Psychiatry 70(2) Summer 2007

85

Looking Back Towards the Future: Is It Time to Change the DSM Approach to Psychiatric Disorders? The Case of Depression
Patrick Luyten and Sidney J. Blatt
With plans underway for the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM), the debate between proponents of an atheoretical, descriptive approach of mental disorders versus an etiologically based classification system has been revitalized. This paper critically reviews findings from three related research areas that have begun to question the atheoretical, descriptive approach to mental disorders of DSM-III and its successors: (a) research concerning the underlying assumptions of DSM, (b) findings from treatment research, and (c) recent biopsychosocial research. Research on depression is used as a paradigmatic example, but the extent to which these findings may generalize to other disorders is also discussed. Strengths and weaknesses of the current DSM approach versus a more etiologically based classification system are considered. This review shows that although an etiologically based classification system of mental disorders seems more promising than a purely descriptive approach, the time seems not yet ripe for such a radical shift in psychiatric diagnosis. Further research oriented towards developing a more etiologically based approach to diagnosis is urgently needed, taking into account research, clinical, and policy implications that such a change in the diagnosis of mental disorders would bring.
DESCRIPTION: THE BOON AND BANE OE CLASSIEICATION?

Historically, the current atheoretical, descriptive and categorical approach to mental disorders was introduced by DSM-III in 1980 (APA, 1980) to halt a confusing plethora of etiologically based classification systems that often had little empirical support. A descriptive approach was adopted with the

aim of fostering research on psychiatric disorders and facilitating the future development of a more empirically and etiologically based classification system (APA, 1980). This descriptive approach was particularly clear in the classification of depressive disorders, which eschewed traditional etiologically based classifications such as the distinction between endogenous versus reactive depression. Instead, with some minor exceptions (e.g., mood disorders due to a general medical

Patrick Luyten, PhD, is affiliated with the Department of Psychology at the University of Leuven in Leuven, Belgium. Sidney]. Blatt, PhD, is affiliated with the Departments of Psychiatry and Psychology at Yale University, New Haven, Connecticut. Preparation of this paper was supported by a grant, from the International Psychoanalytical Association Research Advisory Board. First received on November 28, 2005. Accepted for publication on June 26, 2006. Address correspondence to Patrick Luyten, PhD, Department of Psychology, Tiensestraat 102, B-3000 Leuven, Belgium. E-mail: patrick.luyten@psy.kuleuven.be.

86

Is It Time to Change the DSM Approach?

condition), various types of depression were assumptions and their rationales, followed by categorized on symptom criteria alone. a discussion of relevant empirical findings. In retrospect, the adoption of a descriptive approach towards mental disorders made 1. One of the basic assumptions of perfect sense and clearly had a number of adDSM-III and its successors is that vantages. It reduced diagnostic confusion by clinical disorders are categorically providing a common language for the psychidistinct from subclinical disorders atric community. In addition, it made and normality (APA, 1994, p. xxi), broad-ranging, systematic research on deand that, with some exceptions, pression and other disorders possible. Hence, disorders can and should be diagclearly, one of the main merits of DSM is its nosed based on symptoms alone dedication to systematic empirical research. (APA, 1994, pp. xvii-xviii). There Yet, several authors have suggested that this are a number of advantages associresearch has begun to question a purely deated with this assumption. A catescriptive approach, and they argue that regorical approach facilitates searchers should start to consider developing research by implying a clear demara more etiologically based approach towards cation between cases and the classification of depression and perhaps to non-cases. It is also congruent with other disorders as well (e.g., Clark, 2005; the preferences of clinicians, insurMcHugh, 2005; Parker, 2005; Van Praag, de ance companies, and the public Kloet, & van Os, 2004; Watson, 2005). In (APA, 1994; First, 2005). Finally, it what follows, we critically review three sets of was also hoped that using manifest related findings concerning depression that symptoms, and not etiological fachave emerged from the research literature in tors or latent characteristics, as the recent years which have begun to question the foundation of a diagnostic system current descriptive approach of mood disorwould lead to greater diagnostic ders, in addition, as noted, we also discuss the reliability (Blatt &c Levy, 1998; extent to which these findings may or may not Westen et al., 2002). generalize to other disorders. 2. Although the main purpose of the introduction of a separate axis for personality disorders (Axis II) in SHOULD WE CHANGE THE DSM-III was to stimulate research DESCRIPTIVE APPROACH TO on personality disorders, many reDEPRESSION? searchers subsequently made the (implicit) assumption that Axis I Problems with Key Assumptions and Axis II are independent (for an Underlying the Current DSM overview, see Westen et al., 2002). Approach of Depression Various rationales have been suggested for this assumption, such as It is important to note that, as with any the fact that Axis I disorders like classification system, DSM-III and its succesdepression have more ego-dystonic sors are based on a number of assumptions. characteristics and are more Some of these assumptions are explicit in the state-dependent, while personality DSM; others, however, are more implicit. In disorders are more ego-syntonic addition, although some of these implicit and and could be presumed to be more explicit assumptions have often been incortrait-like, or that biological factors rectly attributed to DSM, they have neverthemight play a greater role in Axis I less had an important influence on research disorders, while personality disorand clinical practice. We discuss each of these ders might be more psychosocial in

Luyten and Blatt been easier to obtain for research on one specific DSM category (e.g., major depression, generalized anxiety disorder) than for research on common etiological factors underlying several disorders (Cuthbert, 2005). Moreover, DSM categories have dominated both research and clinical practice. As First (2005, p. 562) puts it: "Psychology, psychiatry, and social work textbooks are organized using these diagnostic conceptualizations, practice guidelines have been developed based on these categories, and epidemiological data, service use and medical economic data, outcome data, and so forth have been compiled based on these categories. FDA-sanctioned drug indications for psychiatric medications are for the most part expressed in terms of the DSM categories." As a result of this "hegemony of the DSM categorical system" (First, 2005, p. 562), it is hardly surprising that until recently most research concentrated on the etiology and treatment of specific disorders, rather than on shared etiological factors and the efficacy and effectiveness of treatments that may be effective for a range of disorders (Cuthbert, 2005; Van Praag et al., 2004).

