generalizability over time and across situations. The clinician should consider repeating all or part of an assessment battery if the patient's circumstances change, or administering additional tests as new information emerges in therapy. In short, assessment must always inform treatment, but the reverse is also true: Treatment insights should lead the clinician to update and refine an initial assessment battery. Just as the physician administers follow-up tests on the basis of initial medical test data and changes in the patient's health history, the psychologist should administer follow-up tests on the basis of initial psychological test data and changes in the patient's personal and social history. Assess the System, Not Just the Person As Caspi, Bern, and Elder (1989) noted, because dependent people deliberately seek out and cultivate nurturant, supportive relationships, depen126
CLINICAL APPLICATIONS
dency can only be understood completely when analyzed within the interpersonal milieu in which it is exhibited. Thus, it is important that the clinician assess the system, not just the person. Individual testing (especially that related to the patient's perceptions of self and others and perceived social support) is helpful in this process, but nothing can replace direct observation of the system in action. If possible, the clinician should meet key people in the patient's life and observe the patient interacting with these people during an assessment session. If this cannot (or should not) take place during the initial phases of the assessment process, it might still be useful later, after treatment has begun. It is not sufficient merely to assess and observe the system; the clinician must also conceptualize the patient's difficulties in system terms. The clinician should be aware of the major family systems models (Bowen, 1978; Haley, 1976; Minuchin, 1974) and how they enhance understanding of psychopathology, resistance, coping, and defense. Conceptualizing the dependent patient from a systems perspective need not mean forgoing other conceptual frameworks (e.g., cognitive, psychodynamic); instead, the patient's situation can—and should—be analyzed from multiple viewpoints and on multiple levels. Just as it is important to scrutinize discontinuities in test results when assessing the individual, it is important to pay particular attention to discontinuities when assessing the system. In formal testing, this involves analyzing discrepancies between patient self-reports and information gathered from life records and knowledgeable informants. Discontinuities may also emerge in patient reports of social support (e.g., in the form of markedly divergent PSS-Fr and PSS-Fa scores) and in QSDOR data (e.g., in the form of contrasting perceptions of different family members). When interacting with key individuals in the patient's life, discontinuities may show themselves in terms of divergent perceptions and descriptions (either recognized or unrecognized by the actors) and via conflicts between verbal reports and their affective and nonverbal correlates. Regardless of how they are manifest, discontinuities within the patient or the system are not simply obstacles to be overcome, but starting points for increased insight and enhanced functioning.6 Consider the Impact of Culture, Gender, and Age As was true for diagnosis, these variables can affect test performance and assessment outcome, often in subtle ways. While the clinician cannot 6
As several assessment researchers have noted, it is sometimes helpful to introduce role-play scenarios into the interview process to supplement the patient's self-reports (e.g., McCallum & Piper, 1990; McFall & Marston, 1970). These are most useful when the clinician has formed an initial impression of the patient's situation, so role-play can focus on key problem areas. An additional advantage of role-playing in assessment of the dependent patient is that it sets the stage for implementation of roleplaying techniques during the later stages of therapy. As discussed in chapters 8, 9, and 10, such techniques can be an effective means of altering problematic dependent behavior (Overholser, 1987; Overholser & Fine, 1994). ASSESSMENT
127
eliminate culture, gender, and age effects, she should be aware of their potential impact on test results. Unfortunately, self-report tests of personalityincluding self-report tests of dependency—rarely have good cross-cultural norms and tend to overpathologize individuals raised in non-Western societies or in subcultures outside the mainstream (Dana, 1994, 1997). Only two self-report dependency tests—Sinha's (1968) Dependence Proneness (DP) Scale and Kamitani's (1993) ]iritsu Scale—have adequate non-Western norms, and these are limited to Indian and Japanese adults, respectively.7 As discussed in chapter 1, gender affects self-reports of dependency (and other variables as well) and, given the strong and consistent findings in this area, it is reasonable to assume that many male patients will be reluctant to acknowledge dependent feelings and urges (see Gilbert, 1987, for a discussion of this issue). Creating a nonthreatening environment and establishing rapport with a patient prior to testing can help minimize gender biases in self-reported dependency, but these biases cannot be eliminated completely. Because high masculine gender role scores are associated with underreporting of dependency in both women and men (see Table 1.1), administration of a gender role orientation scale (e.g., the Bern, 1974, Sex Role Inventory) can be helpful in this domain. Finally, many psychological tests are written with younger patients in mind, and while these tests can be used with older adults, the clinician should be alert to the possibility of age-related confounds (Baltes, 1996; Brink, 1986; Eddington, Piper, Tanna, Hodkinson, & Salmon, 1990). These confounds take two general forms: content-based and norm-based. With respect to the former issue, test item content is often geared toward persons in early and middle adulthood, and some experiences described in these items may be unfamiliar to older adults (as discussed in chap. 3, late-life dependency differs from early- and mid-life dependency, both behaviorally and phenomenologically). With respect to the latter issue, older adult norms are scanty or nonexistent for many personality tests—including all widely used tests of interpersonal dependency—and as a result, the clinician must make do with normative data from a different age cohort. Use Assessment to Delineate Goals and Track Progress As several clinicians and clinical researchers have noted, the very act of delineating treatment goals can facilitate change (Jones & Price, 1998; 'Neither of these scales is listed in Table 7.2 because neither has been widely used. However, Sinha's (1968) DP Scale is psychometrically sound (Singh & Ojha, 1987; Sinha & Pandey, 1972) and relatively brief (20 Likert scale items). Kamitani's (1993) Jiritsu (socially sensitive independence) Scale is also psychometrically sound (Kamitani, 1993; Yamamoto, 1989), though not as brief (49 Likert scale items). Clinicians who seek to assess dependency in patients raised in India or Japan should consider using these scales in addition to—or in lieu of—other self-report measures. All 20 DP Scale items are provided in Sinha's original (1968) validation study, and all 49 Jiritsu Scale test items are provided by Kamitani (1993). 128
CLINICAL APPLICATIONS
Ritzier, 1998). This is especially true for dependent patients, who are acutely sensitive to the therapist's verbal and nonverbal communications (Lorr & McNair, 1964b; Masling, 1986). Thus, test feedback can be an important component of therapeutic work with dependent patients. The clinician should think of assessment feedback as the bridge between initial informationgathering and long-term growth. In fact, when feedback is communicated clearly and empathically, it is not unusual for patients to report that psychological testing and the feedback that followed it constituted a key therapeutic event—a "trigger" that set in motion some fundamental change processes that unfolded throughout therapy (see Finn, 2003; Finn & Tonsager, 2002). Just as assessment can be used to facilitate change, it can be used to track progress and gauge the patient's movement toward mutually agreedupon goals. Studies show that shifts in psychological test scores often precede patient reports of change (see Blatt, Auerbach, & Levy, 1997; Gruen & Blatt, 1990). Moreover, these score shifts usually occur first in subtle tests (e.g., the RIM or QSDOR), and only later in tests with high face validity (e.g., the DSQ or DMI). In this respect, periodic assessment during the course of therapy can help the practitioner determine the degree to which ongoing interventions are (or are not) producing the anticipated results. Such ongoing feedback is always helpful, but it may be especially useful when working with dependent patients. As discussed in chapter 8, a broad array of therapeutic techniques—psychodynamic and cognitive, behavioral and experiential—can be used in psychotherapy with dependent patients. This creates what might seem like an overwhelming array of choices for the practitioner, but these disparate intervention techniques can be integrated into a cohesive therapeutic program, tailored to the strengths and needs of each dependent patient.
ASSESSMENT
129
8 APPROACHES TO TREATMENT
During the past century there has been a tremendous amount of writing on psychotherapy, with dependent patients. Until the 1940s, much of this work was based on Freud's (1905/1953) classical psychoanalytic model, but in recent years traditional Freudian writings on dependency have been overshadowed by contributions from object relations theory and self psychology (e.g., Kernberg, 1975; Kohut, 1971). Behavioral intervention techniques for treating problematic dependency began to receive increasing attention during the 1950s, around the same time object relations models gained influence, but it took another decade before the behavioral perspective played a significant role in this area. During the 1970s, behavioral techniques led to the development of cognitive strategies for treating problematic dependency (e.g., Beck, 1976; D'Zurilla &. Goldfried, 1971); once both models were established they evolved in synchrony, with considerable mutual influence and exchange. The 1970s also saw increased interest in humanistic and existential conceptualizations of dependency (Bugental, 1976,1978), some of which evolved into experiential treatment models that combined an overarching humanistic perspective with object relations principles (e.g., Bonnano & Castonguay, 1994; Cashdan, 1988). In this chapter, I discuss traditional approaches to treatment of the dependent patient, focusing on four therapeutic modalities: psychodynamic, 131
behavioral, cognitive, and humanistic-experiential. Within each domain I discuss underlying assumptions and therapeutic goals, then present the basic elements of an exemplary intervention program derived from that theoretical perspective. These intervention programs can be effective in and of themselves, but they also represent the building blocks of the integrated psychotherapeutic framework discussed in chapter 9.
THE PSYCHODYNAMIC PERSPECTIVE Contemporary approaches to psychoanalytic psychotherapy are discussed by Blatt and Ford (1994), Crits-Christoph and Barber (1991), Luborsky (1984), Messer and Warren (1995), and Weiss and Sampson (1986). As these reviews illustrate, psychodynamic treatment models have become increasingly diverse in recent years, incorporating ideas and findings from an array of domains within and outside psychology (Paris, 1998; Sperling, Sack, & Field, 2000). There has also been a shift toward structured, time-limited psychodynamic therapy—a significant departure from earlier psychoanalytic treatment approaches (e.g., Brenner, 1973). Assumptions and Goals: Unconscious Conflict and Insight A core assumption of psychoanalytic theory is that many features of conscious experience are rooted in unconscious conflicts, which take two general forms (Eagle, 1984). Some unconscious conflicts reflect clashes between incompatible beliefs, fears, wishes, and urges (e.g., a wish to be cared for versus an urge to compete). Other unconscious conflicts emerge as compromise formations—the disguised, distorted end-products of underlying impulses and defenses against those impulses (e.g., when hostile humor reflects sublimated aggression). The concept of unconscious conflict is useful in understanding the etiology and dynamics of many personality traits, and it is particularly relevant for dependency. The myriad rules and restrictions of mid- to late childhood— coupled with society's expectation of increased self-reliance—almost invariably cause girls and boys to experience intense ambivalence regarding autonomy and dependency and invoke an array of defenses to manage "unacceptable" dependency-related urges (see chap. 3 for a discussion of this process). Some dependency-related conflicts may be conscious; others are at least partially hidden and inaccessible to conscious awareness (Blatt, 1991; Kantor, 1992, 1993; Ryder & Parry-Jones, 1982). The aim of psychoanalytic therapy is not to ameliorate these conflicts, but to make them accessible to consciousness, where they can be examined critically and acted upon mindfully (Brenner, 1973; Eagle, 1984). Thus, a primary goal of psychoanalytic treatment is insight—increased awareness of 132
CLINICAL APPLICATIONS
dependency-related thoughts, feelings, and motives that previously operated outside of awareness. For many dependent patients—especially those with unacknowledged dependency needs—insight is a prerequisite to therapeutic change (Bornstein, 2004; Hopkins, 1986; Snyder, 1963). Once the patient has gained insight into motives and conflicts that previously existed outside awareness, the process of working through—that is, applying newfound insights to current relationships—may begin (Weiss & Sampson, 1986). Though insight by definition must precede working through, these processes are not separate, but synergistic: Insight is necessary for working through to begin, but as working through proceeds, patients gain increased insight as well (Weissmark & Giacomo, 1998). For most patients this means moving beyond superficial awareness of how their dependency needs have affected past and present relationships and gaining a more sophisticated understanding of how these relationships have influenced (and in some instances, helped propagate) their dependency-related feelings, motives, and fears (see Bruch, Rivet, Heimberg, Hunt, & Mclntosh, 1999).
Luborsky and Crits-Christoph's CCRT Method With these overarching psychodynamic principles in mind, Luborksy and Crits-Christoph (1990) developed the Core Conflictual Relationship Theme (CCRT) method, which has been used to treat a variety of Axis I and Axis II disorders, and may be particularly helpful for dependent patients. The basic elements of CCRT can be divided into four categories: • The underlying context: Supportive-Expressive (S-E) therapy. Derived from the seminal writings of Lowenstein (1951) and Greenson (1965), S-E therapy combines psychoanalytic interpretation with a milieu specifically designed to enhance the therapeutic alliance. The first task in S-E therapy is to build a collaborative working relationship through empathic communication on the part of the therapist (Crits-Christoph & Connolly, 1998). The "holding environment" of S-E therapy may have curative value in and of itself, but it also helps minimize anxiety and defensiveness, especially in patients with limited insight into their underlying dependency needs (CritsChristoph & Barber, 1991). Interpretations in S-E therapy are based on object relations principles and framed in what Mayman (1976) termed "experience-near" language (i.e., language the patient can easily relate to personal experiences and past and current relationships). • Insight through analysis of Core Conflictual Relationship Themes (CCRTs). CCRTs are derived from patient narratives that cenAPPROACHES TO TREATMENT
133
ter on relationship episodes—memorable, meaningful interactions with other people (Luborsky & Crits-Christoph, 1990). As patterns emerge in a patient's relationship episodes, these are analyzed in three broad areas: (a) the patient's wishes, intentions, and fears; (b) the response of the other person; and (c) the patient's reaction to the other person's response. By exploring consistencies in CCRTs across different relationships, the patient's dominant needs and defenses are made explicit, and the trait-like aspects of dependency become clear. By examining inconsistencies in CCRTs across different relationships, the contextual specificity of a patient's behavior can be understood, and the situational features of dependency become apparent (see Crits-Christoph & Barber, 1991; Crits-Christoph, Demorest, Muenz, & Baranackie, 1994). • Obstacles to progress: Ambivalence in the therapeutic alliance. Obstacles to progress in S-E therapy can originate in a number of areas. For the dependent patient, ambivalence is common: As the patient becomes increasingly attached to the therapist, anxiety regarding rejection and abandonment increase and behaviors designed to minimize the possibility of relationship disruption begin to dominate (Kantor, 1992; Lower, 1967; Van Sweden, 1995). Dependency-related resistance is not limited to the patient, however; it can also originate in the therapist (see Ryder & Parry-Jones, 1982). The therapist may fear that the patient's dependency will become increasingly intense over time (the "fantasy of insatiability") and that the patient's dependency will make termination impossible, so treatment can never end (the "fantasy of permanence"). If not managed properly, the patient's and therapist's fears may feed on each other and worsen as therapy progresses: The patient becomes increasingly anxious about the risks and responsibilities of autonomy, and the therapist becomes increasingly anxious about the negative impact of the patient's dependency. • The emotional undercurrent: Transference and countertransference. One way to prevent dependency-related fears from undermining treatment is to explore the patient's transference reaction and the therapist's countertransference response (Bornstein, 1994c, 1998a). Common transference patterns in dependent patients include idealization (maintained through denial of therapist imperfections); possessiveness (which may have a strong narcissistic component or involve feelings of jealousy and competitiveness); and projective identification (wherein the patient unconsciously adopts the therapist's language and mannerisms). Common therapist responses to these transference reactions 134
CLINICAL APPLICATIONS
include frustration at the patient's insatiable neediness; hidden hostility (often accompanied by passive-aggressive acting out); overindulgence (ostensibly to protect the "fragile" patient); and pleasurable feelings of power and omnipotence (which can, on occasion, lead to exploitation or abuse). The key to managing these negative reactions is to seek supervision and guidance from other mental health professionals. This is particularly important for those forms of countertransference that are associated with significant risk of patient harm (Abramson, Cloud, Keese, & Keese, 1994; Coen, 1992).1
THE BEHAVIORAL PERSPECTIVE Reviews of contemporary behavioral treatment techniques are provided by Bellack and Hersen (1993), Hayes (1989), Kazdin (1989), Van Houten and Axelrod (1993), and Wolpe (1990). Although early behavioral interventions aimed at altering problematic dependency were based exclusively on operant conditioning procedures, recent models have combined operant techniques with classical conditioning strategies to maximize treatment effectiveness. Assumptions and Goals: Acquired Behavior and Contingency Change As noted in chapter 1, a basic premise of the behavioral view is that people exhibit particular behaviors—even self-defeating, maladaptive behaviors—because these behaviors are (or were) rewarded. In this context, dependency can be conceptualized as a set of responses aimed at obtaining help and support that are acquired and maintained through a combination of conditioning and learning processes. These include (a) direct reinforcement (sometimes continuous, but more commonly intermittent); (b) vicarious reinforcement (through observation of others' dependency-based rewards); and (c) modeling (including symbolic modeling). Studies show that dependent persons are particularly responsive to subtle social cues (Masling, O'Neill, & Katkin, 1982) and more easily conditioned than nondependent persons in a variety of contexts and settings (Burton, McGregor, & Berry, 1979; Exline & Messick, 1967). Thus, behavioral principles may be particularly useful for understanding the persistence of dependent behavior, even in situations where the rewards for this behavior are not apparent. 'Rather than becoming clingy and needy, some patients cope with dependent feelings by becoming counterdependent, putting forth a veneer of rigid self-sufficiency (Colgan, 1987). This response—which is more common among men than women—creates an additional layer of distortion and defense that must be disentangled before the patient can gain insight into the impact of his dependency within and outside therapy. APPROACHES TO TREATMENT
13 5
Although many behaviorally oriented clinicians conceptualize dependency in terms of positive reinforcement of dependent responding, studies suggest that negative reinforcement also plays a role (see Mowrer, 1950). A two-step process is involved: • Step 1: Acquisition of a fear response. Many children are anxious around unfamiliar people, but in some children this anxiety is especially intense and persistent. Just as certain infantile temperament variables (e.g., high arousal, low soothability) shape parents' responses to the child (Bornstein, 1993), temperament variables may help predict which children are likely to become anxious in the presence of unfamiliar people (Kantor, 1993). • Step 2: Avoidance and anxiety reduction. The overanxious child will tend to avoid unfamiliar people in favor of those who are familiar and predictable. Insofar as avoidance of these interactions reduces the child's anxiety level, this behavior becomes part of her characteristic response pattern (Alden, Laposa, Taylor, & Ryder, 2002). As Ainsworth (1969, 1989) noted, later in life these avoidant responses often persist, although in most cases substitute protectors (e.g., supervisors, friends, romantic partners) replace the parents (see Pincus &. Wilson, 2001; Sperling & Herman, 1991). Turkat's Integrated Behavioral Approach Turkat's (1990) behavioral treatment model is based on the premise that dependent responses persist because they are (a) positively reinforced, in at least some relationships, and (b) negatively reinforced insofar as they enable the patient to avoid anxiety-producing situations. Four techniques, used in combination, are useful in altering the contingencies that maintain this pattern: • Extinguishing problematic dependency. Although extinction techniques by themselves are of limited value in effecting long-term behavior change (Martin &. Pear, 1996), they can be useful in altering problematic dependency. To begin, therapist and patient must identify specific behaviors to be reduced or eliminated. This requires that the components of a patient's selfdefeating dependency be broken into discrete responses, so the contingencies that support these responses can be identified. Using this information, a behavior management program is created aimed at decreasing the frequency of undesired dependencyrelated responses. This process will be enhanced if contingency change first takes place within the context of the patienttherapist relationship, and is then attempted in vivo 13 6
CLINICAL APPLICATIONS
(Overholser, 1997). Studies also indicate that these strategies are most effective when therapist and patient discuss the link between contingency change and behavior change early in the process: When the patient understands the rationale underlying a behavior management program, the likelihood that contingency change will lead to a reactive increase in undesired behavior diminishes (Bellack&Hersen, 1993;Linehan, 1993).2 Replacing dependency with autonomy. At the same time dependency-related responding is reduced, efforts should be made to increase the frequency of alternative responses that are incompatible with the undesired behaviors. For the dependent patient, this means increasing the frequency of autonomous responding (Turkat & Carlson, 1984; Turkat & Maisto, 1985). Just as dependent behaviors that are extinguished must be broken into discrete components, autonomous behaviors that are rewarded must be specific, identifiable, and within the patient's behavioral repertoire (McKeegan, Geczy, & Donat, 1993). When novel or unfamiliar target behaviors are involved, it may be necessary to shape these behaviors incrementally, through a series of narrower subgoals (Deitchman, 1978). To facilitate this process, therapist and patient first identify potentially problematic situations (e.g., being assigned an important project at work), then delineate adaptive responses (e.g., doing background research, seeking feedback from more experienced colleagues). Role-play techniques can be used to increase patient confidence and maximize the likelihood that the newly acquired responses will produce the desired consequences in vivo. Using desensitization to facilitate behavior change. To the degree that a patient's dependent behavior is exacerbated by concerns regarding embarrassment, abandonment, or rejection, systematic desensitization techniques should be implemented to help manage this anxiety and facilitate behavior change. Use of desensitization techniques may be particularly important for dependent patients with co-occurring symptoms of avoidant PD and/or social phobia (Alden, 1989; Alden et al., 2002). For these patients, the high levels of autonomic arousal that accompany social interactions interfere with effective carryover 2 Behavioral treatment of dependency can be used in a variety of contexts, but it is particularly effective with school-age children, hospitalized psychiatric patients, patients in rehabilitation settings, and older adults in long-term care (see Bakes, 1996; Kilbourne & Kilbourne, 1983; McKeegan et al., 1993). Because these individuals are in environments where contingencies are easily managed, effecting behavior change through manipulation of reinforcers is comparatively straightforward (Donat, McKeegan, & Neal, 1991). Dependent patients undergoing behavioral therapy in vivo must be highly motivated for treatment to be effective, and—as discussed in chapter 10—involvement of the patient's family can be invaluable in this regard.
APPROACHES TO TREATMENT
13 7
of desensitization gains from therapy to real-world settings (Overholser, 1987). Detailed discussions of fear hierarchy construction and deep muscle relaxation techniques for use in this context are provided by Martin and Pear (1996). Maintaining behavior change posttreatment. To the degree that autonomous behavior becomes self-reinforcing, the likelihood that this new behavior pattern will be maintained increases (Linehan, 1993; Wasson & Linehan, 1993). Thus, autonomous behaviors that are targeted early in therapy should be those most likely to bring rewards, especially social rewards (Turkat & Carlson, 1984; Turkat & Maisto, 1985). Four techniques are useful in this context. These are (a) choosing target behaviors that lead to positive outcomes in the patient's natural environment; (b) doing in vivo training in settings that resemble those wherein the newly acquired behaviors must be exhibited; (c) varying training conditions to reinforce different expressions of the target behavior and increase generalizability; and (d) gradually reducing the frequency of reinforcement during the latter stages of therapy so reward dynamics approximate those of the patient's social milieu.3
THE COGNITIVE PERSPECTIVE Contemporary cognitive treatment approaches are discussed by Beck (1994), Ellis and Dryden (1997), Fleming and Pretzer (1990), Freeman, Simon, Beutler, and Arkowitz (1989), and Young (1994). Although these approaches differ with respect to certain principles and techniques, they share an emphasis on effecting behavior change by altering the patient's characteristic manner of thinking about, perceiving, and interpreting the world. Assumptions and Goals: Dysfunctional Thinking and Cognitive Restructuring Cognitive theorists conceptualize dependency as the product of maladaptive schemas (i.e., self-defeating beliefs about the self and other people) that cause patients to doubt their abilities, denigrate their skills, and exaggerate the imagined consequences of less-than-perfect performance (Ball & 3 Linehan's (1993) dialectical-behavior therapy (DBT), which combines traditional behavioral and cognitive intervention techniques with a Zen-like attitude of acceptance and impettutbability on the patt of the thetapist—even in the face of patient acting out—has great promise in the treatment of problematic dependency. Although DBT has been applied most extensively in work with borderline and eating-disordered patients, many DBT principles can be applied to the dependent patient as well (see Linehan, 1993; Wiser & Telch, 1999).
138
CLINICAL APPLICATIONS
Young, 2000; Overholser, 1997). Maladaptive schemas not only decrease selfesteem and increase anxiety, they also lead to an array of cognitive distortions that strengthen the person's pre-existing negative views. Three cognitive distortions are particularly salient in the etiology and dynamics of dependency: (a) automatic thoughts (i.e., reflexive negative thoughts that are cued by perceived or anticipated failure); (b) negative self-statements (i.e., self-blaming statements that undermine the person's self-confidence); and (c) negative attributional bias (i.e., a distorted interpretation of causality wherein the person punishes herself for perceived imperfections but cannot accept credit for successes). A primary goal of cognitive therapy is cognitive restructuring—altering dysfunctional thought patterns that foster self-defeating behavior. In the case of dependency, cognitive restructuring focuses on strengthening the patient's self-efficacy beliefs, especially those related to social situations. At the same time, the therapist works to detoxify flawed performance (so the patient does not perceive adequate but imperfect efforts as evidence of incompetence), and provide alternative ways of managing negative feedback (so the impact of everyday criticism is not overwhelming). To accomplish these goals, therapist and patient explore (a) the development of the patient's maladaptive dependency-related schemas; (b) the processes that maintain these schemas over time; (c) the avoidance strategies used by the patient to escape schematriggered anxiety; and (d) the compensatory strategies used to manage this anxiety when it cannot be avoided (Ball, 1998; Young, 1994). Cognitive restructuring techniques run the gamut from gentle reframing of biased perceptions (Ball 6k Young, 2000) to aggressive challenging of irrational beliefs (Ellis & Dryden, 1997). Some cognitive therapies focus primarily on dysfunctional thought patterns (Fleming 6k Pretzer, 1990); others also address problematic emotional responses and self-defeating behaviors (Overholser, 1987, 1997). Increasingly, cognitive therapists have utilized concepts from different treatment models so interventions may be tailored to the patient's overall level of functioning. Thus, several cognitive theorists have delineated specific intervention strategies for use with personalitydisordered patients (Linehan, 1993; Fleming 6k Pretzer, 1990) and other patients with longstanding, change-resistant thought and behavior patterns (Alden, 1989; Ball, 1998). Overholser and Fine's Four-Stage Cognitive Therapy Model Overholser and Fine's (1994) four-stage model is based on the premise that problematic dependency is rooted in active avoidance of autonomy that stems from the patient's belief that she is incompetent, ineffective, and doomed to fail without the guidance and protection of others. Overholser and Fine's model seeks to build patient confidence, teach rudimentary social problem-solving skills, and maximize treatment effectiveness by anticipating APPROACHES TO TREATMENT
] 39
potential roadblocks and pitfalls. Intervention techniques are implemented in stages to guide the patient through a process of cognitive and behavior change. • Stage 1: Active guidance. To facilitate change and foster a collaborative alliance, the therapist in Overholser and Fine's (1994) framework takes an active approach early in treatment, providing considerable feedback and structure. Patients are taught behavioral skills that enable them to make manageable but meaningful changes quickly, thereby increasing motivation and commitment. During the initial sessions, the therapist takes a more active approach than usual in helping the patient delineate long-term therapeutic goals. Among the techniques used at this stage are (a) assertiveness training; (b) behavioral assignments; and (c) stimulus control (e.g., avoidance of dependency "triggers"). Because dependent patients are highly motivated to obtain approval from figures of authority (Bornstein, 1992, 1993), including the therapist (Overholser, 1996, 1997), reassurance, praise, and encouragement can be effective in helping the patient alter longstanding dysfunctional thought and behavior patterns. • Stage 2: Enhancement of self-esteem. Because the dependent patient's help- and approval-seeking result in part from low selfesteem (Overholser, 1993; Tripathi, 1982), Stage 2 focuses on building self-confidence. This begins with psychosocial exploration aimed at uncovering the roots of the patient's negative self-view and gradually incorporates various cognitive restructuring techniques designed to change this dysfunctional thought pattern (e.g., scrutiny and challenging of maladaptive schemas, logical analysis of biased perceptions and beliefs). Patients are provided with coping self-statements that bolster their selfefficacy and enable them to manage negative affect on their own. Reframing techniques may be used to help patients see dependency-related challenges as opportunities for personal growth (Dryden & Trower, 1989; Marlatt & Gordon, 1985). • Stage 3: Promotion of autonomy through problem-solving training. As patients begin to show evidence of enhanced self-esteem and self-efficacy, the focus of therapy shifts to increasing autonomous behavior within and outside therapy and reducing the patient's dependence upon the therapist. Problem-solving training is used to help the patient deconstruct each challenge into five components: problem definition, problem source, generation of alternative solutions, solution implementation, and verification (D'Zurilla, 1988; D'Zurilla & Goldfried, 1971). As this pro140
CLINICAL APPLICATIONS
cess proceeds, the therapist becomes less active and encourages the patient to take increasing responsibility for structuring the interaction. To facilitate this shift, the therapist may also use the Socratic method—active, guided questioning—which enables the patient to generate her own solutions and insights (Overholser, 1993, 1997). Self-control strategies (e.g., selfmonitoring, self-reinforcement) provide the patient with the skills needed to inhibit reflexive (i.e., "mindless") dependent behavior, even in stressful situations (Ball & Young, 2000; Young & Lindeman, 1992). • Stage 4: Relapse prevention. To maximize the stability of behavior change and prevent minor setbacks from undermining progress, relapse prevention strategies are introduced in Stage 4. The patient is taught to anticipate potential problems and reframe setbacks so that they are not magnified into global failure experiences (Marlatt & Gordon, 1985; Meichenbaum, 1985; Young & Lindeman, 1992). High-risk situations are identified, and patients are taught alternative ways of responding to these situations. If a backslide occurs within therapy, this is used as an opportunity to introduce strategies for moving beyond minor obstacles. Because many dependent patients experience comorbid depression and anxiety (Birtchnell, 1984; Bornstein, 1994a; Overholser & Freiheit, 1994), techniques for managing mood and anxiety level can help maintain therapeutic gain (Overholser, 1997). Studies show that just as increasing patient autonomy diminishes chronic anxiety, using cognitive techniques to dampen anxiety facilitates the acquisition of autonomous behavior (Black, Monahan, Wesner, Gabel, & Bowers, 1996). THE HUMANISTIC-EXPERIENTIAL PERSPECTIVE Although the humanistic—experiential model has been less frequently used than other approaches in treating problematic dependency, useful reviews of humanistic (Bohart & Greenberg, 1997), existential (Schneider & May, 1995), and experiential strategies (Cashdan, 1988) are available. There is considerable overlap in the principles underlying these frameworks, and many practitioners use a combination of techniques derived from all three models when working with dependent patients. Assumptions and Goals: Falsehood, Defense, and Responsibility Beginning with the writings of Rogers (1951, 1961), a key tenet of the humanistic—experiential perspective is that various familial and societal facAPPROACHES TO TREATMENT
141
tors—most notably parents' conditional positive regard for the child—can cause the developing person to construct a "false" (or inauthentic) self. This false self is created to comply with the perceived expectations of other people and obtain their approval and love. Once in place, the false self leads the individual to deny feelings and urges that are incompatible with parental and societal norms (Maslow, 1968). To the degree that the parents' conditional positive regard was contingent upon the child's obeying rules without question and complying passively with external demands, the child comes to view autonomy as unacceptable and creates a false self centered on pleasing other people. Eventually, dependency is no longer experienced as a choice, but as a given. Defenses aimed at obviating alternative ways of perceiving the world become firmly entrenched, and the dependent person's experiences narrow to the point that other-centered behavior is the sole means of managing anxiety and gaining approval (Bonanno &Castonguay, 1994; Cashdan, 1988; Hassenfeld, 1999).4 The existential perspective on dependency—though rooted in humanistic and experiential theory—shifts the focus from the constricting effects of early experience to the core motivating power of existential dread (Bugental, 1976; May, 1972, 1981). As Becker (1973) noted, awareness of death and eventual nonexistence can be overwhelming, and as a result people devote enormous energy (and considerable psychological resources) to denying their own mortality (Pysczynski, Greenberg, & Solomon, 2000). One key strategy in this effort involves externalizing responsibility for choices: The person comes to see himself as a powerless entity controlled by outside forces (e.g., society's rules, other people, a higher power, luck), rather than an autonomous, freely choosing, responsible being (May, 1981; Yalom, 1980). To the degree that an individual becomes committed to this way of coping, he will tend to exhibit a pattern of dependent behavior (e.g., advice-, support-, and protection-seeking) that both reflects and reifies the externalization strategy. May and Yalom (2000, p. 287) described this process well: [A] major mechanism of defense that serves to block death awareness is our belief in a personal omnipotent servant who eternally guards and protects our welfare A hypertrophy of this particular defense mechanism results in a character structure displaying passivity, dependency, and obsequiousness. Often such individuals dedicate their lives to locating and appeasing an ultimate rescuer.5 4
Although Rogers' (1951, 1961) speculations regarding the dependency-producing effects of conditional positive regard were derived almost entirely from theory and clinical observation, these speculations have since been supported by findings which indicate that parental authoritarianism plays a significant role in the etiology of dependency (Baker, Capron, & Azorlosa, 1996; Head, Baker, & Williamson, 1991; Sroufe, Fox, & Pancake, 1983). 5 Recent research derived from terror management theory suggests that while some people cope with death anxiety through dependent behavior, others cope by distorting their perceptions of self and other people (Pysczynski, Greenberg, & Solomon, 1997). Thus, when individuals are made anxious about their own mortality, they engage in an array of cognitive distortions designed to bolster their
142
CLINICAL APPLICATIONS
Schneider and May's Existential-Integrative Approach In Schneider and May's (1995) Existential-Integrative (E-I) framework, therapy is seen as a route to liberation. The goals of treatment are to help the person envision new possibilities, experience aspects of the self that have been distorted or denied, and accept responsibility for choices large and small. To initiate this process, the therapist creates an environment wherein the patient can express feelings openly—even troubling, "unacceptable" feelings. At the same time, the patient is encouraged to experiment within and outside therapy, incorporating new ways of being into real-world relationships. Four strategies are used to guide this process: • Unconditional positive regard. Like Rogers (1961), Schneider and May (1995) see unconditional positive regard as key to helping the patient loosen his defenses and regain access to unacknowledged aspects of the self. In practical terms, unconditional positive regard requires that the therapist avoid being critical or directive. By being empathic and nonjudgmental—but not flawless or omniscient—the therapist can help the dependent patient gain trust without seeing the therapist as a pseudoparental "guru" (Yalom, 1980). When the dependent patient experiences a relationship with an accepting but imperfect figure who communicates empathy and models existential living, he can begin to move beyond his fruitless search for an omnipotent savior-rescuer. • Guilt and responsibility. Whereas narcissistic people are preoccupied with fantasies of greatness, dependent people are preoccupied with fantasies of smallness (Schneider, 1990). A key correlate of the dependent person's smallness fantasy is the belief that without an omnipotent protector standing by, she will be overwhelmed and annihilated (May & Yalom, 2000). The dependent patient's rumination about impending disaster is defensive: It helps protect her from experiencing guilt regarding inauthenticity and abrogation of responsibility. By helping the dependent patient focus on the here-and-now (rather than obsessing about the future) and on immediate emotional experience (rather than what might eventually occur), the thera-
sense that the world is predictable and controllable (e.g., underestimating health risks, overestimating the degree to which other people share their views and values; see Pysczynski, Greenberg, & Solomon, 2000). One recent finding is particularly relevant to the existential perspective on dependency: When college students undergo an anxiety-producing mortality salience manipulation, they compensate by increasing their estimates of their romantic partner's commitment to the relationship (Florian, Mikulincer, & Hirschberger, 2002). This result echoes Simpson and Gangestand's (1991) finding (discussed in chap. 4) that dependent people may overestimate their romantic partner's commitment to allay abandonment fears.
