The Search for Biomedical Referents
As indicated above in our discussion of the Cambridge Model of Symptom Formation, the DSM qualifies mental symptoms as signs that point to underlying disorders. Whereas in the DSM-I (1952) and II (1968) no epistemological statements concerning the precise nature of mental disorders can be found, its use of concepts like “neurosis” or “psychological reaction” was highly criticized in the 1970s. Biomedically inspired psychiatrists believed that these labels bore witness of underlying psychological presumptions and, since the DSM-III (1980), succeeded in having them replaced (Decker 2013). In the introduction to the DSM- III, the authors state that “the approach taken in DSM-III is atheoretical with regard to etiology or pathophysiological process” (DSM-III 1980, p. 7). Most interpretations of this phrase focus on the idea that the DSM is neutral in terms of differences between therapeutic schools of thought, like psychoanalysis or cognitive therapy. However, if read to the letter it is a most remarkable phrase. If the statement had been: “atheoretical with regard to etiology and pathophysiological process,” it would stress theoretical neutrality at both levels. Yet the statement says: “atheoretical with regard to etiology or pathophysiological process.” In other words, the statement seems to equate etiology and pathophysiology and thus implies that pathophysiological processes, first and foremost, make up mental disorders. What it also says is that at the level of specific neuropsychiatric models and theories, the DSM-III assumes no firm position: no claims are made concerning specific neurobiological disturbances associated to disorders. Such a prioritization of biomedical reasoning is not surprising. After all, most task force members of the DSM-III leaned toward a biomedical approach to psychiatry (Decker 2013).
However, in order to “sell” the DSM to the field of psychiatry, strong epistemological claims on the nature of mental disorders were avoided in the DSM-III, as well as in later editions of the manual. With the advent of the DSM-5 this changed somewhat. The DSM-5 continues to hold on to descriptive criteria; however, disorders are presumed to be biomedical realities more than ever before. While this change is obvious in its preparatory documents, it is subtler in the manual itself. For example, with reference to the motivation to change the DSM-IV-TR, the DSM-5 (p. 5) only mentions cognitive neuroscience, brain imaging, epidemiology, and genetics as relevant background disciplines. Indeed, in line with the basic tenets the so-called neo-Kraepelinian movement formulated in the 1970s (see Chap. 2), basic tenets of biological psychiatric thinking (see Berrios and Markova 2002) guided the development of the DSM-5, like the idea that mental disorders are disorders of the brain caused by biological factors, or the idea that only biologically focused quantitative studies provide relevant scientific insight into the nature of mental disorders.
The process of developing the DSM-5 started in 1999. Key figures involved often indicated that they would drop the descriptive approach and gradually switch to a diagnostic system that was based on a causal understanding of disorders. For example, in 2011 the then president of the American Psychiatric Association, Carol A. Bernstein, wrote: “diagnoses in the DSM-III, DSM-III-R, and DSM-IV are best understood as useful placeholders, based on careful description, but not on deeper understanding” (Bernstein 2011). Guiding the revisions was the attitude that the new handbook should be more strongly informed by scientific innovation—an aim no one can object to—and that the “deeper understanding” that decades of psychiatric research provided would lead to a more objective delineation of the disorder categories. However, the only research addressed during discussions of the new scientific basis for the DSM-5 is, unfortunately, biomedical.
What is iconic in this respect is the edited book A Research Agenda for DSM-V. This book intended to be programmatic for all actual revisions: “Those of us who have worked for several decades to improve the reliability of our diagnostic criteria are now searching for new approaches to an understanding of etiological and pathophysiological mechanisms— an understanding that can improve the validity of our diagnoses and the consequent power of our preventive and treatment interventions” (Kupfer et al. 2002, p. xv). The editors, which include DSM-5 chair David Kupfer and vice-chair Darrel Regier, aimed to devise “a research and analytic agenda that would facilitate the integration of findings from research and experience in animal studies, genetics, neuroscience, epidemiology, clinical research, and cross-cultural clinical services—all of which would lead to the eventual development of an etiologically based, scientifically sound classification system.” Indeed, key to the DSM-5 revisions was the intention that groundbreaking biomedical findings with respect to etiology and pathophysiology would finally inform diagnostic decision-making. While the DSM-III and the DSM-IV were still merely based on convention, the DSM-5 task force foresaw a near future in which diagnosis would at last be truly valid. In making decisions about inclusion and exclusion criteria they aimed “to carve nature at its joints” (Regier et al. 2009, p. 648). This ambitious claim, which uses a metaphor from Plato’s Phaedrus, makes clear that in its classification the DSM-5 assumes that mental disorders are natural kinds, meaning irreducible entities that are rooted in natural laws (Bird and Tobin 2010; Keim Campbell et al. 2011). Thus, the laws thought to be operative are purely biomedical and far removed from psychotherapeutic or social psychiatric assumptions about the processes that make up psychopathology (Bracken et al. 2012; Kirshner 2009; Vanheule 2011).
