Mental Disorders, Collective Observation, and the International Classification of Diseases
Over four decades ago, the distinguished epidemiological psychiatrist Norman Sartorius wrote, “the history of psychiatric classification is in fact a history of psychiatry.”1 During the 1960s and 1970s, Sartorius had been at the center of research by the World Health Organization (WHO) on the international classification and prevalence of mental disorders. During that era, the organization significantly transformed its classificatory manual, the International Statistical Classification of Diseases and Related Health Problems (ICD), releasing the ICD–9 in 1977. The ICD is the standard international manual for recording mortality and morbidity data for insurance and epidemiological purposes. WHO is currently in the final stages of completing its latest update to the text, ICD–11.
Blog Series: Learning by the Book
Join the conversation on Twitter with the hashtag #lbtb18 or on this blog with a post of your own. Tweet or email us links to related discussions. Read more posts in this series, and check out the conference website.
WHO assumed responsibility for producing the ICD upon its founding seventy years ago, the same year it released ICD–6. This was the first edition to include global disease morbidity alongside mortality data, and the first to introduce a detailed descriptive classification for mental disorders, described in Section V of the manual. The ICD is the most important international manual for the coding of death and disorder. It is used by hospitals to record mortality and morbidity data, physicians to bill insurance companies, and statisticians and others undertaking epidemiological research. It is a crucial document in the production of global epidemiology.
Prior to the 1960s, there had been numerous attempts to specifically quantify mental disorders in many parts of the world, but such efforts lacked globally standardized methodologies. Based as they were on different practices in the labeling and collection of mental health data, such efforts could not provide reliable knowledge regarding the global distribution of psychiatric disorders. The inclusion and subsequent expansions of Section V in the ICD followed pioneering work to overcome this lack of standardization in international psychiatric classification. Since the 1960s, Section V of the ICD has been at the center of ongoing efforts to reliably document the global prevalence of mental disorders. Yet the tidy symptom descriptions of Section V masked the much more contentious process of standardizing the perceptions of the professionals using the manual.
The ICD organized the classification and ordering of mental disorders according to their dominant symptomology. Emphasis on patient symptoms was reflected in the numerical organization of the manual. Editions six through nine listed mental disorders in three-digit codes with a possible fourth digit at the first decimal place to distinguish possible subcategories. ICD–9, for example, coded “Schizophrenic psychoses” as 295, listing possible sub-forms at the decimal place, for example, “295.2 Catatonic type.”
In the early Cold War years of the manual’s reorganization, figures within the WHO claimed that this classificatory arrangement was to maintain neutrality in the debates between different schools of psychiatric thought.2 Yet a classification by symptom profile produced its own forms of disagreement.
Between 1965 and the release of ICD–9 in 1977, the WHO held annual weeklong seminars of psychiatrists from around the world, ultimately drawing on experts from thirty-five countries and from both blocs in the Cold War. Physicians from “developing” countries were rare, although Latin American psychiatrists were an exception, their representatives flying to Europe from such countries as Brazil, Colombia, Chile, Argentina, Peru, and Guatemala. Each of these meetings was dedicated to a specific disorder within Section V. It was in these meetings that two new critical instruments in psychiatric practice were developed.
The first was a standardized clinical psychiatric interview that drew on the work of British psychiatrist John Kenneth Wing. The WHO developed Wing’s Present State Examination (PSE) with the aim of taming the idiosyncrasies of the psychiatric interview and dissecting the observation and transcription of symptoms. It did so by dividing observation of the patient’s mood and behavior into a list of items that the interviewer was to move through consecutively, recording the presence or absence of each.
Videotaped patient interviews constituted the second instrument. Film had long been used in psychiatry to present the features of a disorder, but the newly available techniques of video recording made the process of reproducing patient interviews all the easier, facilitating their global circulation. During their week together, psychiatrists would be shown a series of videotaped patient interviews, and they would individually complete a PSE. Disagreements in their findings would then be discussed, appropriate sections of the video re-watched and, ideally, consensus reached on a final description and definition of the disorder. The aim was to create greater uniformity across psychiatric observations by “testing observer variation [and] breaking down diagnostic problems so that variations could be measured.”3
Videotaped interviews and the PSE were both central to the labor of expanding and reorganizing Section V across the 1960s and seventies. Yet even when based on a collective viewing of videotaped interviews, efforts to observe and classify patients according to the tidy classificatory structure of the manual soon demonstrated the difficulty of reaching observational agreements in psychiatry.