87

origin. Whatever the assumptions researchers made, research on Axis I and Axis 11 disorders has been relatively isolated, except perhaps for research on comorbidity between Axis I and Axis 11 disorders (Clark, 2005). 3. Finally, several authors (e.g., Frances, First, & Pincus, 1995; Kupfer et al., 2002) have pointed out that much research on DSM has been inspired by the (often implicit) assumption that each disorder has its own relatively unique etiology and that one therefore also can and should develop a relatively specific treatment for each disorder. Although this may not have been the original intent of DSM-III and its successors, Cuthbert (2005) has noted that many researchers seem to have taken this assumption for granted. This has led to a wealth of research on the etiology and pathogenesis of specific disorders while, until recently, relatively few studies have addressed the question of whether several disorders may share important etiological factors. Part of the rationale for trying to identify relatively specific biological and/or psychosocial vulnerability markers was hope that this would enable the development of specific treatments for each disorder (Kupfer et al., 2002). Accordingly, most treatment studies and treatment guidelines have focused on the efficacy and effectiveness of disorder-specific treatments, and not on the effect of a particular treatment for different disorders (Westen et al., 2002) or particular aspects of treatment processes (Zuroff & Blatt, 2006). This tendency to focus on the etiology of specific disorders and disorder-specific treatments may also have to do with the fact that funding has

Empirical research, however, increasingly points to problems with each of these assumptions (e.g., Blatt & Levy, 1998; Luyten, Blatt, Van Houdenhove, 8c Corveleyn, 2006; McHugh, 2005; Parker, 2005; Widiger & Samuel, 2005). Much of this research is in fact based on DSM categories. As Cuthbert (2005, p. 565) formulates it: "Although the current system has been quite successful for communicating diagnostic information . . . experience with the system has also revealed serious flaws that are becoming increasingly problematic for research and clinical use." Many of these limitations seem to be the consequence of a descriptive and categorical approach.

Is It Time to Change the DSM Approach?

Let us first consider empirical data relevant to these underlying assumptions in the current DSM approach to depression. Then, we will discuss findings relevant to other disorders. First, concerning the categorical approach, the bulk of research evidence suggests that a dimensional approach, with depression situated on a continuum ranging from mild dysphoria to full-blown clinical depression, is more valid than the categorical approach currently espoused by DSM (Blatt, 1974; Has1am, 2003; Kendler & Gardner, 1998; Ruscio & Ruscio, 2000; Solomon et al., 2001). This does not necessarily imply that a dimensional approach should be adopted nor that no discrete depression categories (e.g., melancholic depression) would exist (Parker, 2000). It does imply, however, that the arbitrary consensus-based cut-off criteria used in DSM-III and its successors should be re-evaluated (see also Brown & Barlow, 2005; First, 2005). This could be a gradual process, with initially little change to the current DSM system. For instance, instead of adopting a categorical approach with major depression as the main depression category, one could adopt a dimensional view of depressive disorder, ranging from normal sadness to severe clinical depression. Specifiers such as mild, moderate, and severe with and without psychotic features could then be used to define meaningfulas opposed to consensus-basedcut-offs on this dimension. The validity of these cut-offs could then be investigated by examining their value for predicting course, etiopathogenetic factors involved, and treatment response (Parker, 2005; Widiger & Clark, 2000). Likewise, the reliance on "objective" symptoms in an attempt to improve the reliability of depression diagnosis may have resulted in poor vahdity (Blatt & Levy, 1998; Westen, Shedler, & Bradley, 2006; Widiger & Samuel, 2005). For instance, it is well known that patients with the same DSM diagnosis are often heterogeneous in terms of etiology and pathogenesis, which has hampered research on the identification of etiopathogenetic factors (Nemeroff et al., 2003; Van Praag et al., 2004). Furthermore, this "count/cutoff" ap-

proach also does not match with they way clinicians make diagnoses in clinical practice, which may partly account for the fact that clinicians rarely follow diagnostic procedures prescribed by DSM (Westen et al., 2006). This limits the clinical utility of DSM, an important criterion to evaluate the value of any classification system (First et al., 2004). Second, regarding the assumption of orthogonality between Axis I and Axis II, studies have clearly shown that depression is not independent from personality. Evidence is accruing that temperament (e.g., Clark, 2005) and both broad (e.g., neuroticism) (Kendler, Kuhn, & Prescott, 2004; Ormel, Oldehinkel, & Brilman, 2001) and specific (e.g., interpersonal dependency and perfectionism) (Blatt, 2004; Cox, McWilhams, Enns, & Clara, 2004; Zuroff et al., 2004) personality dimensions play a role in the etiopathogenesis of depression (Westen, Novotny, & Thompson-Brenner, 2004). These findings are also congruent with studies on gene-environment correlations, which strongly suggest that individuals vulnerable to depression in part create their own (stressful) environments (Moffit, Caspi, & Rutter, 2005). They are also congruent with findings of high comorbidity between depression and Axis II disorders, particularly dependent, borderline, and obsessive-compulsive personality disorder, making it highly unlikely that depression and personality disorders are independent (Clark, 2005; Westen et al., 2004). Hence, together, these findings suggest that it may be time to reconsider the distinction between depression and personality (disorders). This may also have important implications for treatment, as some depressed patients could be more likely to benefit from treatments developed for personality disorders, such as Dialectical Behavior Therapy (Linehan, 1993), Mentalization-Based Treatment (Bateman & Fonagy, 2004) or Transference-Focused Psychotherapy (Levy et al., 2006) rather than from more narrow, depression-focused treatments. Future research is needed to address this issue. Third, research has consistently shown that comorbidity between depression and other Axis I disorders (e.g., between mood

Luyten and Blatt

89

and anxiety disorders) is the rule rather than the exception, arguing against the assumption that depression and other disorders are relatively distinct (Luyten, Blatt, & Corveleyn, 2005; Nemeroff, 2002; Parker, 2005). This had led researchers to consider the validity of mixed categories, such as a depressed-anxious disorder (e.g., Phillips et al., 2003). Congruent with these suggestions, studies have shown that both clinical and subclinical depressive disorders share many psychosocial (e.g., Ormel et al., 2001) as well as biological (Claes, 2003; Heim, Plotsky, & Nemeroff, 2005) etiological factors with other disorders, indicating that research may have been overly concerned with identifying the "unique" etiological factors implied in depression (Van Praag et al., 2004) instead of looking for common mechanisms across different disorders (Nemeroff et al., 2003). The fact that various pharmacological agents are often equally effective in treating depression and anxiety disorders points in the same direction (Parker, 2005).
Findings Concerning Other Disorders

