Efforts to categorize and describe “madness” date back to ancient Greece; however, interest in psychiatric categorization has intensified since the turn of the twentieth century (Kirk & Kutchins, 1992). One marked event in this history was the release of the American Psychiatric Association's (APA) third revision of the Diagnostic and Statistical Manual (DSM-III) in 1980. This manual and its successive revisions have been based on a biomedical formulation of “mental disorders,” positioning them as equivalent to medical conditions (Kleinman, 1988; Wilson, 1993). This approach to understanding and categorizing distress dominates mental health practice and has infiltrated popular discourse. Despite its dominance, the DSM has been subject to extensive critique, charged with pathologizing everyday experiences, medicalizing distress, and acting as a tool for social control and a means of maintaining professional and corporate interests.
The DSM is one of the dominant diagnostic systems in the world for classifying “mental disorders” (alongside the World Health Organization's International Classification of Diseases (ICD)) and is widely adopted by mental health professionals. Some examples of diagnoses outlined in the DSM include major depressive disorder, attention-deficit/hyperactivity disorder, schizophrenia, specific phobia, and anorexia nervosa. Each psychiatric diagnosis is accompanied by a list of diagnostic criteria, as well as descriptive information such as prevalence and course. The DSM does not include information on treatment.
Diagnosis; disorder; psychiatry; anti-psychiatry; mental illness; classification; madness
The development of the DSM grew out of a need in the United States to gather statistical information on psychiatric patients (American Psychiatric Association [APA], 2000). Thus, it emerged as a means of collecting census data rather than a tool for clinical work. Multiple approaches to classification emerged in the early twentieth century, variously emphasizing etiology and phenomenology and defining characteristics (APA, 2000). In 1949, for the first time, the ICD (6th edition) included a section on “mental disorders.” Three years later, the APA published its own version, which was released as the first edition of the DSM (APA, 2000).
DSM-I (1952). The first edition of the DSM spanned 130 pages and contained a total of 106 diagnostic categories, each of which was briefly described. Influenced by the prevailing psychoanalytic and psychodynamic approaches of the time, diagnoses were framed as “reactions” (e.g., “schizophrenic reaction”).
DSM-II (1968). At 134 pages, the DSM-II was negligibly longer than the first but contained a significantly expanded set of diagnoses (182 in total). Although the term “reaction” was eliminated in an effort to avoid suggestions of etiology, it continued to reflect a Freudian perspective.
DSM-III (1980). The two previous editions of the DSM had been critiqued for their alignment with psychoanalytic and psychodynamic approaches and for the vague and ambiguous way in which diagnoses were defined. As a result, inter-rater reliability was poor such that the same individual could be given different diagnoses by different psychiatrists. The APA aimed to rectify these problems with the third edition by being atheoretical with respect to etiology and introducing explicit diagnostic criteria (APA, 2000). This extensive revision (with 265 diagnostic categories across 494 pages) involved a dramatic move away from Freudian formulations and toward a scientific/biomedical approach based on Kraepelin's work (Bentall, 2004; Shorter, 1997). Indeed, the “DSM-III is commonly declared to be the most significant factor in promoting what has been called the ‘remedicalization’ of American Psychiatry” (Wilson, 1993, p. 399). In repositioning the manual in line with medical science, the DSM was “transformed from an obscure desk reference – a peripheral clinical tool – into an omnipresent, huge compendium” with broad reaching influence (Kirk & Kutchins, 1992, p. 6).
DMS-III-R (1987). According to the American Psychiatric Association (2000), a revision of the DSM-III was released in order to correct inconsistencies in the system and clarify a number of diagnostic criteria. Largely consistent with the previous edition, it contained 292 diagnoses and spanned 567 pages.
DSM-IV (1994). The fourth edition maintained the theoretical and structural orientation found in the DSM-III but involved an expanded number of diagnostic categories. It contained 297 diagnoses (about three times the number found in DSM-I) across 886 pages.
DSM-IV-TR (2000). A “text revision” of the DSM-IV was released that included no substantive changes or new diagnoses. Rather, the central revisions involved updated information such as descriptions of prevalence, associated features, and culture, age, and gender characteristics. This edition spanned 943 pages, more than seven times the length of the first edition.
DSM-5 (2013). The fifth edition of the DSM involved extensive revisions, including a revised organizational structure, an expanded set of diagnostic categories, and the introduction of some dimentional assessment (APA, 2013). At 947 pages, it is the longest edition of the DSM to date.
Two issues recur in mainstream debates about the DSM. The first is whether “mental disorders” are best understood as categorical or continuous entities. The second revolves around the reliability and validity of diagnostic classification.
