Chapter 1 Thomas A. Widiger and Whitney L. Gore In DSM-IV, there [was] no assumption that each category of mental disorder is a completely discrete entity with absolute boundaries dividing it from other mental disorders or from no mental disorder” (American Psychiatric Association [APA], 2000, p. xxxi). This carefully worded disclaimer, however, was somewhat hollow, as it was the case that “DSM-IV [was] a categorical classification that divides mental disorders into types based on criterion sets with defining features” (APA, 2000, p. xxxi). The categorical model of classification is consistent with a medical tradition in which it is believed (and often confirmed in other areas of medicine) that disorders have specific etiologies, pathologies, and treatments (Guze, 1978; Guze & Helzer, 1987; Zachar & Kendler, 2007). Clinicians, following this lead, diagnosed and conceptualized the conditions presented in DSM-IV-TR as disorders that are qualitatively distinct from normal functioning and from one another. DSM-IV-TR provided diagnostic criterion sets to help guide clinicians toward a purportedly correct diagnosis and an additional supplementary section devoted to differential diagnosis that indicated “how to differentiate [the] disorder from other disorders that have similar presenting characteristics” (APA, 2000, p. 10). The intention of the manual was to help the clinician determine which particular mental disorder provides the best explanation for the symptoms and problems facing the patient. Clinicians devote initial time with a new patient to identify, through differential diagnosis, which specific disorder best explains a patient’s presenting complaints. The assumption is that the person is suffering from a single, distinct clinical condition, caused by a specific pathology for which there will be a specific treatment (Frances, First, & Pincus, 1995). Authors of the diagnostic manual devote a considerable amount of time writing, revising, and researching diagnostic criteria to improve differential diagnosis. They buttress each disorder’s criterion set, trying to shore up discriminant validity and distinctiveness, following the rubric of Robins and Guze (1970) that the validity of a diagnosis rests in large part on its “delimitation from other disorders” (p. 108). “These criteria should…permit exclusion of borderline cases and doubtful cases (an undiagnosed group) so that the index group may be as homogeneous as possible” (Robins & Guze, 1970, p. 108). Scientists may devote their careers to attempting to identify the specific etiology, pathology, or treatment for a respective diagnostic category. Under the assumption that the diagnoses do in fact refer to qualitatively distinct conditions, it follows that there should be a specific etiology, pathology, and perhaps even a specific treatment for each respective disorder. The theories, hypotheses, findings, and disputes regarding the specific etiology, pathology, and/or treatment of a respective mental disorder largely inform the respective chapters of professional, graduate, and undergraduate texts on psychopathology, such as this current edition of Adult Psychopathology and Diagnosis. However, the question of whether mental disorders are, in fact, discrete clinical conditions or arbitrary distinctions along continuous dimensions of functioning has been a long-standing issue (Kendell, 1975) and its significance is escalating with the growing recognition of the limitations of the categorical model (Hyman, 2010; Widiger & Clark, 2000; Widiger & Samuel, 2005). The principal model for the validation of mental disorder diagnostic categories was provided by Robins and Guze (1970), who articulated five fundamental phases: clinical description, laboratory study, delimitation from other disorders, follow-up, and family studies. However, the research that has accumulated to date has not supported the validity of the delimitation of the disorders from one another. “Indeed, in the last 20 years, the categorical approach has been increasingly questioned as evidence has accumulated that the so-called categorical disorders like major depressive disorder and anxiety disorders, and schizophrenia and bipolar disorder seem to merge imperceptibly both into one another and into normality…with no demonstrable natural boundaries” (First, 2003, p. 661). As expressed by the vice chair of DSM-5, “the failure of DSM-III criteria to specifically define individuals with only one disorder served as an alert that the strict neo-Kraepelinian categorical approach to mental disorder diagnoses advocated by Robins and Guze (1970), Spitzer, Endicott, & Robins (1978), and others could have some serious problems” (Regier, 2008, p. xxi). As acknowledged by Kendell and Jablensky (2003), “it is likely that, sooner or later, our existing typology will be abandoned and replaced by a dimensional classification” (p. 8). In 1999, a DSM-5 Research Planning Conference was held under joint sponsorship of the APA and the National Institute of Mental Health (NIMH), the purpose of which was to set research priorities that would optimally inform future classifications. One impetus for this effort was the frustration with the existing nomenclature. In the more than 30 years since the introduction of the Feighner criteria by Robins and Guze, which eventually led to DSM-III, the goal of validating these syndromes and discovering common etiologies has remained elusive. Despite many proposed candidates, not one laboratory marker has been found to be specific in identifying any of the DSM-defined syndromes. Epidemiologic and clinical studies have shown extremely high rates of comorbidities among the disorders, undermining the hypothesis that the syndromes represent distinct etiologies. Furthermore, epidemiologic studies have shown a high degree of short-term diagnostic instability for many disorders. With regard to treatment, lack of treatment specificity is the rule rather than the exception (Kupfer, First, & Regier, 2002, p. xviii). DSM-5 Research Planning Work Groups were formed to develop white papers that would set an effective research agenda for the next edition of the diagnostic manual. The Nomenclature Work Group, charged with addressing fundamental assumptions of the diagnostic system, concluded that it will be “important that consideration be given to advantages and disadvantages of basing part or all of DSM-V on dimensions rather than categories” (Rounsaville et al., 2002, p. 12). The white papers developed by the DSM-5 Research Planning Work Groups were followed by a series of international conferences whose purpose was to further enrich the empirical data base in preparation for the eventual development of DSM-5 (a description of this conference series can be found at www.dsm5.org). The first conference was devoted to shifting personality disorders to a dimensional model of classification (Widiger, Simonsen, Krueger, Livesley, & Verheul, 2005). The final conference was devoted to dimensional approaches across the diagnostic manual, including substance use disorders, major depressive disorder, psychoses, anxiety disorders, and developmental psychopathology, as well as the personality disorders (Helzer, Kraemer, et al., 2008). In the introduction to DSM-5 (APA, 2013), the apparent failure of the categorical model of classification is duly noted. “The historical aspiration of achieving diagnostic homogeneity by progressively subtyping within disorder categories no longer is sensible; like most common human ills, mental disorders are heterogeneous at many levels, ranging from genetic risk factors to symptoms” (APA, 2013, p. 12). It was a major purpose of the authors of DSM-5 to have this new edition of the diagnostic manual be “central to the development of dimensional approaches to diagnosis that will likely supplement or supersede current categorical approaches in the coming years” (APA, 2013, 13). However, as will be discussed herein, the diagnoses of DSM-5 remain largely categorical. The purpose of this chapter is to review the DSM-IV-TR and DSM-5 categorical diagnostic approach. The chapter begins with a discussion of the problematic boundaries among the DSM-IV-TR and DSM-5 categorical diagnoses. We then focus in particular on depression, alcohol abuse and dependence, personality disorders, and intellectual disability. We conclude with a discussion of the shifts within DSM-5 toward a dimensional classification. In an effort to force differential diagnosis, a majority of diagnoses in DSM-III (APA, 1980) contained exclusionary criteria specifying that a respective disorder could not be diagnosed if it occurred in the presence of another disorder. These exclusions by fiat did not prove to be effective (Boyd et al., 1984) and many were deleted in DSM-III-R (APA, 1987). As expressed at the time by Maser and Cloninger (1990), “it is clear that the classic Kraepelinian model in which all psychopathology is comprised of discrete and mutually exclusive diseases must be modified or rejected” (p. 12). Many DSM-5 diagnostic criterion sets, however, continue to include exclusionary criteria that attempt to force clinicians to make largely arbitrary choices among alternative diagnoses (APA, 2013), and it is also evident that there will likely continue to be a highly problematic rate of diagnostic co-occurrence (Kessler, Chiu, Demler, & Walters, 2005; Krueger & Markon, 2006; Maser & Patterson, 2002; Widiger & Clark, 2000). The term comorbidity refers to the co-occurrence of distinct disorders, apparently interacting with one another, each presumably with its own etiology, pathology, and treatment implications (Feinstein, 1970). If one considers the entire diagnostic manual (which has not yet been done by any epidemiological study), it would likely be exceedingly rare for any patient to meet the criteria for just one disorder, and the comorbidity rises even further if one considers lifetime co-occurrence. Brown, Campbell, Lehman, Grisham, and Mancill (2001) reported that 95% of individuals in a clinical setting who meet