Shells and Pebbles

Lessons from the history of the DSM-III

In 1973, in cities across the United States, eight healthy pseudo-patients walked into unsuspecting psychiatrist’s offices. Following a brief interview, they were all admitted to different psychiatric wards. The instructions the experimenters gave were simple: each pseudo-patient should report a single symptom to psychiatrists, namely that they occasionally heard a disembodied voice say “thud”. As required by the experimenters, all participants stopped pretending to hear this “thud” once they were admitted, yet all of them were kept in the ward for several days. Displaying a brief, isolated symptom was all it took for these otherwise healthy people to become diagnosed schizophrenics.

This experiment, conducted by psychologist David L. Rosenhan and published as the infamous article On Being Sane in Insane Places, is emblematic of a revolution within psychiatry during the second half of the 20th century.1 During this time, psychiatry found itself in a state of crisis, following multiple distressing developments which signaled the need for a paradigmatic shift.

Before the 1970s, psychiatry had not put great emphasis on the importance of diagnoses.2 Mental illnesses and their symptoms were not seen as discretely demarcated categories, but instead as personal reactions to one’s specific life story and the difficulties encountered therein. Consequently, the need for a universal classification of disease was minimal.3 This changed during the 1970s when several studies were conducted that signaled the need for a radical change in practices. Not only did the Rosenhan experiment described above illustrate that psychiatrists could not reliably distinguish between sanity and insanity, but several research projects comparing diagnostic standards across the world showed immense discrepancies in the prevalence and diagnosis of mental illness between countries.4 For example, it seemed that American psychiatrists were diagnosing schizophrenia at much higher rates than their British counterparts. To make matters worse, the diagnostic methods employed by these American psychiatrists failed to show any reliability and were proven to be unsystematic.5 This raised a concerning question: how could mental illnesses be scientific matters of fact if psychiatrists across the globe failed to define and diagnose these disorders in any objective way?

Given this context, it stands to reason that during the 1970s, psychiatry was fighting hard to retain its status as an ‘objective’ science. To combat the crisis within their field, a task force of influential American psychiatrists was assembled to create a new, more objective and scientific methodology for psychiatric diagnosis. The result of this meeting was the creation of the third edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III), a manual that persists as one of the ‘bibles’ of modern psychiatry. In the following, I will illustrate the history of the creation of this manual, and delve deeper into the philosophical foundations underlying the development of the DSM-III. This will then lead me to conclude with some remarks about how the history of the creation of the DSM-III currently still impacts our (mis)-understanding through a discussion of the difficulties surrounding the diagnosis of ADHD in women.  Even though the DSM-III was initially created to make psychiatry a ‘more objective’ science, the epistemological principles behind its creation are now complicit in a different, but still deeply subjective difficulty, namely psychiatry’s blindness to personal and social contexts in diagnosis.

The emergence of the operationalist paradigm or a short history of the DSM-III

To create this new and improved diagnostic manual, a task force was mobilized led by influential psychiatrists from Washington University in St Louis, Missouri. Their mission was to devise a new diagnostic system that would improve the reliability of psychiatric diagnoses, and make global communication between mental-health professionals feasible.6 After the DSM-III’s publication in 1980, mental diseases were characterized by measurable symptoms that pointed toward an objective diagnosis. The DSM-III created a shared language amongst psychiatrists and researchers alike, which legitimized the existence of psychiatric ailments. Diagnosis and classification of disease were at the heart of this new paradigm which increased reliability and facilitated research by introducing standards that made reproducibility possible.7 It seemed like psychiatrists finally had their ultimate diagnostic tool, not unlike the somatic doctor’s X-ray or thermometer.8

How was this new diagnostic tool to be used? The DSM-III provided every mental health professional with a checklist of diagnostic criteria. Under this new paradigm, structured interviews specifically targeting these criteria became the gold standard for psychiatric diagnosis. A patient displaying a sufficient number of symptoms on the checklist could now reliably be diagnosed with the corresponding disorder.9 These checklists were designed to be as atheoretical as possible to avoid explicit commitment to any specific psychological theory while devising the manual. Supposedly, the manual could be used by any mental health practitioner regardless of their psychological school of thought. The criteria given in this new tool aimed to relate solely to objectively observable facts, not tinted by any theoretical preconceptions, and in principle, observable by anyone.10 Furthermore, a glossary was provided to unambiguously define the terms used in the criteria.11 Whereas the DSM-II conformed to the highly personalized approach to psychiatry still deeply influenced by psychotherapy, the DSM-III focused on behaviors and outwardly observable factors.12

An example of one of these checklists can be seen in Figure 1, which shows some of the DSM-III diagnostic criteria for a manic episode.13 The criteria read as a checklist, or an algorithm that can be worked through step-by-step, as opposed to an exploration of a patient’s personality and personal history.

