To the Editor: We are concerned by Dennis Gorman's suggestion in the Alcohol & Drug column in the August issue (1) that prevention researchers should not look at broad multiple outcomes from their interventions. We urge Dr. Gorman to read Reducing Adolescent Risk (2). The final chapter summarizes conclusions of the conference of scientists on which the book is based. In it he will find the statement, "It has long been known that risky behaviors covary…. Interventions designed to influence one or more risk behaviors should assess effects on other behaviors."

Dr. Gorman calls this "the adjustable outcome." We disagree. Interventions based on theories of social development and social learning are hypothesized to have effects on broad outcomes. It would be irresponsible for investigators not to collect, analyze, and report data on outcomes of covarying risk behaviors after interventions are made.

Dr. Hawkins is Kozmetsky professor of prevention and director of the social development research group in the School of Social Work at the University of Washington in Seattle. Dr. Catalano is professor and associate director of the group.