APPROACHES TO TREATMENT
] 43
pist can help break the patient's self-defeating cycle of defense and denial. A focus on metacommunication. Like psychoanalysts, existential therapists deconstruct hidden material in the patient's verbalizations and nonverbal behaviors (Bohart & Greenberg, 1997; Ottens & Hanna, 1998). However, unlike psychoanalysts, whose interpretations focus on unconscious motives and defenses, existential therapists focus on metacommunications: pervasive life themes that reflect the patient's core fears (or "dreads"). Key dependency-related dreads in the E-I framework include functioning autonomously, taking risks on one's own, and being overwhelmed by unmanageable responsibility. These dreads both reflect and contribute to the dependent person's narrowed experience of self and prevent her from envisioning alternative choices and actualizing unexplored potentials. While virtually all patients describe their key dreads in negative terms at the outset of treatment, one goal of E-I therapy is to help patients understand the role these dreads play in protecting them from other, more fundamental fears (e.g., death anxiety). Experimenting within and outside therapy. A core component of E-I therapy is the use of exercises designed to short-circuit entrenched defenses, increase emotional awareness, and set the stage for new experiences that help patients reinvent themselves and the world. Schneider and May (1995) provide detailed instructions for a broad array of such exercises, including writing assignments, skill-building tasks, and role-play scenarios designed to increase insight and interpersonal sensitivity. Once the patient becomes comfortable practicing these exercises in therapy, it is important to consolidate gain by applying newfound skills and perspectives in vivo. As Schneider and May (1995, pp. 164-165) noted, "While experimentation within the therapeutic setting is invaluable, experimentation outside therapy is even more beneficial. . . . Experimenting with the actual gives clients new opportunities to live. Although they may not always seize these opportunities, they are invariably vibrant and edifying."
EFFECTIVE USE OF TRADITIONAL TREATMENT MODELS: A FRAMEWORK FOR THE PRACTITIONER Table 8.1 summarizes the core elements of each psychotherapy model discussed in this chapter. To maximize treatment effectiveness using these models, the therapist must do two things. At the "micro" level, effective 144
CLINICAL APPLICATIONS
TABLE 8.1
Psychotherapy With Dependent Patients: Core Elements of Traditional Treatment Models Model Psychoanalytic
Behavioral
Cognitive
HumanisticExperiential
Assumptions Dependency is a product of early experiences and unconscious conflicts Dependency is a consequence of learning and conditioning
Goals
Techniques
Insight followed by Interpretation; working through corrective object within and outside relations therapy Modify contingencies to decrease dependent behavior and increase autonomous behavior Alter self-defeating beliefs and enhance selfefficacy
Operant conditioning to alter behavior; classical conditioning to manage anxiety
Dependency is Cognitive rooted in restructuring; dysfunctional problem-solving thoughts and training maladaptive schemas Dependency stems Re-experience Empathic from neglected aspects connection; inauthenticity/ of self; accept experimentation externalization of responsibility for within and outside responsibility choices therapy
treatment requires that the therapist implement thoughtfully the techniques that are central to whichever theoretical model is guiding her work at that moment, adapting these techniques to accommodate the needs and strengths of each patient (see Bernstein, 1994c, 1998a, for discussions of this issue). At the "macro" level, effective treatment requires that the therapist develop an overarching framework to decide when and how to combine interventions derived from different models. Thus far, our review of traditional treatment approaches for use with dependent patients has focused on the "micro" level application of techniques from each theoretical school. In the following sections, five guidelines are offered to help the practitioner develop the overarching framework necessary to apply this information effectively. Tailor Treatment to the Problem Different disorders demand different interventions, and two considerations are key in tailoring treatment to the problem: • Problem specificity. Narrow problems require focused interventions; pervasive problems demand broader treatment strategies. APPROACHES TO TREATMENT
145
Thus, when patients present with domain-specific dependencyrelated difficulties (e.g., anxiety regarding performance evaluation at work), behavioral and cognitive therapies may be the treatments of choice. When patients describe dependencyrelated difficulties that affect functioning across many areas of life (e.g., generalized lack of assertiveness, dissatisfaction with a broad array of relationships), psychodynamic and humanistic-experiential interventions may be warranted. • Comorbidity. Although some patients seek treatment to deal with problematic dependency, most dependent patients present with other difficulties (e.g., depression, anxiety, relationship conflict; see Joffe & Regan, 1988; Overholser & Freiheit, 1994). Obtaining valid diagnostic information and a detailed clinical history is critical in this context: In some patients problematic dependency was a diathesis for their presenting difficulties, but in other patients dependency only became an issue following Axis I symptom onset (see chap. 6). Even within an individual patient, it is possible for dependency to be primary to certain disorders (e.g., depression) and secondary to others (e.g., agoraphobia). The bottom line: Problem specificity and comorbidity are important considerations when deciding upon a treatment approach, and choice of intervention must be guided in part by the unique combination of difficulties presented by each patient. When problematic dependency predates the onset of symptom-based disorders, the clinician should focus on dependency issues early in therapy, even if this means shifting the focus from present symptoms to longstanding relationship patterns. When problematic dependency followed symptom onset, the clinician should begin by addressing the symptoms themselves; in many patients, dependency levels diminish as symptoms abate (Birtchnell, 1984; Bornstein, 1993; Mavissakalian & Hamann, 1987).6 Tailor Treatment to the Patient Studies show that, in general, dependent patients prefer therapeutic approaches that involve disclosure of personal information and a reasonable amount of feedback from the therapist (Blatt, 1992; Blatt & Ford, 1994; Helweg & Gaines, 1977; Juni & LoCascio, 1985). These findings suggest that—all other things being equal—the dependent patient will respond more 'Beyond these considerations, patients in crisis warrant a different approach than patients with longstanding, relatively stable difficulties. Regardless of which therapeutic modality is used, crises demand structure and a more active stance on the part of the therapist (see Overholser, 1997, and Schneider & May, 1995, for discussions of this issue from the cognitive and existential viewpoints).
146
CLINICAL APPLICATIONS
positively to cognitive therapy, humanistic-experiential therapy, or psychodynamic psychotherapy than to behavioral treatment or traditional psychoanalysis. However, some dependent patients are interested primarily in symptom reduction and behavior change and may be more receptive to treatments that focus on current difficulties than those that emphasize internal dynamics and exploration of past and present relationships. Patients with co-occurring dependency and alexithymia tend to fall into this latter category (O'Neill & Bornstein, 1996). Assessment data—especially those derived from the IDI, ROD Scale, QSDOR, and DSQ or DMI—can help determine which therapeutic modality may be best suited to a particular dependent patient (see chap. 7). Although the patient's preferences should not be the primary determinant of psychotherapeutic modality, studies show that in both psychological and medical settings, treatment compliance and patient satisfaction increase when patients are given some say in which interventions are used (see Weissmark & Giacomo, 1998). Moreover, early involvement in determining the nature of treatment will have a particularly positive impact on the dependent patient: It compels the patient to take responsibility at the outset of therapy, disrupts the patient's usual pattern of looking to others for structure, and sends a message to the patient that she—not the therapist—is the ultimate decision maker (Bornstein, 1994c; Overholser, 1997). To the degree that the patient is able to verbalize her feelings regarding the increased autonomy that decision-making entails, this strategy can also provide useful material for the initial stages of therapy. Link Autonomy With Healthy Dependency When working with a dependent patient—especially one who is extremely clingy and insecure—it is easy to overemphasize autonomy at the expense of healthy dependency. Although increasing autonomous functioning is an important goal of clinical work with dependent patients, autonomy is most adaptive when it is expressed in flexible, situation-appropriate ways and combined with a willingness to seek help and support from other people (Bornstein & Languirand, 2003; Colgan, 1987). This has different implications for different treatment approaches: • Psychodynamic. As Blatt's (1990) dependency-autonomy dialectic (Figure 3.2) illustrated, healthy adult development is an ongoing process of integrating strivings for relatedness and selfdefinition. As patients gain insight into the negative impact of their dependency, they may overcompensate by moving too far toward self-definition. To counter this, the therapist must emphasize relatedness concerns throughout treatment. The initial therapy sessions represent an opportunity to discuss these isAPPROACHES TO TREATMENT
147
sues in the context of the patient-therapist relationship. Impending termination provides another useful context for a discussion of relatedness issues. • Behavioral. Because behavioral treatment of problematic dependency focuses on increasing autonomous responding and extinguishing dependent behavior, it is important that the therapist help the patient acquire healthy dependency skills even as she becomes more independent. Western stereotypes notwithstanding, adaptive functioning in adulthood does not mean doing everything on one's own, no matter how challenging the task (Bakes, 1996; Behrens, 2004; Colgan, 1987). The dependent patient must learn to ask for help and support when appropriate, and use this help and support to function more effectively (see chap. 5). • Cognitive. Healthy dependency is implicit in various aspects of Overholser and Fine's (1994) model (e.g., in the therapist's initial tolerance of the patient's dependent behavior), and it is important to structure problem-solving training so that autonomous and healthy dependent functioning are integrated, and the patient does not construct a false dichotomy such that all forms of help-seeking become equated with failure (Ball & Young, 2000). The therapist must distinguish adaptive and maladaptive help-seeking and help the patient understand the cognitive and behavioral differences between these two patterns. • Humanistic-Experiential. As discussed, the goal of most humanistic and experiential treatment approaches is to increase the patient's autonomy and personal responsibility while simultaneously enhancing her capacity for intimacy and connectedness. The therapist can ensure that this balance is achieved by structuring experiments within and outside therapy so the patient experiences autonomy and intimacy as synergistic, not incompatible. Schneider (1990) and Yalom (1980) provide practical suggestions along these lines from an existential perspective; Bornstein and Languirand (2003) provide additional suggestions for interventions to strengthen the intimacy—autonomy link in individuals, couples, and families. Explore Your Own Feelings About Dependency Two patient-therapist dynamics are particularly vexing for clinicians: dependency and control (Barth, 2001). Most therapists are well aware of the ways in which a struggle for control can undermine therapy (especially when
J 48
CLINICAL APPLICATIONS
this struggle revolves around missed sessions or unpaid bills). Fewer therapists are sensitive to the ways in which their feelings about dependency can contaminate the therapeutic relationship. However, studies indicate that many therapists become quite anxious when a patient seems overly dependent (Lower, 1967; Perry, 1989). Not surprisingly, therapists who are themselves dependent are particularly upset by patients who seem helpless, clingy, and needy (Abramson et al., 1994; Browne & Dolan, 1991). Certain reactions to a patient's expressed dependency urges can create difficulties in therapy. Understanding one's reflexive responses to dependency cannot ensure that these reactions will disappear, but in this situation selfawareness is the best defense against subtle (and not-so-subtle) forms of acting out that undermine therapy and harm the patient. Three reflexive responses to patient dependency are particularly problematic: • Infantilization. This occurs when the therapist perceives the dependent patient as childlike and immature. The primary risk here is that the therapist will become fearful of overwhelming the "fragile" patient and fail to set appropriate limits (Hopkins, 1986). Not only does this virtually ensure limited progress in therapy, but it recapitulates the same destructive dynamic that occurs in many (perhaps most) of the patient's other relationships—the very dynamic that helps propagate the patient's dependent behavior in vivo. • Authoritarianism. Some therapists respond to patient dependency by becoming inflexible rather than indulgent (Perry, 1989; Ryder & Parry-Jones, 1982). These therapists take an overly active stance in therapy, set rigid rules, and may inadvertently steer therapeutic dialogue toward issues with which they—not the patient—are concerned. This authoritarian stance is always counterproductive and, like infantilization, it is particularly harmful if it reiterates a dependency-fostering dynamic that is occurring in other areas of the patient's life. • Denigration. In many ways the most problematic response to patient dependency is denigration. Some therapists are so conflicted regarding their own dependency struggles that they reflexively distance themselves from dependent patients by devaluing and belittling them. Even if the therapist does not reveal these feelings to the patient (though many inadvertently do), denigration sets the stage for manipulation, exploitation, and abuse (Gregory & Gilbert, 1992). Oftentimes denigration cooccurs with another problematic therapist response to dependency, forming the subtext for infantilization and/or authoritarianism.
APPROACHES TO TREATMENT
149
Integrate Mindfully Some therapists are strongly committed to a single therapeutic modality, but surveys indicate that eclecticism is now the most common therapeutic approach among practitioners; between 25% and 50% of clinicians identify themselves as eclectic (Norcross, Karg, & Prochaska, 1997; Watkins & Watts, 1995). Today's emphasis on eclecticism stands in stark contrast to survey data from 3—or even 2—decades ago, when the vast majority of therapists adhered to a single therapeutic approach. The burgeoning of eclecticism is in part a product of contemporary market forces that demand efficiency and cost-effectiveness in treatment. Increased reliance on formal guidelines for empirically validated treatments has also encouraged practitioners to be flexible in their implementation of different intervention strategies (Beutler, Moliero, & Talebi, 2002; Ramsay, 2001; Sperling et al., 2000). Integration of different therapeutic techniques should be planned carefully; without an overarching framework, it can instead be haphazard. The clinician should be prepared to vary intervention strategies within a single therapy case, adhering to a "baseline" therapeutic modality while judiciously blending aspects of other models into treatment as circumstances dictate. To accomplish this, the practitioner must (a) link each assimilated intervention technique to a specific therapeutic goal and (b) monitor the impact of each intervention on patient functioning as therapy progresses. I discuss this integration process in chapter 9.
150
CLINICAL APPLICATIONS
9 AN INTEGRATED TREATMENT MODEL
Given the voluminous clinical literature on dependency, there has been surprisingly little research assessing the impact of psychotherapy on dependency-related behavior. Table 9.1 summarizes the available findings in this area, and as this table shows, extant data are decidedly mixed: Although some investigations found significant decreases in patient dependency following a course of psychotherapy (e.g., Alexander & Abeles, 1968; Rathus, Sanderson, Miller, & Wetzler, 1995), most studies reported no change in dependency levels during treatment (e.g., Maling, Gurtman, & Howard, 1995), or a nonsignificant decrease in dependency over time (e.g., Black, Monahan, Wesner, Gabel, & Bowers 1996; Moore & Blackburn, 1996). Scrutiny of Table 9.1 confirms that no traditional treatment model has proven consistently superior to other approaches across different populations and outcome measures. Paralleling the findings summarized in Table 9.1, studies assessing the moderating effect of patient dependency on treatment outcome have been scanty and inconclusive. The results of these investigations are summarized in Table 9.2. As this table shows, some studies have found a positive impact of dependency on treatment response (e.g., Gurtman, 1996; Hoffart & Martinsen, 1993), whereas others have reported a negative impact (e.g., Klein, 1989; Wonderlich, Fullerton, Swift, & Klein, 1994). The majority of investi151
TABLE 9.1 Effects of Psychotherapy on Dependency-Related Responding Study
Participants
Alexander & Abeles (1968)
20 outpatients (mixed diagnoses) in time-limited, insight-oriented therapy
Black et al. (1996)
Outcome Significant decrease in expressed dependency on family members; significant increase in expressed dependency on the therapist Nonsignificant decrease in PDQ dependency scores from preto posttreatment No change in self-reported dependency across 38 therapy sessions
44 outpatients with panic disorder receiving 8 weeks of cognitive therapy Maling et al. 307 outpatients (mixed (1995) diagnoses) receiving longterm therapy in various modalities Moore & 119 inpatients and outpatients Nonsignificant decrease in SAS Blackburn receiving 16 weeks of sociotropy (dependency) (1996) cognitive therapy for scores from pre- to postdepression treatment Nelson-Gray 9 depressed outpatients Significant decrease in SAS et al. (1996) receiving 12 sessions of dependency scores from precognitive therapy test to 3-month follow-up; no change in IDI scores during this period Rathus et al. 18 agoraphobic outpatients Significant decrease in MCMI-II (1995) receiving 12 weeks of dependency scores from precognitive-behavioral therapy to posttreatment Rector et al. 51 depressed outpatients Nonsignificant decrease in DEQ (2000) receiving 20 weeks of dependency scores from preto posttreatment cognitive therapy Winder et al. 23 outpatients receiving Significant increase in (1962) insight-oriented therapy at dependency-related a child guidance clinic statements only when these statements were verbally reinforced by the therapist
Note. PDQ = Personality Diagnostic Questionnaire (Hyler et al., 1988); SAS = Sociotropy-Autonomy Scale (Beck et al., 1983); IDI = Interpersonal Dependency Inventory (Hirschfeld et al., 1977); MCMI-II = Millon Clinical Muliaxial Inventory-ll (Millon, 1987); DEQ = Depressive Experiences Questionnaire (Blatt et al., 1976). Effect size estimates are not included in this table because the majority of investigations did not provide enough data to calculate a psychotherapy outcome effect size.
gations either found no significant effect of dependency on treatment outcome (e.g., Chambless, Renneberg, Goldstein, & Gracely, 1992) or obtained mixed results, with patient dependency facilitating treatment efficacy in some therapeutic modalities and inhibiting treatment efficacy in others (e.g., Blatt, 1992; Zettle, Haflich, & Reynolds, 1992). Given the modest effects produced by traditional therapeutic approaches, the future of dependency treatment may lie in integration: Clinicians and clinical researchers must find ways to blend aspects of different dependency treatment strategies, combining the most useful elements of each. In this chapter 1 discuss the basic elements of an integrated treatment model for
152
CLINICAL APPLICATIONS
TABLE 9.2 Moderating Effects of Dependency on Psychotherapy Outcome Study Blatt (1992)
Participants
Outcome
42 outpatients (mixed diagnoses) receiving psychoanalysis or psychoanalytic therapy
Dependent (anaclitic) patients showed better results in psychoanalytic therapy than in psychoanalysis; introjective (self-critical) patients showed the opposite pattern Blatt & Ford 90 inpatients (mixed Fewer positive changes on a broad (1994) diagnoses) receiving range of measures in dependent long-term, intensive (anaclitic) than in self-critical psychoanalytic therapy (introjective) patients; positive effects in dependent patients manifested most strongly in measures of object relations and interpersonal functioning Chambless et 51 agoraphobic outpatients No relationship of SCID-II DPD al. (1992) receiving intensive 2diagnoses to treatment outcome week group behavioral treatment plus individual multimodal treatment Clark et al. 148 depressed outpatients High SNAP dependency scores were (2003) receiving 12-14 weeks of associated with less positive cognitive therapy treatment outcome Gurtman 104 outpatients receiving Self-reported dependency scores (1996) long-term therapy in predicted positive outcome across various modalities an array of therapist-related dimensions Hoffart & 77 depressed inpatients MCMI-II dependency scores predicted Martinsen receiving intensive positive treatment outcome (1993) multimodal treatment Klein (1989) 163 depressed outpatients DEQ dependency (anaclitic) scores being treated for a variety predicted less positive outcome of disorders Poldrugo & 404 alcoholic outpatients DPD diagnoses were associated with Forti(1988) receiving group therapy increased therapy responsiveness following intensive shortand follow-through term treatment Rathus et al. 18 agoraphobic outpatients Nonsignificant trend toward more (1995) receiving 12 weeks of positive outcome in patients with cognitive-behavioral SCID-II DPD diagnoses therapy Rector et al. 51 depressed outpatients DEQ dependency (anaclitic) scores (2000) receiving 20 weeks of did not predict treatment outcomes cognitive therapy Wonderlich et 30 female outpatients Self-reported dependency predicted al. (1994) receiving multimodal poor outcome 3 years posttreatments for eating treatment disorders Zettle et al. 30 outpatients receiving SAS sociotropy (dependency) scores (1992) 12 weeks of individual predicted positive response to or group cognitive group therapy but not individual therapy therapy Note. SCID-II = Structured Clinical Interview for DSMPersonality Disorders (Spitzer et al., 1990); SNAP = Schedule for Nonadaptive and Adaptive Personality (Clark, 1993); MCMI-II = Millon Clinical Multiaxial Inventory-ll (Millon, 1987); DEQ = Depressive Experiences Questionnaire (Blatt et al., 1976); SAS = Sociotropy-Autonomy Scale (Beck et al., 1983). Effective size estimates are not included in this table because the maioritv of psychotherapy outcome effect size.
AN INTEGRATED TREATMENT MODEL
153
therapeutic work with dependent patients, focusing on four domains of patient functioning: cognitive, motivational, behavioral, and emotional (see chap. 1 for detailed discussions of these four dependency domains).1
ASSIMILATIVE INTEGRATION OF DEPENDENCY TREATMENT MODELS Messer (1992) introduced the concept of assimilative integration as a method for combining treatment strategies to maximize therapeutic outcome. Messer's framework entails selecting an overarching ("baseline") therapeutic modality on the basis of patient needs and therapist expertise, then supplementing this approach with interventions derived from other treatment models. Messer's framework builds flexibility into the therapeutic relationship, allowing the clinician to (a) begin treatment using whatever modality seems best suited to the patient's presenting problem, coping style, and underlying personality structure; then (b) introduce additional interventions in response to new insights, evolving patient-therapist dynamics, and other changes that occur as therapy moves forward.2 As Beitman (1992) pointed out, the timing of new interventions can be as important as the interventions themselves: A poorly timed therapeutic shift can undermine treatment progress, even though this shift might have facilitated change if introduced earlier or later. Conceptualizing therapy as a series of four sequential stages—engagement, pattern search, change, and termination—Beitman argued that for most patients the optimal point for introduction of new interventions is during stage three. In other words, the therapist's overarching therapeutic strategy should be used to engage the patient (Messer, 1992), provide her with a conceptual framework within which she may understand and describe her current difficulties (Masling & Cohen, 1987), and elucidate longstanding dysfunctional patterns in cognition, be1
Although virtually any combination of therapeutic approaches can be used, certain combinations have been employed more often than others. As Norcross and Newman (1992) noted, the most popular combinations of therapeutic orientations have traditionally been cognitive and behavioral (12% of clinicians), humanistic and cognitive (11%), and psychoanalytic and cognitive (10%). The least common combination in Norcross and Newman's survey was behavioral and psychoanalytic (4%), although Wachtel (1977, 1991) and others (e.g., Linehan, 1993; Messer, 1986) have argued that these ostensibly incompatible frameworks actually have considerable untapped integrative potential. 2 The distinction between eclecticism and integration is elusive, and it is not always possible to separate these two approaches, conceptually or practically. In general, eclecticism involves combining elements of different psychotherapy models, whereas integration involves blending elements of these models and emerging with a qualitatively different therapeutic approach. As Wachtel (1991, p. 44) noted, however: The habits and boundaries associated with the various schools are hard to eclipse, and for most of us integration remains more a goal than a constant daily reality. Eclecticism in practice and integration in aspiration is an accurate description of what most of us in the integrative movement do much of the time.
154
CLINICAL APPLICATIONS
havior, and emotional responding (Wasson & Linehan, 1993). Once these goals have been achieved, the stage is set for therapeutic shifts to be introduced as needed. In applying Beitman's (1992) framework to psychotherapeutic work with dependent patients, it is important to separate therapeutic targets from the strategies used to address them (see Beutler, Moliero, &Talebi, 2002; Ramsay, 2001; Safran & Messer, 1997). In other words, even if a central goal of therapy is to alter dysfunctional thought patterns, this does not mean that cognitive techniques are invariably the treatment of choice. For some patients, cognitive change can be effected most fully using experiential strategies (Schneider & May, 1995) or psychodynamic interventions that make explicit the patient's biased perceptions of self and others (Luborsky & Crits-Christoph, 1990). Similarly, problematic dependency-related emotional responses can be altered using psychodynamic (Coen, 1992), cognitive (Overholser & Fine, 1994), behavioral (Alden et al., 2002), and experiential strategies (Cashdan, 1988).3 Finally, as several clinicians have pointed out, integration can take place in many different ways and need not be limited to the introduction of new therapeutic techniques within a single course of treatment (Lampropoulos, Spengler, Dixon, & Nicholas, 2002; Norcross & Goldfried, 1992; Strieker & Gold, 1993). Integration can be prompted by the therapist, by the patient, or by the situation (e.g., in response to a therapeutic impasse). Psychotherapy researchers now group integration strategies into three general categories (Gold, 1994; Lazarus, 1992). These are summarized in Table 9.3, and may be described as follows: • Integration within a single session. This occurs when patient and/ or therapist discover an opportunity to approach an ongoing issue from a new perspective. It requires the greatest flexibility on the part of both patient and therapist, and often takes place when patient or therapist feel "stuck" working within a single therapeutic modality (Linehan, 1993; Wasson & Linehan, 1993). • Integration across sessions. Less spontaneous than integration within a session, integration across sessions is usually initiated by the therapist and planned ahead of time (Lampropoulos, Sprengler, Dixon, &Nicholas, 2002). This form of integration— which provides an opportunity for patient and therapist to get a fresh perspective on a key problem or issue—comes closest to capturing the spirit and method of Messer's (1992) assimilative integration model. 3
Even within Beitman's (1992) critical third change stage, there are choices to be made. This stage can itself be divided into three substages: giving up old patterns, starting to build new patterns, and ensuring the continuance of newly acquired behavior. Each substage represents a potentially fruitful point for introducing new therapeutic techniques (see Gold, 1994).
AN INTEGRATED TREATMENT MODEL
15 5
TABLE 9.3 Characteristics of Different Types of Psychotherapy Integration Primary impetus
Type of integration Withinsession
Situation
Betweensession
Therapist
Across treatment episodes
Patient
Purpose/goal Overcome resistance; maximize therapeutic opportunity Facilitate change processes; increase insight Revisit key issues in context of changing life circumstances
Transition process
Advantages Spontaneity; adaptation; enhanced insight
Brief/cursory
Fresh perspective on problem Reworking of issues with incubation/ experiential learning between treatments
Review and place problem in new context Patient-initiated; development of new strategies and goals; enhanced selfawareness
Note. Detailed discussions of these integration strategies are provided by Beitman (1992), Gold (1994), Norcross and Newman (1992), and Ramsay (2001).
• Integration across treatment episodes. This form of integration is almost always initiated by the patient, who may deliberately seek different types of therapy during different stages of coping with a problem. Integration across treatment episodes sometimes reflects a patient's dissatisfaction with an earlier therapy experience, but it can also stem from a conscious decision to reconceptualize a longstanding problem from a new vantage point (see Arkowitz & Messer, 1984; Strieker & Gold, 1993).
INTEGRATING STRATEGIES ACROSS DOMAINS OF PATIENT FUNCTIONING The interactionist model summarized in Figure 1.1 suggests four entry points for treatment of dependent patients: cognitive, emotional, behavioral, and motivational. A primary advantage of this model is that it not only identifies potential therapeutic targets but also specifies how intervening at a given level will affect other, related processes. Thus, the interactionist model enables the clinician to formulate an integrated approach to therapy with dependent patients and tailor this approach to each patient's needs, strengths, and unique life circumstances. Within each target domain, an array of complementary and mutually supporting strategies may be used. Cognitive Change Strategies As Castillo (1997) pointed out, the patient's cultural background should help shape all forms of psychotherapeutic intervention (see also Dana, 1997;
156
CLINICAL APPLICATIONS
Triandis, 1996). In the context of problematic dependency, cultural considerations are particularly important when cognitive treatment strategies are used. Studies show that individuals raised in communitarian, sociocentric cultures construct self-representations that are more permeable—and defined more strongly by social connections and interpersonal ties—than are the self-representations of people raised in individualistic, achievement-oriented cultures (Andersen & Chen, 2002). Thus, patients with different cultural backgrounds may experience and respond to therapeutic interventions somewhat differently—especially interventions aimed at exploring and modifying aspects of the self (Benfari, 1969; Gjerde, 2001; Neki, 1976; Sato, 2001). To maximize the effectiveness of interventions designed to alter the patient's biased perceptions of self and others, the unique aspects of these biases should be made explicit during the initial stages of treatment. As noted in chapter 7, Blatt, Chevron, Quinlan, and Wein's (1981) QSDOR is ideally suited to this purpose for two reasons. First, based on preliminary discussion of the patient's situation, the therapist can identify key figures to be described (e.g., parent, spouse, colleague at work) so that the QSDOR can be personalized for each patient. Second, the patient's open-ended descriptions of self and other people not only inform the therapist, but can also be used to facilitate discussion during the first few therapy sessions. Although the QSDOR is very useful for assessing the patient's key cognitive distortions, other conceptual frameworks can also help the clinician formulate treatment interventions in this area. In particular, Markus and Nurius's (1986) concept of possible selves may be useful in this context. As Markus and Nurius noted, behavior, motivation, and emotional responding are not only influenced by past and present views of the self, but by potential future selves as well. Three possible selves are key: the hoped-for self (i.e., the self we wish to become), the feared self (i.e., the self we want to avoid becoming), and the expected self (i.e., the self we anticipate we will become if life goes as planned). Psychometrically sound measures of possible selves are available for use in clinical settings (see Carver, Reynolds, & Scheier, 1994). The concept of possible selves not only broadens the traditional psychological view of the self, but also provides a conceptual framework linking the psychodynamic, cognitive, and humanistic perspectives (which have traditionally focused on past, present, and future selves, respectively).4 Once the therapist has assessed a patient's core beliefs regarding self and others, three strategies are useful in changing problematic dependencyrelated thought patterns: ""Exploration of the patient's possible selves can be valuable, but as Andersen and Chen (2002) noted, several other conceptual frameworks are also useful in understanding the complexity of the patient's self-representation. In particular, Higgins' (1987) concept of self'discrepancies—internal inconsistencies in the patient's experienced self—may be useful in elucidating hidden aspects of the dependent patient's dysfunctional beliefs regarding the self and other people (see also Strauman, 1992).
AN INTEGRATED TREATMENT MODEL
15 7
Altering automatic thoughts and extinguishing negative self' statements. Various strategies can be employed here, including refraining (Beck, 1994), challenging (Ellis & Dryden, 1997), and using paradoxical interventions to illuminate the selfdefeating nature of negative self-statements (Linehan, 1993). These direct approaches can be supplemented with indirect strategies aimed at modifying self-perceptions (e.g., self-esteem building, development of problem-solving skills; see Overholser, 1997; Overholser & Fine, 1994). In applying these techniques, it is important to remember that mental models of the self and perceptions of other people evolve synergistically: Changes in one area invariably spill over into the other (Andersen & Chen, 2002; Dana, 1998). Thus, even if the therapist focuses on altering negative statements regarding the self, changes in the patient's perceptions of significant others are likely to ensue. Rather than allowing these issues to remain in the background, the therapist should periodically refocus the conversation to explore the patient's changing attitudes and beliefs regarding key figures, past and present. Cognitive restructuring. Ball and Young's (2000) framework is useful in exploring the development and maintenance of maladaptive schemas as well as the avoidance and compensatory strategies used by the patient to cope with schema-triggered anxiety. In addition to providing recommendations in this area, Ball and Young (2000) include detailed lists of core topics and elective (secondary) topics that facilitate this restructuring process. Having the patient describe various possible selves—either orally or in writing—can facilitate cognitive restructuring and open up new avenues for discussion and exploration. Transference analysis. By exploring the patient's beliefs regarding the therapist, characteristic distortions in the patient's processing of information regarding figures of authority and potential caregivers are made explicit (see Andersen & Chen, 2002; Andersen, Chen, & Miranda, 2002). This process can be enhanced by combining the psychodynamic strategy of transference interpretation with Yalom's (1980) technique of gradually revealing to the patient minor therapist flaws and imperfections. In this way the dependent patient not only gains insight into her biased perceptions of self and other people, but at the same time experiences a trusting relationship with a valued, caring other who is neither omniscient nor omnipotent. Once the patient gains insight within the therapeutic milieu, it
158
CLINICAL APPLICATIONS
is important to help the patient generalize this insight to other relationships.5 Motivational Change Strategies The key motivational goal for the dependent psychotherapy patient is to replace mindless help- and reassurance-seeking with healthy connectedness—a desire to develop intimate relationships with others while continuing to value self-directedness and personal responsibility. As Bornstein and Languirand (2003) noted, healthy connectedness (or healthy dependency) requires that the patient seek help and support when appropriate, blend intimacy and autonomy in friendships and romantic relationships, and maintain a strong sense of self even when reaching out to others (see chap. 5). As was true for cognitive interventions, strategies aimed at altering dependencyrelated motivations should be tailored to the patient's cultural background and present-day social milieu: As Sato (2001) and Shimizu (2000) noted, persons raised in individualistic societies tend to be more conflicted regarding connectedness and healthy dependency than are persons raised in sociocentric societies. In this respect, a sociocentric background and communitarian value system may facilitate motivational shifts in the dependent patient. The movement from unhealthy overdependence to healthy dependency is complicated by the fact that patients have different degrees of insight regarding their underlying urges and motivational states. In this context, it is useful for the clinician to distinguish self-attributed (i.e., conscious) motives from implicit (i.e., unconscious) motives (see chap. 2). As discussed in chapter 7, this distinction is more complex than researchers initially believed (see also Bomstein, 1998c), and for many patients self-awareness regarding dependency needs varies markedly across different situations and settings. Moreover, dependent patients with comorbid character pathology often show wide fluctuations in insight, even within a single therapy session. Thus, assessment of self-attributed and implicit dependency strivings early in therapy is essential. Once these strivings have been assessed, three principles can be used to replace maladaptive dependency-related urges with healthy dependent motives: • Setting limits on contact. As Overholser (1997) and others (e.g., Emery & Lesher, 1982) noted, "mindless" (i.e., reflexive) helpseeking persists in part because it is intermittently reinforced in various contexts. By setting limits on between-sessions conthan focusing on mental representations of self and others, developmental psychologists have invoked the construct of the internal working model to describe internalized interaction patterns akin to "relationship templates" or scripts (Horowitz, 1991; Main et al., 1985). As Osofsky (1995) noted, internal working models not only represent a bridge between attachment theory and psychodynamic treatment, but also help link object relations models with family systems frameworks.
