Psychiatric classification: Merely repetitive and problematic?

Starcevic has made a series of useful conceptual and empirical contributions to the literature, and his piece on behavioural addictions is another thoughtful paper (Starcevic, 2016). He elegantly describes the emergence of the concept of behavioural addiction, emphasizes that the addiction framework is only one approach and calls for more research on these conditions. I am in agreement with most of his views, although as someone who has slogged away on various revisions of our psychiatric nosology, and with a particular interest in ‘repetitive and problematic behaviours’, I would have hoped that the field not be solely characterized as having ‘ongoing uncertainty, arbitrariness, and ambivalence’.

Indeed, an immediate question is whether this characterization applies not only to conditions characterized by repetitive and problematic behaviours, but also to psychiatric disorders as a whole. Certainly, there has been no recent shortage of critics of the Diagnostic and Statistical Manual of Mental Disorders (DSM) and International Classification of Diseases (ICD) classification systems; clinical neuroscientists emphasize the lack of diagnostic validity, public mental health researchers emphasize the lack of clinical utility, dyed-in-the wool researchers argue for and against various nosological decisions and more radical critics argue that the classifications merely reflect the biases of the pharmaceutical industry and other pernicious Western practices.

Is psychiatric classification itself merely a repetitive and problematic behaviour? van Praag coined the term, ‘nosologomania’ and provided a conceptual framework for biological research on psychiatric disorders that is reminiscent of today’s Research Domain Criteria (RDoC) approach (Van Praag, 2000). Starcevic’s characterization of some work in psychiatric nosology as displaying ‘uncertainty, arbitrariness, and ambivalence’ might even suggest that such efforts bear more than a passing resemblance to the compulsive, impulsive and addictive behaviours that they are attempting to engage with. My aim in this short commentary is to provide some additional context, which I think supports a more forgiving perspective.

A non-essentialist approach

There are some who hold that sooner or later we will be able to carve nature at her joints, and so delineate psychiatric entities in a way that readily distinguishes them from normality and from one another. Those who take this perspective are perhaps more likely to emphasize the importance of necessary and sufficient operational criteria for depicting psychiatric disorders, and to hold that ultimately biosignatures will be useful in psychiatric diagnosis and evaluation. In opposition to such views, a range of critics have emphasized that our constructs of psychopathology are not only theory-bound but also value-laden, and that they differ from time to time and place to place. Such critics may well feel that psychiatric classification is repetitive and problematic, and that other approaches to social deviance and suffering have more value.

My own view is that psychiatric disorders are complex conditions, reflecting multiple cross-level causal mechanisms, and so require a non-essentialist approach to their conceptualization (Stein, 2013). In accordance with this view, any particular categorical classification system will have both pros and cons. This does not, however, mean that psychiatric classification is merely arbitrary; the relevant theories and values can be debated, and progress can be made, with iterative classifications embodying greater diagnostic validity and clinical utility (Stein, 2008). Such progress may not be always be seen in the relatively short time spans of revisions of DSM and ICD, but contemporary classifications can readily be judged to have more diagnostic validity and clinical utility than those from a century ago.

Similarly, when it comes to the behavioural addictions, this nosological construct has both pros and cons. Starcevic clearly and correctly emphasizes many of the cons. At the same time, his essay is of course a response to work that has been done in the area (and in that sense such work has been fertile), and where he might have asked for the construct of behavioural addictions to be entirely scrapped, it is noteworthy that he ends by calling for ‘further studies of behavioural addictions’; an idea that proponents of the construct would undoubtedly agree with. The construct of behavioural addiction may well be useful in understanding aspects of a number of different disorders characterized by repetitive and problematic behaviour, but it is less clear, at least at the current time, how deserving it is of independent nosological status (Grant et al., 2014).

Asking too much of nosology

My own view is that repetitive and problematic behaviours will exist for the foreseeable future (in this sense, psychiatry is different from other fields of medicine, where certain conditions, such as smallpox, can sometimes be completely eradicated). I am also of the view that drawing a line between such behaviours and normal behaviour will always be a contentious exercise, requiring clinical judgment. Contentiousness is particularly likely to be present in areas such as sexual behaviour and aggression (none of us like to be seen as oversexed or over-aggressive). I am fairly convinced that addictions will always be stigmatized compared to more typical conditions (in keeping with the reality that in many ways they are not typical conditions) (Stein, 2013).

Furthermore, it seems clear to me that any particular repetitive behaviour (say working too hard on psychiatric classifications) may reflect a range of different bio-behavioural mechanisms (e.g. academic narcissism and hypomanic energy). I would also acknowledge that at times, nosology swings from one point of view to another, with no apparent progress, particularly over the short haul. In addition, at any particular point in time, nosologists are faced with conflicts that are difficult to resolve (e.g. imaging data may suggest one classification, while genetic data may reflect another). And certainly, as Starcevic alludes to, we need ongoing discussion of the conceptual frameworks we use to make decisions in the face of conflicting data and different points of view (Stein, 2008).

At the same time, I am of the view that we should not ask too much of our psychiatric nosologies; they should guide us but they cannot simply dissolve psychiatry’s core and perennial problems in evaluation and assessment (Stein, 2013). Furthermore, history suggests that we can make progress in psychiatric conceptualizations and classifications, particularly over the long haul. As we learn more about psychiatric conditions, our nosologies embody more and more sophisticated articulations of diagnostic validity and clinical utility. Thus, I would argue that a 21st-century clinician, with a thorough understanding of the advantages and limitations of the construct of behavioural addiction, is well placed to approach his or her patient with repetitive and problematic behaviour, to obtain a comprehensive evaluation and to use this construct as well as a range of other pertinent frameworks to develop an appropriate treatment plan.