At the very first of these meetings, held in London over a week in October 1965, participants were shown nine videos and asked to fill in the PSE according to their observations. Such research aimed to assess the degree of reliability among psychiatrists’ descriptions of standardized videotaped interviews. “While there was substantial agreement on many items,” the report from the meeting noted, “the exercise demonstrated clearly the problem of observer differences in psychiatric diagnosis.” Perhaps most troublingly, it was not merely disorder categories over which physicians disagreed, but the symptoms said to constitute them.
Following observation of one videotaped interview, for example, ten of the psychiatrists noted the presence of “patient arm-swinging,” whereas the other nine marked it absent. In another exercise, seven of seventeen participating clinicians described the patient’s posture as “catatonic.” When the video was re-watched, however, it was found that those seven psychiatrists based their observations of catatonia “on different aspects of behaviour.” In another exercise one-half of the psychiatrists diagnosed the patient as schizophrenic, the other as depressive. While all agreed that the patient in the video manifested delusions, they disagreed “as to the nature of these delusions according to their diagnostic formulation.”4
These observer discussions stimulated efforts to create more detailed descriptions in the ninth edition of the manual, released at the end of a decade of meetings, and contributed to parallel efforts in the American Psychiatric Association’s radical revisions published in the third edition (1980) of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III). On the advice of psychiatrists involved in these meetings, ICD–9 included a descriptive paragraph below each subcategory of disorder, further elaborating its precise features. The code for the paranoid subcategory of schizophrenia (295.3) included the following description:
The form of schizophrenia in which relatively stable delusions, which may be accompanied by hallucinations, dominate the clinical picture. The delusions are frequently of persecution but may take other forms (for example of jealousy, exalted birth, Messianic mission, or bodily change). Hallucinations and erratic behaviour may occur, in some cases conduct is seriously disturbed from the outset, thought disorder may be gross, and affective flattening with fragmentary delusions and hallucinations may develop.5
The discussions and disagreements in these meetings among otherwise internationally dispersed psychiatrists informed the expansion of the descriptions of disorders within Section V of ICD–9. Yet the persistent use of “may” (there are five instances in the three sentences above) and the reliance on descriptive terms such as “relatively” and “gross” underscored the limitations of a descriptive approach to the classification of mental disorders. By collectively observing identical patient interviews, recording their findings on a standardized interview, and discussing divergences in their findings, psychiatrists involved in the renovation of the ICD ultimately achieved a degree of consensus on the description of psychiatric disorders. Yet such agreements were based on a considerable amount of scientific labor among small groups of physicians. Producing consensus among a handful of global specialists viewing identical patient interviews was always going to be a far cry from achieving it at a global level among the diverse users of the manual.
ICD’s Section V nonetheless provided an unprecedented foundation for international standards in psychiatric classification. In addition to expanding the descriptive phenomenology of mental disorders, it has contributed to the production of new instruments and practices for standardizing psychiatric observation, and it has provided global standards in the collection of psychiatric data. Yet the globalization of psychiatric classification and observation will always come up against the essentially incommensurable perceptions of observing physicians. The successes of the ICD themselves demonstrate that reaching observer agreement is as much a question of training as it is of clarifying descriptions. Psychiatrists did not simply see or hear the manifestation of symptoms listed in Section V. They had to learn to recognize them by matching them to the content of collectively discussed videotaped interviews. No matter how tidy and precise the organization of mental disorder codes becomes with the release of ICD–11 later this year, the inescapable differences that exist between observers ensures that complete standardization in matching the expressions of patients’ disorders to ICD codes will always remain unattainable.
David Robertson is a PhD candidate in the Program in the History of Science at Princeton University. His Twitter handle is @davidjr9711
- Norman Sartorius, “Classification: An International Perspective,” Psychiatric Annals 6, no. 8 (1976): 25. ↩︎
- See, for instance, E. Stengel, “Working Paper No. 6: Classification of Mental Disorders,” in Epidemiology of Mental Disorders (Geneva: World Health Organization, 1959), 37. ↩︎
- “First WHO Seminar on Psychiatric Diagnosis, Classification and Statistics, London,” (World Health Organization, 1965), 3. ↩︎
- Ibid., 13. ↩︎
- World Health Organization, International Classification of Diseases: A Manual of the International Statistical Classification of Diseases, Injuries, and Causes of Death, vol. 1 (Geneva: World Health Organization, 1977), 184. ↩︎
OpenEdition suggests that you cite this post as follows:
David Robertson (May 31, 2018). Mental Disorders, Collective Observation, and the International Classification of Diseases. History of Knowledge. Retrieved November 4, 2024 from https://historyofknowledge.hypotheses.org/8106