At the time Kendell wrote The Role of Diagnosis in Psychiatry he had already published an important monograph showing that the separation of depression into neurotic and endogenous types was wrong; they merged. He extended this into all aspects of psychiatric diagnosis, and in the space of a few short chapters in The Role of Diagnosis in Psychiatry he dissected and analysed all the elements of diagnosis that are still the hot topics of today. The elements of diagnosis, the nature of disease and entities in psychiatry, reliability and validity, the process of diagnostic decisions, statistical methods, and the values of categories and dimensions – all were described in sparse accurate prose. Kendell's perspicacity at this time constitutes original thinking, not in the Einstein sense, but in identifying the shallow thinking that was current about diagnosis in psychiatry and adding penetrating new insights.
The two elements of what I think were original thinking at that time were:
• diagnosis is a useful form of communication, it is not a fact
• diagnosis is best considered as a dimension but may be categorised for decision-making.
Kendell did not confine this view to psychiatry. A few years before, there was a vigorous debate very similar to those in psychiatry today, about the classification of hypertension between the giants of the subject, George Pickering and Robert Platt. Platt was convinced that severe or malignant hypertension was a genetically distinct disease; Pickering said blood pressure was a dimensional variable with no separation of hypertensive groups. Pickering was right (Pickering Reference Mayou1960), even though most physicians at the time supported Platt. But this did not mean diagnosis was unimportant. My blood pressure at the age of 18 was the same as it was 50 years later. But because of advances in drug treatment what was originally normotensive became hypertensive, so my diagnosis changed.
The same applies in psychiatry. If we have better ways of treating depression the diagnostic threshold can fall. It should never be fixed. So in psychiatry we should not fall into the Szaszian trap that because there are no laboratory tests or other clinical tests to confirm our diagnoses they do not exist.
Clinical utility
‘Clinical utility’ comprises the two watchwords of the ICD-11 classification of personality disorder due to be published in January 2022. Bob Kendell had it right on the button here, clinical utility was always his focus. A clever classification such as as the Research Domain Criteria (RDoC), which cuts across current diagnostic categories and examines the primary behavioural functions of the brain (Cuthbert Reference Cuthbert and Insel2013), may have traction in some quarters, but it is not clinically useful and so will remain a subject ‘for investigation’ only at present.
The notion of clinical utility first described in Kendell's book was amplified later in an important article co-authored with Assen Jablensky:
‘At present there is little evidence that most contemporary psychiatric diagnoses are valid because they are still defined by syndromes that have not been demonstrated to have natural boundaries. This does not mean, though, that most psychiatric diagnoses are not useful concepts. In fact, many of them are invaluable. But because utility often varies with the context, statements about utility must always be related to context, including who is using the diagnosis, in what circumstances, and for what purposes’ (Kendell Reference Kendell2003, p. 8).
Diagnosis and decision-making
The decision-making chapter in the book is a fascinating one. As was pointed out in an original review, Kendell illustrates the ‘lack of relationship between “diagnostic accuracy” and experience’ (Mayou Reference Kendell and Jablensky1976). The implications of this have been noted by many authorities since, most critically by Allen Frances in his assault on the primitive nature of much DSM thinking, the tick box mentality of the newly fledged psychiatrist who ticks five operational criteria out of nine after a cursory examination and so hits the diagnostic jackpot, from which all then follows without further thought (Frances Reference Frances2013).
Yes, but. In a series of neat videotape experiments, Kendell showed that a core diagnosis is made very quickly, irrespective of the experience of the psychiatrist. Then he asks and answers the key question,
‘If an accurate diagnosis can be generally made after five or 10 minutes what is the point of going on longer? The answer is that, at least in ordinary clinical practice, the so-called diagnostic interview is not concerned with establishing diagnosis in the restricted sense. The interviewer wants to find out what sort of person his patient is, why he has become ill, whether he has been ill before and what treatment he has received’ (Kendell Reference Kendell and Jablensky1975, p. 57).
It is this second line of inquiry that leads to the diagnostic formulation, the fully informed statement about the nature of a clinical problem, not just the skeletal outline. Too often in this age of instant decisions, we go straight for the skeleton.
This leads me to the last of the Kendell aphorisms in the Role of Diagnosis in Psychiatry:
• the easier it is to make a psychiatric diagnosis, the more likely it is to be misused.
That's not a bad series of messages to receive from one book.
Acknowledgements
I thank Ann Kendell and Kate Davidson for their help in the preparation of this review.
Funding
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Declaration of interest
None.
eLetters
No eLetters have been published for this article.