Figure 1: DSM diagnostic checklist for the diagnosis of a schizophrenic disorder

To illustrate the contrast between the DSM-III approach and its predecessors, let us take a look at one of the common criteria for the diagnosis of schizophrenia before the creation of the DSM-III. Since the 1940s, a phenomenon dubbed ‘the praecox feeling’ was considered as one of the core markers of this disease. Psychiatrists admitted that it was difficult for them to indicate why they diagnosed a patient with schizophrenia, but would often point to a particular feeling that the patient would illicit within the clinician during their interactions. This ‘praecox feeling’ was what all schizophrenic patients had in common, and if a psychiatrist felt strangely isolated or uncomfortable after a consultation, this was enough to justify their diagnosis. Contrastingly, the application of such a subjective and vague criterion is completely excluded by the ‘objective’ approach of the DSM-III. Diagnostic criteria now had to be externally observable by any rational observer, making personal feelings elicited within the diagnosing clinician completely irrelevant to any assessment.14

Although the creators of the DSM-III strived to improve the scientific credibility of psychiatry with the exclusion of such subjective criteria as the ‘praecox feeling’, they still warned users against using the manual as their sole diagnostic tool. However, despite these cautionary remarks, the DSM-III became wildly popular, and by the time its successor, the DSM-IV was published, it had become the core of every psychiatric curriculum in the United States.15 A similar development occurred in Europe in 1992, when the International Classification of Disease-10 (ICD-10) was published in which a similar approach to the DSM-IV was taken in its classification of mental illness, cementing this new approach to the classification on a global scale.16

Operationalism as neo-positivism

The shift in diagnostic criteria that occurred within psychiatry during the second half of the 20th century is oftentimes referred to as the “operational revolution”.17 The term ‘operational’ here refers to a philosophical movement inspired by pragmatism that situates the meaning of a concept in the operations necessary to measure said concept.18 For example, instead of referring to a subjective feeling when defining ‘hunger’, an operationalist could refer to ‘the time since last feeding’ as the appropriate operationalist definition. This model of conceptual meaning arguably found its way from philosophy into psychiatry in 1959, through the neopositivist Carl Hempel’s introductory lecture at a conference centering around the issue of the classification of mental disorders. During his address, Hempel argued that if one wants to define a term objectively, say a mental disorder, a standard test should be developed, for example, a standardized set of interview questions. This test should subsequently yield identical results, regardless of who administers it. In the same way that to an operationalist, the term ‘hunger’ would only refer to the measured time since feeding, the definitions of psychiatric illnesses should maximally refer to outwardly observable behaviors over subjectively communicated experiences.19

It is not difficult to see that such an operationalist view of objective definitions found its way into the DSM-III. Of course, as opposed to other types of medicine, psychiatry oftentimes deals with non-observable symptoms. Feelings and beliefs are only directly observable by the patients themselves. However, the operationalist’s commitment to operational definitions of psychiatric categories meant that the diagnostic checklists in the DSM-III favored outwardly observable behaviors over subjectively communicated experiences.20

This concern for safeguarding scientific language from subjective or metaphysical influence is an aspect that further connects the operationalist revolution with not only Hempel’s operationalism but with neo-positivism more generally. Here, Neopositivism (or logical positivism) refers to a movement within philosophy that emerged in the beginning of the 20th century. Its main proponents, which included Hempel himself, were interested in purifying the language of science from any metaphysical or subjective influence, through redefining scientific language in terms of pure observation and logical deduction. Neo-positivists believed that the meaning of scientific terms should ultimately refer back to pure observation sentences, whose content contained nothing more than pure sensory information. These observation sentences would then be completely free of any theoretical charge, seeing as they were the building blocks with which scientific theories were to be built in the first place.

