DSM-5,ICD-11 and ‘pathologization of normal conditions’

ICD Insights

The most frequent criticism directed at the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) is that it goes even further than its predecessors in pathologizing conditions which are in the range of normality (Frances, 2013). This criticism actually consists of two components: (a) some of the conditions included in the manual may not qualify as mental disorders, and (b) the threshold for the diagnosis of some conditions which do qualify as mental disorders may be too low in the manual, so that also normal states are included. Is it possible to prevent this criticism in the development of the International Classification of Diseases, 11th Revision (ICD-11)?

In order to prevent the first component of the criticism, we should ideally be able to define unequivocally what a mental disorder is, or at least we should develop a pragmatic set of inclusion and exclusion criteria to apply explicitly and consistently when a new diagnostic entity is proposed. Unfortunately, we lack both the unequivocal definition and the pragmatic criteria.

The DSM-5 does provide, as does predecessors, a definition of mental disorder (American Psychiatric Association, 2013: 20), but this definition contains an element of tautology (a mental disorder is ‘a syndrome characterized by clinically significant disturbance …’), an element whose real ascertainment is difficult or even impossible at the current state of knowledge (‘a dysfunction in … processes underlying mental functioning’), an element which is present in several mental conditions that are not necessarily mental disorders (causing significant distress or disability), and an element whose assessment is fraught with a variety of possible confounders and which is of doubtful reliability (the syndrome is not ‘an expectable or culturally approved response to a common stressor or loss’). Furthermore, there is no specification anywhere of the level of research evidence (concerning, for instance, prevalence and course) and the degree of diagnostic reliability requested in order to accept a new condition for inclusion in the diagnostic system.

In this situation, it is not surprising that several decisions concerning the inclusion of new diagnostic entities in the DSM-5 are vulnerable to criticism. For instance, disruptive mood dysregulation disorder has been accepted for inclusion despite a ‘questionable’ test-retest reliability in field trials (pooled kappa 0.25, with values as low as 0.06 and 0.11 at two of the three sites) (Regier et al., 2013) and although the research evidence largely emanated from a single American research group, with insufficient information about prevalence in the community (American Psychiatric Association, 2013: 157) and separation from oppositional defiant disorder (Leibenluft et al., 2012: 78). The inclusion was explained by the need to ‘address concerns about potential overdiagnosis and overtreatment of bipolar disorder in children’ (American Psychiatric Association, 2013: 810), but this rationale, perhaps acceptable in the United States, appears questionable from an international perspective. In fact, in most countries of the world, such overdiagnosis and overtreatment do not exist, while the inclusion of the new category in the diagnostic system may itself generate an overdiagnosis of mental disorder and an overuse of medications in children, especially since no validated treatment for the condition is available.

On the other hand, attenuated psychosis syndrome has not been accepted for inclusion in the system, although acknowledging that ‘it is a disorder based on the manifest pathology and impaired function and distress’ (American Psychiatric Association, 2013: 783). This was explained by ‘concerns about reliability in the field trials’ (American Psychiatric Association, 2012). However, the reliability was not actually found to be low in those trials (the kappa was 0.46). The estimate was regarded as inaccurate due to the small sample size (Regier et al., 2013), but this situation was shared by hoarding disorder, which was instead included in the system.

The need to base similar decisions, in the ICD-11, on explicit criteria, applied consistently for the various proposed diagnostic entities, seems obvious.

In order to prevent the second component of the criticism addressed to the DSM-5, we should ideally rely on an unequivocal research evidence supporting our diagnostic thresholds for those mental disorders obviously lying on a continuum with normality, such as depression. Unfortunately, this evidence is not available. For instance, the thresholds fixed for the diagnosis of major depression by the DSM-III and its successors – in terms of number of depressive symptoms, duration of these symptoms and functional impairment – have not been supported by available research (e.g. Kendler and Gardner, 1998). The DSM-5 field trials have not addressed this issue, thus making the system vulnerable to criticism. Furthermore, the threshold for the diagnosis of major depression has been lowered in the DSM-5 by eliminating the bereavement exclusion criterion, a decision not supported by research evidence (Wakefield and First, 2012).

Further studies, formally comparing alternative thresholds for the diagnosis of each mental disorder, especially with respect to their clinical utility, should certainly be encouraged. But meanwhile, should those non-validated thresholds be kept in the name of reliability? Or should we more explicitly acknowledge, in the ICD-11, the role of clinical judgment in evaluating the ‘goodness of fit’ or ‘extent of match’ of each individual case to the prototype of each mental disorder described in the system (Westen, 2012), avoiding pseudo-precise criteria?

These are, of course, conceptual issues that should be addressed at a general level, rather than left to the deliberations of the workgroups in charge of the individual sections of the diagnostic system. Within the process of development of the DSM-5, the establishment of a workgroup for conceptual issues was indeed proposed (Kendler et al., 2008), but this proposal was not implemented. Appointing such a workgroup may represent a useful step, especially if experts in transcultural psychiatry are included and various categories of relevant stakeholders are represented.