AN INTEGRATED TREATMENT MODEL
15 9
tact, the therapist can begin to turn mindless help-seeking into mindful problem-solving. The dependent patient may still attempt to initiate contact between sessions, but if the therapist sets clear limits early in therapy and then responds to contactseeking behaviors firmly but empathically, these can be important learning experiences rather than frustrating conflict situations. When reflexive help-seeking is replaced by mindful help-seeking, the patient's subjective experience is altered in a fundamental way: Help-seeking is no longer experienced as an "irresistible" impulse, but as one (of many) possible responses to an anxiety-producing situation (Bornstein & Languirand, 2003). Habit and reflex gradually give way to conscious choice and responsibility. • Using dependency to facilitate autonomy. As noted in chapter 8, the patient's dependency on the therapist can lead to certain difficulties, especially if the therapist's countertransference response undermines treatment (see Abramson, Cloud, Keese, & Keese, 1994; Strean, 1986). There is a positive side to excessive patient dependency as well, however: The dependent patient is highly motivated to please the therapist and strengthen the relationship. Although this may sometimes lead to conflict as termination nears (a topic I take up later in the chapter), during the early and middle phases of treatment, the patient's motivation to please the therapist can be used to facilitate autonomous behavior outside therapy (Bomstein, 1994c; Bornstein & Bowen, 1995). Dependent patients adhere more conscientiously than do nondependent patients to homework assignments and other tasks recommended by the therapist (Poldrugo & Forti, 1988). By challenging the patient to make changes outside therapy and reinforcing these efforts within the treatment session, the clinician can use the patient's desire to please the therapist to facilitate growth and positive change. • Bridging to connectedness. Developing healthy dependency is a challenge for everyone, but especially for patients who are highly dependent or rigidly detached (Hetherington, 1999; Kantor, 1993). The overdependent patient's lack of experience in behaving autonomously can cause him to overcompensate and move too far toward independence. Because of this, many patients first learn to connect with others in a balanced way by experiencing a healthy dependent relationship with the therapist. Luborksy and Crits-Christoph's (1990) SupportiveExpressive (S-E) therapy model is very useful in this regard, insofar as S-E techniques help create the sort of "holding environment" that facilitates connectedness and intimacy without 160
CLINICAL APPLICATIONS
sacrificing self-directedness and self-awareness. Schneider and May (1995) provide several suggestions for experimentation outside therapy that help the patient extend this pattern to real-world relationships. Emotional-Change Strategies The key emotional goals in psychotherapeutic work with dependent patients are to minimize and manage insecurity, performance anxiety, rejection-abandonment fears, and fears of negative evaluation by others. Although core relationship themes and expressed concerns may differ from patient to patient, dependency-related insecurity almost always centers on fear of negative evaluation (especially by figures of authority) and anxiety regarding abandonment or rejection (most often centering on romantic partners, but also involving parents, friends, and colleagues at work). In romantic relationships, the patient may believe that modest conflict will invariably lead to relationship disruption; reality-testing by the therapist can be useful here, to let the patient know that some degree of conflict is normal and unlikely to undermine an otherwise stable relationship. At work, the patient may believe that less-than-perfect performance on a project or task will inevitably bring serious negative consequences (i.e., a permanent rift in her relationship with her supervisor); again, reality-testing can be helpful in putting workrelated performance in perspective (if only to remind the patient that a competent employee is a valuable commodity, and a solid relationship with a supervisor is unlikely to be ruined by a single performance glitch).6 Challenging the patient's untested inferences is central to altering problematic emotional patterns. Three other techniques, alone and in combination, facilitate this process: • Preventing social "crumbling." In discussing the core emotional challenge of Cluster C personality disorders, Alden (1989) noted that when an individual is extremely anxious regarding social interaction, the resulting physiological response can lead to persistent autonomic arousal. This arousal (a) produces an array of visible symptoms (e.g., blushing, stammering, sweating) and (b) may ultimately cause the individual to "crumble" and lose the ability to engage in minimally appropriate social interaction. Social crumbling cannot be prevented by reframing and challenging (though these strategies are helpful for certain people). Instead, the patient must be provided with a set of 'Strong emotional reactions to minor work conflicts are particularly common in patients raised in more traditional Asian families, particularly those in Japan, India, and Korea. Within these societies, professional relationships have tended to be conceptualized in terms of rigid roles and clearly defined hierarchies (Dana, 1997; Doi, 1973; Johnson, 1993). AN INTEGRATED TREATMENT MODEL
161
proximal coping strategies to manage the situation and dampen autonomic arousal before it undermines social functioning. A key strategy here is to help the patient recognize when incipient panic is beginning and short-circuit this response by employing relaxation techniques (e.g., deep breathing) that attenuate anxiety and autonomic arousal. In extreme cases, biofeedback may help the patient develop the skills needed to moderate arousal effectively. Using role play to build anxiety-management skills. Effective use of anxiety-management techniques requires practice, and role play can provide initial rehearsal opportunities within a closed, safe setting. Guided imagery is a useful method for increasing, then tempering, anxiety. Once the patient has developed some facility in moderating anxiety in this way, role play can be used to extend these skills within the therapy session. The major step in solidifying initial gains is to take them outside therapy, to contexts and situations that involve greater risk and less predictability. Group work is very useful in this regard (as discussed in chap. 10), but ultimately the patient must apply these skills in vivo—at social events, at work, and when interacting with spouse, friends, parents, and others.7 Exploring termination concerns. Many patients have concerns regarding termination and wonder whether they will be able to maintain treatment gains without the holding environment that therapy provides. Dependent patients' termination concerns are more intense and begin earlier in treatment than those of less dependent patients (Bornstein, 1994c; Bornstein &. Bowen, 1995). It is common for highly dependent patients to become preoccupied with termination after only a few therapy sessions— sometimes after only one. More often than not, these concerns reflect some longstanding issues that emerge in most—perhaps all—of the patient's relationships. For this reason, they are worthy of exploration early in therapy. If the patient does not verbalize these concerns within the first few sessions, the therapist should raise them at an opportune moment. Discussion of termination concerns can provide insight regarding the patient's abandonment fears and fantasies and is a useful way of exploring past relationship experiences that helped to create (or reinforce) these patterns (Hopkins, 1986; Weiss & Sampson, 1986). 'Role play can help the patient develop strategies for moderating dependency-related anxiety and provide perspective on dependency-related behavior. By having the patient assume the role of protector while the therapist, partner, or member of a group takes on the role of help- or reassuranceseeker, the patient can experience what it is like to be looked to for support. Empathy becomes a route to behavior change.
162
CLINICAL APPLICATIONS
Behavioral-Change Strategies The primary behavioral goal in therapeutic work with dependent patients is to replace rigid, inflexible dependence with mindful, situationappropriate dependency, helping the patient develop interpersonal strategies for getting dependency needs met without undermining her self-confidence and alienating other people. To accomplish this goal, the therapist must be sensitive to the different self-presentation strategies used by dependent patients (Table 1.3) and alert to the possibility that dependency may sometimes be expressed in active—even aggressive—ways (chap. 4). In addition, the therapist should assess—either formally or informally—the degree to which the patient's behavior reflects one or more of the dependency subtypes described by Pincus and Gurtman (1995) and Bornstein and Languirand (2003). Finally, the therapist should determine whether the patient's problematic behavior reflects instrumental (i.e., task oriented) dependency, emotional dependency, or both. Only when the domains of dysfunctional responding have been identified can steps be taken to reshape these behavior patterns. In some ways, altering problematic dependency-related behavior is the most challenging task confronting the therapist. Longstanding interaction patterns can be difficult to modify, and as Caspi, Bern, and Elder (1989) pointed out, many dependent people have constructed social environments that sustain and reinforce their dependency. For this reason, marital and family treatment strategies are often useful in effecting long-term behavioral change in dependent patients (I discuss use of these treatment strategies in chap. 10). Three techniques, implemented in sequence, will help the patient find new, healthier ways of expressing underlying dependency needs: • Confronting inappropriate behavior. For some dependent patients, behavior change occurs through a series of successive approximations, with minor errors and manageable glitches along the way. For other patients—especially those with significant character pathology—behavior change is fraught with risk. Destabilizing a longstanding behavior pattern (even one that is largely dysfunctional) may lead to impulsive, poorly-planned action that creates additional problems at home and at work (see Ball & Young, 2000; Wasson & Linehan, 1993). For such patients, the therapist must be prepared to confront inappropriate behavior—gently, if possible, or more authoritatively if the behavior carries substantial risk. Although therapists are sometimes reluctant to challenge directly a patient's self-defeating patterns, this can be a very productive strategy when working with dependent patients, even during the initial stages of AN INTEGRATED TREATMENT MODEL
163
therapy. Many dependent patients are comforted by the therapist's willingness to intervene in a forthright manner; the key to effective confrontation in this context is to use it sparingly, so the therapist does not become trapped in the role of pseudo-parent. Encouraging healthy dependency. Because many dependent patients incorrectly assume that autonomy entails doing everything on one's own and resisting all external influences, it is important to reinforce the distinction between healthy and unhealthy dependency and between autonomy and independence. This learning process takes time and cannot be rushed, but it can be facilitated if the therapist models healthy dependent behavior during treatment. Occasional self-disclosure by the therapist can be useful in this regard. It is also important to explore the strengths and limitations of the patient's evolving interpersonal strategies as these emerge in vivo. Movement from unhealthy overdependence to healthy connectedness is rarely smooth, especially for patients who have never experienced intimacy and vulnerability in the absence of destructive overdependence. The therapist should warn the patient to expect setbacks, and encourage the patient to bring these experiences into therapy so errors can be discussed and behavioral adjustments made. Relapse prevention strategies. As Overholser (1997) noted, replacing dependency with autonomy requires that the patient acquire problem-solving skills and become confident enough to use these skills mindfully in lieu of reflexive help-seeking. To apply these skills effectively, the patient must not only practice them within and outside therapy, but must also find ways to avoid social "crumbling" in the face of challenge (Alden, 1989; Kantor, 1993). Whatever route therapy has taken, relapse prevention always involves revisiting these two issues as termination nears. Posttreatment setbacks can rarely be avoided completely, so the most useful approach is to strengthen the patient's problem-solving and anxiety-management skills during the latter stages of treatment.
EFFECTIVE USE OF THERAPEUTIC INTEGRATION STRATEGIES: A FRAMEWORK FOR THE PRACTITIONER Although the empirical literature on psychotherapy with dependent patients offers few firm conclusions regarding effective treatment strategies, one set of findings from Table 9.2 stands out: In two separate samples, Blatt 164
CLINICAL APPLICATIONS
(1992) and Blatt and Ford (1994) found that dependent patients did not respond well to traditional psychoanalysis but showed substantial improvement in psychoanalytic psychotherapy. This result—coupled with Juni and LoCascio's (1985) finding that dependent individuals preferred therapeutic interventions that afforded a high level of patient—therapist interaction— suggests that perceived closeness with the therapist is important to the dependent patient (see also Helweg & Gaines, 1977, for additional data bearing on this issue). From a psychodynamic perspective, these findings suggest that a relatively active stance on the part of the therapist (including some degree of self-disclosure) can strengthen the therapeutic alliance and facilitate treatment (Geller, 2003). From a behavioral perspective, these results indicate that therapist feedback can be an effective means of rewarding gains made by the patient within and outside therapy. The clinical literature on dependency—in conjunction with the empirical findings reviewed in earlier chapters—suggests several avenues through which treatment effectiveness can be enhanced. In the following sections, seven guidelines are offered to help the clinician develop and refine integration strategies for use with a broad array of dependent patients. Place the Patient's Current Treatment in the Context of Previous Treatment Experiences Patients re-enter therapy for a variety of reasons, and it is a mistake to assume that initiation of a new course of treatment invariably occurs in response to a flaw or failure in an earlier treatment effort. Many patients who have had one or more earlier treatment experiences actually select new therapeutic modalities with considerable forethought and only after obtaining a good deal of information regarding available treatment options. Gold (1994, p. 140) described this process well: The patient sifts through the positive changes accrued in initial therapies, looks at his or her unfinished issues, and synthesizes these perceptions against the backdrop of stored information about alternate therapeutic choices. The person's final decision is influenced by the felt need to address and correct limited experiences in previous therapy, by the need to expand earlier gains through work in different areas of experience, and by the awareness of assets and potentials that were not tapped by the previous treatment. To be sure, not all patients are so deliberate in their treatment choices, and some dependent patients seek therapy for protection and support rather than growth and positive change (see Bornstein, 1994c, 1998a; Coen, 1992). To optimize treatment effectiveness in therapeutic work with a "veteran" patient, clinician and patient should explore several issues during the assessment process and initial therapy sessions. These include AN INTEGRATED TREATMENT MODEL
165
• the patient's perceived gains in previous treatments • the patient's sense of how previous treatments were limited or incomplete • what the patient hopes to gain in this course of therapy • why these gains did not occur during earlier therapies • why the patient now feels ready to move forward in dealing with unresolved issues This information is not only critical in placing the current treatment episode in context, but may also help the patient recognize heretofore unacknowledged gains in earlier courses of therapy. Information regarding previous treatment successes and failures can also help the therapist (a) gauge the patient's expectations for therapy (both expressed and unacknowledged); (b) determine the adequacy of patient-therapist and patient-therapy fit; and (c) decide whether it is advisable to undertake treatment with this patient at this time. Use Beitman's Four-Stage Process Model to Guide Your Integration Efforts Beitman's (1992) conceptualization of psychotherapy as a four-stage process (i.e., engagement, pattern search, change, and termination) provides a useful framework for integrating therapeutic strategies over the course of treatment. Although it is not possible to draw sweeping generalizations regarding use of different therapeutic models at specific stages of treatment, Table 9.4 summarizes some guiding principles to shape this process. These principles can be adapted to accommodate specific patient needs and therapeutic goals. Note that while there is considerable flexibility in this framework— both with respect to use of different modalities and timing of therapeutic shifts—the model is based on a progression from insight and change to solidification of gains and prevention of relapse. Thus, for many dependent patients, psychodynamic and experiential interventions will be most useful early in therapy; cognitive and behavioral interventions may be particularly helpful during therapy's middle and latter stages (see Weinberger, 1996).8 Be Prepared to Change Expectations as Therapy Progresses As Bornstein and Bowen (1995) noted, it is important that the therapist gradually shift her expectations regarding dependency and autonomy 8
As is true for patients with other personality styles, these therapeutic techniques must be modified when used with character disordered patients. The maladaptive defenses and inflexible coping strategies associated with many forms of character pathology—coupled with the risk that a patient may act out in self-destructive ways—requires that the therapist be particularly sensitive to negative therapeutic effects and prepared to re-engage the regressed patient using a less challenging approach, if necessary.
166
CLINICAL APPLICATIONS
TABLE 9.4 Integrating Over the Course of Therapy: Core Strategies Therapy stage Engagement
Pattern search
Change
Termination
Integration strategy Begin with modality most appropriate for the patient's presenting problem; take into account patient's past therapy experiences; allow the patient to guide—but not dictate— shifts to other therapy models Use multiple therapeutic frameworks (including psychodynamic and humanistic-existential) to enhance insight and deepen experiential learning; shift frameworks within as well as between sessions Begin effecting change using the most productive frameworks from Stage 2; gradually incorporate other models as needed; tie new interventions to specific change processes (i.e., giving up old patterns, building new patterns, maintenance of new behavior) Shift from insight and experiential learning to solidifying gains and enhancing coping/problem-solving skills; incorporate cognitive and behavioral relapse-prevention techniques
during the course of treatment. Because patient dependency can strengthen the therapeutic alliance, the therapist should be relatively permissive regarding expressed dependency early in therapy, while still setting firm limits regarding after-hours contact and other boundary issues (Emery & Lesher, 1982; Leeman & Mulvey, 1975; Lorr & McNair, 1964a). After the patient has come to trust the therapist and begins to experience the therapeutic milieu as safe and predictable, the therapist should encourage the patient to take more risks, both within and outside the clinical setting. At this point the patient's dependency on the therapist can be used to facilitate autonomy. A second area in which therapist expectations should shift during therapy involves domains of patient functioning. Data gathered during the assessment process can provide important preliminary information regarding those areas wherein positive change is likely to be difficult (e.g., romance, parenting, work). Because change takes place within an evolving interpersonal context, however, it is not possible to predict where obstacles may emerge and when resistance—either patient-based or milieu-based—may begin to build. The therapist must be prepared to shift the focus of therapy from one domain to another as insight accrues and behavior change occurs. For many dependent patients, gains in one domain facilitate progress in other areas. For some dependent patients, gains may be limited to one (or a few) domains during a single course of treatment; growth in other areas—if it occurs at all—will come later. Use Change in One Area to Leverage Change in Others As Figure 1.1 illustrated, dependency-related cognitions, motivations, behaviors, and emotional responses interact in predictable ways (see also AN INTEGRATED TREATMENT MODEL
16 7
Bornstein, 1993, 1996a, for detailed discussions of this process). Altering one component of this system will have an impact on other aspects of the system. Thus, therapeutic interventions should be selected using two criteria: (a) how likely they are to effect change in the target area and (b) the ways in which they influence other aspects of dependency. With this framework in mind, two levels of intervention will be particularly fruitful: • Dysfunctional cognitions. Dependency-related cognitions play a key role in driving dependency-related motivations, which in turn shape dependency-related behaviors and emotional responses. Thus, altering the dependent patient's biased beliefs about the self and others will have a particularly powerful "spillover" effect on other aspects of functioning. • Dependency related emotions. As Figure 1.1 shows, increases in dependency-related emotions (e.g., performance anxiety, fear of negative evaluation) activate feedback loops which ultimately lead to increases in dysfunctional thinking, dependency-related motivation, and dependent behavior. Thus, techniques which dampen these emotional responses can indirectly produce positive change in the other three areas. Adapt Treatment Integration Strategies to Dependency Subtypes Because different subtypes of dependency are associated with contrasting self-presentation styles, intrapsychic dynamics, vulnerabilities, and strengths, different integration strategies are needed for different dependent patients. Tables 9.5 and 9.6 offer suggestions for subtype-specific integration strategies within the context of Pincus and Gurtman's (1995) three-vector model (Table 9.5) and Bornstein and Languirand's (2003) four-pattern model (Table 9.6). Note that for each dependency subtype, engagement techniques are adapted to accommodate the patient's preferred coping style and primary self-presentation strategy. Techniques for effecting change (Stage 3) are explicitly linked to the issues uncovered during pattern search (Stage 2). Termination goals—though they differ from subtype to subtype—invariably involve solidifying gains and providing the patient with tools and coping strategies that help maintain new behavior patterns and minimize the possibility of relapse. Anticipate and Counter Resistance to Maintain Growth and Gain Many dependent patients enter therapy feeling conflicted: Although they seek guidance, reassurance, and support, they are fearful of being rejected or abandoned by the therapist. Not all dependent patients verbalize these fears, but virtually all dependent patients experience them. As depen168
CLINICAL APPLICATIONS
TABLE 9.5
Integrating Over the Course of Therapy: Exploitable, Submissive, and Love Dependency Core issues Therapy stage Engagement
Pattern search
Change
Termination
Exploitable dependency
Submissive dependency
Love dependency
Help patient describe Actively challenge Focus on patient's problem history patient's erroneous goals for therapy; and understand assumptions/ challenge/reframe progression/ beliefs; explore external escalation of dysfunctional attributions for dysfunctional relationship problems coping dynamics Analyze dependency- Identify problem Explore sequence of related feelings domains and attachment/love and fears; explore surface reasons for objects, including factors underlying submissive/yielding parents, friends, tolerance of behavior; explore and romantic maltreatment and hidden payoffs in partners; make compulsive carekey relationships explicit core seeking relationship themes Enhance selfMake explicit Explore beliefs and esteem; reduce unacknowledged fantasies regarding self-defeating anger; alter intimacy, love and behavior; focus on passive selfmerging; help choice and presentation style; patient maintain responsibility for build assertiveness self-focus in actions skills emotion-laden situations Emphasize problem- Focus on relapse Strengthen anxietysolving and prevention management skills generation of techniques alternative responses/ solutions
Note. The core characteristics of exploitable, submissive, and love dependency are discussed in chapter 5 and summarized in Table 5.4. Information regarding assessment of these dependency subtypes is in chapter 7.
dent patients gain confidence that the therapist will not abandon them, they tend to become overly trusting and—on occasion—inappropriately intimate or seductive (Hopkins, 1986). Some dependent patients maintain this position over time; others (e.g., those with conflicted dependency) may move back and forth between closeness and distance—even within the same session. For many dependent patients, these behavioral and attitudinal shifts become more pronounced as termination approaches. Although the patient's dependency on the therapist can facilitate progress during the early and middle stages of treatment, it is important to recognize when the patient's trust and intimacy are no longer strengthening
AN INTEGRATED TREATMENT MODEL
169
>. f* s -- i «j l " ? l § I! *i!o I §«s foi ? sc ^ s - »£e-|^|| £
| 1-1 .^p lfi|3Ef c
•S
ed
73
"m -J= tn E 73 ~
c CD £ =!
O~
o O
o
cc E o> >- o
= ~ X: 0) -P g. C
| §,« 1|
c c •»-" £
C D f f i C C D—
^
: IP i!i1igi|& If 111 if
I ill O
c o> -^ t H c E _. c o " c o n c . * o c o c 2 ~ o o 'to — co g-cO.Scoajjjja-a.a.ScOcoo-QjoEcooo 2 LJ 3 DC
e |» gf^a*§1-3^ H -5. - ^,^^
^ ^ ?-§ g
|
0
|
1 5
g ° ^ § f ^ «§'•§&»
go. 2^
CD • s ^ t s f S f e g S I g»i8sfe
5 &T3.S £ § 7 3 ? 8 r a S | ^ c g - g S ~32--£«>,sggS2)lSo--S ^-o x 9 - c o o ^ ^ = Q.-2: co o ) §)'•= o .E a. OT to CD CD-55C^-£^(Da)i:-yg o3oC0(D •fi 3 o S u S Q . E - a - D c o g - S : Q.2 0)73 Q
g
<
cp|(§
|
f
OiO-o
<->>,..~
ujo|
_JO)CO
?Si 0X
^-
.Sco.Eo
§§£«
M ? ^^lilfl ift^i §--^ l ? l l i « » s " Si'-S^^^ ^ssa ^lUllll IpMl ?|iff 1 5*^8&82.a|lifEg 2Sg
^« ? xi
0:g
§ ^^ TO0
g g cl
2
I >,
a> 5" £ .!=
LU
fig coco
a )
.
s S f e?. E ^>-gc §aS§ OJCOQ.S ^"00)2 o co -D a.
(L
g ..1 o, o'-cuo^-£
-£|S-§,8S« § o.£o = - w c S>fflCQ)CD-?:S
"2 g - o « l S 8 ^ ^ f ^ g S g f f i ^ c
II S g-
S
5
?S
">co.&
S§-^
S -Q. c
>°£
I =
.§
-°
i
f1 CO "5
D)° CD X
!E w" — C/5 ^
fI
s| c: T3
1 8O -c^ E§C D|-^S ; .
M-^-o
s a. ;S.-D »- i§=-o •55 r c D•§ . ± a. : S® - QS ) ^x0 -S - s ^3T•* OC ' - £Sc o I Q-«og§°-^c?>-i°-^ogcoS S"a5c^-^ajo(DcOQ.cDEgg75 o S = £ g-.c o c > - ^ - o ( D S c o S
< CD
0)
*. E
CO
£ CD
CO O)
CD JS
LU
D-
I-
c
E 8 ° <« ^ ";« » 'o> ™ o=Q.8gcSxS.^ -jjcooS-o
DC
-C
^_
^
X
Iff!, fill!
CO
S I
I a
170
LU
?§!«
'5'^=cS
8
E
to
CLINICAL APPLICATIONS
a
C CO
.c
O
o *J
l
c E
0) I-
the therapeutic alliance but have instead become a form of resistance. Studies show that many dependent patients consciously or unconsciously subvert therapy as termination nears in an attempt to preserve the relationship (see Greenberg & Bornstein, 1989; Hopkins, 1986; Lower, 1967; Pyke, 1982; Snyder, 1963). Termination may be particularly challenging for the dependent patient, but it can also be an opportunity to reinforce previous gains and set the stage for additional growth and positive change. No single intervention can be used to ameliorate termination-related resistance in dependent patients, but several strategies, alone and in combination, are useful. These include • discussing termination issues periodically throughout therapy to help the patient prepare for the event • exploring the patient's termination fears and fantasies, using reflective statements to illustrate how these fears may be exaggerated • interpreting termination concerns in the context of past relationships and previous therapy experiences • using transitional objects (e.g., business cards) to lessen separation anxiety as termination nears (a strategy that is especially useful with lower-functioning patients) • rehearsing and reinforcing problem-solving strategies that the patient can use to cope posttreatment
AN INTEGRATED TREATMENT MODEL
1 71
10 SPECIALIZED TREATMENT ISSUES
For many dependent patients, individual psychotherapy by itself is sufficient to effect long-term positive change. For other patients, alternative treatment modalities may be needed in addition to—or in lieu of—individual therapy. The clinician's decision regarding use of adjunct or alternative treatments involves many factors, including (a) the nature of the patient's pathology (both type and severity); (b) the patient's underlying personality structure (including defenses and coping style); (c) past treatment history (both successes and failures); (d) personal history (e.g., family dynamics, cultural background); and (e) current life circumstances (e.g., salient stressors, availability of social support). To implement alternative treatment strategies effectively, the clinician must be flexible as therapy proceeds. Changing circumstances during the course of treatment may dictate that supportive interventions be added or removed. In this respect, use of adjunct treatment strategies involves a process similar to that of Messer's (1992) assimilative integration: In both situations, an overarching therapeutic modality is enhanced by the addition of one or more supplementary interventions. In this chapter, I discuss specialized treatment issues that arise in work with dependent patients. I begin by discussing contexts and settings wherein dependency-related concerns may become salient, even for patients who do 173
not enter treatment with strong underlying or expressed dependency needs. I then consider alternative treatment modalities that—alone or in combination with individual psychotherapy—help alter longstanding, problematic dependency. Finally, I discuss some special challenges associated with dependent patients and offer guidelines the practitioner can use to maximize treatment effectiveness and minimize the risk of a negative outcome.
CONTEXTS AND SETTINGS Certain treatment settings foster dependency by reinforcing passivity, helplessness, and other forms of unhealthy dependent behavior. In these settings it is important that the clinician be sensitive to the possibility that well-intentioned caregiving can lead some patients to become overly attached and consciously or unconsciously undermine treatment progress to avoid ending the relationship. Inpatient treatment may be particularly problematic in this regard, because hospital regimens cause many patients to experience a loss of control and a decrease in self-efficacy. Although some patients find this loss of control aversive, others—especially those predisposed to look to other people for guidance and support—find it comforting (Greenberg & Bornstein, 1989; Raps, Peterson, Jonas, & Seligman, 1982). Various forms of rehabilitation (e.g., physical, psychosocial, vocational) and long-term care can also foster unhealthy dependency in patients who are predisposed to assume a helpless stance in the face of challenge or stress (Booth, 1986; Kilbourne & Kilboume, 1983). In the following sections I discuss concerns that arise in treatment of dependent patients in inpatient settings, rehabilitation settings, and time-limited therapy. Inpatient Treatment Several studies have found that dependent patients remain in inpatient treatment longer than do nondependent patients with similar symptom patterns and demographic profiles (Brown & Rawlinson, 1975; Greenberg & Bornstein, 1989; Overholser, Kabakoff, & Norman, 1989). These findings hold for both medical and psychological disorders, and they are not only statistically significant, but clinically significant as well. For example, Greenberg and Bornstein (1989) found a 68% increase in duration of inpatient psychiatric treatment in dependent patients relative to nondependent patients with comparable diagnoses. Brown and Rawlinson (1975) obtained similar patterns in patients recovering from heart disease. These inpatient treatment results contrast with those obtained in studies of dependency and outpatient treatment, wherein researchers typically report no relationship (Hiler, 1959; Lorr & McNair, 1964a) or a modest relationship (Stamler & Palmer, 1971; Salokangas, Rakkolainen, & Lehtinen, 1 74
CLINICAL APPLICATIONS
1980) between level of patient dependency and length of therapy. The stronger dependency-duration relationship for inpatient than outpatient treatment may reflect the fact that terminating inpatient treatment entails a greater shift toward autonomy on the part of the patient than does terminating outpatient treatment. This is especially true when a hospital regimen is behaviorally restricting or involves invasive testing or intervention.1 Rehabilitation and Long-Term Care A similar "learned helplessness" dynamic can lead to increases in dependency during psychosocial, occupational, or physical rehabilitation. As Goldin (1972) noted, this problem is common in such settings, especially among patients who are predisposed to passivity before entering rehabilitation. This dynamic is exacerbated if progress is slow and positive change does not occur within a relatively short time. In this situation, the patient may begin to regard rehabilitation as an enduring support system rather than a time-limited means of increasing adaptation and autonomous functioning (see Alsop, 1984; Cohort, 1980). A parallel dynamic occurs in long-term care facilities, many of which have traditionally (albeit unintentionally) been constructed to encourage passivity, conformity, and compliance among residents (Bakes, 1996; Bornstein & Languirand, 2001). Studies of changes in patient behavior during the course of long-term care suggest that inadvertent reinforcement of patient dependency can undermine progress and mitigate gain (Booth, 1986; Eddington, Piper, Tanna, Hodkinson, & Salmon, 1990). Langer and Rodin's classic (1976) results further indicate that in nursing home settings, reinforcement of dependency can actually lead to increases in depression, illness rates, and mortality rates.2 Time-Limited Therapy Whereas inpatient, rehabilitation, and long-term care may inadvertently foster dependency, a different challenge arises in time-limited therapy. The constraints imposed by managed care organizations may prevent under'ln interpreting these results, it is important to keep in mind that extant studies of the dependencytreatment duration relationship were conducted before the widespread implementation of managed care and other cost containment strategies. Thus, dependency-related differences in treatment duration would likely be smaller today (or obviated completely by external factors). 2 The flip side of Langer and Rodin's (1976) "helplessness effect" is that it is relatively easy to alter behavioral contingencies to foster self-efficacy and autonomy in nursing home residents. Langer and Rodin found that seemingly minor interventions (e.g., having residents listen to a speech on the importance of personal responsibility, giving each resident a houseplant to nurture) enhanced social interaction, elevated mood, improved health, and decreased mortality rates. Other efficacy-promoting changes in nursing homes (e.g., allowing residents to dress as they wish, decorate their personal space, and choose mealtimes rather than adhering to a rigid schedule) have also been shown to produce positive effects on health and behavior (Bornstein & Languirand, 2001). SPECIALIZED TREATMENT ISSUES
175
lying dependency needs from being expressed in a healthy, adaptive way. This can undermine the therapeutic relationship, interfere with treatment, and have the paradoxical effect of increasing the patient's feelings of helplessness and vulnerability even while these feelings remain unverbalized. The negative impact of time-limited therapy on patient dependency stems from two sources: • Therapist concerns. When limitations on treatment length are externally imposed and based on financial rather than therapeutic considerations, the practitioner may be concerned that patient dependency will make termination difficult—even impossible (another manifestation of the "fantasy of permanence" discussed in chap. 7). As a result, the therapist may consciously or unconsciously engage in various distancing behaviors that prevent the patient from expressing feelings of helplessness and vulnerability (Kantor, 1993; Sperling et al., 2000). • Patient defenses. Many patients—especially those who have difficulty trusting others—take time to open up, let down their guard, and form a productive therapeutic alliance (see Bornstein, 1994c, 1998a; Coen, 1992). Even the most dependent patients may take several sessions (sometimes longer) to forge a productive therapeutic bond, particularly if their trust has been violated in one or more earlier relationships. Knowing from the outset that the number of therapy sessions is limited can cause anxious patients to engage in their own distancing maneuvers and avoid dealing with some important dependency-related concerns. ALTERNATIVE TREATMENT MODALITIES Because many dependent individuals construct interpersonal milieus that reinforce and propagate their dependency, adjunct marital or family therapy can often be useful in working with dependent patients. Group therapy provides dependent patients with feedback regarding the effects of their selfpresentation style on other people, and opportunities to test alternative ways of responding in a safe setting. Finally, pharmacotherapy may be necessary for dependent patients with comorbid depression, anxiety, or thought disorder. In the following sections I discuss use of alternative treatment modalities in psychotherapeutic work with dependent patients. Marital Therapy Given the intra- and interpersonal dynamics of dependency, certain problems are likely to arise within a longstanding monogamous relationship. 176
CLINICAL APPLICATIONS
These include pervasive insecurity (Ojha & Singh, 1985), pathological jealousy (Buunk, 1982), and various forms of manipulative acting out (e.g., illness exaggerations, breakdown threats; Bornstein, 1993, 1998d). The first two responses—insecurity and jealousy—reflect the dependent partner's fear of abandonment; the third response—acting out—reflects the dependent partner's frantic efforts to avoid this abandonment. Ironically, all three reactions are likely to worsen—not lessen—over time, as the dependent partner becomes increasingly invested in maintaining the predictable pattern that has come to characterize the relationship. Although specific points of contention and areas of conflict differ from couple to couple, two issues are key in marital therapy when one or both partners are highly dependent: • Insight. Because dependent people have varying degrees of insight regarding their underlying dependency needs, assessing each person's perceptions of their own and their partner's behavior is an important first step in marital therapy. It is tempting to believe that the dependent person's partner will have some insight regarding the impact of their spouse's dependency, but this is not always the case. The partner may be so motivated to perceive the dependent person in a particular way that she is unable to gauge this person's behavior accurately (Mongrain, Vettese, Shuster, & Kendal, 1998). Thus, couples wherein one partner is highly dependent can be classified into three categories with respect to insight. Some couples have considerable shared insight regarding the impact of the partner's dependency (though they may not see how other relationship dynamics contribute to this pattern). Some couples have differential insight, with one partner (usually the less dependent one) having greater awareness than the other. Finally, some couples engage in mutual synergistic denial, in silent (often unconscious) agreement that neither partner will acknowledge the negative impact of dependency on their relationship. • Coping. Individuals react to excessive dependency in myriad ways, and it is necessary to understand how the more autonomous partner copes with the dependent person's efforts to minimize distance and preclude rejection (see Vettese & Mongrain, 2000). Bornstein and Languirand (2003) identified four common responses to a romantic partner's dependency— codependency—enabling, counterdependency, authoritarianism, and denigration-devaluing—each with its own implications for relationship dynamics. By exploring the partner's habitual responses to the spouse's dependent behavior, selfdefeating patterns can be identified. Key features of these four SPECIALIZED TREATMENT ISSUES
177
TABLE 10.1 Common Patterns of Responding to a Romantic Partner's Dependency Response pattern Codependency Counterdependency Authoritarianism
Denigration
Characteristic behaviors Nurturing; caregiving; reinforcement of dependent behavior Distancing; emotional and physical withdrawal Rule-setting; decisionmaking; controlling behavior Criticism; belittling; infantilization
Primary risks Propagation of dependency to maintain ongoing relationship pattern Neglect; infidelity; relationship dissolution Indirect propagation of dependency via imposition of external structure Physical, sexual, or emotional abuse/neglect
Note. Detailed discussion of these four patterns may be found in Bornstein and Languirand (2003).
patterns, along with primary relationship risks, are described in Table 10.1.3 Family Therapy Chapter 3 discussed the impact of family roles and alliances on the development of dependency early in life. This framework can also be used to explore the impact that each individual's dependency has on the current family system. One or more family members may be invested in maintaining other family members' dysfunctional behavior because of the payoffs this behavior provides and the protection it affords (Benjamin, 1996; Bowen, 1978). To understand the place of individual dependency within the family system, three issues must be addressed: • Roles. Sometimes a dependent person assumes an overtly dependent role within the family, presenting himself as weak and vulnerable and openly soliciting support and care from other family members. At other times a dependent person may assume a more subtle, less easily identified dependent role. When this occurs, the person still solicits extra help and care, but does so indirectly, without acknowledging the care-eliciting behavior (e.g., by feigning or exaggerating illness). The person ex3
ln this context, it is important to note that many spouses are ambivalent regarding a partner's dependency, sometimes feeling overwhelmed by the partner's demands and at other times deriving gratification from being adulated and needed (Rathus & O'Leary, 1997). In such couples the partner's reactions may vary considerably, with dependency bringing a negative response in some contexts and a reinforcing response in others. This variability in responding has two negative consequences. First, it is confusing and upsetting to the dependent person, increasing his or her anxiety and making dependent behavior more likely. Second, as noted in chapter 8, this variable reinforcement pattern renders dependent responding highly resistant to extinction, ultimately making behavior change more difficult.
