My apologies to those of you that already know these things, but I don’t, and blogging has become my late life way of thinking about things I don’t yet know. I take as my justification these overly quoted lines from Eliot:
We shall not cease from exploration.
And the end of all our exploring.
Will be to arrive where we started.
And know the place for the first time.
And the end of all our exploring.
Will be to arrive where we started.
And know the place for the first time.
In my third year as a resident, I began to develop specific diagnostic criteria for the affective disorders; and in so doing I discussed with Drs. Robins, Sam Guze, and George Winokur the possibility of expanding these criteria to include the major psychiatric disorders. During my fourth year as a chief resident, I subsequently pursued this more vigorously and with my coauthors set up a Tuesday afternoon committee. At that time I reviewed close to 1,000 articles in the then existing literature and distilled this data into proposed criteria for the various disorders that we were working on at the time…
Certainly, it was my idea and initial energy that started this committee to work, but without the astute, competent, and highly informed contributions of the other authors, it would never have been possible to complete the task that was done in 1969-1970…
Certainly, it was my idea and initial energy that started this committee to work, but without the astute, competent, and highly informed contributions of the other authors, it would never have been possible to complete the task that was done in 1969-1970…
John Feighner 1989
Psychiatric Residency programs start and end promptly at the end of June/beginning of July. So John Feighner ended his program around June 30, 1970. The article Diagnostic Criteria for Use in Psychiatric Research [full text] was accepted for publication on April 9, 1971, 10 months later, and it opens with:
Diagnostic criteria for 14 psychiatric illnesses [and for secondary depression] along with the validating evidence for these diagnostic categories comes from workers outside our group as well as from those within; it consists of studies of both outpatients and inpatients, of family studies, and of follow-up studies. These criteria are the most efficient currently available; however, it is expected that the criteria be tested and not be considered a final, closed system. It is expected that the criteria will change as various illnesses are studied by different groups. Such criteria provide a framework for comparison of data gathered in different centers, and serve to promote communication between investigators.This communication presents specific diagnostic criteria for those adult psychiatric illnesses that have been sufficiently validated by precise clinical description, follow-up, and family studies to warrant their use in research as well as in clinical practice. These criteria are not intended as final for any illness. The criteria represent a distillation of our clinical research experience, and of the experiences of others cited in the references.
I’ve apparently been misreading this article. I assumed that the criteria, follow-up, family studies, etc. actually came from the St. Louis Group and Barnes Hospital. I gather they mean that these things came from the literature search, including follow-up, family studies, etc.:
While no psychiatric syndrome has yet been fully validated by a complete series of steps, a great deal of work has been published indicating that substantial validation is possible. This communication is a summary of that work in the form of specific diagnostic criteria. The studies of validation for each illness are cited.
And the discussion of the five phases for diagnostic validity were rhetorical, a conceptual repeat from their earlier paper on Schizophrenia. That leaves this:
In addition, we in this department have carried out a study of inter-rater reliability and validation of reliability with an 18-month follow-up study of 314 psychiatric emergency room patients [to be published] as well as a seven-year follow-up study of 87 psychiatric inpatients [to be published], each of whom was interviewed personally and systematically. There were four different raters in the emergency room study. Agreement ranged from 86% to 95% about diagnosis with diagnostic criteria similar to those outlined in this report. There were two different raters in the inpatient study; reliability between those raters was 92%. In the emergency room study and in the inpatient study, validity, as determined by correctly predicting diagnosis at follow-up by criteria such as those of this report, was 93% and 92%, respectively.
There’s no way either of these follow-up studies could fit into the time frame of Feighner’s creation of these criteria and publication – so they must be pre-existing St. Louis studies. If they were ever published, I can’t find them – nor were they part of Spitzer’s later meta-analysis of reliability. Likewise, the comments "with diagnostic criteria similar to those outlined in this report" and "at follow-up by criteria such as those of this report" would lead me to believe that they related to different criteria sets, antedating Feighner’s compilations. And, as noted in a previous comment, among the criteria in this article, only Schizophrenia includes family history [also from the earlier 1970 paper]. So this article that introduced the concept of diagnosis validated by Clinical Description, Laboratory Studies, Delimitation from Other Disorders, Follow-up Studies, and Family Study was, in fact, a literature review by a resident with assistance from staff containing reliability data derived from pre-existing studies using other criteria and were never published. A rather remarkable story.