criteria for lifetime major depression or dysthymia also meet criteria for a current or past anxiety disorder. In the case of psychopathology, comorbidity may be saying more about the invalidity of existing diagnostic distinctions than the presence of multiple coexisting conditions (Krueger, 2002; Widiger & Edmundson, 2011). Diagnostic comorbidity has become so prevalent that some researchers have argued for an abandonment of the term comorbidity in favor of a term (e.g., co-occurrence) that is more simply descriptive and does not imply the presence of distinct clinical entities (Lilienfeld, Waldman, & Israel, 1994). There are instances in which the presence of multiple diagnoses suggests the presence of distinct yet comorbid psychopathologies, but in most instances the presence of co-occurring diagnoses does appear to suggest the presence of a common, shared pathology and, therefore, a possible failing of the current diagnostic system (Krueger & Markon, 2006; Widiger & Clark, 2000). “Comorbidity may be trying to show us that many current treatments are not so much treatments for transient ‘state’ mental disorders of affect and anxiety as they are treatments for core processes, such as negative affectivity, that span normal and abnormal variation as well as undergird multiple mental disorders” (Krueger, 2002, p. 44). Diagnostic criteria have traditionally been developed and modified in order to construct a disorder that is as homogeneous as possible, thereby facilitating the likelihood of identifying a specific etiology, pathology, and treatment (Robins & Guze, 1970). However, the typical result of this effort is to leave many cases unaccounted for. In addition, despite the best effort to construct homogeneous and distinct syndromes, DSM-IV-TR was still replete with heterogeneous conditions with overlapping boundaries (Smith & Combs, 2010). New diagnostic categories are added to the nomenclature in large part to decrease clinicians’ reliance on the nonspecific, wastebasket label of “not otherwise specified” (NOS). NOS was among the most frequent disorders within clinical populations (Widiger & Edmundson, 2011). The function of many of the new disorders that have been added to recent editions of the manual have not involved the identification of uniquely new forms of psychopathology. Their purpose was generally instead to fill problematic gaps. Notable examples for DSM-IV included bipolar II (filling a gap between DSM-III-R bipolar and cyclothymic mood disorders), mixed anxiety-depressive disorder (a gap between anxiety and mood disorders), depressive personality disorder (personality and mood disorders), postpsychotic depressive disorder of schizophrenia (schizophrenia and major depression) (Frances et al., 1995). When new diagnoses are added to fill gaps, they have the ironic effect of creating additional boundary problems, thereby making differential diagnosis even more problematic (Phillips, Price, Greenburg, & Rasmussen, 2003; Pincus, Frances, Davis, First, & Widiger, 1992; Pincus, McQueen, & Elinson, 2003). One must ask, for instance, whether it is really meaningful or useful to determine whether mixed anxiety-depressive disorder is a mood or an anxiety disorder, whether schizoaffective disorder is a mood disorder or a form of schizophrenia (Craddock & Owen, 2010, whether postpsychotic depressive disorder of schizophrenia is a form of depression or schizophrenia, whether early onset dysthymia is a mood or a personality disorder (Widiger, 2003), whether acute stress disorder is an anxiety or a dissociative disorder (Cardena, Butler, & Spiegel, 2003), whether hypochondriasis is an anxiety disorder or a somatoform disorder, whether body dysmorphic disorder is an anxiety, eating, or somatoform disorder, and whether generalized social phobia is an anxiety or a personality disorder (Widiger, 2001a). In all these cases the most accurate answer is likely to be that each respective disorder includes features of different sections of the diagnostic manual. Yet the arbitrary and procrustean decision of which single section of the manual in which to place each diagnosis must be made by the authors of a categorical diagnostic manual, and a considerable amount of effort and research is conducted to guide this decision, followed by further discussion and research to refute and debate whatever particular categorical decision was made. There are comparable examples of what might be arbitrary splitting of categories in DSM-5 (APA, 2013). DSM-5 split out from reactive attachment disorder a new diagnosis of disinhibited social engagement disorders. Binge eating disorder (which was originally included within the diagnosis of bulimia nervosa) obtained official recognition. However, for the most part, changes that occurred in DSM-5 were consistent with the intention to shift the manual more closely to a dimensional model. For example, there are cases in which previously “distinct” diagnoses were lumped together rather than split apart. For example, DSM-5 autism spectrum disorder