second edition of the DSM (DSM-II) was published in 1968 (APA, 1968) and was designed to be consistent with the new ICD-8 (WHO, 1968), the development of which had been heavily influenced by US psychiatrists. Both used the term ‘reaction’ much less, and anxiety and depression were categorised as neuroses rather than ‘reactions’. By the end of the 1950s most psychiatrists worked in outpatient clinics and private practice (Cromby, Harper and Reavey, 2013), so DSM-II covered a broader range of problems than the first edition, including a new section specifically devoted to ‘behaviour disorders of childhood and adolescence’. Although some categories were still influenced by causal theories, the foreword to DSM-II noted that, where there was some controversy about the nature or cause of a disorder, ‘the Committee has attempted to select terms which it thought would least bind the judgment of the user’ (APA, 1968, p. viii).
Neurosis A psychiatric term that refers to a psychological state that causes distress but is not characterised by being out of touch with reality. Depression and anxiety are common examples.
In summary, the DSM arose out of the need to collect statistics on the prevalence and demographics of various disorders, and to develop a classification system consistent with the ICD for use across the US. The system was influenced by ideas about the causes of certain problems and by the populations with which it was to be used.
2 Challenges to psychiatry’s legitimacy: the road to DSM-III
As discussed in Chapters 1 and 2, the 1960s and 1970s saw psychiatry face a number of challenges to its legitimacy, both within its own ranks (e.g. psychiatrists such as Thomas Szasz and R.D. Laing) and from scholars and activists in civil rights movements. Questions were being raised about links between mental health and social conditions, the notion of mental illness itself and its relationship with restrictive social norms. Some researchers took up the challenge to examine the scientific basis of psychiatry, chiefly the validity and reliability of psychiatric diagnosis.
Validity The utility of a diagnostic system (does it do what it intends to do?); the degree to which the classification system provides a way of conceptualising problems that corresponds to service-users’ experiences and provides a means of accurately classifying them.
Reliability The degree to which clinicians agree on a diagnosis for a service user.
2.1 Empirical challenges: the validity and reliability of diagnosis
Psychologists sought to apply to diagnosis the criteria they used when evaluating the validity of psychometric tests. For example, did diagnosis predict outcomes? Were interview-based diagnoses corroborated by other methods? In a review of research relevant to the different types of validity, psychologist Joseph Zubin noted that ‘just now, diagnosis is at its lowest ebb’ (Zubin, 1967, p. 395). He took a middle position between those who saw the diagnostic system as excellent, requiring only better training of diagnostic interviewers, and those who thought that diagnosis was not ‘possible or even desirable’ (Zubin, 1967, p. 395). Instead, he concluded that the system simply represented a ‘good starting point from which to improve approaches to classification’ (Zubin, 1967, p. 395) and he recommended the adoption of a dimensional rather than categorical approach and more reliable interview methods.
Psychologists Frederick Thorne and Peter Nathan (1969) reviewed studies that had compared patients’ DSM-II diagnoses with their responses to a 100-item symptom checklist administered by a psychiatrist. They found that the distribution of symptoms generally did ‘not conform to patterns postulated in the official classification system’ since they were ‘distributed across the whole range of disorders in mixed patterns’ (Thorne and Nathan, 1969, p. 382). Only ‘functional psychosis’ was clearly differentiated, while the symptoms of ‘psychoneurosis’ were found in all diagnostic categories and ‘personality disorder’ was ‘essentially undifferentiable’ from other categories (Thorne and Nathan, 1969, p. 382). Moreover, some of the supposedly classical symptoms of categories were so rare or occurred in such mixed patterns that they had ‘little or no diagnostic predictive power’ (Thorne and Nathan, 1969, p. 382).
However, not only did the DSM categories lack validity, other studies suggested that psychiatrists could not use them reliably. In a joint US–UK study of the psychiatric diagnosis of schizophrenia, Kendell et al. (1971) found that US psychiatrists had a much broader construct of schizophrenia than their UK colleagues, and so were much more likely to diagnose it. Spitzer and Fleiss’s review of studies of diagnostic reliability acknowledged the ‘obvious unreliability of psychiatric diagnosis’ (Spitzer and Fleiss, 1974, p. 344) and argued that reliability could be improved through the use of detailed, specific criteria and structured diagnostic interview schedules.
2.2 Campaigning and diagnosis: the fall of one category and the rise of another
The 1970s saw two different campaigns by activist groups: one that aimed to remove homosexuality from the DSM, and another that aimed to introduce war trauma.
The DSM and homosexuality
Homosexuality had long been regarded as pathological within psychoanalysis and psychiatry, and it had been included as a ‘sexual deviation’ in the personality disorder category in both DSM-I and DSM-II. Many psychologists and behaviour therapists used sexual aversion therapy with gay men, which involved the use of electric shocks (Cromby, Harper and Reavey, 2013). However, the gay rights movement gathered momentum following the 1969 Stonewall riots in New York against homophobic policing, after which lesbian and gay activists disrupted several APA conventions. Kutchins and Kirk (1999) reported that Robert Spitzer, a psychiatrist and professor of psychiatry at Columbia University in New York, met gay psychiatrists and apparently became convinced of the need to change the classification, though he proposed retaining a category for those who were unhappy with their homosexuality. Following further discussions within the APA, in 1973 its board of trustees voted to remove homosexuality from DSM-II but also to introduce a new category of ‘sexual orientation disturbance’ for those unhappy with their sexuality. Attempts by psychoanalysts and other psychiatrists to overturn the decision failed when they lost an APA referendum in 1974. DSM-III relabelled the category as ‘ego-dystonic homosexuality’. Apart from the obvious issue that such ‘disturbances’ of sexuality were only being applied to homosexuals, this category also ignored the impact of widespread prejudice and discrimination. While ‘ego-dystonic homosexuality’ was dropped, the category of ‘sexual disorder not specified’ (including ongoing and significant distress about one's sexual orientation) was included until the publication of DSM-5.
Vietnam veterans, trauma and the DSM
Following the Vietnam war, many campaigners and sympathetic psychiatrists felt that veterans were not being adequately diagnosed or treated for the psychological effects of combat, particularly distress resulting from a war that had been the subject of so much debate in the US. A new diagnostic category would be a means to provide the treatment needed, and to secure this treatment from the Veterans Administration in the US. Once the process to produce DSM-III had begun in 1974, psychiatrists and Vietnam veteran activists lobbied Spitzer (who was leading the effort to develop DSM-III) to introduce a category that might adequately address the needs of these veterans. As a result, Spitzer set up the Committee on Reactive Disorders, and the new category of ‘post-traumatic stress disorder’ (PTSD) was included in DSM-III (Kutchins and Kirk, 1999).
Pause for reflection
Why