Returning to John Feighner’s later narrative [1989] looking back over his residency days:
Also, with the progressive use of lithium and other more specific pharmacological treatments at that time, it seemed imperative to me that we refine our diagnostic criteria to assist us in selecting specific treatments for specific patients and to improve communication between research centers.
There’s nothing wrong with that statement. Who wouldn’t want such things? precise diagnoses keyed to treatments? diagnoses that could be shared among doctors? even other research centers? But it also expresses a very familiar feeling, one felt by any young physician in early encounters with the world of clinical medicine. I remember it well. It came right after "Help!" And it reminded me of another remembrance of things past in The Dictionary of Disorder, an article about Robert Spitzer:
Spitzer first came to the university as a resident and student at the Columbia Center for Psychoanalytic Training and Research, after graduating from N.Y.U. School of Medicine in 1957. He had had a brilliant medical-school career, publishing in professional journals a series of well-received papers about childhood schizophrenia and reading disabilities…
At Columbia Psychoanalytic, however, Spitzer’s career faltered. Psychoanalysis was too abstract, too theoretical, and somehow his patients rarely seemed to improve. “I was always unsure that I was being helpful, and I was uncomfortable with not knowing what to do with their messiness,” he told me. “I don’t think I was uncomfortable listening and empathizing — I just didn’t know what the hell to do.” Spitzer managed to graduate, and secured a position as an instructor in the psychiatry department [he has held some version of the job ever since], but he is a man of tremendous drive and ambition—also a devoted contrarian—and he found teaching intellectually limiting…
I expect any doctor or mental health clinician reading these comments has similar memories from their training days. I have a cascade of my own. The information from the Basic Sciences is stored by subject, but that’s not how clinical medicine operates. People come with stories that translate into signs and symptoms, then gets retrieved as diseases or conditions – a completely different way of thinking than before. And then there’s the question of what to do in response. It’s a shock. I said cascade because each new setting brings its own version: medical student meeting patients for the first time; intern being the doctor in charge for the first time; practitioner all alone for the first time; then for me, starting over with psychiatry, and later psychoanalysis. No matter how smart I got the time before, I was rendered dumb as a post at the next level. What follows is a desperate attempt to construct a grand map of the new place for orientation. Medical training is structured such that there’s plenty of help around, libraries full of books, but everybody has to go through the internal process nonetheless. The first mapping is to master your own feelings of confusion at being a stranger in a land you though would be familiar but wasn’t. The resulting first mapping is almost by definition reductionistic, but orienting and comforting. It allows one to tolerate the chaos enough to begin to create an atlas of atlases that makes you into a clinician down the road. And it’s never complete, because there’s always a case around the corner that takes you to a page in your mind that’s blank.
So it appears that the dreams of our fathers were dreams well known to many of us from earlier times in our own histories. John Feighner, a psychiatry resident in St. Louis, dreamt of a simple diagnostic system that would tell him which treatment to use for what patient – finding that system in the writings of others. Robert Spitzer, a psychiatrist who had fled the messiness and confusion of clinical psychoanalysis for the precision of Biometrics had a similar dream – a diagnostic system that tidied up the chaos and told him “what the hell to do.” He found his in the work of neophyte John Feighner. While there’s something almost quaint and innocent in this pairing ["out of the mouth of babes…"], one would think that the profession as a whole could use something with more nuance and experiential depth than the first pass, second-hand system of a trainee co-opted by a escaping clinician – in both cases relying on reported rather than personally observed validity.
I received an email from psychiatric times regarding the highlights of the 2012 APA conference. These were about opiate dependence, professional ‘burnout’ and Trichotillomania. Not one mention of the DSM field trials. Oh well. Not significant then…
How did Feighner attain such high reliability (86 to 92%) when such results have never been achieved since? I mean nothing later even comes close. I don’t trust the figures he reported.
It is remarkable. At first I thought maybe he had a smaller cohort to compare, but they limited the number of diagnoses in the more recent studies. I guess it was a time of giants…
The buckets were so large the toss couldn’t miss.