Although the extent to which these findings generalize to disorders other than depression is still largely unknown, it can be assumed that there will be wide differences for different disorders. Indications are clear, however, that some of these findings may not be isolated to the case of depression. First, taxometric studies suggest that, besides depression, several anxiety disorders, including posttraumatic stress disorder (PTSD), but also borderline personality disorder, and even severe psychiatric disorders, such as psychotic and bipolar disorders (Haslam, 2003; Ruscio & Ruscio, 2000; Tsuang, Stone, Tarbox, & Faraone, 2003), are best understood as situated on a continuum from normality to subclinical pathology to manifest clinical disorders, contradicting a categorical approach. As Widiger and Clark (2000, p. 950) have pointed out, the goal of future DSM editions may not lie in distinguishing between normality and disorders; "a more realistic goal might be to develop arbitrary but reasonable and

meaningful quantitative points of demarcation along more continuous distributions" because currently "[t]he failure to conduct pilot studies of a criterion set is uncomfortably comparable to releasing a psychological test for publication in the absence of validation data" (Widiger & Clark, 2000, p. 949), and just like any psychological test, the psychometric properties of using these cut-off criteria need to be investigated. Such cut-off criteria would not be arbitrary, but rather based on empirical data, and could be different for different purposes (e.g., for research versus insurance). These studies could also shed more light on the issue of comorbidity. For example, studies show that about 75% of patients with dysthymic disorder have a lifetime history of major depression (Keller et al., 1995). Based on such findings, Widiger and Clark (2000) argue that it seems unreasonable to assume that these individuals have two distinct disorders. Rather, they seem to suffer from one disorder, with more chronic versus more episodic manifestations, much like someone with diabetes may have episodic increases in symptoms. Hence, concerns have been raised about a descriptive, symptom-based approach of mental disorders in general (Charney et al., 2002; Kupfer et al., 2002; McHugh, 2005), emphasizing the fact that the high comorbidity between many disorders suggests a regrouping of DSM categories based on common etiological and pathogenetic factors (see Widiger & Samuel, 2005 for an overview). Or as Kupfer and colleagues (2002, p. xviii) formulate it: "Epidemiologic and clinical studies have shown extremely high rates of comorbidities among the disorders, undermining the hypothesis that the syndromes represent distinct etiologies." For example, it has been argued that research evidence suggests the existence of a category of obsessive-compulsive spectrum disorders (including obsessive-compulsive disorder, Tourette Syndrome and other tic disorders, hypochondriasis, trichotillomania, and body dysmorphic disorder) and a category of stress-related disorders (including PTSD and acute stress disorders), based on a shared pathogenesis and common

90

Is It Time to Change the DSM Approach.'

etiological factors (Philips et al., 2003). In addition, hierarchical models of mental disorders have been proposed, with disorders in a particular cluster presumably sharing etiopathogenetic factors (Watson, 2005). Hence, there is increasing evidence that this shared vulnerability hypothesis (Widiger & Clark, 2000) could inform a new classification system and therefore deserves to be further investigated. Second, concerning the relationship between personality and Axis I disorders, personality factors such as neuroticism and perfectionism, for example, have been related to several other disorders besides depression, such as eating disorders (Westen & Harnden-Fisher, 2001), opiate addiction (Blatt, Rounsaville, Eyre, & Wilber, 1984), cardiovascular and immunological disease (Blatt, Cornell, & Eshkol, 1993), bipolar disorder (Lam, Wright, & Smith, 2004), psychotic disorders (Tsuang et al., 2003), and several anxiety disorders (e.g., Shafran & Mansell, 2001). In addition, there is increasing evidence for the role of temperamental dimensions, such as positive and negative affect, and disinhibition, in a variety of mental disorders (Weinstock & Whisman, 2006). As discussed in more detail below, and as argued by others (e.g., Claes, 2003; Cuthbert, 2005; Heim et al., 2005; Osuch et al., 2004), research in the neurosciences and psychobiological stress research, in particular, may be a central factor in reconsidering the descriptive approach of DSM of depression and psychiatric disorders in general. Again, however, much more research is needed. Moving away from studying the etiology of specific disorders to studying etiological pathways involved in different disorders, however, has important implications for future research. Cuthbert (2005, p. 566), for instance, notes that "a common approach to studying the pathophysiology or genetics of DSM disorders has been to select patients who are diagnosed only with the disorder of interest, excluding patients with comorbid disor-

ders However, this strategy could be counterproductive" (Cuthbert, 2005, p. 567). Instead, he argues, "given the increasing realization that most DSM diagnoses do not represent homogeneous categories, reliable genetic associations or biomarkers are much more likely to be established when better definitions and delineation of disorders are achieved."! The same is probably true for psychosocial factors as well. Finally, there is a wide consensus that many DSM diagnoses have little predictive power towards treatment response: "[L]ack of treatment specificity is the rule rather than the exception" (Kupfer et al., 2002, p. xviii), and thus seem not to be limited to the mood disorders alone. This lack of treatment specificity also limits the clinical utility of the current DSM system (First et al., 2004). Hence, future research, including research on an etiologically based classification system, might increase the relevance of diagnoses for treatment response, an issue we will now discuss in more detail. TREATMENT RESEARCH The introduction of DSM-III has not only facilitated fundamental research on the etiology and pathogenesis of depression but has also facilitated the identification of effective treatments of this disabling disorder (e.g., Hollon et al., 2002). Because guidelines have almost invariably adopted DSM categories, however, the current DSM approach may also be hampering the further development of more effective treatments. Widiger and Samuel (2005, p. 494), for example, conclude that "DSM-IV routinely fails in the goal of guiding the clinician in the presence of one specific disorder" precisely because of the excessive diagnostic comorbidities and unresolvable boundary disputes in the current DSM system. As noted, one of the main problems with the current approach is that treatment

1. (For a recent example concerning substance use disorder, see Beseler et al., 2006.)

Luyten and Blatt

91

studies tend to focus on isolated DSM categories, assuming that, as for any other disease, a specific treatment for each psychiatric disorder should he developed (Kupfer et al., 2002). Yet, studies increasingly suggest that this strategy might be counterproductive (Cuthhert, 2005). For example, research has consistently shown that non-specific therapeutic factors (such as the therapeutic relationship) are often as important as, or even more important than, specific psychotherapeutic techniques or biological agents in predicting treatment outcome in depression (Blatt & Zuroff, 2005; Zuroff & Blatt, 2006). These findings suggest that more research efforts should be devoted to examining these factors. In addition, the fact that patients with a DSM diagnosis of depression are often heterogeneous with respect to etiology and pathogenesis may have created considerable difficulty in establishing the specificity of treatments, confirming the famous "dodo bird verdict," namely that all so-called bona fide treatments have similar efficacy for many disorders (Blatt & Zuroff, 2005). This dodo bird verdict, however, might be in part a result of the fact that patients with the same DSM diagnosis are heterogeneous in terms of etiology and pathogenesis, which limits attempts to find evidence for specificity of both biological (Parker, 2000; Zimmerman, Mattia, & Posternak, 2002) and psychosocial treatments (Nemeroff et al., 2003). Comparisons of etiologically different subgroups of patients, however, may facilitate the emergence of differences in the effectiveness of specific treatments or specific dimensions in the treatment process. For example, several studies have shown that anachtically depressed patients (i.e., patients whose problems are mainly focused around relational issues and conflicts) are particularly responsive to the more supportive elements in psychotherapy, whereas introjective depressed patients (i.e., patients who are mainly concerned with achievement, autonomy and identity) are primarily responsive to insight and explanation (e.g., Blatt et al., 1998). Another compelling example illustrating the need to adopt dif-

ferent treatment approaches for different patients is provided by a re-analysis of data from a large randomized clinical trial comparing the efficacy of 12-week psychotherapy (Cognitive Behavioral Analysis System of Psychotherapy [CBASP]), antidepressant treatment with nefazodone, and a combination of the two in a sample of 681 patients suffering from chronic forms of depression. This study initially found that a combination of CBASP and medication was associated with better outcome as compared to either monotherapy (Keller et al., 2000). In a re-analysis of these data, however, Nemeroff and colleagues (2003) found that in patients with a history of early childhood trauma, CBASP was superior to antidepressant treatment, both in terms of mean post-treatment depression severity scores and remission rates. In fact, in patients with a history of early trauma (but not in patients without such history), the combination of CBASP with nefazodone was not superior to treatment with CBASP alone. Because most of these and similar studies are based on post hoc analyses of existing studies, more systematic research is needed to replicate these findings. Nevertheless, if replicated, findings such as these indicate that patients may show a differential response to different treatments as a function of etiological and/or pathogenetic factors, and not primarily as a function of currently adopted DSM diagnoses (Blatt &c Shahar, 2004; McHugh, 2005; Nemeroff et al., 2003). BIOPSYCHOSOCIAL FINDINGS ON DEPRESSION A third and final reason why it may be time to abandon the descriptive approach of DSM is that research evidence is increasingly challenging the DSM assumption that an etiologically based classification system of psychiatric disorders is not yet feasible. At the time of the introduction of DSM-III, neurobiological research was still in its infancy, and psychosocial research often lacked methodological rigor, making integration between these approaches difficult or even im-

92

Is It Time to Change the DSM Approach?

possible. However, it is clear that progress has al., 1993). These studies suggest that early Ufe been made in each of these areas, and develop- stress may play a central role in explaining ments in the neurosciences in particular may later vulnerability (as well as resilience toin the future be able to bridge the gap between wards stress), leading to disturbances in various theoretical approaches towards psy- stress-reactivity (Gunnar, 2002; Van chiatric disorders, particularly as these devel- Houdenhove et al., 2005), and in disturbances opments increasingly provide evidence for re- in basic dispositions such as anxiety-pronecursive interactions between biological, ness and the regulation of aggression that play psychological, and social factors (Claes, a role in depression and other psychiatric 2003; Gunnar, 2002; Heim et al., 2005; (Osuch et al., 2004) and functional disorders Moffit et al., 2005; Ursano, 2004). (Van Houdenhove, Fgle, &c Luyten, 2004). Yet, the extent to which research findIn addition, studies on depression sugings are available, and thus integration be- gest that genetic factors in psychiatric disortween psychosocial and biological ap- ders most likely consist of both proaches is possible, varies widely for "within-the-skin" and "outside-the-skin" different disorders. Concerning depression, effects (Kendler, 2001), with the former referalthough much work still lies ahead, several ring to direct physiological effects of genes on attempts to integrate biological and neuroregulatory mechanisms implied in mood psychosocial approaches already exist (e.g., regulation and/or stress responsivity (Claes & Clark, 2005; Heim et al., 2005; Gold & Nemeroff, 2005), while the latter refer to the Chrousos, 2002; Kendler, Gardner, & effect of genes on the environment (and vice Prescott, 2002; Luyten et al., 2006). In all of versa). Several studies, for example, have prothese models, (recursive) interactions between vided evidence for gene-environment interacgenetic and neurobiological processes, early tions, in that a functional polymorphism of and current life stress, and relatively stable the 5-HTT gene is associated with depression personality dimensions or cognitive affective only in the face of adverse life experiences schemas of self and others play a pivotal role, (Caspi et al., 2003; Kaufman et al., 2004; suggesting considerable room for dialogue Kencller et al., 2005; for a review, see and integration between these approaches. Levinson, 2006). Interestingly, the functional In fact, together with research on polymorphism of the 5-HTT gene has also PTSD, the study of depression may play a been associated with subclinical depression, paradigmatic role in the further development again pointing to continuity between etiologiof a valid classification system of psychiatric cal factors implied in clinical and subchnical disorders because of its focus on the interplay depression (Gonda et al., 2005) and the need between environment and psychological and for a dimensional rather than categorical biological factors (see also Ursano, 2004). For view. These gene-environment interactions example, growing evidence from studies in ro- work both ways, emphasizing the reciprocal dents, non-human primates, and humans in- nature of associations between biological and dicates the important role of interactions be- psychosocial factors. Kaufman and colleagues tween early and current life stress and genetic (2004), for example, found that maltreated factors in Hypothalamic Pituitary Adrenal children with the 5-HTT gene polymorphism (HPA) axis dysfunction in depression and sev- had elevated levels of depression compared to eral psychiatric disorders that show high normal controls, but only when they also had comorbidity with depression, such as PTSD low levels of social support, providing strong and borderline personality disorder, in vari- evidence for the impact of environmental facous functional disorders that may share com- tors on gene expression and shedding light on mon etiopathogenetic processes with depres- potential mechanisms implicated in resihence sion, such as Chronic Fatigue Syndrome and and recovery. Hence, these findings challenge fibromyalgia (Claes, 2003), and in cardiovas- earher models that made a neat distinction becular and immunological disorders (Blatt et tween nature and nurture (Rose, 2001) and

Luyten and Blatt

93

that were aimed at identifying specific genes as a disorder that shares etiological and for specific disorders (Cuthbert, 2005; Moffit pathogenetic factors with other disorders. et al., 2005). These findings also clearly sug- Moreover, these findings suggest that a degest that there probably is no "true" genetic scriptive and a categorical approach may also load for depression, but that genetic liability be hampering further research from a for depression is different for different types psychosocial as well as from a biological point of individuals under different environmental of view on other disorders. For instance, both circumstances (Kendler, 2001). In addition, Osuch et al. (2004) and Friedman and Harris they suggest the possible existence of other (2004) discuss the fact that post-mortem gene-environment interactions implied in studies of the brain are extremely difficult bedepression and other disorders (Moffit et al., cause of the high comorbidity of disorders in the current DSM. Many brains in the depres2005). Finally, these studies also provide evi- sion brain bank, for example, probably come dence for the mediating role of personality or from individuals with PTSD. Another way to cognitive affective schemas of self and other be- put this is that this problem is the result of the tween biological and psychosocial processes fact that brain banks have been established for (e.g., Fonagy, Gergely, Jurist, & Target, 2004). specific DSM diagnoses and have insuffiFor example, several studies have shown that ciently included information concerning secure attachment representations play a cru- comorbidity and possible etiological factors cial role in the normal development of the HPA (see also Bonne & Charney, 2004). Hence, as Charney and colleagues axis and thus resilience in the face of adversity, protecting against the adverse effects of early (2002) have argued that instead of focusing iife stress (Gunnar, 2002). Broad personality on (longitudinal) studies of disorders, the defactors such as neuroticism, in turn, have been velopment of an etiologically based classificaassociated with the generation of life stress, in- tion system should start with developmentally creasing the probability for depression based investigations of etiological and (Kendler, Kuhn, & Prescott, 2004), while more pathogenetic pathways from infancy to adultspecific personality dimensions such as de- hood and how these pathways give rise to sevpendency and perfectionism have been shown eral disorders. In addition, our increasing to influence the interpretation of life stress knowledge of the biological mechanisms in(Zuroff et al., 2004), which in turn influences volved in the successful treatment of psychiatthe neurobiological effects of adversity (Luyten ric disorders (Ftkin, Pittenger, Polan, & et al., 2006). Studies in this context suggest that Kandel, 2005) is also likely to transform our both psychophysiological and biochemical conception of prevention and treatment of stress might be contingent on the "fit" between these disorders. As noted, identifying such depersonality and the nature of the stressor velopmental processes will probably allow (Allen, Horne, & Trinder, 1996; Ewart, tailoring treatment to particular individuals, Jorgensen, &C Kolodner, 1998; Gruen et al., both from a neurobiological (Aitchison, Basu, 1997; Sauro et al., 2001). Gruen and col- McGuffin, & Craig, 2005) as well as from a leagues (1997), for instance, found that expo- psychotherapeutic perspective (Blatt & sure to an induced failure-stressor was associ- Zuroff, 2005), and is likely to shed more light ated with changes in plasma homovanillic acid on the etiological and pathogenetic factors (HVA), the primary dopamine metabolite in that underlie psychopathology. humans, in highly perfectionistic, but not in Several of these etiologically based thecontrol subjects. oretical models that focus on temperament Taken together, these findings from re- and personality already exist (e.g., see Widiger cent stress research are likely to alter our con- & Simonson, 2005a, 2005b for an overview). ception and classification of depression in Various other models explaining comorbidity many ways. They suggest that we should not between disorders have been developed, inconsider depression as an isolated disease, but cluding, but not limited to, the tripartite

94

Is It Time to Change the DSM Approach?

model of depression and anxiety which has developed into a more comprehensive hierarchical model (Clark, 2005; Watson, 2005; for an alternative see Lara, Pinto, Akiskal, & Akiskal, in press), and theoretical models examining gene-environment interactions (e.g., Askland & Parsons, in press; Kendler, Prescott, Myers, & Neale, 2003; Krueger, Caspi, Moffitt, & Sliva, 1998). Yet, it is also clear that much more research in this area is needed, particularly concerning the interrelationship between these various theoretical models (Brown & Barlow, 2005; First, 2005). CONCLUSIONS AND FUTURE PERSPECTIVES It is clear that we are on the brink of a new era in psychiatric diagnosis. Just as the introduction of DSM-III introduced a new era in psychiatry and psychology, findings reviewed in this paper might lead to a radical shift in the way we classify mental disorders (Brown & Barlow, 2005). Research findings are increasingly challenging the way we conceptualize, classify, and treat psychiatric disorders, moving away from a descriptive towards an etiologically based approach. This is true in research on depression, but also for a number of other disorders (e.g., Charney et al., 2002), although there are wide differences between the various disorders. Thus, an important task for further research will be to disentangle the many etiological pathways from infancy to adulthood in the genesis of psychiatric disorders, and to investigate, rather than accept a priori, basic assumptions concerning the nature of psychiatric disorders (Widiger &c Clark, 2000). Yet, although some researchers have argued that the available data suggest the adoption of a more etiologically based classification system in the next edition of DSM (e.g., Watson, 2005), we believe the time is not yet ripe for such a radical shift, mainly because, as

noted, existing findings are still relatively poorly integrated and it is highly unlikely that a consensus could soon be reached among proponents of various theories (First, 2005). Moreover, much relevant research is simply lacking. As Brown and Barlow (2005, p. 555) recently put it: "[Although] a clear consensus has emerged that we must move beyond description and back to a consideration of etiologic theory . . . this approach is so radically different that we are unlikely to achieve consensus in time for the publication of the DSM-V." In addition, the consequences of such a radical shift on clinical practice, research, and public policy should first be carefully evaluated (First, 2005). In the meantime, however, data reviewed in this paper clearly suggest that research in this direction, comparing different alternatives of classification systems, is urgently needed if we are to overcome limitations of our current classification system. This might entail the adoption of hybrid systems in future editions of DSM, which combine dimensional with categorical, and descriptive with etiological considerations (Westen et al., 2006), or it might entail using a different classification system for research and for clinical practice (Krueger et al., 2005). Hence, a daunting task lies ahead of us, which will necessitate research and consensus on basic dimensions and/or endophenotypes (First, 2005). This will also inevitably entail a multidisciplinary approach, spanning expertise from the fields of psychiatric genetics to developmental psychopathology, and from cell research to epidemiological and sociocultural studies (Ursano, 2004). Ultimately, however, these studies may lead to a more comprehensive, etiologically based system of psychiatric disorders that could then lead to more fully informed evidence-based prevention and treatment strategies with greater clinical utility (Watson, 2005).

Luyten and Blatt

95

REFERENCES Aitchison, K. J., Basu, A., McGuffin, P., & Craig, I. (2005). Psychiatry and the "new genetics": Hunting for genes for behaviour and drug response. British Journal of Psychiatry, 186, 91-92. Allen, N. B., Horne, D. J., & Trinder, J. (1996). Sociotropy, autonomy, and dysphoric emotional responses to specific classes of stress: A psychophysiological evaluation. Journal of Abnormal Psychology, 105, 25-33. American Psychiatric Association (1980). Diagnostic and Statistical Manual of Mental Disorders, Third Edition. Washington, DC: American Psychiatric Association. American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Disorders Fourth Edition. Washington, DC: American Psychiatric Association. Askland, K., & Parsons, M. (in press). Toward a biaxial model of "bipolar" affective disorders: Spectrum phenotypes as the products of neuroelectrical and neurochemical alterations. Journal of Affective Disorders. clinical course in immunological and cardiovascular disease. Clinical Psychology Review, 13, 421-450. Blatt, S. J., & Levy, K. N. (1998). A psychodynamic approach to the diagnosis of psychopathology. In J. W. Baron (Ed.), Making Diagnosis Meaningful. Enhancing Evaluation and Treatment of Psychological Disorders (pp. 73-109). Washington, DC: American Psychological Association. Blatt, S. J., Rounsaville, B. J., Eyre, S., & Wilber, C. (1984). The psycho-dynamics of opiate addiction. Journal of Nervous and Mental Disease, 172,342-352. Blatt, S. J., & Shahar, C. (2004). Psychoanalysis: For what, with who, and how: A comparison with psychotherapy. Journal of the American Psychoanalytic Association, 52, 393-447. Blatt, S. J., & Zuroff, D. C. (2005). Empirical evaluation of the assumptions in identifying evidence-based treatments in mental health. Clinical Psychology Revietv, 25, 459-486.

Bateman, A., & Fonagy, P. (2004). Psychotherapy for Borderline Personality Disorder. Mentalization-Based Treatment. Oxford: Oxford University Press.

Blatt, S. J., Zuroff, D. C , Bondi, C. M., Sanislow, C. A., & Pilkonis, P. A. (1998). When and how perfectionism impedes the brief treatment of depression: Further analyses of The National Institute of Mental Health Treatment of Depression Collaborative Research Program. Beseler, C , Jacobson, K. C , Kremen, W. S., Ly- Journal of Consulting and Clinical Psychology, ons, M. J., Clatt, S. J., Faraone, S. V., Cillespie, 66, 423-428. N. A., & Tsuang, M. T. (2006). Is there heterogeneity among syndromes of substance use dis- Bonne, O., & Charney, D. S. (2004). order for illicit drugs. Addictive Behaviors, 31, Neuropathology of stress: Prospects and caveats. Psychiatry, 67, 407-411. 929-947. Beutler, L. E., & Malik, M. L. (Eds.). (2002). Rethinking the DSM. A Psychological Perspective. Washington, DC: American Psychological Association. Brown, T. A., & Barlow, D. H. (2005). Dimensional versus categorical classification of mental disorders in the Fifth Edition of the Diagnostic and Statistical Manual of Mental Disorders and beyond: Comment on the special section. JourBlatt, S. J. (1974). Levels of object representa- nal of Abnormal Psychology, 114, 551-556. tion in anaclitic and introjective depression. PsyCaspi, A., Sugden, K., Moffitt, T. E., Taylor, A., choanalytic Study ofthe Child, 29, 107-157. Craig, I. W., Harrington, H., McClay, J., Mill, Blatt, S. J. (2004). Experiences of Depression: J., Martin, J., Braithwaite, A., & Poulton, R. Theoretical, Clinical and Research Perspectives. (2003). Influence of life stress on depression: Washington DC: American Psychological Moderation by a polymorphism in the 5-HTT Association. gene. Science, 301, 386-389. Blatt, S. J., Cornell, C. E., & Eshkol, E. (1993). Charney, D. S., Barlow, D. H., Botteron, K., CoPersonality style, differential vulnerability and

96

Is It Time to Change the DSM Approach?

hen, J., Goldman, D., Gur, J. D., Ling, K.-M., Lopez, J. F., Meador-Woodruff, J. H., Moldin, S. O., Nestler, E. J., Watson, S. J., & Zalcman, S. J. (2002). Neuroscience Research Agenda to guide development of a pathophysiologically based classification system. In D. J. Kupfer, M. B. First, & D. A. Regier (Eds.), A Research Agenda for DSM-V (pp. 31-83). Washington, DC: American Psychiatric Association.

liams, J. B., Ustun, B., & Peele, R. (2004). Clinical utility as a criterion for revising psychiatric diagnoses. American Journal of Psychiatry, 161, 946-954. Fonagy, P., Gergely, G., Jurist, E. L., & Target, M. (2004). Affect Regulation, Mentalization, and the Development of the Self. London/New York: Karnac.

Claes, S. (2003). Corticotropin-releasing hor- Frances, A. J., First, M. B., & Pincus, H. A. mone (CRH) in psychiatry: From stress to (1995). DSM-IV Guidebook. Washington, DC: psychopathology. Annals of Medicine, 35,1-12. American Psychiatric Press. Claes, S. J., & Nemeroff, C. B. (2005). Friedman, M. J., & Harris, W. W. (2004). ToCorticotropin-Releasing Factor (CRF) and ma- ward a national PTSD brain bank. Psychiatry, jor depression: Towards an integration of psy- 67, 384-390. chology and neurobiology in depression Gold, P. W., & Chrousos, G. P. (2002). Organiresearch. In J. Corveleyn, P. Luyten, &c S. J. Blatt zation of the stress system and its dysregulation (Eds.), The Theory and Treatment of Depres- in melancholic and atypical depression: High vs. sion: Towards a Dynamic Interactionism Model low CRH/NE states. Molecular Psychiatry, 7, (pp. 227-252). Leuven/Mahwah, NJ: Leuven 254-275. University Press/Lawrence Erlbaum. Gonda, X., Juhasz, G., Laszik, A., Rihmer, Z., & Clark, L. A. (2005). Temperament as a unifying Bagdy, G. (2005). Subthreshold depression is basis for personality and psychopathology. linked to the functional polymorphism of the Journal of Abnormal Psychology, 114, 5HT transporter gene. Journal of Affective Dis505-521. orders, 87, 291-297. Cox, B. J., McWilliams, L. A., Enns, M. W., & Clara, I. P. (2004). Broad and specific personality dimensions associated with major depression in a nationally representative sample. Comprehensive Psychiatry, 45, 246-253. Cuthbert, B. N. (2005). Dimensional models of psychopathology: Research agenda and clinical utility. Journal of Abnormal Psychology, 114, 565-569. Etkin, A., Pittenger, C , Polan, H. J., & Kandel, E. (2005). Toward a neurobiology of psychotherapy: Basic science and clinical applications. Journal ofNeuropsychiatry and Clinical Neuroscience, 17, 145-158. Gruen, R. J., Silva, R., Ehrlich, J., Schweitzer, J. W., & Friedhoff, A. J. (1997). Vulnerability to stress: Self-Criticism and stress-induced changes in biochemistry. Journal of Personality, 65, 33-47. Gunnar, M. R. (2002). Integrating neuroscience and psychological approaches in the study of early experiences. Annals of the New York Academy of Sciences, 1008, 238-247. Haslam, N. (2003). Categorical versus dimensional models of mental disorder: The taxometric evidence. Australian and New Zealand Journal of Psychiatry, 37, 696-704.

Heim, C , Plotsky, P. M., & Nemeroff, C. B. Ewart, C. K., Jorgensen, R. S., &C Kolodner, K. B.(2005). Importance of studying the contribu(1998). Sociotropic cognitions moderates blood tions of early adverse experiences to pressure response to interpersonal stress in neurobiological findings in depression. high-risk adolescent girls. International Journal Neuropsychopharmacology, 29, 641-648. of Psychophysiology, 28, 131-142. Hollon, S. D., Munoz, R. F., Barlow, D. H., First, M. B. (2005). Clinical utility: A prerequi- Beardslee, W. R., Bell, C. C , Bernal, G., Clarke, site for the adoption of a dimensional approach G. N., Franciosi, L. P., Kazdin, A. E., Kohn, L., in DSM. Journal of Abnormal Psychology, 114, Linehan, M. M., Markowitz, J. C , Miklowitz, 560-564. D. J., Persons, J. B., Niederehe, G., &C Sommers, First, M. B., Pincus, H. A., Levine, J. B., Wil- D. (2002). Psychosocial intervention develop-

Luyten and Blatt

97

ment for the prevention and treatment of depres- Kendler, K. S., Prescott, G. A., Myers, J., & sion: Promoting innovation and increasing Neale, M. G. (2003). The structure of genetic and environmental risk factors for common psyaccess. Biological Psychiatry, 52, 610-630. chiatric and substance use disorders in men and Kaufman, J., Yang, B.-Z., Douglas-Palumberi, women. Archives of General Psychiatry, 60, H., Houshyar, S., Lipschitz, D., Krystal, J. H., & 929-937. Gelernter, J. (2004). Social support and serotonin transporter gene moderate depression in Krueger, R. F., Gaspi, A., Moffit, T. E., & Silva, maltreated children. Proceedings of the Na- P. A. (1998). The structure and stability of comtional Academy of Sciences of the USA, 101, mon mental disorders (DSM-III-R): A longitudinal-epidemiological study. Journal of 17316-17321. Abnormal Psychology, 107, 216-227. Keller, M. B., Klein, D. N., Hirschfeld, R. M. A., Kocsis, J. H., Mc-Cullough, J. P., Miller, I., Krueger, R. F., Watson, D., & Barlow, D. H. First, M. B., Holzer, C. P., Ill, Keitner, G. I., (2005). Introduction to the special section: ToMarin, D. B., & Shea, T. (1995). Results of the ward a dimensionally based taxonomy of DSM-IV mood disorders trial. American Jour- psychopathology. Journal of Abnormal Psynal of Psychiatry, 152, 843-849. chology, 114, 491-A93. Keller, M. B., McCuUough, J. P., Klein, D. N., et al. (2000). A comparison of nefazodone, the cognitive behavioral analysis system of psychotherapy, and their combination for the treatment of chronic depression. Neto England Journal of Medicine, 342, 1462-1472. Kupfer, D. J., First, M. B., & Regier, D. A. (2002). A Research Agenda for DSM-V. Washington, DG: American Psychiatric Association. Lam, D., Wright, K., & Smith, N. (2004). Dysfunctional assumptions in bipolar disorder. Journal of Affective Disorders, 79, 193-199.

Kendler, K. S. (2001). Twin studies of psychiat- Lara, D. R., Pinto, O., Akiskal, K., & Akiskal, ric illness. An update. Archives of General Psy- H. S. (2006). Toward an integrative model of the chiatry, 58,1005-1014. spectrum of mood, behavioral and personality Kendler, K. S. (2002). Hierarchy and disorders based on fear and anger traits: I. Gliniheritability: The role of diagnosis and modeling cal implications. Journal of Affective Disorders, in psychiatric genetics. American Journal of Psy- 94, 67-87. chiatry, 259,515-518. Levy, K. N., Glarkin, J. F., Yeomans, F. E., Scott, Kendler, K. S., & Gardner, G. O. (1998). Boundaries of major depression: An evaluation of DSM-IV criteria. American Journal of Psychiatry, 155, 172-177. L. N., Wasserman, R. H., &c Kernberg, O. F. (2006). The mechanisms of change in the treatment of borderline personality disorder with transference-focused psychotherapy. Journal of Clinical Psychology, 62, 481-501.

Kendler, K. S., Gardner, G. O., & Prescott, G. A. (2002). Toward a comprehensive developmen- Levinson, D. F. (2006). The genetics of deprestal model for major depression in women. Amer- sion: A review. Biological Psychiatry, 60, 84-92. ican Journal of Psychiatry, 159, 1133-1145. Linehan, M. M. (1993). Cognitive-Behavioral Kendler, K. S., Kuhn, J., & Prescott, G. A. Treatrnent of Borderline Personality Disorder. (2004). The interrelationship of neuroticism, New York: Guilford. sex, and stressful life events in the prediction of Luyten, P., Blatt, S. J., & Gorveleyn, J. (2005). episodes of major depression. American Journal Towards integration in the theory and treatment of Psychiatry, 161, 631-636. of depression? The time is now. In J. Gorveleyn, Kendler, K. S., Kuhn, J. W., Vittum, J., Prescott, G. A., & Riley, B. (2005). The interaction of stressful life events and a serotonin transporter polymorphism in the prediction of episodes of major depression. Archives of General Psychiatry, 62,529-535. P. Luyten, & S. J. Blatt (Eds.), The Theory and Treatment of Depression: Towards a Dynamic Interactionism Model (pp. 2 6 5 - 2 9 6 ) . Leuven/Mahwah, NJ: Leuven University Press/Lawrence Eribaum. Luyten, P., Blatt, S. J., Van Houdenhove, B., &

98

Is It Time to Change the DSM Approach?