I 78
CLINICAL APPLICATIONS
hibiting this behavior may be completely unaware of it, and on occasion a hidden dependent role can be assumed by a person who is ostensibly highly autonomous (e.g., a successful executive who is "helpless" when it comes to parenting and household tasks; see Kaplan, 1983). A common family dynamic in these situations involves denial: Other members of the system do not label the hidden behavior accurately in order to protect the family member's hidden dependent role.4 Alliances. A family member's dependency—whether overt or hidden—usually brings rewards. More often than not, these rewards extend beyond the dependent person. When other family members derive benefit from one person's dependency, they may form an alliance to propagate the reward-producing behavior. For example, a mother's indirect expression of dependency (e.g., feigned illness) can allow her adolescent daughter to shift into a caregiving role, thereby avoiding some difficult challenges related to her own burgeoning sexuality. The daughter may eventually become as invested as the mother is in maintaining the mother's dependent behavior. At the same time, other family members benefit (e.g., father may avoid confronting marital difficulties by devoting himself to his work, son may gain freedom he would not otherwise have because mother and father are preoccupied with their dysfunctional roles). Loci of resistance. To the degree that family members derive benefit from one person's dependency, loci (or "pockets") of resistance are likely to form, each of which serves to maintain the status quo and undermine change (see Sloman, Atkinson, Milligan, & Liotti, 2002). This resistance may be largely (or even completely) unconscious, and a key task for the therapist is to make hidden loci of resistance explicit. Once this occurs, it becomes easier to identify hidden alliances that are working to maintain one or more family members' dysfunctional behavior and easier to alter aspects of the system to facilitate growth and change. Loci of resistance can be brought into the open by exploring the benefits that different family members derive from one person's dysfunctional behavior. As change begins to occur within the system, resistance will increase, and the strategies used by different members of the system can be interpreted (Minuchin, 1974).
4
Some writers have suggested that men's expressions of dependency within the family are often hidden and indirect, and therefore not labeled as dependency (Cadbury, 1991; Kaplan, 1983). This may account in part for the frequently observed gender differences in expressed—but not implicit— dependency strivings among adults in Western societies (Bornstein, 1995c). SPECIALIZED TREATMENT ISSUES
179
Group Therapy Of two extant studies assessing the impact of patient dependency on response to group therapy, one (Zettle, Haflich, & Reynolds, 1992) reported a positive effect, and the other (Chambless, Renneberg, Goldstein, & Gracely, 1992) reported no effect. These contrasting results may be due in part to differences in therapeutic modality: Zettle et al. examined the effect of patient dependency on response to cognitive group therapy, whereas Chambless et al. assessed the effect of dependency on response to behavioral group therapy (see Table 9.2 for descriptions of these investigations). Group therapy can have a positive impact on the dependent patient, but it involves certain risks as well. By being aware of these issues, the clinician can work to maximize benefit and minimize the possibility of unintended negative effects. • Benefits of group treatment for dependency. As noted earlier, a primary benefit of group treatment is that other group members provide feedback regarding the dependent person's interpersonal style (Chazan, 2001). Therapy groups are ideal in this regard because the same strategies used by patients to get their dependency needs met in vivo tend to be recapitulated in the group setting. Groups also provide opportunities for role play and the development of social skills to facilitate behavior change. In this context, the low-risk environment of the group may help certain shy or timid patients attempt new behaviors that would be impossible in social or work settings. Group members can provide support—and peer pressure—to help a recalcitrant patient take steps to bring about change. • Risks of group treatment for dependency. A major risk of group treatment for dependency is the formation of unhealthy alliances that undermine change (Ammon, 2002). In certain instances, these alliances result from the transfer or extension of dependency from a real-world caregiver to a caregiver within the group. In other instances, two (or more) dependent patients may ally to propagate each other's longstanding behavior patterns. A second major risk of group treatment for dependency is that one or more dependent patients may begin to compete for the attention and favor of the therapist. A sibling rivalry dynamic can ensue, with group members undermining each other's efforts to further their status in the eyes of the therapist. If other members of the group do not interpret this behavior once its purpose becomes apparent, the therapist should. There is no ideal composition of personality and pathology configurations within a group aimed at ameliorating problematic dependency. The 180
CLINICAL APPLICATIONS
advantages of a heterogeneous group are that the dependent patient will have the opportunity to observe and model other interaction styles and obtain feedback from nondependent as well as dependent group members. Advantages of a focused group composed entirely of dependent patients include increased empathy, greater opportunities for support, and inspiration that comes from observing progress in group members who overcome similar problems and challenges. More important than the composition of the group with respect to dependency may be the composition of the group with respect to overall level of adjustment (Ammon, 2002; Schermer & Pines, 1999). In a therapy group with a small number of character-disordered patients, the higher-functioning patients can—with considerable help from the therapist—usually set limits and keep the group focused. When character-disordered patients outnumber higher-functioning patients—especially if the character-disordered patients have significant borderline, antisocial, or narcissistic traits—so much energy may be spent containing these patients' pathology that little energy remains to effect therapeutic change. Pharmacotherapy
Many patients enter outpatient therapy having already begun a regimen of psychotropic medication, often initiated by their primary-care physician. In inpatient settings, O'Neill and Bornstein (2001) found that dependent psychiatric patients receive nearly 50% more medication prescriptions than do nondependent patients with similar demographic and diagnostic profiles. These differences occurred for antidepressants, anxiolytics, and neuroleptics, but not lithium (which was prescribed for dependent and nondependent patients at equal rates). It may be that the increased number of psychotropic medication prescriptions received by dependent patients reflects increased symptom severity in these patients (after all, longstanding psychological or physical symptoms often lead to increases in expressed dependency). Alternatively, the dependent patient's generalized help-seeking tendencies may cause physicians to prescribe medications more frequently, in response to the patient's persistent complaints (see Tyrer, Mitchard, Methuen, & Ranger, 2003). Whatever the cause of dependent patients' increased medication use, it is important that the clinician be aware of the impact of drug treatment on dependent responding and the impact of patient dependency on drug efficacy. Table 10.2 summarizes the available data on the effects of psychotropic medication on dependency. Although results in this area have been somewhat inconsistent, findings suggest that both tricyclic antidepressants and selective serotonin reuptake inhibitors (SSRIs) can diminish selfreported dependency levels in inpatients and outpatients. These results doveSPECIALIZED TREATMENT ISSUES
181
TABLE 10.2 Effects of Psychotropic Medication on Dependent Responding Study
Participants
Black et al. (1996)
44 outpatients with panic disorder receiving 8 weeks of daily fluvoxamine Fava et al. 83 outpatients receiving 8 (1994) weeks of daily fluoxetine for major depression Lauer (1976) 74 inpatients and outpatients receiving daily imipramine or nortriptyline for 3-60 months (mixed diagnoses) Moore & 119 inpatients and outpatients Blackburn receiving 16 weeks of daily (1996) (unspecified) antidepressant treatment Peselow et al. 47 hypomanic outpatients (1995) receiving daily lithium for 4-8 weeks (some patients received other medications in addition to lithium) Rector et al. 51 depressed outpatients (2000) receiving 12 weeks of daily treatment with an MAO inhibitor, tricyclic antidepressant, or SSRI Zaretsky et al. 142 depressed outpatients (1997) receiving 8 weeks of daily fluoxetine
Outcome No change in PDQ dependency scores from pre- to posttreatment Nonsignificant trend toward reduction in DPD symptoms during the course of treatment Significant decrease in selfreported dependency during the course of treatment Nonsignificant decrease in SAS sociotropy (dependency) scores from pre- to posttreatment No change in dependency levels based on patient and informant reports Nonsignificant decrease in DEQ dependency scores from preto posttreatment Significant decrease in DAS dependency scores from preto posttreatment
Note. PDQ = Personality Diagnostic Questionnaire (Hyler et al., 1988); SAS = Sociotropy-Autonomy Scale (Beck et al., 1983); DEQ = Depressive Experiences Questionnaire (Blatt et al., 1976); DAS = Dysfunctional Attitudes Scale (Weissman & Beck, 1978). Effect size estimates are not included in this table because the majority of investigations did not provide enough data to calculate a medication efficacy effect size.
tail with those of myriad investigations that have shown that dependency and depression levels covary, with dependency levels increasing as depression worsens and decreasing as depression remits (Joffe & Regan, 1988; Klein, Harding, Taylor, & Dickstein, 1988). The observed effects of antidepressant medications on patients' dependency levels may be mediated by changes in mood. Although there have been numerous studies of the effects of medication on patient dependency, there has been only one investigation of the moderating effects of patient dependency on psychotropic medication response: Peselow, Robins, Sanfilipo, Block, and Fieve (1992) found that dependent outpatients showed a significantly worse response than matched autonomous outpatients following 3 to 6 week regimens of imipramine, fluoxetine, paroxetine, or clovomaxine. These results contrast with studies of the moderating effects of dependency on psychotherapy outcome (sum182
CLINICAL APPLICATIONS
marized in Table 9.1), which found no strong or consistent impact of dependency on treatment efficacy.5
SPECIAL TREATMENT CHALLENGES Most clinicians do not associate patient dependency with lethality or other forms of self-destructive acting out. However, studies indicate that dependent patients may be at elevated risk for abuse (both victimization and perpetration), illness (both minor and major), and suicide. In the following sections, I discuss these special treatment challenges. Dependency and Abuse In chapter 4,1 described the links between dependency and abuse, noting that this relationship is complex and somewhat counterintuitive: Not only are dependent persons at increased risk for being victimized (Rusbult & Martz, 1995; Watson et al, 1997), but they are also at increased risk for perpetrating abuse themselves (Holtzworth-Monroe, 2000; HoltzworthMonroe et al., 1997). Several dependency—abuse pathways are germane in this context: • Victimisation pathways. High levels of trait dependency lead to increased risk for physical and sexual abuse, in part because dependent individuals fear relationship disruption and have difficulty setting limits on their partner's inappropriate behavior (Bergman, Larsson, Brismar, & Klang, 1988; Kalmuss & Straus, 1982). Although studies to date have focused exclusively on male perpetrators and female victims, it is likely that highly dependent men are at increased risk for victimization as well. • Perpetration pathways. Dependent women are at increased risk for engaging in child abuse (de Young & Lowry, 1992; Kertzman, 1980), because they (a) have difficulty modulating negative affect and (b) fear the marital conflict that might result from a misbehaving child. Men with high levels of trait dependency are at increased risk for perpetrating spouse abuse, although in this context the key dynamic involves jealousy and control (Holtzworth-Monroe, 2000; Holtzworth-Monroe, Stuart, & Hutchinson, 1997). Threatened by their partner's autonomy 'Although there has been only one study assessing the impact of patient dependency on response to electroconvulsive therapy (ECT) for severe, intractable depression, the results of this investigation suggested that dependency is unrelated to ECT efficacy immediately posttreatment and 6 and 12 months later (Casey & Butler, 1995). SPECIALIZED TREATMENT ISSUES
183
and anxious regarding the possibility of rejection or abandonment, dependent men may physically abuse their partners as a means of intimidating and controlling them. Stress and Illness Bernstein's (1998e) meta-analytic data (discussed in chap. 4) confirm that dependent persons are at increased risk for illness and disease. Studies further indicate that the dependency-illness link is mediated by interpersonal stress and relationship disruption. Bornstein (1995b, Study 1) tracked illness rates in dependent and nondependent college students over a 3-month period and found that only those dependent students who reported high levels of interpersonal stress showed significantly increased illness rates. In a modified replication of this investigation, Bornstein (1995b, Study 2) found that social support attenuated the negative effects of interpersonal stress in dependent individuals: Dependent students with high levels of interpersonal stress and low levels of social support showed increased illness rates over 3 months, whereas dependent students with high levels of interpersonal stress and high levels of social support showed illness rates that were comparable to those of the nondependent students. For the clinician working with healthy adolescents or adults, these dependency-illness findings may be of theoretical—but not practical— interest. However, for the clinician working with ill or elderly patients, these results have important practical implications. They suggest that relationship conflict or disruption can have a significant negative impact on dependent patients' health status, exacerbating existing illnesses (e.g., heart disease, cancer, AIDS, autoimmune disorders) and—if interpersonal stress is sustained or severe—potentially initiating new disease processes. The clinician working with dependent ill or elderly patients should assess the patient's interpersonal stress levels periodically (see Bornstein, 1995b, for a discussion of procedures in this area). For those dependent patients who are severely or chronically stressed, use of stress-management techniques may be warranted.6 Dependency and Suicide As Table 10.3 shows, there have been 11 studies of the dependencysuicide link during the past several decades. These investigations used a variety of participant samples (e.g., inpatients, outpatients, college students), dependency scales (e.g., self-report, interview), and suicide indices (e.g., sui'Dependency may also have a negative impact on the course of those chronic medical conditions (e.g., diabetes) where patient self-care plays a key role in symptom management. This negative impact reflects the dependent patient's longstanding habit of looking to others for support, but it may also be due in part to the increased stress (and decreased immunocompetence) that result from minor relationship conflict and the perceived threat of relationship disruption.
184
CLINICAL APPLICATIONS
cide attempt, completed suicide). Despite these differences, studies in this area produced fairly consistent results, suggesting that dependency is associated with moderately elevated suicide risk. These findings are particularly compelling because they are not limited to retrospective and concurrent investigations, but also hold for prospective studies wherein dependency levels were assessed before suicide data were collected (Epstein, Thomas, Shaffer, & Perlin 1973).7 It is noteworthy that the only sample wherein dependency levels were assessed via projective test (Bornstein & O'Neill, 2000) was also the only investigation to yield an inverse relationship—albeit a very modest one— between dependency and suicide. Moreover, when self-report dependency levels were assessed in the same participants, a significant, positive dependency-suicide link was obtained (r = .23). Apparently, the dependencysuicide relationship holds for self-attributed—but not implicit—dependency needs. Bornstein and O'Neill's (2000) results point to the importance of distinguishing expressed and underlying dependency urges when predicting dependency-related behavior in clinical settings.
EFFECTIVE USE OF ALTERNATIVE TREATMENT STRATEGIES: A FRAMEWORK FOR THE PRACTITIONER Effective use of alternative treatment techniques in clinical work with dependent patients requires that the practitioner monitor changes in the patient's functioning and interpersonal milieu throughout treatment, adjusting intervention strategies as circumstances dictate. This monitoring process can include periodic formal testing, informal observation, or a combination of the two. Six guidelines help maximize the effectiveness of alternative treatment techniques. Be Alert for Signs of Dependogenic Parenting As discussed in chapter 3, parenting practices can play a key role in fostering dependent behavior in children. Certain "dependogenic" (dependency-fostering) practices are particularly likely to be exhibited by parents who are themselves dependent (Byng-Hall, 2002; Kochanska, Friesenberg, Lange, & Martel, 2004). Thus, the clinician working with a dependent patient (or couple) should be alert for signs of dependogenic parenting and 'Studies to date have not distinguished unsuccessful suicide attempts from manipulative suicide gestures, so it is impossible to know the degree to which dependent patients engage in this form of self-destructive behavior as a way of intimidating and controlling others. Borderline patients with pronounced underlying dependency needs may be more likely than other patients to make suicide gestures (Steele, van der Hart, & Nijenhuis, 2001; Wasson & Linehan, 1993); patients with hostile or conflicted dependency are also at increased risk (Bornstein & Languirand, 2003). SPECIALIZED TREATMENT ISSUES
185
TABLE 10.3 Studies of Dependency and Suicide Study
Sample
Birtchnell (1981) Blatt et al. (1996) Bornstein & O'Neill (2000)
1 94 female psych outpts 670 high school students 1 52 psych inpts
1 52 psych inpts Brent et al. 86 community (1994) adults Cantor 199 female (1976) college students Clark et al. 56 psych (1993) inpts Epstein et 33 medical students al. (1973) Overholser 106 psych inpts etal. (1989) Pallis & 369 psych Birtchnell outpts (1976) 1 89 psych Paykel & Dienelt inpts (1971) Westen et al. 296 psych (2003) inpts and outpts
Design
Dependency measure
Suicide index
Effect size (i)
Retro
MMPI Dy
Suicide attempt
.22
Concur
DEQ
Suicidal ideation
.23
Concur
MMPI Dy
Suicide attempt
.33
Concur
ROD
-.05
Retro Retro
DPD symptoms EPPS
Suicide attempt Completed suicide Suicide attempt
Concur
SNAP
.34
Pro
KAS
Concur
MCMI
Suicidal ideation Completed suicide Suicidal ideation
Concur
MMPI Dy
Suicide attempt
.18
Pro
LK orality
Suicide attempt
.14
Retro
DPD symptoms
Suicide attempt
.17
.02 .63
.57 .05
Note, Retro = retrospective design; Concur = concurrent design; Pro = prospective design; MMPI Dy = MMPI Dependency Scale (Navran, 1954); ROD = Rorschach oral Dependency Scale (Masling etal., 1967); DEQ = Depressive Experiences Questionnaire Dependency Scale (Blatt et al., 1976); EPPS = Edwards Personal Preference Scale (Edwards, 1954); SNAP = Schedule for Nonadaptive and Adaptive Personality (Clark, 1993); KAS = Katz Adjustment Scales (Katz & Lyerly, 1963); MCMI = Millon Clinical Multiaxial Inventory (Millon, 1977); LK Orality = Lazare-Klerman Orality Scale (Lazare, Klerman, & Armor, 1966). Effect sizes represent the magnitude of the relationship between dependency level and suicide risk.
prepared to provide feedback to the parents regarding the effects of this behavior on their children's development. In situations wherein parenting practices are having a strong negative impact, it may be necessary to recommend adjunct family or child therapy. Several parenting styles are linked to the development of dependency in children: • Overprotectiveness. Overprotectiveness fosters dependency by teaching the child that he is vulnerable and weak and cannot 186
CLINICAL APPLICATIONS
survive without the protection of others. The overprotected child is likely to develop a "helpless self-concept" (see Figure 1.1) and a habit of looking to other people for guidance and support. Numerous studies have documented the link between parental overprotectiveness and dependency in children (e.g., Baker, Capron, & Azorlosa, 1996; Berg & McGuire, 1974; Richman& Flaherty, 1987). • Authoritarianism. Like overprotectiveness, parental authoritarianism contributes to the child's perception of himself as weak and ineffectual. Authoritarian parenting teaches the child that the way to survive is to accede to others' expectations and demands rather than taking the initiative and doing things on one's own. The links between parental authoritarianism and dependency in children have been documented by McCranie and Bass (1984), Vaillant (1980), and Whiffen and Sasseville (1991). • Neglect/mistreatment. Neglect and mistreatment have a powerful effect on the development of dependency. Insofar as the neglected or mistreated child comes to see the world as a dangerous place with frequent negative events she cannot control, a generalized sense of helplessness is likely to result. Low selfefficacy and a pessimistic outlook often follow (Egeland & Sroufe, 1981), leading to hostile or conflicted dependency (Bornstein & Languirand, 2003). • Premature independence pressure. When parents overemphasize independent functioning before the child is cognitively and emotionally prepared to deal with such demands, the child is placed under considerable pressure to perform behaviors that are beyond her capabilities. Children in this situation have difficulty coping, develop a view of themselves as inadequate and ineffectual, look to others for protection and support and, ultimately, develop the helpless self-concept that forms the core of a dependent personality (Birtchnell, 1980; Bornstein, 1993). Use Assessment Data to Structure Marital Therapy Just as assessment of individual functioning is critical for effective treatment of that patient, assessment of dyadic functioning is critical for effective treatment of a couple (see Carlson & Sperry, 1998). Two assessment techniques can set the stage for effective marital therapy with couples wherein one or both partners are highly dependent: • Assessment of both partners' implicit and self-attributed dependency strivings. This information helps the clinician understand the SPECIALIZED TREATMENT ISSUES
187
genesis and maintenance of dysfunctional dependency-related interactions (see chap. 7). • Assessment of each partner's dependency style. In those couples wherein both partners are at least moderately dependent, this information can be useful in disentangling complex dyadic coping patterns. Detailed descriptions of these dependency styles are provided in chapter 5 and summarized in Tables 5.4 and 5.5. Whatever assessment tools are used to develop a framework for marital therapy, it is important to interpret test results within a cultural context. Individualistic and sociocentric societies differ with respect to gender roles and gender-based norms for marital relationships (Fukuda & Ogawa, 1988; Minuchin, 2002), and studies show that assimilation and acculturation often create conflict (both intra- and interpersonal) in couples who are integrating individualistic norms into a longstanding sociocentric value system (Tseng & Streitzer, 2001). In this situation, the practitioner should periodically refocus the discussion on assimilation-acculturation issues; many couples will be reluctant to verbalize these concerns themselves and might not see such issues as appropriate topics for therapy. Do Not Ignore the Symbolic Meaning of Medication As several theorists and researchers have noted, medication—especially psychotropic medication—has tremendous symbolic meaning (Brinkley, 1993; Castillo, 1997; Fisher & Greenberg, 1996). Medication can represent a physical connection to the physician (i.e., a transitional object), an identifiable confirmation of helplessness and disability, a social stigma (or status symbol), or even a negotiating chip the patient uses to acquire power within the family (see Sotsky, 1992; Todd-Bazemore, 1999). For the dependent patient, the symbolic power of medication is particularly salient. Although research indicates that dependent patients are not at increased risk for substance use disorders (Calsyn & Saxon, 1990; Nace, Davis, & Gaspari, 1991), their somatizing tendencies, suggestibility, and generalized helpless stance can lead them to believe that withdrawal from a medication regimen will have intolerable negative effects. This fear—coupled with the connection to the clinician that a medication prescription represents—may cause some dependent patients to become so preoccupied with drug treatment that they have difficulty benefiting from psychotherapy. When extreme, the patient's focus on medication and its effects can escalate into a form of resistance, undermining other treatment interventions. This risk is especially pronounced for patients with hidden dependency (Bornstein & Languirand, 2003) and for dependent patients with pronounced mood or anxiety disorder symptoms (for whom psychotropic medications are likely to bring about more rapid symptom change than traditional psychotherapy). 188
CLINICAL APPLICATIONS
Distinguish Unhealthy Dependency From Compensatory Dependency In working with elderly patients or patients who are physically challenged or seriously ill, it is important to distinguish unhealthy dependency (i.e., dependent behavior aimed at externalizing responsibility and avoiding challenge) from compensatory dependency (i.e., dependent behavior in one domain that facilitates autonomous functioning in other areas). In these patient populations, flexible, situation-appropriate help-seeking can be far more adaptive than rigid independence and refusal to accept help (Nagurney, Reich, & Newsom, 2004). As Bornstein and Languirand (2003) and Tseng and Streitzer (2001) pointed out, however, our Western cultural emphasis on self-reliance and self-sufficiency can sometimes lead clinicians to misinterpret adaptive, compensatory dependency as unhealthy, maladaptive dependency. Several strategies are useful in distinguishing these contrasting forms of dependency: • Monitor the patient's motives and goals. In unhealthy dependency the patient seeks help to avoid a challenge and maintain a helpless position. In compensatory dependency the patient seeks help to confront a challenge and enhance autonomous functioning. Thus, the patient who asks his spouse to drive him to the bank because he is afraid to go alone is exhibiting unhealthy dependency; the patient who asks his spouse to drive him to the bank so he can maintain control over his finances is exhibiting compensatory dependency. • Explore the patient's emotional responses to receiving help. A telltale sign of unhealthy dependency is the feeling of relief that results from avoiding a frightening situation. Many dependent patients also experience guilt and/or shame at having given in to their fear (though they might be unwilling or unable to acknowledge this aspect of their response). In contrast, compensatory dependency is typically associated with feelings of satisfaction, empowerment, and increased self-efficacy after receiving help; in this situation the patient has asked for help mindfully and used it adaptively. • Scrutinize the behavioral consequences of receiving help. Just as patients' emotional reactions are important, their behavioral responses can help distinguish unhealthy from compensatory dependency. In unhealthy dependency, receiving help is generally followed by withdrawal and passivity. If a patient is experiencing guilt or shame, receiving help may also be followed by selfdenigrating statements and an increase in depressive symptoms. A very different pattern occurs when help-seeking is done in the service of compensatory dependency. In this situation reSPECIALIZED TREATMENT ISSUES
189
ceiving help is usually followed by increased activity, more positive affect, and a series of positive self-statements. Consider Financial as Well as Trait Dependency When Evaluating Abuse Risk Dependency leads to increased likelihood of victimization within romantic relationships because dependent people have difficulty setting limits on their partner's behavior and are willing to go to great lengths to preserve the status quo. Beyond trait dependency, studies show that financial dependency and physical dependency (e.g., that resulting from illness or disability) also lead to increased risk for abuse and neglect (Comijs, Smit, Pot, Bouter, & Jonker, 1998; Reay & Browne, 2001). Although the nature of the dependency may differ, the process is similar: Like trait dependency, financial and physical dependency both cause the individual to fear relationship disruption and tolerate a greater degree of inappropriate behavior in the partner because they perceive few viable alternatives to their present situation— even if the relationship is highly dysfunctional (see Rusbult & Martz, 1995). A very different dependency-abuse dynamic involves the adult child's financial dependency on an ill or frail parent. Studies show that adult children in a caregiving role are more likely to engage in elder abuse when they are financially dependent on the care-receiver than when they are financially self-sufficient (see Pillemer & Wolf, 1986). Although at first this pattern seems counterintuitive (and inconsistent with the economic dependencespouse abuse link described earlier), it appears to reflect the caregiver's frustration at being "trapped" in a stressful, demanding role. In contrast to caregivers who are financially self-sufficient, caregivers who rely on the carereceiver for financial support have difficulty expressing anger directly and setting appropriate limits on the care-receiver's behavior. Hidden resentment may eventually build to the point that abuse or neglect occurs.8 In Settings That Reinforce Dependency, Structure Treatment With Termination in Mind Chapter 9 outlined strategies the practitioner can use to counter termination-related resistance during outpatient treatment. Similar strategies may 8
Beyond these results, it is important to remember that victimization, in and of itself, can lead to increased feelings of helplessness, powerlessness, and dependency (Curry & Stone, 1995; Watson et al., 1997). Just as the neglected or maltreated child develops a kind of learned helplessness, with pervasive feelings of weakness and vulnerability, the spouse or aging parent who is abused or neglected is likely to show increased dependency on the caregiver. Ironically, studies suggest that parents who neglected or mistreated their children are particularly likely to be victimized if their children ever assume a caregiving role (Pillemer & Wolf, 1986; Reay & Browne, 2001). When this occurs, the dependency-abuse relationship has come full circle: Anger resulting from earlier mistreatment remains unexpressed—or indirectly expressed—for many years, finally erupting when roles are reversed, and the adult child becomes caregiver for an ill or aging parent.
190
CLINICAL APPLICATIONS
be utilized in inpatient, rehabilitation, and long-term care settings, although the dynamics of termination-related resistance differ somewhat in these contexts (Bornstein & Languirand, 2001). Unlike outpatient therapy, inpatient treatment, rehabilitation, and long-term care are based upon a perceived need for additional structure and more intensive intervention. Moreover, these settings tend to develop unique patient cultures that lead to predictable changes in behavior as new patients are socialized into the existing milieu. Institutional patient cultures can take many forms (e.g., competitive, hypochondriacal, sexualized), but in the present context, the milieu is most problematic when it propagates and reinforces dependency. A dependency-fostering dynamic need not be explicit (in fact, it is often hidden), but the clinician must be alert to the possibility that institutional culture is undermining positive change. This process may be exacerbated by the dependent patient's care-seeking tendencies: Some dependent patients find inpatient treatment so rewarding that they consciously or unconsciously work to undermine progress as termination nears. Because these two processes can combine to obviate gain, effective treatment of dependent patients in inpatient, rehabilitation, and long-term care settings not only requires that the practitioner counter dependency-related resistance in individual patients, but also create a treatment milieu wherein positive change is supported and autonomy is reinforced.
SPECIALIZED TREATMENT ISS UES
191
REFERENCES Abelson, R. P. (1981). Psychological status of the script concept. American Psychologist, 36, 715-729. Abraham, K. (1927). The influence of oral erotism on character formation. In C. A. D. Bryan &. A. Strachey (Eds.), Selected papers on psycho-analysis (pp. 393-406). London: Hogarth. Abramson, L., Seligman, M. E. P., & Teasdale, J. (1978). Learned helplessness in humans: Critique and reformulation, journal of Abnormal Psychology, 87,49-74Abramson, P. R., Cloud, M. Y., Keese, N., & Keese, R. (1994). How much is too much? Dependency in a psychotherapeutic relationship. American journal of Psychotherapy, 48, 294-301. Agrawal, K., & Rai, S. N. (1988). Post-noise frustration tolerance as a function of controllability of noise and dependence proneness. Indian Journal of Psychometry and Education, 19, 85-89. Ainsworth, M. D. S. (1969). Object relations, dependency, and attachment: A theoretical review of the infant-mother relationship. Child Development, 40, 9691025. Ainsworth, M. D. S. (1972). Attachment and dependency: A comparison. In J. L. Gewirtz (Ed.), Attachment and dependency (pp. 97-137). New York: Wiley. Ainsworth, M. D. S. (1989). Attachments beyond infancy. American Psychologist, 44, 709-716. Alam, K. (1986). An experimental study of retention in relation to approval motive and dependence proneness. journal of Psychological Research, 30, 144-149. Alden, L. E. (1989). Short-term structured treatment for avoidant personality disorder, journal of Consulting and Clinical Psychology, 57, 756-764Alden, L. E., Laposa, J. M., Taylor, C. T., & Ryder, A. G. (2002). Avoidant personality disorder: Current status and future directions, journal of Personality Disorders, 16, 1-29. Alexander, J. F., & Abeles, N. (1968). Dependency changes in psychotherapy as related to interpersonal relationships. Journal of Consulting and Clinical Psychology, 32, 685-689. Allard, G., Butler, J., Faust, D., & Shea, M. T. (1995). Errors in hand-scoring objective personality tests. Pro/essiona/ Psychology: Research and Practice, 26,304-308. Allen, N. B., Home, D. J., &.Trinder, J. (1996). Sociotropy, autonomy, anddysphoric responses to specific classes of stress: A psychophysiological evaluation. Journal of Abnormal Psychology, 105, 25-33. Almgren, P. E., Nordgren, L., & Skantze, H. (1978). A retrospective study of operationally defined hysterics. British journal of Psychiatry, 132, 67—73. Alnaes, R., & Torgerson, S. (1988). DSM-III symptom disorders (Axis I) and personality disorders (Axis II) in an outpatient population. Acta Psychiatrica Scandinavica, 78, 348-355.
