A biometric evaluation of the somatotherapies in schizophrenia.

Although the need for sound research on the outcome of mental disease has long been recognized, and although probably no other medical specialty has accumulated such an abundance of statistics as psychiatry, yet the dearth of adequately designed investigations especially in the field of the evaluation of therapy is conspicuous. The literature describing and appraising the results of insulin, metrazol, and electroshock treatment is voluminous. Reports on lobotomy results have also mounted steadily. In spite of these studies the satisfactory evaluation of therapy still remains one of the most troublesome problems of psychiatric research. In reviewing the vast literature of therapeutic results one finds conflicting reports ranging from severe skepticism of the various therapies to inordinate enthusiasm for them. Consequently every clinician can cite a study to support his particular viewpoint on any given therapy. This state of affairs suggests the need for a critical study of the methodology of these evaluation studies with a view towards the improvement of research design in this area in the future.