Indeed, in different publications key figures in the development of the DSM-5 voiced the idea that mental disorders should be thought of as neurobiological entities. According to Stein et al. (2010, p. 1760) the term “mental disorder” is actually misleading and could be replaced by “psy?chiatric disorder” “insofar as it emphasizes that these conditions are not purely ‘mental’ and that the line between ‘psychiatric disorder’ and ‘other medical disorders’ is not distinct.” In a collaborative article on the future of psychiatry, the DSM-5 task force chair David Kupfer and colleagues even go a step further and use the term “complex brain disorder” instead of “mental disorder” (Reynolds et al. 2009, my italics). In their view, psychiatry is above all a “clinical neuroscience,” with the help of which “a deeper understanding of causal pathways to major neuropsychiatric illness is evolving, thus rendering artificial the boundary between psychiatry and neurology” (Reynolds et al. 2009, p. 2), which indeed points to biological psychiatric assumptions (Berrios and Markova 2002). More critically, such a line of reasoning could be seen as illustrative of genetic essentialism and neuroessentialism (Dar-Nimrod and Heine 2011).
However, as the publication date of the DSM-5 approached the tone changed somewhat. In a 2011 commentary in the American Journal of Psychiatry the chairs noted the following: “we anticipated that these emerging diagnostic and treatment advances would impact the diagnosis and classification of mental disorders faster than what has actually occurred” (Kupfer and Regier 2011, p. 672). The message they then gave was that biological evidence helped in distinguishing disorders, grouping them into spectra, but did not make up the criteria sets: “While not central to the criteria themselves, this information is nonetheless useful and informative for helping DSM provide a more precise picture of the clinical realities of psychiatric diagnosis” (Kupfer and Regier 2011, p. 672). However, neither in the DSM-5 itself nor in related publications did the DSM task force make clear which neurobiological evidence was actually used in making decisions about disorder criteria. Without such detailed information on how decades of psychiatric research has informed conclusions about diagnostic categories, the claim that advances in research helped provide a “more precise picture of the clinical realities” is rhetorical. Above all, such lines of reasoning bear witness to what Berrios and Markova (2002) call the technology alibi, a rhetorical strategy used since the early nineteenth century in order to support further belief in biomedical illness assumptions:
“As far as we know, the first person to use the technology alibi was Georget (1820) to explain the failure of biological psychiatry to find any sustainable and replicable connections between brain and madness. Then, as now, it was only a ‘matter of months’ before the pathology and genetics of dementia praecox were sorted out; then, as now, the expectation was kept alive by the release of premature data; and then, as now, there was never the scientific honesty of publishing a denial saying that what had been claimed earlier was nonsense” (Berrios and Markova 2002, p. 5).
Despite all rhetoric of grounding the DSM-5 on neurobiological concepts, the manual strongly resembles the previous versions of the handbook and does not read like a rupture with the past: the DSM-5 and the DSM-IV-TR (2000) largely contain the same disorder categories and diagnostic criteria for specific disorders remain mainly the same. This most probably indicates that in the end, empirical research has not been as conclusive as first speculated. After all, the diagnostic categories of the DSM-5 still only contain lists of characteristic symptoms and complaints, which Bernstein (2011) mockingly qualified as “useful placeholders.” In other words, despite decades of biologically inspired rhetoric on scientific innovation in psychiatry, not a single “hard” indicator could be included.
Interestingly, other new ideas that had also been launched with reference to “science” were not granted, like the project to assemble existing disorders into larger clusters and spectra. Whereas at first it was indicated that such grouping was “suggested by the scientific evidence” (Bernstein 2011), in the end it was not retained in the handbook. Scientific (or other) arguments as to why the envisioned change was not retained were not communicated.
The net result is that the diagnostic criteria in the DSM-5 are as descriptive as ever before. Nonetheless, the authors of the DSM-5 make the following claim about these criteria: “they are intended to summarize characteristic syndromes of signs and symptoms that point to an underlying disorder with a characteristic developmental history, biological and environmental risk factors, neuropsychological and physiological correlates, and typical clinical course” (DSM-5, p. 19, my italics). Without validity studies actually proving that the polythetic symptom profiles from the DSM correspond to underlying disorders, the above claim is, indeed, no more than an intention. An intention is something one aims or hopes for, not something that can be taken for granted. Indeed, straightforward references to validity studies that would substantiate disorder categories are lacking in the DSM-5 and related publications.11 This means that validity is as much a problem in the DSM-5 as it was in the previous two editions of the handbook.
Why then all the buzz about the scientific basis of the DSM-5 (e.g., Kupfer et al. 2002)? I conclude that, above all, the DSM-5 officials’ frequent references to the idea of scientific evidence aimed to polish up the credibility of the DSM. The message they seemingly want to sell is that DSM diagnosis should be thought of as rooted in sound research, although this is far from clear. Most probably the objective to find a specific and coherent neurobiological structure underlying specific disorders is far too ambitious, except perhaps for neurocognitive disorders (Frances 2013; Kagan 2012; McNally 2011; Van Os 2016).
-  In Chap. 2 I discuss the fact that this hard work to improve the reliability of psychiatric classification was not particularly successful.
-  Obviously, these reflections gave rise to the appendix on cultural formulation in the DSM-5.
-  In a 19-page document entitled Highlights of Changes from DSM-IV-TR to DSM-5 (http://www.dsm5.org/documents/changes%20from%20dsm-iv-tr%20to%20dsm-5.pdf) all changes arereviewed. Nevertheless, several authors, like Allen Frances (2013), correctly suggest that for severaldisorders inclusion and exclusion criteria were subtly relaxed, thus predicting waves of inflatingdiagnosis. For example, whereas DSM-IV-TR indicates that a diagnosis of ADHD can be madeonly if symptoms are present before the age of seven, the DSM-5 indicates that the diagnosis canbe made if symptoms are present before the age of12. This “small” criterion change actually impliesthat the population eligible for the diagnosis actually doubled.