« PoprzedniaDalej »
AN ANALYSIS OF THE ACCURACY OF PSYCHO
PATHIC HOSPITAL DIAGNOSES.*
BY LAWSON GENTRY LOWREY, A. M., M.D., Instructor in Neuropathology and in Psychiatry, Harvard Medical School;
Temporary Investigator of Brain Syphilis, Massachusetts Commission on Mental Diseases; Chief Medical Officer, Psychopathic Department, Boston State Hospital. Studies of the outcome in psychiatric cases are obviously very important, especially where, as is true at the Psychopathic, diagnoses are based upon symptoms and the longitudinal section of the patient's life before admission. Only in a few cases do we have opportunity to see the final outcome and so check the diagnosis ourselves. The Kraepelinian conception of mental disease, to which we attempt to adhere, was largely founded upon a study of the terminus of pathological states. Accordingly, one important zone of psychiatric advance, for us at least, lies in a study of the outcome of the cases we see here.
The best method available to us for doing this is to follow those cases (about 60 per cent of our admissions) which are committed to the state institutions. This gives us information concerning outcome and also gives us an opinion independently formed; often, as can easily be shown, with diagnostic standards quite different from our own.
Accordingly, we have devised a follow-up scheme by which we secure from each institution its diagnosis and a brief note concerning the condition of the patient three months, six months and one year after commitment. Of course, in the organic cases and those already demented when seen by us, one note is usually sufficient unless there be disagreement in diagnosis. But in the active, acute cases, it is best to secure the full series. In this way conflicting diagnoses in the same case are sometimes given by the institution or institutions.
* A contribution from the Psychopathic Hospital, series of 1918. Read at the seventy-fourth annual meeting of the American Medico-Psychological Association, Chicago, June 4-7, 1918.
Advance in psychiatry can only come, as I have pointed out elsewhere,' if we carefully study the whole patient; make correct symptomatic diagnoses and then check such diagnoses against outcome. Furthermore, the real test of our diagnostic skill lies in applying such a rigorous system of inquiry regarding the further history of our patients. An additional value to such follow-up studies is that they reveal errors in working technique which need modification. They keep the workers keyed up to do the best possible work and tend to establish the habit of careful analysis. Furthermore, they show which groups of cases are most difficult of diagnosis; tend to establish causes for errors; lead to a wider co-operation and understanding between institutions; lead to more uniform standards of diagnoses. The application of uniform standards of diagnoses is really of much greater value than the selection of a uniform statistical grouping for patients.
Two previous studies of Psychopathic Hospital diagnoses have appeared. In 1914 Southard and Stearns published a report dealing with the accuracy of Psychopathic Hospital Diagnoses in 1913. The study was carried out by following the patients committed from the Psychopathic Hospital to other state institutions and ascertaining the diagnosis of the institution to which the patient was sent. They found that about one case in five got no diagnosis at the Psychopathic, and that of those cases that had received a diagnosis, one in four had the diagnosis altered in the next state hospital. They found that a residuum of about 6 per cent remained unclassified. They considered that the most difficult field of diagnosis was shown to be that of dementia præcox and manic-depressive psychoses, and offered some abstracts of the more interesting individual patients. They were struck by the few changes made in the Psychopathic diagnosis of manic-depressive.
Recently the writer' has published a paper dealing with the accuracy of early diagnoses within the Psychopathic Hospital. This was done by checking the diagnosis in the admission office against the rounds, or staff meeting, or discharge diagnosis in the same patient. Of course this represents the checking of one diagnostic standard against itself and not against another standard, as is obtained when the Psychopathic Hospital diagnosis is checked against the diagnosis of some other state institution. It is really a study in the accuracy of snap diagnoses in psychiatry, and it was shown that a high percentage of early diagnostic accuracy depends upon accurate observation, careful interpretation and sufficient information.
In this paper, data are presented dealing with the diagnoses in 419 cases committed to some state hospital, after a residence in the Psychopathic Hospital for from a few days to a month or more. The patients forming this group were committed during the period from November 1, 1916, to June 1, 1917, and were
reported on by the other institutions once, twice or three times. All cases with change in diagnosis were reported on at least twice in order to make sure whether the institution would change its diagnosis. So the group has been followed for from a year to a year and a half. I wish here to express our thanks to the superintendents of the various state hospitals who, by their reports, made this study possible.
Of the 419 patients, 23, or 5.5 per cent, received no definite diagnosis at the Psychopathic, i. e., were left “ unclassified ”; leaving 396 patients receiving a definite diagnosis.
Table I shows for each institution the number of patients sent; the number in which the diagnosis agreed and the number in which the diagnosis disagreed with the Psychopathic. The table shows that in 91, or 23 per cent, the diagnosis was changed, leaving 305 in which the diagnoses agreed. This figure is very near that found by Southard and Stearns.
In Table II are found by diagnosis and by institution the agreements in diagnosis so that an idea may be had of the type of cases sent to each institution.
1 One case first called manic-depressive, but on second inquiry the diagnosis agrees.
One case first called presenile dementia, but on second inquiry the diagnosis agrees. 3 Two cases first called dementia præcox, but on second inquiry the diagnosis agrees. *One case first called dementia præcox, but on second inquiry the diagnosis agrees. 6 One case first called dementia præcox, but on second inquiry the diagnosis agrees. One case first called psychopathic, but on second inquiry the diagnosis agrees.
Four cases first called manic-depressive, and one infection-exhaustion psychosis, all eventually called dementia præcox.
Table III presents all of the data concerning diagnosis in readily accessible form. This shows, for each psychopathic diagnosis, the diagnoses made at the other institutions. The interesting features will be pointed out in the discussion below.
Table IV is a summary in which the results in all psychopathic groups having more than 10 cases are brought together. It will be seen that the error in dementia præcox is low, while the error in manic-depressive is high; this in contrast to the findings of Southard and Stearns. In certain smaller groups the error is