The value of DSM-III for psychotherapy. A feasibility study

Establishing the usefulness of DSM-III for psychotherapy will probably require numerous studies. This paper reports a feasibility study for one kind of investigation: the use of clinical records to survey the DSM-III diagnoses and therapy outcome of 30 patients who had outpatient psychotherapy. We were interested in what range of DSM-III diagnoses was represented by patients in outpatient therapy; whether judgments of diagnosis and outcome could be made reliably from clinical records; how well patients in the therapy fit DSM-III diagnoses; if patients did not fit well, why not? We found that an exploration of the relationship between DSM-III diagnosis and outcome from clinical records is feasible; a relatively wide range (13) of DSM-III diagnoses was represented by the 30 patients; 80% of the patients fit well or moderately well into a DSM-III diagnosis; most of the 20% who did not fit well represent the class of problems of living, which does not mean, however, that their problems were minor or unimportant; in some cases the diagnosis--even when it fit the patient well--did not express the essence of the problem for which he or she was to be treated; there was a small correlation (.19) between the patient's rating on axis V and therapy outcome. The findings are discussed in terms of the pro and con arguments that have been made about DSM-III.