Corveleyn, J. (2006). Depression research and treatment: Are we skating to where the puck is going to be? Clinical PsycPjology Review, 26, 985-999. McHugh, P. R. (2005). Striving for coherence: Psychiatry's efforts over classification. Journal of the American Medical Association, 293, 2526-2528. Moffitt, T. E., Caspi, A., & Rutter, M. (2005). Strategy for investigating interactions between measured genes and measured environments, Archives of General Psychiatry, 62, 473-481. National Institute of Mental Health. (2003). Breaking Ground, Breaking Through: The Strategic Plan for Mood Disorders Research. Retrieved September 3, 2 0 0 4 , from http://www.nimh.nih.gov/strategic/mooddisorders.pdf. Nemeroff, C. B. (2002). Editorial. Comorbidity of mood and anxiety disorders: The rule, not the exception? American Journal of Psychiatry, 159, 3-4. Nemeroff, C. B., Heim, C. M., Thase, M. E., Klein, D. N., Rush, J. A., Schatzberg, A. F., Ninan, P. T., McCullough, J. P., Jr, Weiss, P. M., Dunner, D. L., Rothbaum, B. O., Kornstein, S., Keitner, G., & Keller, M. B. (2003). Differential responses to psychotherapy versus pharmacotherapy in patients with chronic forms of major depression and childhood trauma. Proceedings of the National Academy of Sciences of the USA, 100, 14239-14296.

PhiUips, K. A., First, M. B., & Pincus, H. A. (Eds.). (2003). Advancing DSM. Dilemmas in Psychiatric Diagnosis. Washington, DC: American Psychiatric Association. Rose, S. (2001). Moving on from old dichotomies: Beyond nature-nurture towards lifeline perspective. British Journal of Psychiatry, 178 (Suppl. 40), s3-s7. Ruscio, J., & Ruscio, A. M. (2000). Informing the continuity controversy: A taxometric analysis oi depression. Journal of Abnormal Psychology, 109, 473-487. Sauro, M. D., Jorgensen, R. S., Larson, C. A., Frankowski, j . J., Ewart, C. K., & White, J. (2001). Sociotropic cognition moderates stress-induced cardiovascular responsiveness in college women. Journal of Behavioral Medicine, 24, 423-439. Shafran, R., & Mansell, W. (2001). Perfectionism and psychopathology: A review of research 3ind treatment. Clinical Psychology Review 21 879-906. Solomon, A., Haaga, D. A. F., & Arnow, B. A. (2001). Is clinical depression distinct from subthreshold depressive symptoms? A review of the continuity issue in depression research./owrnal of Nervous and Mental Disease, 189, 498-506.

Tsuang, M. T., Stone, W. S., Tarbox, S. I., & Faraone, S. V. (2003). Insights from neuroscience for the concept of schizotaxia and the diagnosis of schizophrenia. In K. A. Phillips, M. B. Ormel, J., Oldehinkel, A. J., & Brilman, E. I. First, &c H. A. Pincus (Eds.), Advancing DSM. (2001). The interplay and etiological continuity Dilemmas In Psychiatric Diagnosis (pp. of neuroticism, difficulties, and life events in the 105-127). Washington, DC: American Psychietiology of major and subsyndromal, first and atric Association. recurrent depressive episodes in later life. AmeriUrsano, R. J. (2004). Editor's note. Psychiatry, can Journal of Psychiatry, 158, 885-891. 67, 309. Osuch, E., Ursano, R., Li, H., Webster, M., Hough, C , Fullerton, C , & Leskin, G. (2004). Van Houdenhove, B., & Egle, U. T. (2004). Brain environment interactions: Stress, Fibromyalgia: A stress disorder? Piecing the Posttraumatic Stress Disorder, and the need for a biopsychosocial puzzle together. Psychotherapy postmortem brain collection. Psychiatry, 67, and Psychosomatics, 73, 267-275. 353-383. Van Houdenhove, B., Egle, U., & Luyten, P. Parker, G. (2000). Classifying depression: (2005). The role of life stress in fibromyalgia. Should paradigms lost be regained? American Current Rheumatology Reports, 7, 365-370. Journal of Psychiatry, 157, 1195-1203. Van Praag, H. M., de Kloet, R., & van Os, J. Parker, G. (2005). Beyond major depression. (2004). Stress, the Brain and Depression. Cambridge: Cambridge University Press. Psychological Medicine, 35, 467-474.

Luyten and Blatt

99

Watson, D. (2005). Rethinking the mood and psychopathology. Psychological Bulletin, 126, anxiety disorders: A quantitative hierarchical 946-963. model for DSM-V. Journal of Abnormal PsyWidiger, T. A., Sc Samuel, D. B. (2005). Diagchology, 114, 522-536. nostic categories or dimensions? A question for Weinstock, L. M., & Whisman, M. A. (2006). the Diagnostic and Statistical Manual of Mental Neuroticism as a common feature of the depres- Disorders-Fifth Edition. Journal of Abnormal sive and anxiety disorders: A test of the revised Psychology, 114, 494-504. integrative hierarchical model in a national sam- Widiger, T. A., & Simonsen, E. (Eds.) (2005a). ple. Journal of Abnormal Psychology, 115, Special section: A research agenda for the devel68-74. opment of a dimensional classification system of Westen, D., & Harnden-Fischer, J. (2001). Personality profiles in eating disorders: Rethinking the distinction between axis I and axis II. American Journal of Psychiatry, 158, 547-562. Personality disorders. Part I. Journal of Personality Disorders, 19(1).

Widiger, T. A., & Simonsen, E. (Eds.) (2005b). Special section: A research agenda for the develWesten, D., Heim, A. K., Morrison, K., opment of a dimensional classification system of Patterson, M., &C Campbell, L. (2002). Simplify- Personality disorders. Part 11. Journal of Personing diagnosis using a prototype-matching ap- ality Disorders, 19(3). proach: Implications for the next edition of the Zimmerman, M., Mattia, J. I., & Posternak, M. DSM. In L. Beutler &C M. Malik (Eds.), RethinkA. (2002). Are subjects in pharmacological ing the DSM: A Psychological Perspective (pp. treatment trials of depression representative of 2 2 1 - 2 5 0 ) . Washington, DC: American patients in routine clinical practice? American Psychological Association. Journal of Psychiatry, 159, 469-473. Westen, D., Novotny, C. M., & ThompZuroff, D. C , & Blatt, S. J. (2006). The therason-Brenner, H. (2004). The empirical status of peutic relationship in the brief treatment of deempirically supported psychotherapies: Aspression: Contributions to clinical improvement sumptions, findings, and reporting in controlled and enhanced adaptive capacities. Journal of clinical trials. Psychological Bulletin, 130, Consulting and Clinical Psychology, 74, 631-663. 130-140. Westen, D., Shedler, J., &c Bradley, R. (2006). A Zuroff, D. C , Mongrain, M., &C Santor, D. prototype approach to personality disorder di- (2004). Conceptuahzing and measuring personagnosis. American Journal of Psychiatry, 163, ality vulnerability to depression: Comment on 846-856. Coyne and Whiffen( 1995). ^ / / / Z Widiger, T. A., & Clark, A. (2000). Toward fw, 130,489-511. DSM-V and the classification of