subsumes within one diagnosis DSM-IV-TR autistic disorder, Asperger’s disorder, childhood disintegrative disorder, and pervasive developmental disorder not otherwise specified. The archaic subtypes of schizophrenia were deleted. “Instead a dimensional approach to rating severity of core symptoms of schizophrenia is included in DSM-5 Section III” (APA, 2013, p. 810). The DSM-IV-TR diagnoses of somatization disorder, hypochondriasis, and pain disorder were all subsumed within one diagnosis of somatic symptom disorder. Pathological gambling was lumped with substance use disorders within a new section concerning addictive disorders, and substance abuse and dependence are no longer conceptualized as categorically distinct conditions. Included in Section III of DSM-5 is a proposed dimensional trait model that would subsume all of the existing personality disorder categories. Mood disorders is a section of the APA diagnostic manual for which the presence of qualitatively distinct conditions is particularly difficult to defend, especially for the primary diagnoses of dysthymia and major depressive disorder. Discussed here will be early onset dysthymia, the continuum of depression, and subthreshold major depression, along with more general points concerning the boundary between mood and personality disorder. There is no meaningful distinction between early-onset dysthymia, an officially recognized mood disorder diagnosis, and depressive personality disorder, a diagnosis proposed for DSM-IV but included within its appendix (APA, 2000). In fact, much of the empirical and conceptual basis for adding dysthymia to the DSM-III (APA, 1980) came from research and clinical literature concerning depressive personality (i.e., Keller, 1989). As acknowledged by the principal architects of DSM-III, dysthymia is “roughly equivalent to the concept of depressive personality” (Spitzer, Williams, & Skodol, 1980, p. 159). Depressive personality disorder was included within the mood disorders section of DSM-III despite the recommendations to recognize its existence as a disorder of personality (Klerman, Endicott, Spitzer, & Hirschfeld, 1979) because it resembled the symptomatology of other mood disorders (i.e., depressed mood) more than it resembled the symptoms of other personality disorders (e.g., schizoid). However, whereas mood disorders are defined largely by similarity in content (i.e., mood being the predominant feature; APA, 2013), the personality disorders are defined largely by form (i.e., early onset, pervasive, and chronic) often with quite different content (e.g., schizoid also shares little resemblance to histrionic personality disorder). After DSM-III was published, it became evident that many of the persons who were consistently and characteristically pessimistic, gloomy, cheerless, glum, and sullen (i.e., dysthymic) had been that way since childhood and that in many cases no apparent or distinct age of onset could be established. In other words, its conceptualization as a personality disorder became apparent. DSM-III-R, therefore, added an early-onset subtype (APA, 1987) and acknowledged that “this disorder usually begins in childhood, adolescence, or early adult life, and for this reason has often been referred to as a Depressive Personality” (APA, 1987, p. 231). Personality disorder researchers proposed again for DSM-IV to include a depressive personality disorder diagnosis. They were told that in order for it to be included, it needed to be distinguished from the already established diagnosis of early-onset dysthymia, a task that might be considered rather difficult, if not unfair, given that the latter construct was based in large part on the former construct. Nevertheless, the DSM-IV Personality Disorders Work Group developed a proposed diagnostic criterion set that placed relatively more emphasis on cognitive features not currently included within the criterion set for dysthymia (including early-onset), as well as excluding somatic features (Task Force on DSM-IV, 1991). This criterion set was provided to the DSM-IV Mood Disorders Work Group to include within their DSM-IV field trial to determine empirically whether it was indeed possible to demarcate an area of functioning not yet covered by early-onset dysthymia, or at least identify persons not yet meeting diagnostic criteria for early-onset dysthymia. The proposed criterion set was successful in reaching this goal (Phillips et al., 1998), which, perhaps, should not be surprising because no criterion set for a categorical diagnosis appears to be entirely successful in covering all cases. However, the Mood Disorders Work Group was equally impressed with the potential utility of the depressive personality diagnostic criteria for further describing and expanding the coverage of dysthymia (Keller et al., 1995) and, therefore, incorporated much of the proposed criteria for depressive personality into their proposed revisions for dysthymia, including early-onset (Task Force on DSM-IV, 1993). The DSM-IV Task Force recognized that it might be problematic to now require the personality disorder researchers to further redefine depressive personality to distinguish it from this further revision of dysthymia. Therefore, the DSM-IV Task Force decided instead to include both criterion sets in the appendix to DSM-IV (along with the original criterion set for dysthymia within the mood disorders section), with the acknowledgment that there may not be any meaningful distinction between them (APA, 1994; Frances et al., 1995). However, depressive personality disorder was not even included within the appendix for DSM-5. Dysthymia and chronic major depressive disorder are now collapsed within persistent depressive disorder in DSM-5. The common view is that many instances of sadness (or even depression) do not constitute a mental disorder. Persons can be very sad without having a mental disorder (Horwitz & Wakefield, 2007). However, a simple inspection of the diagnostic criteria for major depressive disorder would not lend confidence to a conceptualization of this condition as being qualitatively distinct from “normal” depression or sadness (Andrews et al., 2008). Persons who are just very sad will have most of the same attributes (if not all of them) but just at a lesser degree of severity. The diagnostic criteria for major depressive disorder include depressed mood, loss of interest or pleasure, weight loss (or gain), insomnia (or hypersomnia), psychomotor retardation (or agitation), loss of energy, feelings of worthlessness, and/or diminished capacity to make decisions (APA, 2013). Each of these diagnostic criteria is readily placed along a continuum of severity that would shade imperceptibly into what would be considered a “normal” sadness or depression. DSM-5, therefore, includes specific thresholds for each of them, but they are clearly arbitrary thresholds that simply demarcate a relatively higher level of severity from a lower level of severity (e.g., “nearly every day” or “markedly diminished,” and at least a “2-week” period; APA, 2013, p. 188). The diagnosis requires five of these nine criteria, with no apparent rationale for this threshold other than it would appear to be severe enough to be defensible to be titled as a “major” depressive episode, as distinguished from a “minor” depressive episode, which is then distinguished from “normal” sadness (APA, 2013). Depression does appear to shade imperceptibly into “normal” sadness (Andrews et al., 2008). Üstün and Sartorius (1995) conducted a study of 5,000 primary-care patients in 14 countries and reported a linear relationship between disability and number of depressive symptoms. Kessler, Zhao, Blazer, and Swartz (1997) examined the distribution of minor and major symptoms of depression using data from the National Comorbidity Survey. They considered the relationship of these symptoms with parental history of mental disorder, number and duration of depressive episodes, and comorbidity with other forms of psychopathology. Respective relationships increased with increasing number of symptoms, with no clear, distinct break. Sakashita, Slade, and Andrews (2007) examined the relationship between the number of symptoms of depression and four measures of impairment using data from the Australian National Survey of Mental Health and Well-Being, and found that the relationship was again simply linear, with no clear or natural discontinuity to support the selection of any particular cutoff point. Taxometrics refers to a series of related statistical techniques to detect whether a set of items is optimally understood as describing (assessing) a dimensional or a categorical construct (Beauchaine, 2007; Ruscio & Ruscio, 2004). Other statistical techniques, such as cluster or factor analyses, presume that the construct is either categorical or dimensional (respectively) and then determines how best to characterize the variables or items in either a categorical or dimensional format (respectively). Taxometric analyses are uniquely intriguing in providing a direct test of which structural model is most valid in characterizing the set of items or variables. A number of taxometric studies have been conducted on various symptoms and measures of depression. The first was provided by Ruscio and Ruscio (2000) in their taxometric analyses of items from the Beck Depression Inventory and, independently, items from the Zung Self-Rating Depression Scale in a sample of 996 male veterans who had received a diagnosis of post-traumatic stress disorder but also had a high prevalence rate of major depressive disorder, as well as a sample of 8,045 individuals from the general population (60% female) who completed the items from the Depression scale of the Minnesota Multiphasic Personality Inventory. They indicated that “results of both studies, drawing on three widely used measures of depression, corroborated the dimensionality of depression” (Ruscio & Ruscio, 2000, p. 473). The taxometric findings of Ruscio and Ruscio (2000) have been subsequently replicated, including taxometric analyses of (a) structured interview assessments of DSM-IV-TR major depressive disorder symptoms and, independently, items from the Beck Depression Inventory in a sample of 960 psychiatric outpatients (Slade, 2007), (b) major depressive disorder diagnostic criteria assessed in the 1,933 persons who endorsed at least one criterion in the Australian National Survey of Mental Health and Well-Being (Slade & Andrews, 2005), (c) self- and parent-reported depressive symptoms in 845 children and adolescents drawn from the population-based Georgia Health and Behavior Study (Hankin, Fraley, Lahey, & Waldman, 2005), (d) responses to MMPI-2 depression scales completed by 2,000 psychiatric inpatients and outpatients (Franklin, Strong, & Greene, 2002), (e) epidemiologic survey of depressive symptoms within 392 college students (Baldwin & Shean, 2006), (f) Beck Depression Inventory items reported by 2,260 college students (Ruscio & Ruscio, 2002), and (g) depression items in the Composite International Diagnostic Interview as administered in the National Comorbidity Survey to 4,577 participants who endorsed the item concerning a lifetime occurrence of sad mood or loss of interest (Prisciandoro & Roberts, 2005). However, in contrast to the findings from these eight taxometric studies, three taxometric studies have supported a latent class taxon, including semistructured interview assessments of DSM-IV-TR major depressive disorder symptoms in 1,800 psychiatric outpatients (Ruscio, Zimmerman, McGlinchey, Chelminski, & Young, 2007), interview and self-report assessments of depression in 1,400 high school students (Solomon, Ruscio, Seeley, & Lewinsohn, 2006), and self-report and interview data on depression in 378 adolescents receiving treatment for depression (Ambrosini, Bennett, Cleland, & Haslam, 2002). In sum, the bulk of the evidence does appear to support a dimensional understanding of depression, but there is some ambiguity and inconsistency in the taxometric findings (Beach & Amir, 2003; Beauchaine, 2007; Widiger, 2001b). Depression is a section of the diagnostic manual that does have considerable difficulty identifying or defining a clear boundary with “normal” sadness. Subthreshold cases of depression (i.e., persons with depressive symptoms below the threshold for a DSM-5 mental disorder diagnosis) are clearly responsive to pharmacologic interventions, do seek treatment for their sadness, and are often being treated within primary care settings (Judd, Schettler, & Akiskal, 2002; Pincus et al., 2003). These facts contributed to the proposal to include within an appendix to DSM-IV a diagnosis of “minor depressive disorder,” which it is acknowledged “can be difficult to distinguish from periods of sadness that are an inherent part of everyday life” (APA, 2000, p. 776). Wakefield (2007) has been critical of the criteria for major depressive disorder for including an inconsistently applied exclusion criterion. The DSM-IV-TR excluded most instances of depressive reactions to the loss of a loved one (i.e., uncomplicated bereavement). Depression after the loss of a loved one could be considered a mental disorder if “the symptoms persist for longer than 2 months” (APA, 2000, p. 356). Allowing persons just 2 months to grieve before one is diagnosed with a mental disorder does appear to be rather arbitrary. More importantly, it is also unclear if depression in response to other losses should not also then be comparably excluded, such as depression secondary to the loss of a job or physical health (Wakefield, Schimtz, First, & Horwitz, 2007). Why the loss of a person is treated so differently from the loss of health or a job is not clear. On the other hand, one could argue alternatively that all exclusion criteria should be removed. Perhaps the problem is not that depression in response to a loss of a job or physical disorder should not be a disorder, analogous to bereavement (Wakefield, 2007); perhaps the problem is that bereavement should be a mental disorder (Bonanno et al., 2007; Forstmeier & Maercker, 2007; Widiger & Miller, 2008). What is currently considered to be a normal depression in response to the loss of a loved one does often, if not always, include pain and suffering, meaningful impairment to functioning, and is outside of the ability of the bereaved person to fully control, the essential hallmarks of a mental disorder (Widiger & Sankis, 2000). The depression is a reasonable response to the loss of a loved one, a psychological trauma, but many physical disorders and injuries are reasonable and understandable responses to a physical trauma. The loss is perhaps best understood as part of the etiology for the disorder, not a reason for which a disorder is not considered to be present (Widiger, 2012a). One of the major revisions for DSM-5 was indeed to weaken the distinction between normal bereavement and a mental disorder of depression. DSM-5 no longer excludes the diagnosis of a major depressive disorder if the depression is secondary to the loss of a