193
Alperin, R. M. (2001). Barriers to intimacy: An object relations perspective. Psychoanalytic Psychology, 18, 137-156. Alsop, A. E. (1984). Purley Day Hospital: An appraisal with special reference to institutionalized behavior and dependence. Occupational Therapy, 47,306-310. American Psychiatric Association. (1952). Diagnostic and statistical manual of mental disorders (1st ed.). Washington, DC: Author. American Psychiatric Association. (1968). Diagnostic and statistical manual of mental disorders (2nd ed.). Washington, DC: Author. American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders (3rd ed). Washington, DC: Author. American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd ed., rev.). Washington, DC: Author. American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author. American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text rev.). Washington, DC: Author. Ammon, M. (2002). Integration of group dynamic principles in dynamic psychiatric treatment. Archives of Psychiatry and Psychotherapy, 4, 5-14. Andersen, S. M., & Chen, S. (2002). The relational self: An interpersonal socialcognitive theory. Psychological Review, 109, 619-645. Andersen, S. M., Chen, S., & Miranda, R. (2002). Significant others and the self. Self and Identity, 1,159-168. Anderson, K. L. (1986). Androgeny, flexibility, and individualism. Journal of Personality Assessment, 50, 265-278. Arkowitz, H., & Messer, S. B. (Eds.). (1984). Psychoanalytic therapy and behavior therapy: Is integration possible? New York: Plenum. Arvey, R. D. (1992). Constructs and construct validation: Definitions and issues. Human Performance, 5, 59-69. Auerbach, J. S., & Blatt, S. J. (1996). Self-representation in severe psychopathology: The role of reflexive self-awareness. Psychoanalytic Psychology, 13, 297-341. Baker, J. D., Capron, E. W., & Azorlosa, J. (1996). Family environment characteristics of persons with histrionic and dependent personality disorders, journal of Personality Disorders, 10, 82-87. Baldwin, M. W., & Sinclair, L. (1996). Self-esteem and "if.. .then" contingencies of interpersonal acceptance. Journal of Personality and Social Psychology, 71, 11301141. Ball, S. A. (1998). Manualized treatment for substance abusers with personality disorders: Dual Focus Schema Therapy. Addictive Behaviors, 23, 883-891. Ball, S. A., & Young, J. E. (2000). Dual Focus Schema Therapy for personality disorders and substance dependence. Cognitive and Behavioral Practice, 7, 270-281. Baltes, M. M. (1996). The many faces of dependency in old age. Cambridge, England: Cambridge University Press. 194
REFERENCES
Banaji, M. R., Hardin, Q, & Rothman, A. J. (1993). Implicit stereotyping in person judgment. Journal of Personality and Social Psychology, 65, 272-281. Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavior change. Psychological Review, 84, 191-215. Bandura, A. (1978). The self system in reciprocal determinism. American Psychologist, 33, 344-358. Bandura, A. (1991). Human agency: The rhetoric and the reality. American Psychologist, 46, 157-162. Bandura, A., & Walters, R. H. (1963). Social learning and personality development. New York: Holt, Rinehart & Winston. Barber, J. P., & Morse, J. Q. (1994). Validation of the Wisconsin Personality Disorders Inventory with the SCID-II and PDE. Journal of Personality Disorders, 8, 307-319. Bargh, J. A. (1982). Attention and automaticity in the processing of self-relevant information. Journal of Personality and Social Psychology, 43, 425-436. Bargh, J. A., & Chartrand, T. L. (1999). The unbearable automaticity of being. American Psychologist, 54, 462-479. Earth, F. D. (2001). Money as a tool for negotiating separateness and connectedness in the therapeutic relationship. Clinical Soda! Work Journal, 29, 79-93. Bartholomew, K. (1997). Adult attachment processes: Individual and couple perspectives. British Journal of Medical Psychology, 70, 249-263. Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachment as a fundamental human motivation. Psychological Bulletin, 117,497-529. Baumrind, D. (1980). New directions in socialization research. American Psychologist, 35, 639-652. Beasley, R., & Stoltenberg, C. D. (1992). Personality characteristics of male spouse abusers. Pro/esszonal Psychology: Research and Practice, 23, 310-317. Beck, A. T. (1976). Cognitive therapy and the emotional disorders. New York: International Universities Press. Beck, A. T., Brown, G., Steer, R. A., & Weissman, A. N. (1991). Factor analysis of the Dysfunctional Attitudes Scale in a clinical population. Psychological Assessment, 3, 478-483. Beck, A. T., Epstein, N., Harrison, R. P., &. Emery, G. (1983). Development of the Sociotropy-Autonomy Scale: A measure of personality factors in psychopathology. Unpublished manuscript, University of Pennsylvania School of Medicine. Beck, A. T., & Freeman, A. (1990). Cognitive therapy of the personality disorders. New York: Guilford. Beck, J. S. (1994). Cognitive therapy: Basics and beyond. New York: Guilford. Becker, D. F., Grilo, C. M., Edell, W. S., & McGlashan, T. H. (2001). Applicability of personality disorder criteria in late adolescence: Internal consistency and criterion overlap two years after psychiatric hospitalization. Journal of Personality Disorders, 15, 255-262. REFERENCES
195
Becker, E. (1973). The denial of death. New York: Free Press. Behrens, K. Y. (2004). A multifaceted view of the concept of amae: Reconsidering the indigenous Japanese concept of relatedness. Human Development, 47,1-27. Beitman, B. D. (1992). Integration through fundamental similarities and useful differences among the schools. In J. C. Norcross & M. R. Goldfried (Eds.), Handbook of psychotherapy integration (pp. 202-230). New York: Basic Books. Bellack, A. S., & Hersen, M. (Eds.). (1993). Handbook of behavior therapy in the psychiatric setting. New York: Plenum Press. Bern, S. (1974). The measurement of psychological androgeny. Journal of Consulting and Clinical Psychology, 42, 155-162. Bender, L. (1938). A visual motor gestalt test and its clinical use. American Orthopsychiatric Association Research Monographs, 3, 1-176. Benfari, R. C. (1969). Relationship between early dependence training and the patient-therapist dyad. Psychological Reports, 25, 552-554. Benjamin, L. S. (1974). A structural analysis of social behavior. Psychological Review, 81,392-425. Benjamin, L. S. (1996). Interpersonal diagnosis and treatment of personality disorders. New York: Guilford. Berg, I. (1974). A self-administered dependency questionnaire for use with mothers of schoolchildren. British journal of Psychiatry, 124, 1-9. Berg, I., & McGuire, R. (1974). Are mothers of school-phobic adolescents overprotective? British Journal of Psychiatry, 124, 10-13. Bergman, B., Larsson, G., Brismar, B., &.Klang, M. (1988). Aetiological and precipitating factors in wife battering. Acta Psychiatrica Scandinavica, 77, 338-345. Berman, S. M. W., & McCann, J. T. (1995). Defense mechanisms and personality disorders: An empirical test of Millon's theory. Journal of Personality Assessment, 64, 132-144. Bernard, H. R. (2001). Social research methods. Thousand Oaks, CA: Sage. Besser, A., Priel, B., & Wiznitzer, A. (2000). High-risk pregnancies: Personality, antenatal attachment, and perceived social support. Paper presented at the 108th annual meeting of the American Psychological Association, Washington, DC. Beutler, L. E., Moliero, C., & Talebi, H. (2002). How practitioners can systematically use empirical evidence in treatment selection. Journal of Clinical Psychology, 58, 1199-1212. Bhogle, S. (1983). Antecedents of dependency in children of low social class. Ps;ychological Studies, 28, 92-95. Bhogle, S. (1978). Child rearing practices among three cultures. Sociai Change, 5, 813. Birren, J. E., 6k Schaie, K. W. (Eds.). (1996). Handbook of the psychology of aging. San Diego, CA: Academic Press. Birtchnell, J. (1980). Women whose mothers died in childhood: An outcome study. Psychological Medicine, JO, 699-713. 196
REFERENCES
Birtchnell, J. (1981). Some familial and clinical characteristics of female suicidal psychiatric inpatients. British Journal of Psychiatry, 138, 381-390. Birtchnell, J. (1984). Dependence and its relationship to depression. British Journal of Medical Psychology, 57, 215-225. Birtchnell, J. (1988). Defining dependence. British Journal of Medical Psychology, 61, 111-123. Birtchnell, J. (1991). The measurement of dependence by questionnaire. Journal of Personality Disorders, 5, 281-295. Birtchnell, J., & Kennard, J. (1983). What does the MMPI dependency scale really measure? Journal of Clinical Psychology, 39, 532-543. Birtchnell, J., & Shine, J. (2000). Personality disorders and the interpersonal octagon. British Journal of Medical Psychology, 73, 433-448. Black, D. W., Monahan, P., Wesner, R., Gabel, J., & Bowers, W. (1996). The effect of fluvoxamine, cognitive therapy, and placebo on abnormal personality traits in 44 patients with panic disorder. Journal of Personality Disorders, 10, 185-194. Blais, M. A., Hilsenroth, M. J., Castlebury, F., Fowler, J. C., & Baity, M. R. (2001). Predicting DSM-IV-R Cluster B personality disorder criteria from MMPI-2 and Rorschach data: A test of incremental validity. Journal of Personality Assessment, 76, 150-168. Blatt, S. J. (1974). Levels of object representation in anaclitic and introjective depression. Psychoanalytic Study of the Child, 29, 107-157. Blatt, S. J. (1990). Interpersonal relatedness and self-definition: Two personality configurations and their implications for psychopathology and psychotherapy. InJ. L. Singer (Ed.), Repression and dissociation: Implications for personality theory, psychopathology, andhealth (pp. 299-335). Chicago: University of Chicago Press. Blatt, S. J. (1991). A cognitive morphology of psychopathology. Journal of Nervous and Mental Disease, 179, 449-458. Blatt, S. J. (1992). The differential effect of psychotherapy and psychoanalysis with anaclitic and introjective patients. Journal of the American Psychoanalytic Association, 40, 691-724. Blatt, S. ]., Auerbach, J. S., & Levy, K. N. (1997). Mental representations in personality development, psychopathology, and the therapeutic process. Review of General Psychology, 1, 351-374. Blatt, S. J., Bers, S. A., & Schaffer, C. E. (1993). The assessment of self descriptions. Unpublished research manual. Yale University, New Haven, CT. Blatt, S. J., Chevron, E. S., Quinlan, D. M., Schaffer, C. E., & Wein, S. J. (1988). The assessment of qualitative and structural dimensions of object representations (Revised edition). Unpublished research manual. Yale University, New Haven, CT. Blatt, S. J., Chevron, E. S., Quinlan, D. M., & Wein, S. J. (1981). The assessment of qualitative and structural dimensions of object representations. Unpublished research manual. Yale University, New Haven, CT. Blatt, S. J., D'Afflitti, J. P., &. Quinlan, D. M. (1976). Experiences of depression in normal young adults. Journal of Abnormal Psychology, 85, 383-389. REFERENCES
197
Blatt, S. J., & Ford, R. Q. (1994)- Therapeutic change: An object relations perspective. New York: Plenum. Blatt, S. J., Hart, B., Quinlan, D. M., Leadbeater, B., & Auerbach, J. (1993). Interpersonal and self-critical dysphoria and behavior problems in adolescents. Journal of Youth and Adolescence, 22, 253-269. Blatt, S. J., & Homann, E. (1992). Parent-child interaction in the etiology of dependent and self-critical depression. Clinical Psychology Review, 12, 47-91. Blatt, S. J., Rounsaville, B., Eyre, S. L, &. Wilber, C. (1984). The psychodynamics of opiate addiction, journal of Nervous and Mental Disease, 172, 342-352. Blatt, S. J., Schaffer, C. E., Bers, S. A., & Quinlan, D. M. (1992). Psychometric properties of the Depressive Experiences Questionnaire for Adolescents. Journal of Personality Assessment, 59, 82-98. Blatt, S. J., &Shichman, S. (1983). Two primary configurations of psychopathology. Psychoanalysis and Contemporary Thought, 6, 187-254. Blatt, S. J., Stayner, D. A., Auerbach, J. S., & Behrends, R. S. (1996). Change in self and object representations in long-term, intensive, inpatient treatment of seriously disturbed adolescents and young adults. Psychiatry, 59, 82-107. Blatt, S. J., Zohar, A, Quinlan, D. M., Luthar, S., & Hart, B. (1996). Levels of relatedness within the dependency factor of the Depressive Experiences Questionnaire for Adolescents. Journal of Personality Assessment, 67, 52-71. Blatt, S. J., Zohar, A. H., Quinlan, D. M., Zuroff, D. C., & Mongrain, M. (1995). Subscales within the dependency factor of the Depressive Experiences Questionnaire. Journal of Personality Assessment, 64, 319-339. Blatt, S. J., & Zuroff, D. C. (1992). Interpersonal relatedness and self-definition: Two prototypes for depression. Clinical Psychology Review, 12, 527-562. Blum, G. S. (1949). A study of the psychoanalytic theory of psychosexual development. Genetic Psychology Monographs, 39, 3-99. Bodden, ]. C. (1970). Cognitive complexity as a factor in appropriate vocational choice. Journal of Counseling Psychology, 17, 364-368. Bohart, A. C., &. Greenberg, L. S. (Eds.). (1997). Empathy reconsidered. Washington, DC: American Psychological Association. Bond, M., Gardner, S. T., Christian, J., & Sigal, J. J. (1983). Empirical study of selfrated defense styles. Archives of General Psychiatry, 40, 333-338. Booth, T. (1986). Institutional regimes and induced dependency in homes for the aged. The Gerontologist, 26, 418—423. Bonanno, G. A., &. Castonguay, L. G. (1994). On balancing approaches to psychotherapy: Prescriptive patterns of attention, motivation, and personality. Psychotherapy, 31,571-587. Bornstein, R. F. (1992). The dependent personality: Developmental, social, and clinical perspectives. Psychological Bulletin, 112, 3-23. Bornstein, R. F. (1993). The dependent personality. New York: Guilford. Bornstein, R. F. (1994a). Adaptive and maladaptive aspects of dependency: An integrative review. American Journal of Orthopsychiatry, 64, 622-635. 198
REFERENCES
Bornstein, R. F. (1994b). Construct validity of the Interpersonal Dependency Inventory: 1977-1992. Journal of Personality Disorders, 8, 65-77. Bornstein, R. F. (1994c). Dependency in psychotherapy: Effective therapeutic work with dependent patients. In L. VandeCreek, S. Knapp, & T. L. Jackson (Eds.), Innovations in clinical practice (Vol. 13, pp. 139-150). Sarasota, FL: Professional Resources. Bornstein, R. F. (1995a). Active dependency. Journal of Nervous and Mental Disease, 183, 64-77. Bornstein, R. F. (1995b). Interpersonal dependency and physical illness: The mediating roles of stress and social support. Journal of Social and Clinical Psychology, 14, 225-243. Bornstein, R. F. (1995c). Sex differences in objective and projective dependency tests: A meta-analytic review. Assessment, 2, 319-331. Bornstein, R. F. (1996a). Beyond orality: Toward an object relations/interactionist reconceptualization of the etiology and dynamics of dependency. Psychoanalytic Psychology, 13, 177-203. Bornstein, R. F. (1996b). Construct validity of the Rorschach Oral Dependency Scale: 1967-1995. Psychological Assessment, 8, 200-205. Bornstein, R. F. (1997a). Dependent personality disorder in the DSM-/V and beyond. Clinical Psychology: Science and Practice, 4, 175-187. Bornstein, R. F. (1997b). Long-term retest reliability of Interpersonal Dependency Inventory scores in college students. Assessment, 4, 359-364. Bornstein, R. F. (1998a). Conscious and unconscious dependency needs: Implications for assessment and treatment. In L. VandeCreek, S. Knapp, & T. L. Jackson (Eds.), Innovations in clinical practice (Vol. 16, pp. 429^-38). Sarasota, FL: Professional Resources. Bornstein, R. F. (1998b). Dependency in the personality disorders: Intensity, insight, expression, and defense. Journal of Clinical Psychology, 54, 175-189. Bornstein, R. F. (1998c). Implicit and self-attributed dependency strivings: Differential relationships to laboratory and field measures of help seeking. Journal of Personality and Social Psychology, 75, 778-787. Bornstein, R. F. (1998d). Implicit and self-attributed dependency needs in dependent and histrionic personality disorders. Journal of Personality Assessment, 71, 1-14. Bornstein, R. F. (1998e). Interpersonal dependency and physical illness: A metaanalytic review of retrospective and prospective studies. Journal of Research in Personality, 32, 480-497. Bornstein, R. F. (1998f). Reconceptualizing personality disorder in the DSM-V: The discriminant validity challenge. Clinical Psychology: Science and Practice, 5,333343. Bornstein, R. F. (1999). Criterion validity of objective and projective dependency tests: A meta-analytic assessment of behavioral prediction. Psychological Assessment, 11,48-5 7. REFERENCES
199
Bernstein, R. F. (ZOOla). A meta-analysis of the dependency-eating disorders relationship: Strength, specificity, and temporal stability, journal ofPsychopathology and Behavioral Assessment, 23, 151-162. Bomstein, R. F. (ZOOlb). Clinical utility of the Rorschach Inkblot Method: Refraining the debate. Journal of Personality Assessment, 77, 39-47. Bornstein, R. F. (2002). A process dissociation approach to objective-projective test score interrelationships. Journal of Personality Assessment, 78, 47-68. Bornstein, R. F. (2003). Psychodynamic models of personality. In T. Millon & M. J. Lerner (Eds.), Comprehensive handbook of psychology: Personality and social psychology (Vol. 5, pp. 117-134). New York: Wiley. Bornstein, R. F. (2004). Dependent personality disorder. In I. B. Weiner (Ed.), Adult psychopathology case studies (pp. 23-28). Hoboken, NJ: Wiley. Bornstein, R. F, Bonner, S., Kildow, A. M., &. McCall, C. A. (1997). Effects of individual versus group test administration on Rorschach Oral Dependency scores, journal of Personality Assessment, 69, 215-228. Bornstein, R. F., & Bowen, R. F. (1995). Dependency in psychotherapy: Toward an integrated treatment approach. Psychotherapy, 32, 520-534Bornstein, R. F., Bowers, K. S., & Bonner, S. (1996a). Effects of induced mood states on objective and projective dependency scores. Journal of Personality Assessment, 67, 324-340. Bornstein, R. F., Bowers, K. S., & Bonner, S. (1996b). Relationships of objective and projective dependency scores to sex role orientation in college student participants. Journal of Personality Assessment, 66, 555-568. Bornstein, R. F., Bowers, K. S., & Robinson, K. J. (1995). Differential relationships of objective and projective dependency scores to self-reports of interpersonal life events in college students. Journal of Personality Assessment, 65, 255-269. Bornstein, R. F., & Cecero, J. J. (2000). Deconstructing dependency in a five-factor world: A meta-analytic review. Journal of Personality Assessment, 74, 324-343. Bornstein, R. F., Geiselman, K. J., Eisenhart, E. A., & Languirand, M. A. (2002). Construct validity of the Relationship Profile Test: Links with attachment, identity, relatedness, and affect. Assessment, 9, 373-381. Bornstein, R. F., Geiselman, K. J., Gallagher, H. A., Ng, H. M., Hughes, E. E., & Languirand, M. A. (2004). Construct validity of the Relationship Profile Test: Impact of gender, gender role, and gender role stereotype. Journal of Personality Assessment, 82, 104-113. Bornstein, R. F., & Greenberg, R. P. (1991). Dependency and eating disorders in female psychiatric inpatients. Journal of Nervous and Mental Disease, 179, 148152. Bornstein, R. F., Greenberg, R. P., Leone, D. R., & Galley, D. J. (1990). Defense mechanism correlates of orality. Journal of the American Academy of Psychoanalysis, 18, 654-666. Bornstein, R. F., Hilsenroth, M. J., Padawer, J. R., & Fowler, J. C. (2000). Interpersonal dependency and personality pathology: Variations in Rorschach Oral
200
REFERENCES
Dependency scores across Axis II diagnoses. Journal of Personality Assessment, 75, 478-491. Bornstein, R. F., & Kennedy, T. D. (1994). Interpersonal dependency and academic performance. Journal of Personality Disorders, 8, 240-248. Bornstein, R. F., Krukonis, A. B., Manning, K. A., Mastrosimone, C. C., & Rossner, S. C. (1993). Interpersonal dependency and health service utilization in a college student sample, journal of Social and Clinical Psychology, 12, 262-279. Bornstein, R. F., & Languirand, M. A. (2001). When someone you love needs nursing home care. New York: Newmarket. Bornstein, R. F., & Languirand, M. A. (2003). Healthy dependency. New York: Newmarket. Bornstein, R. F., Languirand, M. A., Geiselman, K. J., Creighton, J. A., West, M. A., Gallagher, H. A., & Eisenhart, E. A. (2003). Construct validity of the Relationship Profile Test: A self-report measure of dependency-detachment, journal of Personality Assessment, 80, 64-74. Bornstein, R. F., Leone, D. R., & Galley, D. J. (1988). Rorschach measures of oral dependence and the internalized self-representation in normal college students. journal of Personality Assessment, 52, 648-657. Bornstein, R. F., Manning, K. A., Krukonis, A. B., Rossner, S. C., & Mastrosimone, C. C. (1993). Sex differences in dependency: A comparison of objective and projective measures. Journal of Personality Assessment, 61, 169—181. Bornstein, R. F., & Masling, J. M. (1985). Orality and latency of volunteering to serve as experimental subjects. Journal of Personality Assessment, 49, 306-310. Bornstein, R. F., Masling, J. M., & Poynton, F. G. (1987). Orality as a factor in interpersonal yielding. Psychoanalytic Psychology, 4, 161-170. Bornstein, R. F., & O'Neill, R. M. (1997). Construct validity of the Rorschach Oral Dependency (ROD) Scale: Relationship of ROD scores to WAIS-R scores in a psychiatric inpatient sample. Journal of Clinical Psychology, 53, 99-105. Bornstein, R. F., & O'Neill, R. M. (2000). Dependency and suicidality in psychiatric inpatients. Journal of Clinical Psychology, 56, 463-473. Bornstein, R. F., Poynton, F. G., & Masling, J. M. (1985). Orality and depression: An empirical study. Psychoanalytic Psychology, 2, 241—249. Bornstein, R. F., Riggs, J. M., Hill, E. L., & Calabrese, C. (1996). Activity, passivity, self-denigration, and self-promotion: Toward an interactionist model of interpersonal dependency. Journal of Personality, 64, 637—673. Bornstein, R. F., Rossner, S. C., &. Hill, E. L. (1994). Retest reliability of scores on objective and projective measures of dependency: Relationship to life events and inter-test interval. Journal of Personality Assessment, 62, 398-415. Bornstein, R. F., Rossner, S. C., Hill, E. L., &Stepanian, M. L. (1994). Face validity and fakability of objective and projective measures of dependency. Journal of Personality Assessment, 63, 363-386. Bowen, M. (1978). Famity therapy in clinical practice. New York: Jason Aronson. Bowlby, J. (1969). Attachment. New York: Basic.
REFERENCES
20 J
Bowlby, J. (1973). Separation: Anxiety and anger. New York: Basic. Bowlby, J. (1980). Loss: Sadness and depression. New York: Basic. Brenner, C. (1973). An elementary textbook of psychoanalysis. New York: Anchor. Brent, D. A., Johnson, B. A., Perper, ]., Connolly, ]., Bridge, ]., Bartle, S., &. Rather, C. (1994). Personality disorder, personality traits, and completed suicide in adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 33, 1080-1086. Bringle, R. G., & Buunk, B. (1985). Jealousy and social behavior. In P. Shaver (Ed.), Review of personality and social psychology (Vol. 6, pp. 241-264). Beverly Hills, CA: Sage. Brink, T. L. (Ed.). (1986). Clinical gerontology: A guide to assessment and intervention. New York: Haworth. Brinkley, J. R. (1993). Pharmacotherapy of borderline states. Psychopharmacology, 16, 853-884. Brock, G. W., & Barnard, C. P. (1988). Procedures in family therapy. Boston: Allyn & Bacon. Brown, J. S., & Rawlinson, M. (1975). Relinquishing the sick role following open heart surgery. Journal of Health and Social Behavior, 16, 12-27. Browne, A., & Dolan, B. (1991). Counsellors' responses to clients' non-attendance at counselling sessions. Counselling Psychology Quarterly, 4, 109-118. Bruch, M. A., Rivet, K. M., Heimberg, R. G., Hunt, A., & Mclntosh, B. (1999). Shyness and sociotropy: Additive and interactive relations in predicting interpersonal concerns. Journal of Personality, 67, 373-406. Bucci, W. (1997). Psychoanalysis and cognitive science: A multiple-code theory. New York: Guilford. Bugental, J. (1976). The search for existential identity. San Francisco: Jossey-Bass. Bugental, J. (1978). Psychotherapy and process: The fundamentals of an existentialhumanistic approach. Reading, MA: Addison-Wesley. Bullitt, C. W., &. Farber, B. A. (2002). Gender differences in defensive style. Journal of the American Academy of Psychoanalysis, 30, 35-51. Burton, A., McGregor, H., & Berry, P. (1979). The effects of social reinforcement on dominant and dependent mildly intellectually handicapped school leavers. British Journal of Social and Clinical Psychology, 18, 129-133. Buss, A. H., & Plomin, R. (1984). Temperament: Early developing personality traits. Hillsdale, NJ: Erlbaum. Buunk, B. (1982). Anticipated sexual jealousy: Its relationship to self-esteem, dependency, and reciprocity. Personality and Social Psychology Bulletin, 8,310-316. Byng-Hall, J. (2002). Relieving parentified children's burdens in families with insecure attachment patterns. Family Process, 41, 375-388. Cadbury, S. (1991). The concept of dependence as developed by Birtchnell: A critical evaluation. British Journal of Medical Psychology, 64, 237-251.
202
REFERENCES
Calsyn, D. A., &. Saxon, A. J. (1990). Personality disorder subtypes among cocaine and opioid addicts using the MCMI. International Journal of the Addictions, 25, 1037-1049. Campbell, D. T., & Fiske, D. (1959). Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological Bulletin, 56, 81-105. Cantor, N., Mischel, W., & Schwartz, ]. (1982). A prototype analysis of psychological situations. Cognitive Psychology, 14, 45-77. Cantor, N., Smith, E. E., French, R. D., & Mezzich, J. (1980). Psychiatric diagnosis as prototype categorization, journal of Abnormal Psychology, 89, 181-193. Cantor, P. C. (1976). Personality characteristics found among youthful female suicide attempters. journal of Abnormal Psychology, 85, 324-329. Carlson, J., & Sperry, L. (Eds.). (1998). The disordered couple. Philadelphia: Brunner/ Mazel. Carnes, B. A. (1984). Concept analysis: Dependence. Critical Care Quarterly, 6, 2939. Carver, C. S., Reynolds, S. K., & Scheier, M. F. (1994). The possible selves of optimists and pessimists. Journal of Research in Personality, 28, 133-141Casey, P., & Butler, E. (1995). The effects of personality on response to ECT in major depression. Journal of Personality Disorders, 9, 134-142. Cashdan, S. (1988). Object relations therapy. New York: W. W. Norton. Caspi, A., Bern, D. J., & Elder, G. H. (1989). Continuities and consequences of interactional styles across the life course. Journal of Personality, 57, 375—406. Castillo, R. J. (1997). Culture and mental illness: A client-centered approach. Pacific Grove, CA: Brooks/Cole. Gates, J. A. (1999). The art of assessment in psychology: Ethics, expertise, and validity. Journal of Clinical Psychology, 55, 631-641. Cattell, R. B. (1965). The scientific analysis of personality. Baltimore: Penguin. Cervone, D., & Shoda, Y. (1999). Beyond traits in the study of personality coherence. Current Directions in Psychological Science, 8, 27-32. Chadha, N. K. (1983). A study of dependence proneness and text anxiety in schoolgoing children. Indian Psychological Review, 24, 34-38. Chambless, D. L., Renneberg, B., Goldstein, A., & Gracely, E. J. (1992). MCMIdiagnosed personality disorders among agoraphobic outpatients: Prevalence and relationship to severity and treatment outcome. Journal of Anxiety Disorders, 6, 193-211. Chambless, D. L, Renneberg, B., Gracely, E. J., Goldstein, A. ]., &Fydrich, T. (2000). Axis I and II comorbidity in agoraphobia: Prediction of psychotherapy outcome in a clinical setting. Psychotherapy Research, 10, 279-295. Chantry, K., & Craig, R. ]. (1994). Psychological screening of sexually violent offenders with the MCMI. Journal of Clinical Psychology, 50, 430-435. Chazan, R. (2001). The group as therapist. London: Jessica Kingsley.
REFERENCES
203
Chevron, E. S., Quinlan, D. M., & Blatt, S. J. (1978). Sex roles and gender differences in the experience of depression. Journal of Abnormal Psychology, 87, 680683. Choca, J. P., Shanley, L. A., & Van Denburg, E. (1992). Interpretive guide to the Millon Clinical Multiaxial Inventory. Washington, DC: American Psychological Association. Clark, K. E., & Ladd, G. W. (2000). Connectedness and autonomy support in parent-child relationships: Links to children's socioemotional orientation and peer relationships. Developmental Psychology, 36, 485-498. Clark, L. A. (1993). Manual for the Schedule for Nonadaptive and Adaptive Personality (SNAP). Minneapolis: University of Minnesota Press. Clark, L. A, McEwen, J. L., Collard, L. M., & Hickok, L. G. (1993). Symptoms and traits of personality disorder: Two new methods for their assessment. Psychological Assessment, 5, 81-91. Clark, L. A., Vittengl, J., Kraft, D., & Jarrett, R. B. (2003). Separate personality traits from states to predict depression. Journal of Personality Disorders, 17, 152172. Coen, S. J. (1992). The misuse of persons: Analyzing pathological dependency. Hillsdale, NJ: Analytic. Colgan, P. (1987). Treatment of dependency disorders in men: Toward a balance of identity and intimacy. Journal of Chemical Dependency Treatment, 1, 205-227. Collins, N. L., & Read, S. J. (1990). Adult attachment, working models, and relationship quality in dating couples, journal of Personality and Social Psychology, 58, 644-663. Colton, S. I. (1980). Instrumental dependency in institutionalized schizophrenics. Psychosocial Rehabilitation Journal, 4, 9-18. Comijs, H. C, Smit, J. H., Pot, A. M., Bouter, L. M, &. Jonker, C. (1998). Risk indicators of elder maltreatment in the community. Journal of Elder Abuse and Neglect, 9, 67-76. Coolidge, F. L. (1998). Coolidge Personality and Neuropsychological Inventory for Children manual. Colorado Springs, CO: Author. Coolidge, F. L., Thede, L. L., & Jang, K. L. (2000). Heritability of personality disorders in childhood: A preliminary investigation. Journal of Personality Disorders, 15, 33-10. Corbitt, E. M., & Widiger, T. A. (1995). Sex differences among the personality disorders: An exploration of the data. Clinical Psychology: Science and Practice, 2, 225-238. Costa, P. T. & McCrea, R. R. (1985). The NEO Personality Inventory manual. Odessa, FL: Psychological Assessment Resources. Costa, P. T., & McCrea, R. R. (1992). Revised NEO Personality Inventory (NEO-PIR) and NEO Five-Factor Inventory (NEO-FFI) professional manual. Odessa, FL: Psychological Assessment Resources. Costa, P. T., & Widiger, T. A. (Eds.). (1994). Personality disorders and the five-factor model of personality. Washington, DC: American Psychological Association. 204
REFERENCES
Cramer, P. (2002). The study of defense mechanisms: Gender implications. In R. F. Bornstein & J. M. Masling (Eds.), The psychodynamics of gender and gender role (pp. 81-127). Washington, DC: American Psychological Association. Cramer, P., Blatt, S. J., & Ford, R. Q. (1988). Defense mechanisms in the anaclitic and introjective personality configuration. Journal of Consulting and Clinical Psychology, 56, 610-616. Crits-Christoph, P., & Barber, J. P. (1991). Handbook of short-term dynamic psychotherapy. New York: Basic. Crits-Christoph, P., & Connolly, M. B. (1998). Empirical basis of supportiveexpressive psychodynamic psychotherapy. In R. F. Bornstein & J. M. Masling (Eds.), Empirical studies of the therapeutic hour (pp. 109-151). Washington, DC: American Psychological Association. Crits-Christoph, P., Demorest, A., Muenz, L. R., & Baranackie, K. (1994). Consistency of interpersonal themes. Journal of Personality, 62, 499-526. Cross, S. E., Bacon, P. L., &. Morris, M. L. (2000). The relational-interdependent self-construal and relationships. Journal of Personality and Social Psychology, 78, 791-808. Cross, S. E., & Madson, L. (1997). Models of the self: Self-construal and gender. Psychological Bulletin, 122, 5-37. Cross, S., &. Markus, H. (1991). Possible selves across the life span. Human Development, 34, 230-255. Curry, L. C., & Stone, J. G. (1995). Understanding elder abuse: The social problem of the 1990s. Journal of Clinical Geropsychology, 1, 147-156. Dana, R. H. (1994). Testing and assessment ethics for all persons: Beginning and agenda. Professional Psychology: Research and Practice, 25, 349—354Dana.R. H. (1997). Understanding cultural identity in assessment and intervention. Thousand Oaks, CA: Sage. Dana, R. H. (1998). Personality assessment and the cultural self. In L. Handler & M. J. Hilsenroth (Eds.), Teaching and learning personality assessment (pp. 325346). Mahwah, NJ: Erlbaum. Davis, M. S., & Eichorn, R. L. (1963). Compliance with medical regimens. Journal of Health and Human Behavior, 4, 240-249. Davison, S., Morven, L., & Taylor, P. J. (2001). Examination of the screening properties of the Personality Diagnostic Questionnaire 4+ (PDQ-4+) in a prison population. Journal of Personality Disorders, 15, 180-194. Deitchman, P. S. (1978). Management of the passive dependent client. Journal of Applied Rehabilitation Counseling, 9, 86-90. Dell, P. F. (1998). Axis II pathology in outpatients with dissociative identity disorder. Journal of Nervous and Mental Disease, 186, 352-356. de Ruiter, C., & Greeven, P. G. J. (2000). Personality disorders in a Dutch forensic psychiatric sample: Convergence of interview and self-report measures. Journal of Personality Disorders, 14, 162-170. REFERENCES
205
Devens, M., & Erickson, M. T. (1998). The relationship between defense style and personality disorders, journal of Personality Disorders, 12, 86-93. Devito, A. J., & Kubis, J. K. (1983). Actual and recalled test anxiety, flexibility, rigidity, and self-control. Journal of Clinical Psychology, 39, 970-975. de Young, M., & Lowry, J. A. (1992). Traumatic bonding: Clinical implications for incest. Child Welfare, 71, 165-175. Doi, T. (1973). The anatomy of dependence. Tokyo: Kodansha. Dollard, J., & Miller, N. E. (1950). Personality and psychotherapy. New York: McGrawHill. Donat, D. C., McKeegan, G. F., & Neal, B. (1991). Training inpatient psychiatric staff in the use of behavioral methods: A program to enhance utilization. Psy~ chosocial Rehabilitation Journal, 15, 69-74. Dryden, W., & Trower, P. (Eds.). (1989). Cognitive psychotherapy: Stasis and change. New York: Springer. Dunnington, M. J. (1957). Behavioral differences in sociometric status groups in a nursery school. Child Development, 28, 103-111. Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive approach to motivation and personality. Psychological Review, 95, 256-273. Dworkin, R. H, Burke, B. B., Maher, B. A., & Gottesman, 1.1. (1976). A longitudinal study of the genetics of personality. Journal of Personality and Social Ps^chol0^,34,510-518. D'Zurilla, T. (1988). Problem-solving therapies. In K. S. Dobson (Ed.), Handbook of cognitive—behavioral therapies (pp. 85-135). New York: Guilford. D'Zurilla, T., & Goldfried, M. R. (1971). Problem solving and behavior modification. Journal of Abnormal Psychology, 78, 107-126. Eagle, M. N. (1984). Recent developments in psychoanalysis. New York: McGraw-Hill. Eddington, C., Piper, J., Tanna, B., Hodkinson, H. M., & Salmon, P. (1990). Relationships between happiness, behavioural status, and dependency on others in elderly persons. British Journal of Clinical Psychology, 29, 43-50. Edwards, A. L. (1954). Manual for the Edwards Personal Preference Schedule. New York: Psychological Corporation. Egeland, B., & Sroufe, L. A. (1981). Developmental sequelae of maltreatment in infancy. In R. Rizley & D. Cicchetti (Eds.), New directions for child development (pp. 77-92). San Francisco: Jossey-Bass. Ekselius, L., Lindstrom, E., Knorring, L., Bodlund, O., & Kullgren, G. (1994). Comorbidity among the personality disorders in DSM-III-R. Personality and Individual Differences, 17, 155-160. Elfhag, K., Barkeling, B., Carlsson, A. M., &. Rossner, S. (2003). Microstructure of eating behavior associated with Rorschach characteristics in obesity. Journal of Personality Assessment, 81, 40-50. Ellis, A., & Dryden, W. (1997). The practice of rational emotive behavior therapy. New York: Springer.