Although Hempel was not explicitly cited by the developers of the DSM-III, in Neopositivism and the DSM psychiatric classification: An epistemological history, Massimiliano Aragona explicitly characterizes diagnostic manuals such as the DSM as being neo-positivist systems.21 Aragona argues that although operationalism is not necessarily a tenet of neo-positivism, the operational revolution can be seen as embodying neo-positivist ideals.22 Proof of this more general neo-positivist influence can easily be found within the DSM-III. For example, in its insistence on the sharp distinction between the observational and theoretical levels of psychiatry and psychology. The creators of the DSM-III claimed that through this manual, clinicians could bypass any theoretical assumptions by relying completely on objective and observable symptoms. The observation of “markedly peculiar behavior” (see figure 1) in a patient could therefore function as an atheoretical observation sentence” from which further logical deductions could be made, such as for example “This patient is schizophrenic”. Therefore, it is fair to say that the DSM-III’s insistence on operational definitions of mental illness, is symptomatic of its more general allegiance to neo-positivist ideals as exemplified through its alleged atheoretical and ‘objective’ nature.

Operationalism as a source of issues within psychiatry today

Although the operationalist approach to psychiatry initially appeared to address some of the issues within psychiatry in the mid-20th century, it has become the target of increasing criticism in the last two decades. Calls for more personalized approaches to diagnosis with less focus on classification are emerging, along with the feeling that there is a need for a type of psychiatry that doesn’t ignore social and personal context.23 In tackling these issues within present-day psychiatry it is useful to keep in mind the history of its foundational text. Even though the DSM has been updated several times throughout the years, its basic principles are still the same ones that were used for the DSM-III.24 If we take the idea of logical positivism as the epistemological basis for the creation of the DSM-III seriously, it should come as no surprise that the manual is badly equipped to offer us a more contextualized and ‘socially conscious’ form of psychiatry.

            To make this idea more concrete let us consider an example that is currently a subject of debate in discussions around neurodivergence, namely the difficulty of diagnosing women with ADHD. One of the key issues here is the fact that many women are overlooked by DSM-driven approaches to diagnosis due to a lower likelihood of them displaying externalized symptoms than their male counterparts. Within the epistemology of the DSM, the psychiatric patient is a completely decontextualized, genderless, classless patient, whose external manifestation of markers and symptoms of mental illness make it possible to classify them in an objective and neat little box. This is however a very poor fit for the diagnosis of a demographic such as ADHD women, who due to societal factors tend to comply with a gender role which makes them less likely to display typical externalized symptoms of ADHD.25 Whereas it might, for example, be seen as normal for little boys to display certain disruptive hyperactive traits within the classroom, a little girl with the same tendency might get more heavily reprimanded for similar behaviors and hence learn from a very young age to effectively ‘mask’ what would otherwise be considered symptoms of ADHD.

Without the context of the history and foundational epistemology behind the DSM, an issue like this could seem to be an unlucky coincidence that should and could be remedied with further research and better criteria within the same paradigm. However, when we keep in mind that the epistemological basis of the manual encourages abstraction from societal factors and intentionally favors externalized symptoms, a quick fix within the confined guidelines of the operationalist epistemology seems much harder to achieve. This is why awareness of the history of psychiatry and its epistemological basis is so important when discussing reforms within mental health.26 If not we risk merely putting a band-aid over a psychic bullet-wound. 