loved one. “Responses to a significant loss (e.g., bereavement, financial ruin, losses from a natural disaster, a serious medical illness or disability)” (APA, 2013, p. 161) can now all be diagnosed as a mental disorder. One of the sections of the diagnostic manual for which a categorical model of classification and conceptualization has had a firmly entrenched tradition has been the substance use disorders. Alcoholism in particular has long been conceptualized as a qualitatively distinct disease (Garbutt, 2008; Goodwin & Guze, 1996). A significant change to its diagnosis and conceptualization occurred with DSM-III-R (APA, 1987) when it shifted from being understood as a purely physiological dependence to a broader and less specific behavioral dependence (Carroll, Rounsaville, & Bryant, 1994; Edwards & Gross, 1976). “Dependence is seen as a complex process that reflects the central importance of substances in an individual’s life, along with a feeling of compulsion to continue taking the substance and subsequent problems controlling use” (Schuckit et al., 1999, p. 41). To many, though, the diagnosis does still refer to a disease, but one that is developed through a normal social-learning history (Kandel, 1998). However, the diagnosis has been broadened considerably in DSM-5 wherein it is referred to as a behavioral addiction, and would, therefore, be listed along with pathological gambling (Martin, 2005; Petry, 2006; Potenza, 2006). Pathological gambling has been considered by many substance use and pathological gambling researchers and clinicians to be an addiction, but it could not be included within the substance related disorders section because it does not involve the ingestion of a substance (Bradford, Geller, Lesieur, Rosenthal, & Wise, 1996). This requirement has been deleted in DSM-5, with the section renamed “substance-related and addictive disorders” (APA, 2013). This new class of disorders could eventually contain a wide variety of possible behavioral addictions, including an excessive participation in shopping, sex, or the Internet. As stated at one point on the DSM-5 website, along with pathological gambling, “other addiction-like behavioral disorders such as ‘Internet addiction’…will be considered as potential additions to this category as research data accumulate” (APA, 2010, “Substance Related Disorders,” para. 1). The preface to this section of the diagnostic manual explicitly states that Internet, sex, and shopping addictions are not included because there is currently insufficient evidence to support their validity. However, it is apparent that the broadening of the concept of substance dependence to include behavioral forms of addiction will encourage clinicians to diagnose these additional variants. “This ‘slippery slope’ makes it difficult to know where to draw the line demarcating any excessive behavior as an addiction” (Petry, 2005a, p. 7). Provided within an appendix to DSM-5 for conditions needing further study is Internet gaming disorder (i.e., behavioral addiction on Internet games), including its diagnostic criteria, risk factors, prevalence, and differential diagnosis. Proposed for inclusion in the sex disorders section of DSM-5 was hypersexual disorder, which can indeed be identified as a sex addiction (Kafka, 2010; Ragan & Martin, 2000; Winters, 2010). The distinction between harmful substance use and a substance use disorder is itself unclear and indistinct. Presumably, persons can choose to consume alcohol without being compelled to do so by the presence of a mental disorder. The DSM-5 diagnostic criteria for a substance use disorder are fallible indicators for harmful and dyscontrolled usage (e.g., use more than originally intended, continue to use despite social consequences, and reduction of other activities in preference for the substance; APA, 2013). The more of these indicators of dyscontrol that are present, the more likely that there is in fact dyscontrol, but none can be considered infallible in the identification of dyscontrol and no particular number of them clearly demarcates a boundary between the presence versus absence of dyscontrolled usage. It is not even clear how much purportedly volitional or regulatory control a normal, healthy person has over adaptive, healthy behaviors (Bargh & Ferguson, 2000; Howard & Conway, 1986; Kirsch & Lynn, 2000; Wegner & Wheatley, 2000), let alone the boundary between controlled and dyscontrolled harmful behaviors. Both normal and abnormal human functioning is, at best, the result of a complex interaction of apparent volitional choice with an array of biogenetic and environmental determinants. The distinction between DSM-IV-TR alcohol abuse and dependence was equally fuzzy. Abuse has generally been considered to be simply a residual category and/or a less severe form of dependence (Saunders, 2006). Some of the diagnostic criteria for abuse were contained with the criterion set for dependence (e.g., interference with social, occupational, or recreational activities), which is