206
REFERENCES
Emery, G., &.Lesher, E. (1982). Treatment of depression in older adults: Personality considerations. Psychotherapy, 19, 500-505. Epstein, L. C, Thomas, C. E., Shaffer, ]. W., &. Perlin, S. (1973). Clinical prediction of physician suicide based on medical student data. Journal of Nervous and Mental Disease, 156, 19-28. Erikson, E. H. (1950). Childhood and society. New York: Norton. Exline, R. V., & Messick, D. (1967). The effects of dependency and social reinforcement upon visual behavior during an interview. British journal of Social and Clinical Psychology, 6, 256-266. Exner, J. E. (1993). The Rorschach: A comprehensive system (Vol. 2, 3rd ed.). New York: Wiley. Fairbairn, W. R. D. (1952). An object relations theory of the personality. New York: Basic. Fals-Stewart, W. (1992). Personality characteristics of substance abusers: An MCMI cluster typology of recreational drug users in a therapeutic community and its relationship to length of stay and outcome. Journal of Personality Assessment, 59, 515-527. Fava, M., Bouffidea, E., Pava, J. A., McCarthy, M. K., & Rosenbaum, J. F. (1994). Personality disorder comorbidity with major depression and response to fluoxetine treatment. Psychotherapy andPsychosomatics, 62, 160-167. Feeney, J. A., &Noller, P. (1990). Attachment style as a predictor of adult romantic relationships. Journal of Personality and Social Psychology, 58, 281-291. Feist, J. G. (1994). Personality and working style predictors of integrative complexity: A study of scientists thinking about research and teaching. Journal of Personality and Social Psychology, 67, 474-484. Fenichel, O. (1945). The psychoanalytic theory of neurosis. New York: Norton. Fichman, L., Koestner, R., & Zuroff, D. C. (1996). Dependency, self-criticism, and perceptions of inferiority at summer camp. Journal of Youth and Adolescence, 25, 113-126. Fichman, L., Koestner, R., &. Zuroff, D. C. (1997). Dependency and distress at summer camp. Journal of Youth and Adolescence, 26, 217-232. Fink, D. (1991). The comorbidity of multiple personality disorder and DSM-III-R Axis II disorders. Psychiatric Clinics of North America, 14, 547-566. Finn, S. E. (2003). Therapeutic assessment of a man with "ADD." Journal of Personality Assessment, 80, 115-129. Finn, S. E., & Tonsager, M. E. (2002). How therapeutic assessment became humanistic. Humanistic Psychologist, 30, 10—22. First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1995). The Structured Clinical Interview for DSM-III-R Personality Disorders (SCID-II), Part I: Description. Journal of Personality Disorders, 9, 83-91. Fischer, P. L., &Tomey, J. V. (1976). Influence of children's stories on dependency, a sex-typed behavior. Developmental Psychology, 12, 489-490.
REFERENCES
207
Fisher, G. (1969). Psychological needs of heterosexual pedophiliacs. Diseases of the Nervous System, 30, 419-421. Fisher, J. M., & Fisher, S. (1975). Response to cigarette deprivation as a function of oral fantasy. Journal of Personality Assessment, 39, 381-385. Fisher, P., Winne, P. H., &Ley, R. G. (1993). Group therapy for adult women survivors of child sexual abuse: Differentiation of completers versus dropouts. Psychotherapy, 30, 616-624. Fisher, S. (1970). Body experience in fantasy and behavior. New York: AppletonCentury-Crofts. Fisher, S., &. Greenberg, R. P. (1996). Freud scientifically reappraised: Testing the theories and therapy. New York: Wiley. Fleming, B., & Pretzer, J. L. (1990). Cognitive-behavioral approaches to personality disorders. In M. Hersen, R. M. Eisler, & P. M. Miller (Eds.), Progress in behavior modification (Vol. 25, pp. 119-151). Thousand Oaks, CA: Sage. Florian, V., Mikulincer, M., & Hirschberger, G. (2002). The anxiety-buffering function of close relationships: Evidence that relationship commitment acts as a terror management mechanism. Journal of Personality and Social Psychology, 82, 527-542. Fowler, J. C., Hilsenroth, M. J., & Handler, L. (1996). A multimethod approach to assessing dependency: The Early Memory Dependency Probe. Journal of Personality Assessment, 67, 399-413. Fowler, J. C., Hilsenroth, M. J., & Handler, L. (2000). Martin Mayman's early memories technique: Bridging the gap between personality assessment and psychotherapy. Journal of Personality Assessment, 75, 18—32. Frank, E., Kupfer, D. J., Jacob, M., & Jarrett, D. (1987). Personality features and response to acute treatment in recurrent depression. Journal of Personality Disorders, 1, 14-26. Franz, C. E., &. White, K. M. (1985). Individuation and attachment in personality development: Extending Erikson's theory. Journal of Personality, 53, 224-256. Freeman, A., & Leaf, R. C. (1989). Cognitive therapy applied to personality disorders. In A. Freeman, K. M. Simon, L. E. Beutler, & H. Arkowitz (Eds.), Comprehensive handbook of cognitive therapy (pp. 403-434). New York: Plenum. Freeman, A., Simon, K. M., Beutler, L. E., & Arkowitz, H. (Eds.). (1989). Comprehensive handbook of cognitive therapy. New York: Plenum. Freud, S. (1953). Three essays on the theory of sexuality. Standard edition of the complete, psychological works ofSigmund Freud (Vol. 7, pp. 125-248). London: Hogarth. (Original work published 1905) Freud, S. (1959). Character and anal erotism. Standard edition of the complete psychological works ofSigmund Freud (Vol. 9, pp. 167-176). London: Hogarth. (Original work published 1908) Friedlander, M. L., & Siegel, S. M. (1990). Separation-individuation difficulties and cognitive—behavioral indicators of eating disorders among college women. Journal of Counseling Psychology, 37, 74-78.
208
REFERENCES
Friedman, H. S., & Booth-Kewley, S. (1987). The disease-prone personality: A metaanalytic review of the construct. American Psychologist, 42, 539-555. Fromm, E. (1947). Man for himself. New York: Rinehart. Fu, V. R., Hinkle, D. E., & Hanna, M. A. K. (1986). A three-generation study of the development of individual dependency and family interdependence. Genetic Psychology Monographs, 112, 153-171. Fukuda, A., & Ogawa, K. (1988). Dependency and sex role in interpersonal relationships. Bulletin of the Faculty of Education, Yokohama National University, 28, 2139. Gabriel, S., & Gardner, W. L. (1999). Are there "his" and "hers" types of interdependence? The implications of gender differences in collective versus relational interdependence for affect, behavior, and cognition. Journal of Personality and Social Psychology, 77, 642-655. Galatzer-Levy, R. M., & Cohler, B. J. (1993). The essential other: A developmental psychology of the self. New York: Basic. Geller, J. D. (2003). Self-disclosure in psychoanalytic-existential therapy. Journal of Clinical Psychology, 59, 541-554. Gewirtz, J. L. (Ed.). (1972). Attachment and dependency. New York: Wiley. Gilbert, D. T., & Malone, P. S. (1995). The correspondence bias. Psychological Bulletin, 117,21-38. Gilbert, L. A. (1987). Male and female emotional dependency and its implications for the therapist-client relationship. Professional Psychology: Research and Practice, 18,555-561. Gjerde, P. F. (2001). Attachment, culture, and Amae. American Psychologist, 56, 826-827. Gleser, G. C., &. Ihilevich, D. (1969). An objective instrument for measuring defense mechanisms. Journal of Consulting and Clinical Psychology, 33, 51-60. Glover, E. (1925). Notes on oral character formation. International Journal of Psychoanalysis, 6, 131-154. Gold, J. R. (1994). When the patient does the integrating: Lessons for theory and practice. Journal of Psychotherapy Integration, 4, 133-158. Goldberg, J. O., Segal, Z. V., Vella, D. D., &. Shaw, B. F. (1989). Depressive personality: MCMI profiles of sociotropic and autonomous subtypes. Journal of Personality Disorders, 3, 193-198. Goldin, ]. G. (1972). Behavioral manifestations of dependency. Northwestern University Studies in Rehabilitation, 13, 15-31. Golding, ]. M., & Singer J. L. (1983). Patterns of inner experience: Daydreaming styles, depressive moods, and sex roles. Journal of Personality and Social Psychology, 45, 663-675. Goldman-Eisler, F. (1950). The etiology of the oral character in psychoanalytic theory. Journal of Personality, 19, 189-196.
REFERENCES
209
Goldman-Eisler, F. (1951). The problem of "orality" and its origin in early childhood. Journal of Mental Science, 97, 765-782. Golightly, C, Nelson, D., & Johnson, J. (1970). Children's Dependency Scale. Developmental Psychology, 3, 114-118. Goodstein, R. K. (1985). Common clinical problems in the elderly: Camouflaged by ageism and atypical presentation. Psychiatric Annals, 15, 299-312. Gordon, M., & Tegtemeyer, P.F.(1983). Oral dependent content in children's Rorschach protocols. Perceptual and Motor Skills, 57, 1163-1168. Gottesman, I. I. (1966). Genetic variance in adaptive personality traits. Journal of Child Psychology and Psychiatry, 7, 199-208. Gough, H. G. (1956). California Psychological Inventory administrator's guide. Palo Alto, CA: Consulting Psychologists Press. Greenberg, J. R., & Mitchell, S. J. (1983). Object relations in psychoanalytic theory. Cambridge, MA: Harvard University Press. Greenberg, R. P., & Bornstein, R. F. (1989). Length of psychiatric hospitalization and oral dependency. Journal of Personality Disorders, 3, 199-204. Greenberg, R. P., & Fisher, S. (1977). The relationship between willingness to adopt the sick role and attitudes toward women. Journal of Chronic Disease, 30, 29-37. Greenson, R. (1965). The working alliance and transference neurosis. Psychoanalytic Quarterly, 34, 158-181. Greenspan, S. I. (1989). The development of the ego. New York: International Universities Press. Gregory, B. A., & Gilbert, L. A. (1992). The relationship between dependency behavior in female clients and psychologists' perceptions of seductiveness. Professional Psychology: Research and Practice, 23, 390-396. Grilo, C. M., Anez, L. M., & McGlashan, T. H. (2002). DSM-IV-R Axis II comorbidity with borderline personality disorder in monolingual Hispanic psychiatric outpatients. Journal of Nervous and Mental Disease, 190, 324-330. Groth-Marnat, G. (1999). Handbook of psychological assessment (3rd ed.). New York: Wiley. Grove, W. M., Zald, D. H., Lebow, B. S., Snitz, B. E., & Nelson, C. (2000). Clinical versus mechanical prediction: A meta-analysis. Psychological Assessment, 12,1930. Gruen, R. J., & Blatt, S. J. (1990). Change in self- and object representation during long-term dynamically oriented treatment. Psychoanalytic Psychology, 7, 399— 422. Guntrip, H. (1961). Psychoanalytic theory, therapy, and the self. New York: Basic. Gurtman, M. B. (1992). Construct validity of interpersonal personality measures: The interpersonal circumplex as a nomological net. Journal of Personality and Social Psychology, 63, 105-118. Gurtman, M. B. (1996). Interpersonal problems and the psychotherapy context: The construct validity of the Inventory of Interpersonal Problems. Psychological Assessment, 8, 241-255.
210
REFERENCES
Haaga, D. A. F., Fine, J. A., Terrill, D. R., Stewart, B. L, & Beck, A. T. (1995). Social problem-solving deficits, dependency, and depressive symptoms. Cognitive Therapy and Research, 19, 147-158. Haley, J. (1976). Problem solving therapy. San Francisco, CA: Jossey-Bass. Halmi, K. A. (1997). Models to conceptualize risk factors for bulimia nervosa. Archives of General Psychiatry, 54, 507-509. Hammen, C., Ellicott, A., & Gitlin, M. (1989). Vulnerability to specific life events and prediction of course of disorder in unipolar depressed patients. Canadian journal of Behavioral Science, 21, 377-388. Handler, L., & Meyer, G. J. (1998). The importance of teaching and learning personality assessment. In L. Handler & M. J. Hilsenroth (Eds.), Teaching and learning personality assessment (pp. 3-30). Mahwah, NJ: Erlbaum. Harkness, A. R., Tellegen, A., & Waller, N. (1995). Differential convergence of selfreport and informant data for Multidimensional Personality Questionnaire traits. Journal of Personality Assessment, 64, 185-204. Harrison, F. C. (1987). Dependency-responsibility-morality: A metapsychological synthesis. American Journal of Social Psychiatry, 7, 245-252. Hart, S. D., Dutton, D. G., & Newlove, T. (1993). The prevalence of personality disorder among wife assaulters. Journal of Personality Disorders, 7, 329-341. Hartup, W. W. (1963). Dependence and independence. In H. W. Stevenson (Ed.), Child psychology (pp. 333-363). Washington, DC: National Society for the Study of Education. Hartup, W. W. (1989). Social relationships and their developmental significance. American Psychologist, 44, 120-126. Hartup, W. W. (1999). Friendships and adaptation across the life span. Current Directions in Psychological Science, 8, 76-79. Hassenfeld, I. N. (1999). "Generative caring" psychotherapy for patients who are reluctant to talk. American Journal of Psychotherapy, 53, 495-500. Hayes, S. C. (1989). Rule -governed behavior: Cognition, contingencies, and instructional control. New York: Plenum. Hayward, C., & King, R. (1990). Somatization and personality disorder traits in nonclinical volunteers. Journal of Personality Disorders, 4, 402-406. Head, S. B., Baker, J. D., & Williamson, D. A. (1991). Family environment characteristics and dependent personality disorder. Journal of Personality Disorders, 5, 256-263. Heathers, G. (1955). Acquiring dependence and independence: A theoretical orientation, journal of Genetic Psychology, 87, 277-291. Heiss, G. E., Herman, W. H., & Sperling, M. B. (1996). Five scales in search of a construct: Exploring continued attachment to parents in college students. Journal of Personality Assessment, 67, 102-115. Heller, K., & Goldstein, A. P. (1960). Client dependency and therapist expectancy as relationship-maintaining variables in psychotherapy. Journal of Consulting Psychology, 25, 371-375.
REFERENCES
211
Helweg, G. C., & Gaines, L. S. (1977). Subject characteristics and preferences for different approaches to psychotherapy: A multivariate study, journal of Consulting and Clinical Psychology, 45, 963-964. Hetherington, A. (1999). The development of trust and dependency in an individual through the lifespan and its influence on group dynamics. Social Science International, 15, 1-10. Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect. Psychological Review, 94, 319-340. Hiler, E. W. (1959). The sentence completion test as a predictor of continuation in psychotherapy. Journal of Consulting Psychology, 23, 544-549. Hill, E. L., Gold, S. N., & Bornstein, R. F. (2000). Interpersonal dependency among adult survivors of childhood sexual abuse in therapy. Journal of Child Sexual Abuse, 9,71-86. Hirschfeld, R. M. A., Klerman, G. L., Clayton, P. J., & Keller, M. B. (1983). Personality and depression: Empirical findings. Archives of General Psychiatry, 40,993998. Hirschfeld, R. M. A., Klerman, G. L., Cough, H. G., Barrett, J., Korchin, S. J., & Chodoff, P. (1977). A measure of interpersonal dependency. Journal of Personality Assessment, 41, 610-618. Hirschfeld, R. M. A., Shea, M. T., & Weise, R. (1991). Dependent personality disorder: Perspectives for DSM-IV-R. Journal of Personality Disorders, 5, 135-149. Hoare, P. (1984). Does illness foster dependency? A study of epileptic and diabetic children. Developmental Medicine and Child Neurology, 16, 20-24Hoffart, A., & Martinsen, E. W. (1993). The effect of personality disorders and anxious-depressive comorbidity on outcome in patients with unipolar depression and with panic disorder and agoraphobia. Journal of Personality Disorders, 7,304311. Hollender, M. C., Luborsky, L., & Harvey, R. (1970). Correlates of the desire to be held in women. Journal of Psychosomatic Research, 14, 387-390. Holmbeck, G. N., &. Wandrei, M. L. (1993). Individual and relational predictors of adjustment in first-year college students. Journal of Counseling Psychology, 40, 73-78. Holmes, J. (1997). Attachment, autonomy, intimacy: Some clinical implications of attachment theory. British Journal of Medical Psychology, 70, 231-248. Holtzworth-Monroe, A. (2000). A typology of men who are violent toward their female partners: Making sense of the heterogeneity in husband violence. Current Directions in Psychological Science, 9, 140-143. Holtzworth-Monroe, A., Stuart, G. L., & Hutchinson, G. (1997). Violent versus nonviolent husbands: Differences in attachment patterns, dependency, and jealousy. Journal of Family Psychology, 11,314-331. Hooker, K., & Kaus, C. R. (1994). Health-related possible selves in young and middle adulthood. Psychology and Aging, 9, 126-133. Hopkins, L. K. (1986). Dependency issues and fears in long-term psychotherapy. Psychotherapy, 23, 535-539.
212
REFERENCES
Horney, K. (1945). Our inner conflicts. New York: W. W. Norton. Horowitz, M. J. (Ed.). (1991). Person schemas and maladaptive interpersonal patterns. Chicago: University of Chicago Press. Huffcut, A. I., Conway, J. M., Roth, P. L., & Stone, N. J. (2001). Identification and meta-analy tic assessment of psychological constructs measured in employment interviews. Journal of Applied Psychology, 86, 897-913. Hull, C. L. (1943). Principles of behavior. New York: Appleton-Century-Crofts. Hunsley, ]., &. Bailey, J. M. (1999). The clinical utility of the Rorschach: Unfulfilled promises and an uncertain future. Psychological Assessment, II, 266-277. Hunt, E., Browning, P., &. Nave, G. (1982). A behavioral explanation of dependent and independent mildly mentally retarded adolescents and their mothers. Applied Research in Mental Retardation, 3, 141-150. Huprich, S. K. (2001). Object loss and object relations in depressive personality analogues. Bulletin of the Menninger Clinic, 65, 549-559. Hyler, S. E., Rieder, R. O., Williams, J. B. W., Spitzer, R. L., Hendler, J., & Lyons, M. (1988). The Personality Diagnostic Questionnaire: Development and preliminary results. Journal of Personality Disorders, 2, 229-237. Ihilevich, D., &. Gleser, G. C. (1986). The Defense Mechanisms Inventory. Owosso, MI: DMI Associates. Ihilevich, D., &. Gleser, G. C. (1991). Defenses in psychotherapy. Owosso, MI: DMI Associates. Jackson, H. J., Whiteside, H. L., Bates, G. W., Bell, R., Rudd, R. P., &. Edwards, J. (1991). Diagnosing personality disorders in psychiatric inpatients. Acta Ps^chiatrica Scandinawca, 83, 206-213. Jacobsberg, L., Perry, S., & Frances, A. (1995). Diagnostic agreement between the SCID-II screening questionnaire and the Personality Disorder Examination. Journal of Personality Assessment, 65, 428-433. Jacobson, E. (1964). The self and object world. New York: International Universities Press. Jakubczak, L. F., & Walters, R. H. (1959). Suggestibility as dependency behavior. Journal of Abnormal and Social Psychology, 59, 102-107. Jansen, M. A., Arntz, A., Merckelbach, H., & Mersch, P. P. A. (1994). Personality disorders and features in social phobia and panic disorder. Journal of Abnormal Psychology, 103, 391-395. Joffe, R. T., &. Regan, J. J. (1988). Personality and depression. Journal of Psychiatric Research, 22, 279-286. Johnson, F. A. (1993). Dependency and Japanese socialization. New York: New York University Press. Johnson, J. G., &. Bornstein, R. F. (1992). Utility of the Personality Diagnostic Questionnaire—Revised in a nonclinical population. Journal of Personality Disorders, 6,450-457.'
REFERENCES
213
Johnson, J. G., Bornstein, R. F., & Krukonis, A. B. (1992). Defense styles as predictors of personality disorder symptomatology. Journal of Personality Disorders, 6, 409-416. Johnson, J. G., Rabkin, J. G., Williams, J. B. W., Remien, R. H., &. Gorman, J. M. (2000). Difficulties in interpersonal relationships associated with personality disorders and Axis I disorders: A community-based longitudinal investigation. Journal of Personality Disorders, 14, 42-56. Jones, E. E., & Price, P. B. (1998). Interaction structure and change in psychoanalytic therapy. In R. F. Bornstein & J. M. Masling (Eds.), Empirical studies of the therapeutic hour (pp. 27-62). Washington, DC: American Psychological Association. Jones, M. B. (1957). The Pensacola Z Survey: A study in the measurement of authoritarian tendency. Psychological Monographs, 71, 1-19. Juni, S., & LoCascio, R. (1985). Preference for counseling and therapy as related to preoedipal fixation. Psychological Reports, 56, 431-438. Kagan, J., &. Moss, H. A. (1960). The stability of passive and dependent behavior from childhood through adulthood. Child Development, 31, 577-591. Kagan, J., & Mussen, P. (1956). Dependency themes on the TAT and group conformity. Journal of Consulting Psychology, 20, 29-32. Kalmuss, D. S., & Straus, M. A. (1982). Wife's marital dependency and wife abuse. Journal of Marriage and the Family, 44, 277-286. Kamitami, Y. (1993). The structure of jiritsu (socially sensitive independence) in young Japanese women. Psychological Reports, 72, 855-866. Kantor, M. (1992). Diagnosis and treatment of the personality disorders. St. Louis, MO: Ishiyaku EuroAmerica. Kantor, M. (1993). Distancing. Westport, CT: Praeger. Kaplan, B. (1975). Toward further research on family and health. In B. Kaplan & J. Cassel (Eds.), Family and health: An epidemiological approach (pp. 89-106). Chapel Hill, NC: Institute for Research in Social Science. Kaplan, M. (1983). A woman's view of DSM-III. American Psychologist, 38, 786792. Karliner, R., Westrich, E., Shedler, J., & Mayman, M. (1996). The Adelphi Early Memory Index: Bridging the gap between psychodynamic and scientific psychology. In J. M. Masling & R. F. Bornstein (Eds.), Psychoanalytic perspectives on developmental psychology (pp. 43-67). Washington, DC: American Psychological Association. Katz, M. M., & Lyerly, S. B. (1963). Methods for measuring adjustment and social behavior in the community. Psychological Reports Monographs, 13, 503-535. Kaul, V., Mathur, P., & Murlidharan, R. (1982). Dependency and its antecedents: A review. Indian Educational Review, 17, 35—46. Kazdin, A. E. (1989). Behavior modification in applied settings. Homewood, IL: Dorsey Press. 214
REFERENCES
Keinan, G., & Hobfoll, S. E. (1989). Stress, dependency, and social support: Who benefits from husband's presence in delivery? Journal of Social and Clinical Psychology, 8, 32-44. Kelly, G. A. (1955). The psychology of personal constructs (Vols. 1 & 2). New York: Norton. Kernberg, O. F. (1975). Borderline conditions and pathological narcissism. New York: Jason Aronson. Kernberg, O. F. (1984). Severe personality disorders. New Haven, CT: Yale University Press. Kertzman, D. (1980). Dependency, frustration tolerance, and impulse control in child abusers. Saratoga, CA: Century Twenty-One. Kilbourne, B. K., & Kilbourne, M. T. (1983). Concurrent feelings of power and dependency in a sheltered workshop for chronic neuropsychiatrics. International journal of Social Psychiatry, 29, 173-179. Klein, D. N. (1989). The Depressive Experiences Questionnaire: A further evaluation. Journal of Personality Assessment, 53, 703-715. Klein, D. N. (2003). Patients' versus informants' reports of personality disorders in predicting 7-year outcome in outpatients with depressive disorders. Psychological Assessment, 15, 216-222. Klein, D. N., Harding, K., Taylor, E. B., & Dickstein, S. (1988). Dependency and self-criticism in depression: Evaluation in a clinical population. Journal of Abnormal Psychology, 97, 399-404. Kline, P., &. Storey, R. (1980). The etiology of the oral character, journal of Genetic Psychology, 136, 85-94. Klonsky, E. D., Jane, J. S., Turkheimer, E., & Oltmanns, T. F. (2002). Gender role and personality disorders. Journal of Personality Disorders, 16, 464-476. Kochanska, G., Friesenberg, A. E., Lange, L. A., & Martel, M. M. (2004). Parents' personality and infants' temperament as contributors to their emerging relationship. Journal of Personality and Social Psychology, 86, 744-759. Kohut, H. (1971). The analysis of the self. New York: International Universities Press. Kohut, H. (1977). The restoration of the self. New York: International Universities Press. Kraeplin, E. (1913). Psychiatrie: Ein lehrbuch. Leipzig: Barth. Kuperminc, G. P., Blatt, S. J., & Leadbeater, B. J. (1997). Relatedness, self-definition, and early adolescent adjustment. Cognitive Therapy and Research, 21, 301320. Kwon, P., Campbell, D. G., & Williams, M. G. (2001). Sociotropy and autonomy: Preliminary evidence for construct validity using TAT narratives. Journal of Personality Assessment, 77, 128-138. Lampropoulos, G. K., Spengler, P. M., Dixon, D. N., & Nicholas, D. R. (2002). How psychotherapy integration can complement the scientist-practitioner model. Journal of Clinical Psychology, 58, 1227-1240.
REFERENCES
215
Langer, E. J. (1989). Umdfulness. Reading, MA: Addison Wesley. Langer, E. J., & Rodin, J. (1976). The effects of choice and enhanced personal responsibility for the aged. Journal of Personality and Social Psychology, 34, 191198. Lanyon, R. I., & Goodstein, L. D. (1997). Personality assessment. New York: Wiley. Lapan, R., &. Patton, M. J. (1986). Self-psychology and the adolescent process: Measures of pseudoautonomy and peer-group dependence. Journal of Counseling Psychology, 33, 136-142. Lauer, J. W. (1976). The effect of tricyclic antidepressant compounds on patients with passive-dependent traits. Current Therapeutic Research, 19, 495-505. Lavan, H., & Johnson, J. G. (2002). The association between Axis I and Axis II psychiatric symptoms and high-risk sexual behavior during adolescence. Journal of Personality Disorders, 16, 73-94Lazare, A., Klerman, G. L., &. Armor, D. J. (1966). Oral, obsessive, and hysterical personality patterns. Archives of General Psychiatry, 14, 624—630. Lazarus, A. A. (1992). Multimodal therapy: Technical eclecticism with minimal integration. In ]. C. Norcross & M. R. Goldfried (Eds.), Handbook of psychotherapy integration (pp. 231-262). New York: Basic Books. Leary, T. (1957). Interpersonal diagnosis of personality. New York: Ronald. Lee, R. M., &Robbins, S. B. (1995). Measuring belongingness: The Social Connectedness and Social Assurance scales. Journal of Counseling Psychology, 42, 232241. Lee, R. M., & Robbins, S. B. (1998). The relationship between social connectedness and anxiety, self-esteem, and social identity. Journal of Counseling Psychology, 45, 338-345. Leeman, C. P., & Mulvey, C. H. (1975). Brief psychotherapy of the dependent personality: Specific techniques. Psychotherapy and Psychosomatics, 25, 36-42. Lemery, K. S., Goldsmith, H. H., Klinnert, M. D, &Mrazek, D. A. (1999). Developmental models of infant and childhood temperament. Developmental Psychology, 35, 189-204. Lerner, P. (1991). Psychoanalytic theory and the Rorschach. Hillsdale, NJ: Erlbaum. Lerner, R. M., &. Tubman, J. G. (1989). Conceptual issues in studying continuity and discontinuity in personality development across life. Journal of Personality, 57, 343-374. Levit, D. B. (1991). Gender differences in ego defenses in adolescence. Journal of Personality and Social Psychology, 61, 992-999. Li, S. C. (2002). Connecting the many levels and facets of cognitive aging. Current Directions in Psychological Science, 11, 38-43. Lin, N., Dean, A., & Ensel, W. (1986). Social support, life events, and depression. New York: Academic. Linehan, M. (1993). Cognitive behavioral therapy for borderline personality disorder. New York: Guilford.
216
REFERENCES
Lingiardi, V., Lonati, C., Delucchi, F., Fossati, A., Vanzulli, L., & Maffei, C. (1999). Defense mechanisms and personality disorders. Journal of Nervous and Mental Disease, 187, 224-228. Little, S. A., & Garber, ]. (2000). Interpersonal and achievement orientations and specific stressors predicting depressive and aggressive symptoms in children. Cognitive Therapy and Research, 24, 651—670. Livesley, W. J., Schroeder, M. L., &. Jackson, D. N. (1990). Dependent personality disorder and attachment problems. Journal of Personality Disorders, 4, 131-140. Livesley, W. K., & Jang, K. L. (2000). Toward an empirically based classification of personality disorder, journal of Personality Disorders, 14, 137-151. Livesley, W. K., Jang, K. L., Jackson, D. N., & Vernon, P. A. (1993). Genetic and environmental contributions to dimensions of personality disorder. American journal of Psychiatry, 150, 1826-1831. Loas, G., Corcos, M., Perez-Diaz, F., Verrier, A., Guelfi, J. D., Halfon, O., et al. (2002). Criterion validity of the Interpersonal Dependency Inventory: A preliminary study of 621 addictive subjects. European Psychiatry, 17, 477-478. Loranger, A. W. (1995). International Personality Disorder Examination (IPDE) manual. Unpublished manuscript, Cornell University Medical College. Loranger, A. W., Sartorius, N., Andreoli, A., Berger, P., Buchheim, P., Channabasavanna, S. M., et al. (1994). The International Personality Disorder Examination. Archives of General Psychiatry, 51, 215—224. Lorr, M., &McNair, D. M. (1964a). Correlates of length of psychotherapy. Journal of Clinical Psychology, 20, 497-504. Lorr, M., & McNair, D. M. (1964b). The interview relationship in therapy. Journal of Nervous and Mental Disease, 139, 328-331. Lowenstein, R. M. (1951). The problem of interpretation. Psychoanalytic Quarterly, 20, 1-14. Lower, R. B. (1967). Psychotherapy of neurotic dependency. American Journal of Psychiatry, 124,514-519. Luborsky, L. (1984). Principles of psychodynamic psychotherapy: A manual for supportive-expressive treatment. New York: Basic. Luborsky, L., & Crits-Christoph, P. (1990). Understanding transference: The Core Conflictual Relationship Theme method. New York: Basic. Maffei, C., Fossati, A., Lingiardi, V., Madeddu, F., Borellini, C., & Petrachi, M. (1995). Personality maladjustment, defenses, and psychopathological symptoms in nonclinical subjects. Journal of Personality Disorders, 9, 330-345. Mahon, N. E. (1982). The relationship of self-disclosure, interpersonal dependency, and life changes to loneliness in young adults. Nursing Research, 31, 343-347. Mahoney, M. J. (1976). Scientist as subject. Cambridge, MA: Ballinger. Mahoney, M. J. (1977). Publication prejudices: An experimental study of confirmatory bias in the manuscript review system. Cognitive Therapy and Research, 1, 161-175.