The DSM represents a categorical approach to classification whereby a person is deemed to either have or not have a disorder. In contrast, a dimensional system is based on the quantification of attributes along a continuum (e.g., rating depressed mood on a five-point scale). A central debate in mainstream circles has been whether “mental illnesses” are best understood as discrete categories (differences in kind) or in terms of continuous dimensions whereby health and pathology exist at opposite ends of a spectrum (differences in degree). While noting that a dimensional approach is associated with increased reliability and enhanced clinical communication, the APA has traditionally adopted a categorical approach, indicating that it is more “familiar and vivid” and useful in clinical practice and research (2000, p. xxxii). However, the APA is increasingly moving toward the incorporation of dimensional approaches. For example, in the DSM-5, autistic disorder, Asperger's disorder, and pervasive developmental disorder were controversially consolidated into autism spectrum disorder, and a hybrid dimensional-categorical model has been proposed for the personality disorders (APA, 2013).
Beginning with the DSM-III, a central concern of the APA has been improving the diagnostic reliability of the DSM. The poor inter-rater reliability of the first two editions of the DSM sparked questions about the validity of psychiatry as a whole, fanned by the publicity of the Rosenhan (1973) study in which psychiatrists failed to distinguish confederates from psychiatric patients. Thus, the release of the DSM-III has been described as a means to reshape the nomenclature as well as the status of the discipline (Kirk & Kutchins, 1992). The APA has since been devoted to the project of improving diagnostic reliability, involving the introduction of specified diagnostic criteria and the use of field trial research. However, the degree to which these efforts have been successful has been disputed. In a thorough reanalysis of the methodology and data of the DSM-III, Kirk and Kutchins (1992) concluded that the APA's claims of success “in resolving the reliability problem were flawed, incompletely reported, and inconsistent” (p. 15). Moreover, they argued that the intense focus on reliability as a technical problem that could be addressed through research and increasingly rigid diagnostic criteria served to distract from the deeper issue of validity and the question “Are the experiences described in the DSM really ‘mental disorders’?” After all, a system may be entirely reliable and entirely invalid. Thus, they concluded that “the DSM revolution in reliability has been a revolution in rhetoric, not in reality” (Kutchins & Kirk, 1997, p. 53).
Within mainstream debates, issues of validity have been approached in terms of attempting to identify the correct parameters around diagnostic categories. Two issues have figured prominently in discussions of threats to diagnostic validity. First, comorbidity (the presence of two or more disorders) is the rule rather than the exception (APA, 2012). That is, it is common for individuals to meet criteria for multiple diagnoses. Second, it is common practice to employ “not otherwise specified” criteria, essentially providing a diagnosis in the absence of the required diagnostic criteria (APA, 2012). Thus, the clinical reality is that people's experiences of distress and challenge do not appear to fit neatly into individual diagnostic categories, but rather overlap and fall between them. Efforts to address these problems have focused on refining diagnostic categories and criteria based on research and expert consensus. In contrast to such rather superficial treatments of diagnostic validity, critical scholars have questioned the fundamental validity of diagnostic categorization and the very concept of “mental disorders” (Bentall, 2004; Szasz, 1962).
Although the DSM remains the dominant system for articulating and understanding distress, it has been roundly critiqued by consumers of the mental health system, practitioners, scholars, and activists. The following represent some of their key arguments:
In his classic text, The Myth of Mental Illness, Thomas Szasz (1962) argued that “mental illness” is not the product of the objective observation of bodily signs, but rather of social judgement. That is, “mental illnesses” are not “things” in the same way as infections or fractures, but necessarily involve the moral, ethical, and social evaluations of people's behavior. While Szasz did not deny that people experience significant distress or behave in problematic ways, he argued that these did not reflect “diseases” but “problems of living.”
In framing suffering in terms of “disorders,” the DSM locates problems within individuals, thereby eclipsing the interpersonal, social, and political contexts that reliably give rise to people's suffering, including poverty, violence, discrimination, and oppression (Belle & Doucet, 2003; Caplan & Cosgrove, 2004; Marecek & Hare-Mustin, 2009; Mezzich et al., 1999). As a result, solutions are limited to changing the individual at the cost of redressing social injustices including sexism and racism. Accordingly, mental health practitioners have been charged with helping people adapt to oppressive social conditions (Marecek & Hare-Mustin, 2009). In keeping with this argument, feminist scholars have been particularly critical of the DSM, charging it with bias against women, medicalizing and depoliticizing their misery, regulating their behavior, and constructing women's suffering as individual pathology rather than a response to systemic misogyny and injustice (Caplan, 1995; Caplan & Cosgrove, 2004; Marecek & Hare-Mustin, 2009).
In contrast to the American Psychiatric Association's positioning of the DSM as an atheoretical document, it has been argued that psychiatry and its “bible” are neither neutral nor value-free (Caplan, 1995; Kirk & Kutchins, 1992). Instead, it is a product of the time and place in which it was constructed and reflects the interests and worldviews of the authors, who have been predominantly white, male, and American psychiatrists with ties to pharmaceutical companies (Caplan, 1995; Cosgrove, Bursztajn, & Krimsky, 2009). Thus, the DSM can be understood as a means of “social control” (Szasz, 1962), whereby those in power maintain the status quo by deciding what is (and is not) socially acceptable. For example, homosexuality appeared in earlier editions of the DSM, thereby officially pathologizing same-sex desire as “mental illness” (Kutchins & Kirk, 1997; Shorter, 1997). The APA's referendum and subsequent decision in 1973 to strike homosexuality from the DSM reflected the changing social climate, as well as political pressure from gay rights activists (Kutchins & Kirk, 1997; Shorter, 1997). The homosexuality controversy highlighted the political and subjective nature of psychiatric diagnoses and their role in the social construction of deviance (Wilson, 1993).