always a problem for disorders that would be considered to be qualitatively distinct. It is largely for this reason that the formal distinction between abuse and dependence was abandoned in DSM-5 (APA, 2013). The diagnostic criteria for alcohol dependence were written largely in an effort to describe a prototypic case of the disorder, a practice that is still followed for all but a few of the disorders throughout DSM-5. However, prototypic cases are typically understood to be the most severe cases and/or the cases that involve all possible features or symptoms of the disorder (First & Westen, 2007). The construction of diagnostic criterion sets in terms of prototypic cases does work to an extent, but it also fails to adequately describe many of the actual cases, including the subthreshold cases, and perhaps even the typical cases, depending upon the distribution of features and symptomatology within the population. Constructing criterion sets in terms of prototypic cases can be comparable to confining the description and diagnosis of (for instance) intellectual disability to the most severe variant, and then attempting to apply this description to mild and moderate variants; a method of diagnosis that would obviously be sorely limited. The limitations of this approach are now becoming more closely appreciated in the diagnosis of dyscontrolled substance use and, more specifically, alcohol use disorders, where the existing criterion sets are failing to adequately describe (for instance) dyscontrolled and impairing alcohol usage in adolescents (Crowley, 2006) and other “diagnostic orphans” (Saunders, 2006). The limitation is perhaps most clearly demonstrated in studies using item response theory (IRT) methodology. IRT allows the researcher to investigate the fidelity with which items are measuring a latent trait along the length of its continuum, contrasting, for instance, the amount of information that different diagnostic criteria provide at different levels of the latent trait (Muthen, 2006). Some diagnostic criteria, for instance, might be most useful in distinguishing among mild cases of the disorder, whereas other diagnostic criteria are most useful in distinguishing among the more severe cases of the disorder. A number of IRT analyses have now been conducted for the diagnosis of substance dependence (and other disorders) and the findings are remarkably consistent (Reise & Waller, 2009). The existing diagnostic criterion sets (and/or symptoms currently assessed in existing instruments) cluster around the high end of the disorder as opposed to being spread out across the entire range of the continuum (e.g., Kahler & Strong, 2006; Langenbucher et al., 2004; Muthen, 2006; Proudfoot, Baillie, & Teesson, 2006; Saha, Chou, & Grant, 2006). This consistent pattern of results is in stark contrast to what is traditionally found in cognitive ability testing, where IRT analyses have been largely developed and previously applied (Reise & Waller, 2009). It is evident from the IRT analyses that the existing diagnostic criterion sets are sorely inadequate in characterizing the lower and even middle range of substance use dysfunction, consistent with the DSM-IV-TR and DSM-5 descriptions being confined to a prototypic case (of the presumably qualitatively distinct disorder). If alcohol usage was conceptualized along a continuum, the job of the authors of the diagnostic manual would be to construct a description and measurement of the disorder that adequately represents each of the levels or degrees to which the disorder appears along this continuum rather than attempt to describe the prototypic case. The DSM-IV-TR criterion set was confined to the most severe cases and was not describing well a large proportion of persons with clinically significant alcohol use dysfunction. As a result, clinicians had to rely on the nondescriptive, wastebasket diagnosis of NOS to describe the lower range of the continuum (Saunders, 2006). A step in the direction of recognizing the continuous nature of substance use disorder was incorporated in DSM-5. Along with the abandonment of the distinction between abuse and dependence, DSM-5 also includes a rating of severity for a substance use disorder, depending upon the number of diagnostic criteria that are met. For example, a “mild” substance use disorder is suggested by the presence of just two to three features (APA, 2013). However, the features for the mildest and the most severe cases are still the same. What would be more informative would be to have the different levels be defined by the features that are relatively specific to that level, analogous to how the comparable distinctions are made between the levels of severity for an intellectual disability.
Mental Disorders as Discrete Clinical Conditions: Dimensional Versus Categorical Classification
Diagnostic Boundaries
Depression
The Continuum of Depression
Subthreshold Major Depression
Alcohol Abuse and Dependence
Personality Disorders
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