REFERENCES
217
Mahoney, M. J. (1987). Scientific publication and knowledge politics, journal of Social Behavior and Personality, 2, 165-176. Main, M., Kaplan, N., & Cassidy, J. (1985). Security in infancy, childhood, and adulthood. Monographs of the Society for Research in Child Development, 50, 66104. Maling, M. S., Gurtman, M. B., & Howard, K. I. (1995). The response of interpersonal problems to varying doses of psychotherapy. Psychotherapy Research, 5, 63-75. Marcia, J. E. (1993). The relational roots of identity. In J. Kroger (Ed.), Discussions on ego identity (pp. 118-128). Hillsdale, NJ: Erlbaum. Markus, H. (1977). Self-schemata and processing information about the self. Journal of Personality and Social Psychology, 35, 63-78. Markus, H., &Nurius, P. (1986). Possible selves. American Psychologist, 41,954-969. Marlatt, G. A., & Gordon, J. (Eds.). (1985). Relapse prevention. New York: Guilford. Marshall, H. R., & McCandless, B. R. (1957). Relationships between dependence on adults and social acceptance by peers. Child Development, 28, 413-419. Marshall,]. R., &Neill, J. (1977). The removal of a psychosomatic symptom: Effects on the marriage. Family Process, 16, 273—280. Martin, G., & Pear, J. (1996). Behavior modification. Upper Saddle River, NJ: Simon &. Schuster. Masling, J. M. (1960). The influences of interpersonal and situational variables in projective testing. Psychological Bulletin, 57, 65-85. Masling, J. M. (1986). Orality, pathology, and interpersonal behavior. InJ. Masling (Ed.), Empirical studies of psychoanalytic theories (Vol. 2, pp. 73-106). Hillsdale, NJ: Erlbaum. Masling, J. M., & Cohen, I. S. (1987). Psychotherapy, clinical evidence, and the self-fulfilling prophecy. Psychoanalytic Psychology, 4, 65-79. Masling, J. M., Johnson, C., & Saturansky, C. (1974). Oral imagery, accuracy of perceiving others, and performance in Peace Corps training. Journal of Personality and Social Psychology, 30, 414-419. Masling, J. M., O'Neill, R. M., & Jayne, C. (1981). Orality and latency of volunteering to serve as experimental subjects. Journal of Personality Assessment, 45, 2022, Masling, J. M., O'Neill, R. M., & Katkin, E. S. (1982). Autonomic arousal, interpersonal climate, and orality. Journal of Personality and Social Psychology, 40, 395400. Masling, J. M., Rabie, L., & Blondheim, S. H. (1967). Obesity, level of aspiration, and Rorschach and TAT measures of oral dependence. Journal of Consulting Psychology, 31,233-239. Masling, J. M., Schiffner, J., & Shenfeld, M. (1980). Client perception of the therapist, orality, and sex of client and therapist. Journal of Counseling Psychology, 27, 294-298.
218
REFERENCES
Masling, J. M., & Schwartz, M. A. (1979). A critique of research in psychoanalytic theory. Genetic Psychology Monographs, 100, 257-307. Masling, J. M., Weiss, L. R., & Rothschild, B. (1968). Relationships of oral imagery to yielding behavior and birth order. Journal of Consulting and Clinical Psychology, 32, 89-91. Maslow, A. H. (1958). Emotional blocks to creativity. Journal of Individual Psychology, 14, 51-56. Maslow, A. H. (1968). Toward a psychology of being. New York: Van Nostrand. Mavissakalian, M., &. Hamann, M. S. (1987). DSM-III personality disorder in agoraphobia: II. Changes with treatment. Comprehensive Psychiatry, 28, 356-361. May, R. (1972). Power and innocence. New York: Norton. May, R. (1981). Freedom and destiny. New York: Norton. May, R., & Yalom, I. (2000). Existential psychotherapy. In R. J. Corsini & D. Wedding (Eds.), Current psychotherapies (pp. 273-302). Itasca, IL: F. E. Peacock. Mayman, M. (1968). Early memories and character structure. Journal of Projective Techniques and Personality Assessment, 32, 303-316. Mayman, M. (1976). Psychoanalytic theory in retrospect and prospect. Bulletin of the Menninger Clinic, 40, 199-210. McCallum, M., & Piper, R. E. (1990). The psychological mindedness assessment procedure. Psychological Assessment, 2, 412-418. McCandless, B. R., Bilous, C. B., & Bennett, H. L. (1961). Peer popularity and dependence on adults in pre-school-age socialization. Child Development, 32,511518. McClelland, D. C., Koestner, R., & Weinberger, J. (1989). How do self-attributed and implicit motives differ? Psychological Review, 96, 690-702. McCranie, E. W., & Bass, J. D. (1984). Childhood family antecedents in dependency and self-criticism. Journal of Abnormal Psychology, 93, 3-8. McFall, R. M., & Marston, A. (1970). An experimental investigation of behavioral rehearsal in assertive training. Journal of Abnormal Psychology, 76, 295-303. McGrath, R. E. (2001). Toward more clinically relevant assessment research. Journal of Personality Assessment, 77, 307-332. McKeegan, G. F., Geczy, B., & Donat, D. C. (1993). Applying behavioral methods in the inpatient setting: Patients with mixed borderline and dependent traits. Ps;ychosocial Rehabilitation Journal, 16, 55-64. McMahon, R. C., Kelley, A., & Kouzekanani, K. (1993). Personality and coping styles in the prediction of dropout from treatment for cocaine abuse. Journal of Personality Assessment, 61, 147-155. Meichenbaum, D. H. (1985). Stress inoculation training. Elmsford, NY: Pergamon. Melley, A. H., Oltmanns, T. F., & Turkheimer, E. (2002). The Schedule for Nonadaptive and Adaptive Personality (SNAP): Temporal stability and predictive validity of the diagnostic scales. Assessment, 9, 181-187.
REFERENCES
219
Melnick, B., &. Hurley, J. R. (1969). Distinctive personality attributes of childabusing mothers, journal of Consulting and Clinical Psychology, 33, 746-749. Mersky, H., & Trimble, M. (1979). Personality, sexual adjustment, and brain lesions in patients with conversion symptoms. American Journal of Psychiatry, 136,179182. Messer, S. B. (1986). Behavioral and psychoanalytic perspectives at therapeutic choice points. American Psychologist, 41, 1261-1272. Messer, S. B. (1992). A critical examination of belief structures in integrative and eclectic psychotherapy. In J. C. Norcross &. M. R. Goldfried (Eds.), Handbook of psychotherapy integration (pp. 130-165). New York: Basic. Messer, S. B., & Warren, C. S. (1995). Models of brief psychodynamic psychotherapy: A comparative approach. New York: Guilford. Meyer, B., Pilkonis, P. A., Proietti, J. M., Heape, C. L., & Egan, M. (2001). Attachment styles and personality disorders as predictors of symptom course. Journal of Personality Disorders, 15, 371-389. Meyer, G. J., Finn, S. E., Eyde, L. D., Kay, G. G., Moreland, K. L., Dies, R. R., et al. (2001). Psychological testing and psychological assessment: A review of evidence and issues. American Psychologist, 56, 128-165. Mezzich, J. E., Fabrega, H., & Coffman, G. A. (1987). Multiaxial characteristics of depressive patients. Journal of Nervous and Mental Disease, 175, 339-346. Mihura, J. L., Meyer, G. J., Bel-Bahar, T., &. Gunderson, J. (2003). Correspondence among observer ratings of Rorschach, Big Five Model, and DSM-IV-R personality disorder constructs. Journal of Personality Assessment, 81, 20-39. Miller, D. R., & Stine, M. W. (1951). The prediction of social acceptance by means of psychoanalytic concepts. Journal of Personality, 20, 162-174. Miller, S. M. (1979). Interrelationships among dependency, empathy, and sharing. Motivation and Emotion, 3, 183-199. Millon, T. (1977). Millon Clinical Multiaxial Inventory manual. Minneapolis, MN: National Computer Systems. Millon, T. (1981). Disorders of personality: DSM-lll, Axis II. New York: Wiley. Millon, T. (1987). Millon Clinical Multiaxial Inventory-II manual. Minneapolis, MN: National Computer Systems. Millon, T. (1990). Toward a newpersonology: An evolutionary model. New York: Wiley. Millon, T. (1996). Disorders of personality: DSM-IV-R and beyond. New York: Wiley. Millon, T., Millon, C., & Davis, R. (1994). Millon Clinical Multiaxial Inventory-Ill. Minneapolis, MN: National Computer Systems. Minuchin, P. (2002). Cross-cultural perspectives: Implications for attachment theory and family therapy. Family Process, 41, 546—550. Minuchin, S. (1974). Families and family therapy. Cambridge, MA: Harvard University Press. Mischel, W. (1973). Toward a cognitive social learning reconceptualization of personality. Psychological Review, 80, 252—283. 220
REFERENCES
Mischel, W. (1979). On the interface of cognition and personality: Beyond the person-situation debate. American Psychologist, 34, 740-754. Mischel, W. (1984). Convergences and challenges in the search for consistency. American Psychologist, 39, 351-364. Mischel, W., Shoda, Y., & Mendoza-Denton, R. (2002). Situation-behavior profiles as a locus of consistency in personality. Current Directions in Psychological Science, 11, 50-54. Mongrain, M. (1998). Parental representations and support-seeking-behaviors related to dependency and self-criticism, journal of Personality, 66, 151-173. Mongrain, M., Lubbers, R., & Struthers, W. (2004). The power of love: Mediation of rejection in roommate relationships of dependents and self-critics. Personality and Social Psychology Bulletin, 30, 94-105. Mongrain, M., Vettese, L. C., Shuster, B., & Kendal, N. (1998). Perceptual biases, affect, and behavior in the relationships of dependents and self-critics. Journal of Personality and Social Psychology, 75, 230-241. Moore, R. G., &. Blackburn, I. M. (1996). The stability of sociotropy and autonomy in depressed patients undergoing treatment. Cognitive Therapy and Research, 20, 69-80. Moore, S. G., & Updegraff, R. (1964). Sociometric status of preschool children as related to age, sex, nurturance-giving, and dependence. Child Development, 35, 519-524. Morrison, A. T., & Hunt, H. T. (1996). "Reals," "roles," and demand features: A critical look at interview versus questionnaire measures of subjective states. Canadian Psychology, 37, 112-119. Mowrer, O. H. (1950). Learning theory and personality dynamics. New York: Ronald. Murphy, C. M., Meyer, S. L., & O'Leary, K. D. (1994). Dependency characteristics of partner assaultive men. Journal of Abnormal Psychology, 103, 729-735. Murray, H. A. (1938). Explorations in personality. New York: Oxford University Press. Nace, E. P., Davis, C. W., & Gaspari, J. P. (1991). Axis II comorbidity in substance abusers. American Journal of Psychiatry, 148, 118-120. Nacev, V. (1980). Dependency and ego strength as indicators of patients' attendance in psychotherapy. Journal of Clinical Psychology, 36, 691-695. Nagurney, A. J., Reich, J. W., & Newsom, J. T. (2004). Gender moderates the effects of independence and dependence desires during the social support process. Psychology and Aging, 19, 215-218. Narduzzi, K. J., &. Jackson, T. (2000). Personality differences between eatingdisordered women and a nonclinical comparison sample: A discriminant classification analysis. Journal of Clinical Psychology, 56, 699-710. Narduzzi, K. J., & Jackson, T. (2002). Sociotropy-dependency and autonomy as predictors of eating disturbance among Canadian female college students. Journal of Genetic Psychology, 163, 389-401. Navran, L. (1954). A rationally derived MMPI scale to measure dependence. Journal of Consulting Psychology, 18, 192. REFERENCES
221
Neimeyer, R. A. (1992). Constructivist approaches to the measurement of meaning. In G. J. Neimeyer (Ed.), Handbook of constructivist assessment (pp. 58-103). Newbury Park, CA: Sage. Neki, J. S. (1976). An examination of the cultural relativism of dependence as a dynamic of social and therapeutic relationships: I. Socio-developmental. British journal of Medical Psychology, 49, 1-10. Nelson, D. R., Hammen, C., Daley, S. E., Burge, D., &Davila, J. (2001). Sociotropic and autonomous personality styles: Contributions to chronic life stress. Cognitive Therapy and Research, 25, 61-76. Nelson-Gray, R. O., Johnson, D., Foyle, L. W., Daniel, S. S., & Harmon, R. (1996). The effectiveness of cognitive therapy tailored to depressives with personality disorders. Journal of Personality Disorders, 10, 132-152. Neuberg, S. L., &. Fiske, S. T. (1987). Motivational influences on impression formation: Outcome dependency, accuracy-driven attention, and individuating processes. Journal of Personality and Social Psychology, 53, 431-444. Nietzel, M. T., &. Harris, M. J. (1990). Relationship of dependency and achievement/autonomy to depression. Clinical Psychology Review, 10, 279—297. Norcross, J. C., & Goldfried, M. R. (Eds.). (1992). Handbook of psychotherapy integration. New York: Basic. Norcross, J. C., Karg, R. S., & Prochaska, J. O. (1997). Clinical psychologists in the 1990s. Clinical Psychologist, 50, 4-9. Norcross, J. C., & Newman, C. F. (1992). Psychotherapy integration: Setting the context. In]. C. Norcross & M. R. Goldfried (Eds.), Handbook of psychotherapy integration (pp. 3—45). New York: Basic. O'Boyle, M. (1993). Personality disorder and multiple substance dependence. Journal of Personality Disorders, 7, 342-347. Ojha, H. (1972). The relation of prestige suggestion to rigidity and dependence proneness. Journal of Psychological Research, 16, 70-73. Ojha, H., &. Singh, R. I. P. (1985). Relationship of marriage role attitude with dependence proneness and insecurity in university students. Ps^chologia, 28, 249253. Ojha, R. S. (1978). Reaction time as a function of dependence proneness. Ps^chofogical Studies, 33, 75-79. Oldham, J. M., Skodol, A. E., Kellman, H. D., Hyler, S. E., Doidge, N, Rosnick, L., & Gallagher, P. E. (1995). Comorbidity of Axis I and Axis II disorders. American Journal of Psychiatry, 152, 571-578. O'Neill, F. A., &. Kendler, K. S. (1998). Longitudinal study of interpersonal dependency in female twins. British Journal of Psychiatry, 172, 154-158. O'Neill, R. M., &. Bornstein, R. F. (1990). Oral dependence and gender: Factors in help-seeking response set and self-reported psychopathology in psychiatric inpatients. Journal of Personality Assessment, 55, 28-40. O'Neill, R. M., &. Bornstein, R. F. (1996). Dependency and alexithymia in psychiatric inpatients. Journal of Nervous and Mental Disease, 184, 302-306. 222
REFERENCES
O'Neill, R. M., &. Bornstein, R. F. (2001). The dependent patient in a psychiatric inpatient setting: Relationship of interpersonal dependency to consultation and medication frequencies. ]oumol of Clinical Psychology, 57, 289-298. Osofsky, J. D. (1995). Perspectives on attachment and psychoanalysis. Psychoanalytic Psychology, 12, 347-362. Osofsky, J. D., & O'Connell, E. J. (1972). Parent-child interaction: Daughters' effects upon mothers' and fathers' behaviors. Developmental Psychology, 7, 157168. Ottens, A. J., & Hanna, F. J. (1998). Cognitive and existential therapies: Toward an integration. Psychotherapy, 35, 312-324Overholser, J. C. (1987). Facilitating autonomy in passive-dependent persons: An integrative model. Journal of Contemporary Psychotherapy, 17, 250-269. Overholser, J. C. (1990). Emotional reliance and social loss: Effects on depressive symptomatology, journal of Personality Assessment, 55, 618-629. Overholser, J. C. (1992). Interpersonal dependency and social loss. Personality and Individual Differences, 13, 17-23. Overholser, J. C. (1993). Elements of the Socratic method: I. Systematic questioning. Psychotherapy, 30, 67-74. Overholser, J. C. (1996). The dependent personality and interpersonal problems. Journal of Nervous and Mental Disease, 184, 8-16. Overholser, J. C. (1997). Treatment of excessive interpersonal dependence: A cognitive-behavioral model. Journal of Contemporary Psychotherapy, 27, 283-301. Overholser, J. C., & Fine, M. A. (1994). Cognitive-behavioral treatment of excessive interpersonal dependency: A four-stage psychotherapy model. Journal of Cognitive Psychotherapy, 8, 55-70. Overholser, J. C., & Freiheit, S. (1994). Assessment of interpersonal dependency using the Millon Clinical Multiaxial Inventory-II (MCMI-II) and the Depressive Experiences Questionnaire. Personality and Individual Differences, 17, 7178. Overholser, J. C., Kabakoff, R., & Norman, W. H. (1989). The assessment of personality characteristics in depressed and dependent psychiatric inpatients. Journal of Personality Assessment, 53, 40-50. Oyserman, D., Coon, H. M., & Kemmelmeier, M. (2002). Rethinking individualism and collectivism: Evaluation of theoretical assumptions and meta-analyses. Psychological Bulletin, 128, 3-72. Pallis, D. J., & Birtchnell, J. (1976). Personality and suicidal history in psychiatric inpatients. Journal of Clinical Psychology, 32, 246-253. Paris, J. (1998). Working with traits: Psychotherapy of personality disorders. Northvale, NJ: Jason Aronson. Parker, G., & Lipscombe, P. (1980). The relevance of early parental experiences to adult dependency, hypochondriasis, and utilization of primary care physicians. British Journal of Medical Psychology, 53, 355-363. REFERENCES
223
Paykel, E. S., & Dienelt, M. N. (1971). Suicide attempts following acute depression. Journal of Nervous and Mental Disease, 153, 234-243. Perry, J. C. (1989). Dependent personality disorder. In American Psychiatric Associaion (Ed.), Treatments of psychiatric disorders (pp. 2649-2659). Washington, DC: Editor. Peselow, E. D., Robins, C. J., Sanfilipo, M. P., Block, P., & Fieve, R. R. (1992). Sociotropy and autonomy: Relationship to antidepressant drug treatment response and endogenous-nonendogenous dichotomy. Journal of Abnormal Psychobgy, 101,479-486. Peselow, E. D., Sanfilipo, M. P., & Fieve, R. R. (1995). Relationship between hypomania and personality disorders before and after successful treatment. American journal of Psychiatry, 152, 232-238. Pfohl, B., Blum, N., Zimmerman, M., & Stangl, D. (1989). Structured Interview for DSM-lll-R Personality (SIDP-R). Iowa City, IA: Department of Psychiatry, University of Iowa. Piaget, J. (1954). The construction of reality in the child. New York: Basic. Pillemer, K. A., & Wolf, R. S. (Eds.). (1986). Elder abuse: Conflict in the community. Dover, MA: Auburn. Pincus, A. L. (1994). The interpersonal circumplex and the interpersonal theory: Perspectives on personality and its pathology. In S. Strack & M. Lorr (Eds.), Differentiating normal and abnormal personality (pp. 114—136). New York: Springer. Pincus, A. L. (2002). Constellations of dependency within the five-factor model of personality. In P. T. Costa & T. A. Widiger (Eds.), Personality disorders and the five-factor model of personality (pp. 203—214). Washington, DC: American Psychological Association. Pincus, A. L., & Gurtman, M. B. (1995). The three faces of interpersonal dependency: Structural analysis of self-report dependency measures. Journal of Personality and Social Psychology, 69, 744-758. Pincus, A. L., &. Wilson, K. R. (2001). Interpersonal variability in dependent personality. Journal of Personality, 69, 223-251. Piper, W. E., Ogrodniczuk, J. S., Joyce, A. S., McCallum, M., Weideman, R., & Azim, H. F. (2001). Ambivalence and other relationship predictors of grief in psychiatric outpatients. Journal of Nervous and Mental Disease, 189, 781-787. Poldrugo, F., & Forti, B. (1988). Personality disorders and alcoholism treatment outcome. Drug and Alcohol Dependence, 21, 171—176. Priel, B., & Besser, A. (2000). Dependency and self-criticism among first-time mothers: The roles of global and specific support. Journal of Social and Clinical Psychology, 19, 437-450. Priel, B., & Shahar, G. (2000). Dependency, self-criticism, social context, and distress: Comparing moderating and mediating models. Personality and Individual Differences, 28, 515-525. Procidano, M. E. (1992). The nature of perceived social support: Findings of metaanalytic studies. In C. D. Spielberger &. J. N. Butcher (Eds.), Advances in personality assessment (Vol. 9, pp. 1-26). Hillsdale, NJ: Erlbaum. 224
REFERENCES
Procidano, M. E., &. Heller, K. (1983). Measures of perceived social support from friends and from family: Three validation studies. American Journal of Community Psychology, 11, 1-24. Pyke, J. M. (1982). Dependency issues in long-term treatment of schizophrenia. Issues in Mental Health Nursing, 4, 77-85. Pyszczynski, T., Greenberg, J., & Solomon, S. (1997). Terror management theory of self-esteem and cultural worldviews: Empirical assessment and conceptual refinements. In M. Zanna (Ed.), Advances in experimental social psychology (Vol. 29, pp. 61-139). San Diego, CA: Academic. Pyszczynski, T., Greenberg, ]., &. Solomon, S. (2000). Proximal and distal defense: A new perspective on unconscious motivation. Current Directions in Psychological Science, 9, 156-160. Ramsay, ]. R. (2001). The clinical challenges of assimilative integration. Journal of Psychotherapy Integration, II, 21-42. Raps, C. S., Peterson, C., Jonas, M., & Seligman, M. E. P. (1982). Patient behavior in hospitals: Helplessness, reactance, or both? Journal of Personality and Social Psychology, 42, 1036-1041. Rathus, J. H., & O'Leary, K. D. (1997). Spouse-Specific Dependency Scale: Scale development. Journal of Family Violence, 12, 159-168. Rathus, ]. H., Sanderson, W. C., Miller, A. L, & Wetzler, S. (1995). Impact of personality functioning on cognitive behavioral treatment of panic disorder: A preliminary report. Journal of Personality Disorders, 9, 160-168. Ready, R. E., & Clark, L. A. (2002). Correspondence of psychiatric patient and informant ratings of personality traits, temperament, and interpersonal problems. Psychological Assessment, 14, 39-49. Reay, A. M. C., & Browne, K. D. (2001). Risk factor characteristics in carers who physically abuse or neglect their elderly dependents. Aging and Mental Health, 5, 56-62. Rector, N. A., Bagby, R. M., Segal, Z. V., Joffe, R. T., & Levitt, A. (2000). Selfcriticism and dependency in depressed patients treated with cognitive therapy or pharmacotherapy. Cognitive Therapy and Research, 24, 571—584. Richman, J. A., & Flaherty, ]. A. (1987). Adult psychosocial assets and depressed mood over time: Effects of internalized childhood attachments. Journal of Nervous and Mental Disease, 175, 703-712. Riggio, R. E. (1986). Assessment of basic social skills. Journal of Personality and Social Psychology, 51,649-660. Ritzier, B. A. (1998). Teaching dissemination of personality assessment results in graduate programs. In L. Handler & M. J. Hilsenroth (Eds.), Teaching and learning personality assessment (pp. 413-427). Mahwah, NJ: Erlbaum. Robins, C. J., Bagby, R. M., Rector, N. A., Lynch, T. R., & Kennedy, S. H. (1997). Sociotropy, autonomy, and patterns of symptoms in patients with major depression: A comparison of dimensional and categorical approaches. Cognitive Therapy and Research, 21, 285-300. REFERENCES
225
Robins, C. J., Ladd, ]., Welkowitz, J., Blaney, P. H., Diaz, R., & Kutcher, G. (1994). The Personal Style Inventory: Preliminary validation studies of new measures of sociotropy and autonomy. Journal of Psychopathology and Behavioral Assessment, 16, 277-300. Robins, C. J., & Luten, A. G. (1991). Sociotropy and autonomy: Differential patterns of clinical presentation in unipolar depression. Journal of Abnormal Psychology, 100, 74-77. Rogers, C. R. (1951). Client-centered therapy. Boston: Houghton Mifflin. Rogers, C. R. (1961). On becoming a person. Boston: Houghton Mifflin. Rogers, C. R. (1963). The actualizing tendency in relation to motives and consciousness. In M. R. Jones (Ed.), Nebraska symposium on motivation (Vol. 11, pp. 124). Lincoln: University of Nebraska Press. Rosenthal, R. (1966). Experimenter effects in behavioral research. New York: AppletonCentury-Crofts. Rossman, P. (1984). Assessing different aspects of psychosocial dependency: A new scale and some empirical results. Studia Psycholagica, 26, 317-322. Rude, S. S., & Burnham, B. L. (1995). Connectedness and neediness: Factors of the DEQ and SAS dependency scales. Cognitive Therapy and Research, 19,323-340. Rusbult, C. E., & Martz, J. M. (1995). Remaining in an abusive relationship: An investment model analysis of nonvoluntary dependence. Personality and Social Psychology Bulletin, 21, 558-571. Russo, P. M., Cecero, J. J., & Bornstein, R. F. (2001). Implicit and self-attributed dependency needs in homeless men and women. Journal of Social Distress and the Homeless, 10, 269-277. Ryder, R. D., & Parry-Jones, W. L. (1982). Fear of dependence and its value in working with adolescents. Journal of Adolescence, 5, 71-81. Safran, J. D., & Messer, S. B. (1997). Psychotherapy integration: A postmodern critique. Clinical Psychology: Science and Practice, 4, 140-152. Salokangas, R. K. R., Rakkolainen, V., &. Lehtinen, P. (1980). The psychiatric treatment system as an adaptational treatment structure: A study of dependency in treatment behavior of schizophrenic patients. Psychiatrica Fennica, 7, 99-109. Sandier, J., & Dare, C. (1970). The psychoanalytic concept of orality. Journal of Psychosomatic Research, 14, Ill-Ill. Sanfilipo, M. P. (1994). Masculinity, femininity, and subjective experiences of depression, journal of Clinical Psychology, 50, 144-157. Sans, J., &. Avia, D. (1994). Cognitive specificity in social anxiety and depression: Self-statements, self-focused attention, and dysfunctional attitudes. Journal of Social and Clinical Psychology, 13, 105-137. Sato, T. (2001). Autonomy and relatedness in psychopathology and treatment: A cross-cultural formulation. Genetic, Social, and General Psychology Monographs, 127, 89-127. Saviola, M. E. (1981). Personal reflections on physically disabled women and dependency. Professional Psychology; Research and Practice, 12, 112-117. 226
REFERENCES
Schafer, R. (1954). Psychoanalytic interpretation in Rorschach testing. New York: Grune & Stratton. Schermer, V., & Pines, M. (Eds.). (1999). Group psychotherapy of the psychoses. London, UK: Jessica Kingsley. Schlenker, B. R., & Weigold, M. F. (1990). Self-consciousness and self-presentation: Being autonomous versus appearing autonomous, journal of Personality and Social Psychology, 59, 820-828. Sciuto, G., Diaferia, G., Battaglia, M., Perna, G., Gabriele, A., & Bellodi, L. (1991). DSM-II/-R personality disorders in panic and obsessive-compulsive disorder: A comparison study. Comprehensive Psychiatry, 32, 450—457. Schneider, K. (1923). Die psychopathischen personlichkeiten. Vienna: Deuticke. Schneider, K. J. (1990). The paradoxical self: Toward an understanding of our contradictory nature. New York: Plenum. Schneider, K. J., & May, R. (1995). The psychology of existence: An integrative, clinical perspective. New York: McGraw-Hill. Schore, A. (1994). Affect regulation and the origin of the self: The neurobiology of emotional development. Hillsdale, NJ: Erlbaum. Seligman, M. E. P. (1991). Learned optimism. New York: Random House. Serbin, L. A. (1997). Research on international adoption: Implications for developmental theory and social policy. International journal of Behavioral Development, 20, 83-92. Shaver, P. R., & Hazan, C. (1993). Adult romantic attachment: Theory and evidence. In D. Perlman & W. Jones (Eds.), Advances in personal relationships (Vol. 4, pp. 29-70). London: Jessica Kingsley. Shaver, P. R., & Mikulincer, M. (2002). Attachment-related psychodynamics. Attachment and Human Development, 4, 133-161. Shimizu, H. (2000). Beyond individualism and sociocentrism: An ontological analysis of the opposing elements in personal experiences of Japanese adolescents. Human Development, 43, 195-211. Simpson, J. A., & Gangestad, S. W. (1991). Individual differences in sociosexuality: Evidence for convergent and discriminant validity. Journal of Personality and Social Psychology, 60, 870-883. Singer, J. L. (Ed.). (1990). Repression and dissociation: Implications for personality theory, psychopathology, and health. Chicago: University of Chicago Press. Singer, J. L., & Salovey, P. (1990). Organized knowledge structures and personality: Person schemas, self schemas, prototypes, and scripts. In M. J. Horowitz (Ed.), Person schemas and maladaptive interpersonal patterns (pp. 33-79). Chicago: University of Chicago Press. Singh, R. R., & Ojha, S. K. (1987). Sex difference in dependence proneness, insecurity, and self-concept. Manas, 34, 61-66. Singh, S. (1981). Fear of failure among females as related to personality, locus of control, behavior problems, and childrearing practices. Personality Study and Group Behavior, 1, 96-108. REFERENCES
227
Sinha, B. K., & Watson, D. C. (2001). Personality disorder in university students: A multitrait-multimethod matrix study. Journal of Personality Disorders, 15, 235244. Sinha, J. B. P. (1968). A test of dependence-proneness. Journal of Psychological Research, 12, 66-70. Sinha, ]. B. P., & Pandey, J. (1972). The processes of decision-making in dependence prone persons. Journal of Psychological Research, 16, 35-37. Sloman, L, Atkinson, L., Milligan, K., &. Liotti, G. (2002). Attachment, social rank, and affect regulation: Speculations on an ethological approach to family interaction. Family Process, 41, 313-327. Smith, T. W., O'Keefe, J. L., &. Jenkins, M. (1988). Dependency and self-criticism: Correlates of depression or moderators of the effects of stressful events? Journal of Personality Disorders, 2, 160-169. Snyder, W. U. (1963). Dependency in psychotherapy. New York: MacMillan. Solomon, A., & Haaga, D. A. F. (1993). Sociotropy, autonomy, and self-complexity. Journal of Social Behavior and Personality, 8, 743-748. Sotsky, S. M. (1992). Integration of pharmacotherapy and a psychiatric perspective. Family Process, 31, 114-116. Speed, A., & Oangestad, S. W. (1997). Romantic popularity and mate preferences: A peer-nomination study. Personality and Social Psychology Bulletin, 23, 928935. Sperling, M. B., & Berman, W. H. (1991). An attachment classification of desperate love. Journal of Personality Assessment, 56, 45-55. Sperling, M. B., Sack, A., & Field, C. L. (2000). Psychodynamic practice in a managed care environment. New York: Guilford. Spitzer, R. L., Williams, J. B. W., Gibbon, M., & First, M. B. (1990). Structured Clinical Interview for DSM-111-R Personality Disorders (SCID-II). Washington, DC: American Psychiatric Press. Sroufe, L. A., Fox, N. E., & Pancake, V. R. (1983). Attachment and dependency in developmental perspective. Child Development, 54, 1615-1627. Stamler, C., & Palmer, ]. O. (1971). Dependency and repetitive visits to the nurse's office in elementary school children. Nursing Research, 20, 254-255. Stangl, D., Pfohl, B., Zimmerman, M., Bowers, W., & Corenthal, C. (1985). A structured interview for DSM-III personality disorders. Archives of General Psychiatry, 42, 591-596. Steele, K., van der Hart, O., & Nijenhuis, E. R. S. (2001). Dependency in the treatment of complex posttraumatic stress disorder and dissociative disorders. Journal of Trauma and Dissociation, 2, 79-116. Stein, K. F. (1994). Complexity of self-schema and responses to disconfirming feedback. Cognitive Therapy and Research, 18, 161-178. Steinberg, L., & Silverberg, S. B. (1986). The vicissitudes of autonomy in early adolescence. Child Development, 57, 841-851.
228
REFERENCES
Stern, D. N. (1985). The interpersonal world of the infant. New York: Basic Books. Stewart, S. H., Knize, K., & Pihl, R. O. (1992). Anxiety sensitivity and dependency in clinical and non-clinical panickers and controls, journal of Anxiety Disorders, 6, 119-131. Stokes, G. S., & Cooper, L. A. (2001). Content/construct approaches in life history form development for selection. International Journal for Selection and Assessment, 9, 138-151. Stokes, G. S., Mumford, M. D., & Owens, W. A. (Eds.). (1994). The biodata handbook: Theory, research, and applications. Palo Alto, CA: Consulting Psychologists Press. Stone, A. A., Turkkan, J. S., Bachrach, C. A., Jobe, J. B., Kurtzman, H. S., & Cain, V. S. (Eds.). (2000). The science of self-report: Implications for research and practice. Mahwah, NJ: Erlbaum. Strauman, T. J. (1992). Self-guides, autobiographical memory, and anxiety and dysphoria: Toward a cognitive model of vulnerability to emotional distress. Journal of Abnormal Psychology, 101, 87-95. Strean, H. S. (Ed.). (1986). Countertrans/erence. New York: Haworth. Strieker, G., & Gold, J. R. (Eds.). (1993). Comprehensive handbook of psychotherapy integration. New York: Plenum. Sullivan, H. S. (1947). Conceptions of modem psychiatry. Washington, DC: William Alanson White Institute. Sullivan, H. S. (1953). The interpersonal theory of psychiatry. New York: Norton. Sundin, E., Armelius, B. C., & Nilsson, T. (1994). Reliability studies of scales of psychological capacities: A new method to assess psychological change. Psychoanalysis and Contemporary Thought, J 7, 591-615. Tabin, J. K. (1985). On the way to self. New York: Columbia University Press. Tait, M. (1997). Dependence: A means or an impediment to growth? Britishjoumal of Guidance and Counselling, 25, 17-26. Thompson, R., (Si Zuroff, D. C. (1998). Dependent and self-critical mothers' responses to adolescent autonomy and competence. Personality and Individual Differences, 24,311-324Thompson, R., &. Zuroff, D. C. (1999). Dependent and self-critical mothers' responses to adolescent sons' autonomy and competence, journal of Youth and Adolescence, 28, 365-384. Tjosvold, D., & Fabrey, L. J. (1980). Motivation for perspective-taking: Effects of interdependence and dependence on interest in learning others' intentions. Psychological Reports, 46, 755-765. Todd-Bazemore, B. (1999). Cultural issues in psychopharmacology: Integrating medication treatment with Lakota Sioux traditions. Journal of Clinical Psychology in Medical Settings, 6, 139-150. Torgersen, S., Lygren, S., Oien, P. A., Skre, L, Onstad, S., Edvardsen, ]., et al. (2000). A twin study of personality disorders. Comprehensive Psychiatry, 41, 416—425.
REFERENCES
229
Triandis, H. C. (1996). The psychological measurement of cultural syndromes. American Psychologist, 51, 407-415. Tribich, D., & Messer, S. (1974), Psychoanalytic character type and status of authority as determiners of suggestibility. Journal of Consulting and Clinical Psychology, 42, 842-848. Tripathi, N. K. M. (1982). Approval motive and dependence proneness. Perspectives in Psychological Research, 5, 23-28. Tseng, W. S., & Streitzer, J. (Eds.). (2001). Culture and psychotherapy: A guide to clinical practice. Washington, DC: American Psychiatric Publishing. Turkat, I. D. (1990). The personality disorders: A psychological approach to clinical management. Elmsford, NY: Pergamon. Turkat, I. D., & Carlson, C. R. (1984). Data-based versus symptomatic formulation of treatment: The case of a dependent personality. Journal of Behavior Therapy and Experimental Psychiatry, 15, 153-160. Turkat, I. D., & Maisto, S. A. (1985). Personality disorders: Application of the experimental method to the formulation and modification of personality disorders. In D. H. Barlow (Ed.), Clinical handbook of psychological disorders (pp. 502570). New York: Guilford. Tyrer, P., Mitchard, S., Methuen, C., & Ranger, M. (2003). Treatment rejecting and treatment seeking personality disorders: Type R and Type S. Journal of Personality Disorders, 17,263-268. Vaillant, G. E. (1980). Natural history of male psychological health: VIII. Antecedents of alcoholism and orality. American Journal of Psychiatry, 137, 181-186. Vaillant, G. E. (1994). Ego mechanisms of defense and personality pathology. Journal of Abnormal Psychology, 103, 44-50. Van Houten, R., &. Axelrod, S. (Eds.). (1993). Behavior analysis and treatment. New York: Plenum. Van Sweden, R. C. (1995). Regression to dependence. North vale, NJ: Jason Aronson. Vettese, L. C., & Mongrain, M. (2000). Communication about the self and partner in the relationships of dependents and self-critics. Cognitwe Therapy and Research, 24, 609-626. Viglione, D. J. (1999). A review of recent research addressing the utility of the Rorschach. Psychological Assessment, 11, 251-265. Wachtel, P. L. (197 7). Psychoanalysis and behavior therapy: Toward an integration. New York: Basic. Wachtel, P. L. (1991). From eclecticism to synthesis: Toward a more seamless psychotherapeutic integration. Journal of Psychotherapy Integration, 1, 43-54. Waldo, T. G., & Merritt, R. D. (2000). Fantasy proneness, dissociation, and DSMIV—R Axis II symptomatology. Journal of Abnormal Psychology, 109, 555-558. Walters, R. H., & Parke, R. D. (1964). Social motivation, dependency, and susceptibility to social influence. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 1, pp. 231-276). New York: Academic.
230
REFERENCES
Wasson, E. J., & Linehan, M. (1993). Personality disorders. In A. S. Bellack &. M. Hersen (Eds.), Handbook of behavior therapy in the psychiatric setting (pp. 329353). New York: Plenum. Watkins, C. E., Campbell, V. L, Nieberding, R., & Hallmark, R. (1995). Contemporary practices of psychological assessment by clinical psychologists. Professional Psychology: Research and Practice, 26, 54-60. Watkins, C. E., & Watts, R. E. (1995). Psychotherapy survey research studies: Some consistent findings and integrative conclusions. Psychotherapy in Private Practice, 13, 49-68. Watson, C. G., Barnett, M., Nikunen, L, Schultz, C., Randolph-Elgin, T., & Mendez, C. M. (1997). Lifetime prevalences of nine common psychiatric/personality disorders in female domestic abuse survivors. Journal of Nervous and Mental Disease, 185, 645-647. Watson, P. J., Biderman, M. D., & Boyd, C. (1989). Androgeny as synthetic narcissism. Sex Roles, 21, 175-207. Weinberger, J. (1996). Common factors aren't so common: The common factors dilemma. Clinical Psychology: Science and Practice, 2, 45-69. Weiner, I. B. (2000). Using the Rorschach properly in practice and research. Journal of Clinical Psychology, 56, 435-438. Weiss,]., & Sampson, H. (1986). The psychoanalytic process: Theory, clinical observation, and empirical testing. New York: Guilford. Weiss, L. R. (1969). Effects of subject, experimenter, and task variables on compliance with the experimenter's expectation. Journal of Projective Techniques and Personality Assessment, 33, 247-256. Weiss, L. R., & Masling, J. M. (1970). Further validation of a Rorschach measure of oral imagery: A study of six clinical groups. Journal of Abnormal Psychology, 76, 83-87. Weiss, R. D., Mirin, S. M., Michael, J. L., & Sollogub, A. C. (1986). Psychopathology in chronic cocaine abusers. American Journal of Drug and Alcohol Abuse, 12, 17-29. Weissman, A. N., & Beck, A. T. (1978). Development and validation of the Dysfunctional Attitudes Scale. Paper presented at the American Educational Research Association Meeting, Toronto, Canada. Weissmark, M. S., & Giacomo, D. A. (1998). Doing psychotherapy effectively. Chicago: University of Chicago Press. Welkowitz, J., Lish, J. D., &. Bond, R. N. (1985). The Depressive Experiences Questionnaire: Revision and validation. Journal of Personality Assessment, 49, 89—94Werner, H., & Kaplan, B. (1963). Symbol formation: An organismic-developmental approach to language and the expression of thought. New York: Wiley. Westen, D., & Shedler, J. (1999). Revising and assessing Axis II: Part 1. Developing a clinically and empirically valid assessment method. American Journal of Psychiatry, 156, 258-272.
REFERENCES
231
Westen, D., Shedler, J., Durrett, C., Glass, S., & Martens, A. (2003). Personality diagnoses in adolescence: DSM-IV-R Axis II diagnoses and an empirically derived alternative. American journal of Psychiatry, 160, 952-966. Whiffen, V. E., & Aube, J. A. (1999). Personality, interpersonal context, and depression in couples, journal of Social and Personal Relationships, 16, 369-383. Whiffen, V. E., &Sasseville, T. M. (1991). Dependency, self-criticism, and recollections of parenting: Sex differences and the role of depressive affect, journal of Social and Clinical Psychology, 10, 121-133. White, H. (1986). Damsels in distress: Dependency themes in fiction for children and adolescents. Adolescence, 21, 251-256. Widiger, T. A. (1993). The DSM-III-R categorical personality disorder diagnoses: A critique and an alternative. Psychological Inquiry, 4, 75-90. Widiger, T. A., Frances, A. J., Pincus, H. A., Davis, W. W., & First, M. B. (1991). Toward an empirical classification for the DSM-IV-R. journal of Abnormal Psychology, 100, 280-288. Wiggins, J. S., &. Winder, C. L. (1961). The Peer Nomination Inventory. Psychological Reports, 9, 643-677. Winder, C. L., Ahmad, F. Z., Bandura, A., & Rau, L. C. (1962). Dependency of patients, psychotherapists' responses, and aspects of psychotherapy. Journal of Consulting Psychology, 26, 129-134. Wise, E. A. (1996). Comparative validity of MMPI-2 and MCMI-II personality disorder classifications, journal of Personality Assessment, 66, 569-582. Wiseman, H. (1997). Interpersonal relatedness and self-definition in the experience of loneliness during the transition to university. Personal Relationships, 4, 285299. Wiser, S., & Telch, C. F. (1999). Dialectical behavior therapy for binge-eating disorder, journal of Clinical Psychology, 55, 755-768. Wolpe, J. (1990). The practice of behavior therapy. Elmsford, NY: Pergamon. Wonderlich, S. A., Fullerton, D., Swift, W. J., & Klein, M. H. (1994). Five-year outcome from eating disorders: Relevance of personality disorders. International Journal of Eating Disorders, 15, 233-243. Wood, J. M., Lilienfeld, S. O., Nezworski, T. M., & Garb, H. N. (2001). Coming to grips with negative evidence for the Comprehensive System for the Rorschach: A comment on Ganaco, Loving, and Bodholdt; Ganellen; and Bomstein. Journal of Personality Assessment, 77, 48-70. Wood, J. M., Nezworski, M. T., & Stejskal, W. J. (1996). The Comprehensive System for the Rorschach: A critical examination. Psychological Science, 7, 3-10. Yalom, I. D. (1980). Existential psychotherapy. New York: Basic Books. Yamaguchi, S. (2004). Further clarifications of the concept of amae in relation to dependence and attachment. Human Development, 47, 28-33. Yamamoto, R. (1989). A study on both sides of its own self: An investigation of developmental trends from adolescence to adulthood and sex differences. Japanese Journal of Educational Psychology, 37, 302-311.
232
REFERENCES
Yasunaga, S. (1985). The effects of dependency and strategy patterns on modeling. Japanese Journal of Psychology, 55, 374-377. Young, J. E. (1994). Cognitive therapy for personality disorders: A schema-focused approach. Sarasota, FL: Professional Resources Press. Young, J. E., & Lindeman, M. D. (1992). An integrative schema-focused model for personality disorders. Journal of Cognitive Psychotherapy, 6, 11-23. Zaretsky, A. E., Fava, M., Davidson, K. G., Pava, J. A., Matthews, J., & Rosenbaum, J. F. (1997). Are dependency and self-criticism risk factors for major depressive disorder? Canadian Journal of Psychiatry, 42, 291-297. Zettle, R. D., Haflich, J. L, & Reynolds, R. A. (1992). Responsivity to cognitive therapy as a function of treatment format and client personality dimensions. Journal of Clinical Psychology, 48, 787-797. Zimmerman, M. (1994). Diagnosing personality disorders: A review of issues and research methods. Archives of General Psychiatry, 51, 225-245. Zirkel, S. (1992). Developing independence in a life transition: Investing the self in the concerns of the day. Journal of Personality and Social Psychology, 62,506—521. Zuckerman, M., & Grosz, H. J. (1958). Suggestibility and dependency, journal of Consulting Psychology, 22, 328. Zuroff, D. C., & Fitzpatrick, D. K. (1998). Depressive personality styles: Implications for adult attachment. Personality and Individual Differences, 18, 253-265. Zuroff, D. C., & Mongrain, M. (1987). Dependency and self-criticism: Vulnerability factors for depressive affective states. Journal of Abnormal Psychology, 96,14-22. Zuroff, D. C., Moskowitz, D. S., Wielgus, M. S., Powers, T. A., & Franko, D. L. (1983). Construct validation of the dependency and self-criticism subscales of the Depressive Experiences Questionnaire. Journal of Research in Personality, 17, 226-241.
REFERENCES
233
INDEX Attachment relationship. See also Insecure attachment dependency behaviors in, 76, 76n.3 infants, 46-47 Attributional bias, 13, 74 Authoritarianism in autonomous marital partner, 177178 in overdependent parents, 52 in parenting, 12, 17, 142n.3, 187 Automatic thoughts, 12, 158 Autonomic reactions, in social "crumbling," 161-162 Autonomy dependency needs tension, 44 healthy dependency integration, 147148 problem-solving training, 140-141 reinforcement of, 137-138 therapeutic strategies, 137—138, 160 Avoidant personality disorder, comorbidity, 102-104 Axis I, comorbidity, 100-102, 107-108 Axis II, comorbidity, 102-104, 107 Axis III, 108 Axis IV, clinical implications, 108-109
Abandonment fears interventions, 161-162 in marriage, 177 Abuse-dependency link, 183-184 financial dependence role in, 190 interactionist perspective, 71 in mothers, 67 perpetrator pathways, 183-184 victimization pathways, 183 Adaptive dependency, 76-77 Adjustment disorder, comorbidity, 101 Adolescent dependency, 49-50, 54 Adults, 50-52, 54 childhood dependency correlations, 4445 dependency domains, 50-52, 54 Age effects assessment confound, 127-128 clinical considerations, 106-107 and dependency levels, 98-99 Agoraphobia, comorbidity, 101 Alcohol abuse, comorbidity, 101-102 Ambivalence in adolescence, 49-50 and dependency-detachment dialectic, 44 in therapeutic alliance, 134 Anger turned inward, 80, 117-118 Anorexia nervosa, comorbidity, 102 Antidepressants, 181-182 Antisocial personality disorder, 103—104 Anxiety disorders, comorbidity, 101 Anxiety management, 162 Archival data, 123 Assessment, 21-37, 111-129 clinical application, 115-116,123-129 dependency subtypes, 120-121 of healthy dependency, 81-82 interactionist perspective, 62 interviews, 31—33 projective tests, 28-31,33-37,112-116 self-report measures, 22-28, 33-37, 112-116 situational context factors, 126 tailored approach, 123-125 testing distinction, 112
Behavioral anchoring, 121 Behavioral interventions, 135-138 assumptions and goals, 135-136 core elements, 145 healthy dependency goal, 148,163-164 integrated approach, 136-138,163-164 Behavioral perspective, 6-8 change strategies, 163-164 contextual influences, 58 dependency model, 6-8, 18-19 interactionist model component, 1516 Biodata, 123 Biopsychosocial model, 14-15 Blacky Test Oral Dependency Scale, 29n.5, 36 Borderline personality disorder, comorbidity, 102-103 Boundary issues, 167
235
Career adaptation, in health dependency, 81n Caregivers, elder abuse, 190 Change phase, in psychotherapy, 166-170 Child abuse in dependent mothers, 67 dependogenic effect of, 187 in overdependent parents, 52 perpetration pathways, 183 Child sexual abuse, in dependent men, 67 Children, 47-49, 54 adulthood dependency correlations, 4445 dependency development, 47-49, 54 relationship scripts, 47-49 Codependency, 177-178 Cognitive complexity, 80-81, 8In Cognitive interventions, 138-141, 156-159 assumptions and goals, 138-139 core elements, 145 four-stage model, 139-141 healthy dependency goal, 148 integrated strategies, 156-158 tailored approach, 146-147 Cognitive processes, 11-14 contextual influences, 58 dependency model, 11-13 integrated therapy approach, 156-158, 168 interactional model component, 15-16 Cognitive restructuring, 139, 158 Combined modality therapy. See Integrated treatment Communitarian cultures, 8-9. See also Cultural factors Comorbidity, 99-104 Axis I, 100-102 Axis II, 102-104 clinical impact, 107 diagnostic considerations, 92 DSM-IV information, 100 tailored treatment, 146 Compensatory dependencies, 76-77, 189190 Compliant behavior, 64—65 Compulsive personality disorder, comorbidity, 103 Computer-based Rorschach interpretation, 29n.5 Confirmatory bias, and dependency attributions, 75n Conflicted dependency subtype
236
INDEX
assessment, 121—122 integrated therapy approach, 170 social influence strategies, 85-86 Conformity, passivity association, 59 Confrontation strategy, 163-164 Connectedness bridging to, 160-161 healthy dependence overlap, 76-77 Contextual influences, 57-72 Contingency management, 136-138 Continuous monitoring method, 33n Control issues, patient—therapist dynamic, 148-149 Conversion disorder, comorbidity, 102 Coping style assessment, 117-119, 126 in autonomous marital partner, 177178 Core Conflictual Relationship Theme method, 133-135 Counterdependent response, 135n, 177-178 Countertransference, 134-135 Couples therapy, 177-178, 187-188 Crisis management, 146n Cultural factors, 8-11 in assessment, 127-128 clinical considerations, 106-107 and cognitive change interventions, 156-157 dependency toleration role, 8-9, 45, 45n in marital therapy, 188 and motivational change strategies, 159 Death anxiety, 142n.5 Defense mechanisms/style assessment, 117-119, 126 healthy versus unhealthy dependency, 80 projective versus objective assessment, 115-116 tailored assessment, 125-126 and time-limited therapy, 176 Defense Mechanisms Inventory (DMI), 118119 Defense Style Questionnaire (DSQ), 118 Denigration pattern in marriage, 177-178 in therapist, 149 Dependence Proneness Scale (DPS), 23n.2, 36, 128, 128n "Dependency overestimation bias," 74
Dependent personality disorder assessment, 111-129 comorbidity, 99-104 diagnosis, 91-109 differential diagnosis, 99-104 DSM series approach, 93-96 epidemiology, 96-99 interview measures, 31—33 treatment compliance, 70 Dependent Personality Style Scale (DPSS), 25,114 "Dependogenic" parenting, 185-187 Depression, comorbidity, 100-101 Depressive Experiences Questionnaire (DEQ), 23-24, 36, 114 Depressive Experiences Questionnaire for Adolescents (DEQ-A), 81-82 Desensitization techniques, 137-138 Detachment-dependency dynamic, 42—44 Diagnosis, 91-109 clinical considerations, 104-109 DSM series, 93-96 interactionist perspective, 62 mood states influence on, 91, 91n Dialectical-behavior therapy, 138n Differential diagnosis, 99-104 Disinhibition, older adults, 53 Dissociative disorder, comorbidity, 102 Drug treatment, 181-183, 188 DSM-I, 59n, 93 DSM-II, 93 DSM-III, 93 DSM-IV/DSM-IV-TR, 93-96 clinical implications, 104-109 dependency-passivity problem, 59n, 94 differential diagnosis and comorbidity, 100 empirical evidence, 94—97 external validity problem 94 interactionist perspective implications, 62 threshold model cautions, 105-106 DSM-V, 96-97 Dysfunctional cognition. See Cognitive pro-
Early Memory Dependency Probe (EMDP), 31 Eating disorders, comorbidity, 102 Eclecticism, 154n.2. See also Integrated approach Elderly
dependence-abuse link, 190, 190n functional and emotional dependency, 53n54 pseudodementia, 53 Electroconvulsive therapy, 183n Emotion-related dependency conceptualization, 6-7 instrumental dependency relationship, 6-7 intervention strategies, 161-162 older adults, 53-54 therapeutic techniques, 168 Engagement phase, psychotherapy, 166-170 Enmeshed families, 48-49 Epidemiology, 96-99 Existential perspective, 142-144 Experience sampling method, 33n Exploitable dependency subtype assessment, 120—121 and dependent personality disorder, 105n integrated therapy approach, 169 in three-vector model, 13, 83-84 Exploitation, dependency problem, 64-65 Extinction techniques, 136—137 "False" self, 142 Family system assessment, 127 childhood dependency development, 48-49 relationship scripts in, 48-49 resistance to change, 179 roles and alliances in, 178—179 Family therapy, 178-179 Femininity, dependency correlation, 9-10 Financial dependency, and abuse risk, 190 Five-factor model, dependency traits, 14 Four-pattern dependency model, 84-86, 121-122, 168, 170 Freud, Sigmund, "oral fixation" model, 4-5 Friendship in healthy versus unhealthy dependence, 79 overdependence in, 51 Frustration tolerance, 125-126 Functional dependency, older adults, 53 Fundamental attribution error, dependency perceptions, 74 Gender differences, 8-11 assessment method effects, 9-11, 35 INDEX
237
clinical considerations, 106-107 in dependency, 8-11 dependent personality disorder prevalence, 99 self-report versus projective tests, 35 Gender roles, 8-11 assessment factor, 127-128 dependency attributions factor, 74 dependency etiology, 8-11, 17 Generalized anxiety disorder, comorbidity,
101 Genetic factors, 45-46 Goal setting, assessment use, 128-129 Grief, and dependency levels, 101 Group therapy, 180-181 Guilt, in existential approach, 143-144 Health services utilization, 69n Healthy dependency, 73-87 assessment, 81-82, 125-126 conceptual issues, 75-77 in dependency spectrum, 42 self-presentation in, 77-79 as treatment goal, 147-148, 164 versus unhealthy dependency, 74-82 Help-seeking in compensatory versus unhealthy dependency, 189-190 dependency-related strength, 69 dependency stereotype, 74 instrumental-emotional dependency model, 6-7 limit-setting, 159-160 social reinforcement model, 7-8 Helpless dependency subtype, 84-85, 121122, 170 "Helpless" self-concept clinical implications, 105 in dependent personality disorder, 94, 96-97, 105 dependogenic parenting role, 187 therapy spillover effects on, 168 Heritability, 45-46 Hidden dependency subtype, 85, 121-122, 170 Histrionic personality disorder, comorbidity, 102-104 Holtzman Inkblot Test Dependency Scale, 29n.6, 36 Hospitalization, and dependency, 98,174—175 Hostile dependency subtype, 85, 121-122, 170
238
INDEX
Humanistic-experiential approach, 141-144 assumptions and goals, 141-142 core elements, 145 healthy dependency goal, 148 Idealization, in transference reactions, 134 Immaturity, 65-67 Illness-dependency link, 64-65, 68n, 184 Immunocompetence, 64, 68n, 184n Implicit dependency needs. See also Projective measures projective measures, 34-37, 116-117 therapeutic implications, 159 Impulse control, 80 In vivo behavioral therapy, 136-137, 137n Inauthentic self, 142 Individualistic cultures. See Cultural factors Indian culture. See Cultural factors Infantile dependency, 45- 47, 54 Infantilization, 65-67, 149 Informant report, 123 Inpatients dependency in, 98, 174-175 psychotherapy termination resistance, 190-191 Insecure attachment style dependency-related deficits, 63-64 infants, 46—47 in romantic relationships, 51-52, 6364 Insight documentation of, 125 in healthy versus unhealthy dependency, 77-78 in marriage, 177 psychoanalytic therapy goal, 132-133 Institutional dependency, 98, 191 Instrumental dependency subtype, 6—7 Integrated treatment, 151-171 assimilative model, 154-156 and dependency subtypes, 168-170 four-stage process model, 166—167 framework, 150, 164-171 strategies, 155-164 timing considerations, 154 Interactionist model, 15-19 clinical application, 62-63,70-72,156164 key contributions, 19 components, 15-18 Intermittent reinforcement, 7-8 Internal working models, 159n
Internalized schemas, 45 International Personality Disorder Examination (IPDE), 32-33 Interpersonal Dependency Inventory (IDI), 23, 106, 114 Interpersonal relationships, 77, 79 Interpersonal sensitivity, dependency strength, 68 Interview measures, 31-33 advantages and disadvantages, 31-32 clinical use considerations, 106 Introjects, 47 Japanese culture, 9, 45n, 98 Jealousy, 63-64 Jiritsu Scale, 128, 128n Learned helplessness, 13 Life span perspective, 39—55 Limit setting, 167 Long-term care facilities dependency link, 175 psychotherapy termination in, 190-191 Love dependency. See also Romantic relationships assessment, 120-121 integrated therapy approach, 169 in three-vector model, 13, 84 Maladaptive schetnas, 138-139, 158 Marital therapy, 176-178, 187-188 Masculinity, 9-10 Mature dependency, 76-77 Measurement. See Assessment Medical help-seeking, 69 Medication, 181-183 dependency levels effect, 181-182 symbolic meaning, 188 Metacommunication, in existential approach, 144 Millon Clinical Multiaxial Inventory (MCMI), 27-28, 36, 106, 114 "Mindless" dependency, 78, 78n Minnesota Multiphasic Personality Inventory (MMPI) Dependency scale, 27, 36, 114 Modeling, dependency link, 8 Mood disorders, comorbidity, 100-101 Mortality salience, 142-143, 143n Mother-daughter relationship, 67 Motivation assessment of, 125
integrated strategies for change, 159161 interactionist model component, 15—16 Multiple possible selves concept, 48n Multitrait-multimethod matrix, 34 Narcissistic personality disorder, comorbidity, 103 Negative reinforcement, 136 Negative self-statements, 12, 158 Neuroticism, 14 Nursing home settings, 175, 175n.2 Obesity, 102n.7 Object relations theory dependency conceptualization, 5-6 self psychology differences, 5 Objective measures. See Self-report measures Observational assessment, 33n Obsessive-compulsive disorder, comorbidity, 101 Older adults, 52-54. See also Elderly Openness trait, 14 Oral Dependency Scale, 23n.2 "Oral" stage, 4-5 Outcome studies, 151-153 Outpatient treatment, and dependence, 98, 174-175 Overidentification , in friendships, 51 Overprotective parenting dependency etiology, 12, 17, 186-187 infant temperament interaction, 47 and treatment, 186-187 Overweight, 102n.7 Panic disorder, comorbidity, 101 Paradoxical interventions, 158 Paranoid personality disorder, comorbidity, 103-104 Parental overdependence, impact of, 52 Parenting dependency etiology, 6, 12, 17, 47n, 185-187 dependogenic form of, 185-187 and healthy versus unhealthy dependence, 79 infant temperament interaction, 47, 47n in overdependent parents, 52 Passivity-dependency link, 58-61 clinical cautions, 105 contextual influences, 58-61 INDEX
239
in DSM-I and DSM-IV, 59n, 93-94 psychoanalytic theory, 58-59 Patient role, dependency stereotype, 74 Peer-group dependence, adolescents, 50 Peer ratings, 65-66 Perceived Social Support Scale (PSS), 119120 Permissiveness, in overdependent parents, 52 Personal Style Inventory (PSI), 24-25, 114 Personality Diagnostic Questionnaire (PDQ), 28, 106, 114 Personality traits and dependency models, 13-15, 19 interactionist perspective, 61 theoretical perspective, 19 Pharmacotherapy, 181-183 dependency levels effect of, 181-182 symbolic meaning, 188 Popularity ratings, peers, 66 Positive reinforcement, 136 Possessiveness, in friendships, 51 Possible selves concept, 157, 157n Presenting complaints, 74-75 Prevalence rates, 97-99 Primary dependency, 107-108 Problem-solving training, 140-141, 158 Projective identification, 134 Projective measures, 28-31, 33-37,112-116 advantages and disadvantages, 28-29, 113 clinical use guidelines, 125 criterion validity, 36 gender differences in dependency, 1011 implicit dependency needs measure, 34— 37, 112-116 objective tests relationship, 33-37 integrated approach, 37, 115-116 intercorrelations, 34-36 suicide-dependency link, 185 Pseudodementia, 53-54 Psychoanalysis, 4-5, 165. See also Psychodynamic psychotherapy Psychodynamic model, 4-6, 19, 58-61 Psychodynamic psychotherapy, 132-135 assumptions and goals, 132-133 core elements, 145 healthy dependency goal, 147-148 integrated approach, 165 tailored treatment, 146-147 Psychosocial stress, 108-109
240
INDEX
Psychotherapy, 131-171. See also specific methods Psychotropic medication, 181-183, 188 Qualitative and Structural Dimensions of Object Representations (QSDOR) cognitive distortions measure, 157 description and psychometrics, 116117, 124 Questionnaires. See Self-report measures Reassurance-seeking behavioral and social learning models, 6-8 and emotional dependency, 6 therapeutic strategies, 159-160 Reframing, 158 Regression, 66 in adolescence, 50 in unhealthy dependency, 80 Rehabilitation settings, dependency in, 175 Reinforcement model, 7-8, 135-136 Rejection fears, 161-162 Relapse prevention, 141, 164 Relationship Profile Test (RPT), 26-27, 82, 114 Relationship reactivity, 77 Relationship scripts, 47-49 Resistance ambivalence role in, 134 in families, 179 in termination conflicts, 168-171,190191 Response to treatment, 151-153 Role-play, 127n, 162, 162n Romantic relationships coping patterns, 177-178 in healthy versus unhealthy dependency, 79 overdependency effects, 51-52, 63-64 Rorschach Inkblot Method, 28-30 Rorschach Oral Dependency (ROD) scale, 29-30, 113-114 psychometrics, 28-30 self-report measures intercorrelations, 35 validity coefficient meta-analysis, 36 Schizoid personality disorder, comorbidity, 103 School phobia, lOln Scripts. See Relationship scripts
Secondary dependency, 107-108 Secure attachment, and healthy dependency, 76n.3 Selective serotonin reuptake inhibitors, 181182 Self-concept. See "Helpless" self-concept Self-control strategies, 141 Self-disclosure, 165 Self-esteem, enhancement of, 140 Self-presentation effects projective test advantage, 29-30 self-report measures weakness, 22 Self-presentation styles diagnostic considerations, 106 healthy versus unhealthy dependency, 77-79 interactionist model, 16-18 Self psychology, 5-6 Self-report measures, 22-28,33-37,112-116 advantages and disadvantages, 22, 113 childhood-adult correlations, 44 clinical considerations, 115-116, 124 criterion validity, 36 explicit dependency needs measure, 3437, 112-116 gender differences in dependency, 9-11 overview, 22-28 projective tests relationship, 33-37, 115-116 suicide-dependency link, 185 Self-schema cognitive restructuring, 158 dependency etiology, 12-13, 17-18 DSM criteria, 94, 96-97 interactionist model, 17-18 therapy spillover effects on, 168 Separation problems, infants, 46 Sex differences. See Gender differences Sex roles. See Gender roles Sexual victimization, 65, 183 Smoking, 102n.6 Social "crumbling," prevention of, 161—162 Social learning models, 6-8 Social Network Scale (SNS), 120 Social perception, 116-117 Social phobia, comorbidity, 101 Social reinforcement model, 7-8 Social skills assessment, 119-120, 126 healthy versus unhealthy dependency, 77-78
Social support assessment, 119-120 illness buffer, 184 Sociocentric cultures, 8-9. See also Cultural factors Sociometric status, dependency link, 66 Sociotropy, self-report measure, 24-25 Somatization, comorbidity, 120 Spot sampling method, 33n Spouse abuse-dependency link, 67, 183-184 Strengths-based assessment, 125-126 Stress factors, 108-109, 184 Structured Clinical Interview for DSM Personality Disorders (SCID-II), 32, 106 Structured Interview for Diagnosis of Personality (SIDP), 33, 106 Submissive dependency subtype assessment, 120-121 integrated therapy approach, 169 in three-vector model, 13, 83 Substance abuse, comorbidity, 101-102 Succorance need, 29, 66 Suggestibility dependency-related problem, 64—65 interactionist perspective, 71 Suicide-dependency link, 184-186 Supportive-expressive therapy, 133, 160161 Systems perspective in assessment, 126-127 in families, 48-49, 178-179 Tailored assessment, 123-125 Tailored treatment, 145-147 Temperament, 46-47 Termination dependent patient issues, 134 in four-stage process model, 166-171 in long-term care or inpatient settings, 190-191 management of, 162, 167, 171 Terror management theory, 142n-143n Testing, 112, 112n. See also Assessment Theatrical dependency, 80 Thematic Apperception Test dependency scale, 29-31,36 Therapeutic alliance obstacles, 134 psychodynamic perspective, 165 Therapist—patient dynamic, control issues, 148-149 INDEX
241
3-Vector Dependency Inventory (3VDI), 25-26, 114, 120-121 Three-vector model, 82-84, 168-169 Threshold model, in DSM-ZV, 105-106 Time-limited therapy, 175-176 Tobacco use, 102n.6 Trait models, 13-15,19,61 Transference, 134-135, 158-159 Treatment compliance, 70 Treatment outcome, 151-153 Treatment seeking, 74- See aho Help-seeking Tricyclic antidepressants, 181-182 Tripartite dependency model, 13 Twin studies, 45^6 Unconditional positive regard, 143
242
INDEX
Unconscious conflicts psychoanalytic theory, 132-133 therapy implications, 159 Unhealthy dependency, 73-87, 189-190 Vicarious reinforcement, 8 Victimization, 64-65, 183 Western culture, 8-9, 45. See also Cultural factors Work dependency effect on, 53, 161, 161n in healthy versus unhealthy dependency, 79 Working through, in psychoanalysis, 133
ABOUT THE AUTHOR
Robert F. Bornstein received his PhD in clinical psychology from the State University of New York at Buffalo in 1986, completed a yearlong internship at the Upstate Medical Center in Syracuse, New York, and is a professor of psychology at Gettysburg College. Dr. Bornstein has published more than 150 articles and book chapters on personality dynamics, diagnosis, and treatment. He wrote The Dependent Personality (1993) and Healthy Dependency (2003), coauthored (with Mary Languirand) When Someone You Love Needs Nursing Home Care (2001), and edited seven other volumes of psychological research. Dr. Bornstein is a fellow of the American Psychological Association, American Psychological Society, Pennsylvania Psychological Association, and Society for Personality Assessment. His research has been funded by grants from the National Institute of Mental Health and the National Science Foundation, and he received the Society for Personality Assessment's Walter Klopfer and Martin Mayman Awards for Distinguished Contributions to the Personality Assessment Literature.
243