It has been argued that in positioning the DSM as a scientific document, and “mental disorders” as equivalent to medical disorders, the APA serves to reduce psychiatry's marginalization and defend it as a legitimate medical specialization (Kirk & Kutchins, 1992; Wilson, 1993). Further, the DSM and its ever-expanding list of diagnoses serves the interests of the pharmaceutical industry by shaping which experiences are ascribed the status of “mental illness” (Kutchins & Kirk, 1997). Notably, it has been observed that almost 70 % of the task force members for the DSM-5 are associated with the pharmaceutical industry (an increase of 20 % over the DSM-IV) (Cosgrove et al., 2009). While the APA has attempted to address this conflict of interest with transparency measures, these have been deemed insufficient to address this bias and restore public trust (Cosgrove et al. 2009).
The number of diagnoses has risen dramatically since the first edition of the DSM, resulting in an ever-expanding “bracket creep” whereby everyday experiences are increasingly labeled as forms of pathology (Kutchins & Kirk, 1997; Shorter, 1997). Accordingly, it is increasingly “normal” to be “abnormal,” leading to fundamental questions regarding the validity of psychiatric diagnosis (Caplan, 1995). Moreover, in widening the scope of what is regarded as “disordered,” the mental health professions extend their reach, while the pharmaceutical industry benefits from the rising number of diagnoses in need of treatment (Kutchins & Kirk, 1997).
Far from being a benign and merely descriptive document, the DSM has very real and potentially dangerous consequences for those diagnosed (Caplan & Cosgrove, 2004). The application of psychiatric diagnoses often results in individual shame as well as social and economic costs such as discrimination in legal proceedings for child custody and prejudiced insurance practices. Moreover, these implications are particularly problematic for those on the margins of society, who are most likely to receive more serious diagnoses (Caplan & Cosgrove, 2004; Kutchins & Kirk, 1997).
A central critique of the DSM is that it implicitly promotes a Western understanding of the self and suffering (Caplan & Cosgrove, 2004; Kleinman, 1988; Marecek & Hare-Mustin, 2009). “Mental disorders” are presented as essential entities of the universal human that vary only superficially along culture, age, and gender lines (Mezzich et al., 1999). Reflecting this assumption, the DSM is organized such that disorders are first defined with diagnostic criteria, and then qualified through a listing of specific culture- and gender-related issues that describe variations in presentation. Further, a separate “Glossary of Cultural Concepts of Distress” is presented in the appendix of the DSM, suggesting that the disorders in the main text are “culture-free” (Mezzich et al., 1999). This approach has been critiqued as ignoring the interpenetration of language, culture, and experience and reducing the expressions of distress in cultures outside of white, middle class America to psychiatric exotica (Mezzich et al., 1999). In exporting these ideas, the DSM contributes to the homogenization of expressions and understandings of distress, while its guise of scientific neutrality conceals its ethnocentrism. As such, the DSM can function as a form of cultural imperialism (Bentall, 2004; Marecek & Hare-Mustin, 2009; Watters, 2010).
The adoption of DSM diagnoses can invite an individualized, depoliticized, and pathologized understanding of distress and a limited scope within which to formulate solutions. Given the multiple individual, ethical, and political problems inherent in psychiatric diagnoses, critical scholars have largely eschewed DSM labels in favor of more phenomenological and contextualized framings of people's problems and experiences (Bentall, 2004; Caplan & Cosgrove, 2004). Many have called for a radical shift away from individualistic understandings and toward more political approaches to address human misery (Caplan & Cosgrove 2004; Marecek & Hare-Mustin, 2009). For example, Caplan and Cosgrove (2004) suggest labeling problems using descriptors like “the consequences of poverty” or “the damage done by interpersonal discrimination/demeaning treatment.” In order to help individuals in distress, therefore, change must be social and political as well as individual (Caplan, 1995; Marecek & Hare-Mustin, 2009). It is essential, then, that service provision (including insurance policies) be decoupled from the requirement of a DSM diagnosis.
The APA appears committed to the pursuit of a biomedical framework for “mental disorders,” with focused attention on scientific advancements in molecular genetics, neuroscience, and cognitive and behavioral science (APA, 2012). Meanwhile, critical approaches underscore how all knowledge reflects power and is a product of historical and political circumstance. As such, these contrasting intellectual traditions appear to function as two solitudes, with the latter having had rather limited effect on the former. The challenge for critical scholars and activists will be to exert greater influence in professional and popular formulations of distress.
The fifth edition of the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders was released in May 2013. The main
The American Psychiatric Association (APA) provides an official definition of “mental disorder” in the DSM as: “A clinically significant behavioral
The Diagnostic and Statistical Manual of Mental Disorders (DSM) classification system for mental disorders, developed by the American Psychiatri