  1. David L. Rosenhan, “On Being Sane in Insane Places,” Science Vol. 179, No. 4070. (1973): 250-254. ↩︎
  2. Rick Mayes and Allan V. Horwitz, “DSM-III and the Revolution in the Classification of Mental Illness,” Journal of the History of the Behavioural Sciences, Vol. 41(3), 249. ↩︎
  3. Ibid., 250. ↩︎
  4. N. C. Andreasen, “DSM and the Death of Phenomenology in America: An Example of Unintended Consequences,” Schizophrenia Bulletin 33, no. 1 ( 2006): 110, https://doi.org/10.1093/schbul/sbl054. ↩︎
  5. Ibidem. ↩︎
  6. Julie Nordgaard, Louis A. Sass, and Josef Parnas, “The Psychiatric Interview: Validity, Structure, and Subjectivity,” European Archives of Psychiatry and Clinical Neuroscience 263, no. 4 (2013): 354, https://doi.org/10.1007/s00406-012-0366-z. ↩︎
  7. Ibid. ↩︎
  8. Julie Nordgaard, Louis A. Sass, and Josef Parnas, “The Psychiatric Interview,” 354. ↩︎
  9. Ibid., 354-355. ↩︎
  10. Massimiliano Aragona, “Neopositivism and the DSM Psychiatric Classification. An Epistemological History. Part 1: Theoretical Comparison,”History of Psychiatry 24, no. 2 (2013): 176, https://doi.org/10.1177/0957154X12450142. ↩︎
  11. N. C. Andreasen, “DSM and the Death of Phenomenology in America,”111. ↩︎
  12. Josef Parnas and Pierre Bovet, “Psychiatry Made Easy: Operation(al)Ism and Some of Its Consequences,” In Philosophical Issues in Psychiatry iii : The Nature and Sources of Historical Change., Vol. 1. Eds. Kendler, K. S., & Parnas, J. (Oxford: Oxford University Press, 2015), 196, https://doi.org/10.1093/med/9780198725978.003.0023 . ↩︎
  13. American Psychiatric Association, “Diagnostic and Statistical Manual of Mental Disorders (Third Edition),” (APA: 1980), 189. ↩︎
  14. Pallagrosi, Mauro and Laura Fonzi, “On the Concept of Praecox Feeling,” psychopathology , 51 (2018) 353-361.  ↩︎
  15. N. C. Andreasen, “DSM and the Death of Phenomenology in America,” 111. ↩︎
  16. Josef Parnas and Pierre Bovet, “Psychiatry Made Easy,” 198 ↩︎
  17. Ibid., 190. ↩︎
  18. Hasok Chang, “Operationalism,” The Stanford Encyclopedia of Philosophy (Fall 2021 Edition), Edward N. Zalta (ed.), https://plato.stanford.edu/archives/fall2021/entries/operationalism. ↩︎
  19. Josef Parnas and Pierre Bovet, “Psychiatry Made Easy,” 194. ↩︎
  20. Josef Parnas and Pierre Bovet, “Psychiatry Made Easy,” 194. ↩︎
  21. Massimiliano Aragona, “Neopositivism and the DSM Psychiatric Classification. An Epistemological History. Part 1: Theoretical Comparison,”History of Psychiatry” 24, no. 2 (2013): 166, https://doi.org/10.1177/0957154X12450142. ↩︎
  22. Ibid., 174. ↩︎
  23. Examples: Stijn Vanheule, Pieter Adriaens, Ariane Bazan, Piet Bracke, Ignaas Devisch, Jean-Louis Feys, Brenda Froyen, et al. “Belgian Superior Health Council Advises against the Use of the DSM Categories.” The Lancet. Psychiatry 6, no. 9 (2019): 726. https://doi.org/10.1016/S2215-0366(19)30284-6; Philip Thomas, “The Limits of Evidence-Based Medicine in Psychiatry,” Philosophy, Psychiatry, & Psychology, Vol. 19, no. 4 (2012), 299. ↩︎
  24. Athanasios Koutsoklenis and Juho Honkasilta, “ADHD in the DSM-5-TR: What Has Changed and What Has Not,” Frontiers in Psychiatry 13 (January 10, 2023): 1-6 https://doi.org/10.3389/fpsyt.2022.1064141. ↩︎
  25. Example: Patricia O. Quinn and Manisha Madhoo, “A Review of Attention-Deficit/Hyperactivity Disorder in Women and Girls: Uncovering This Hidden Diagnosis,” in The Primary Care Companion For CNS Disorders, 2014, https://doi.org/10.4088/PCC.13r01596; ↩︎
  26. Athanasios Koutsoklenis and Juho Honkasilta, “ADHD in the DSM-5-TR: What Has Changed and What Has Not,” Frontiers in Psychiatry 13 (January 10, 2023): 1-6 https://doi.org/10.3389/fpsyt.2022.1064141. ↩︎

Posted

in

by

Tags: