Research criteria for the diagnosis of depression

Four sets of research criteria for depression are compared: the Feighner, Spitzer, Texas Actuarial, and Composite methods. Of 80 carefully selected depressed patients, 68% satisfied Feighner's criteria, 71% Spitzer's, 75% the Texas Actuarial, and 94% the Composite criteria. Not only did the criteria sets select different proportions of the total, but they also selected substantially different patients, suggesting that results obtained from studies using one set of criteria for diagnosis or selection might not be generalizable to other groups of "depressed" patients. The authors recommend further studies, comparing relative validity as well as reliability, of alternative research diagnostic criteria.