[Skip to Navigation]
December 1980

DSM-III vs DSM-II: How Much More Reliable?

Author Affiliations

Department of Psychiatry University of Iowa Iowa City, IA 52242

Arch Gen Psychiatry. 1980;37(12):1426. doi:10.1001/archpsyc.1980.01780250112016

To the Editor.—  Cantwell and his colleagues presented a series of four articles1-4 comparing DSM-II and DSM-III in the diagnosis of childhood psychiatric disorders. Summarizing the results, they conclude that "DSMIII, with additional refinement, gives promise of being a very usable and reliable classification system for the diagnosis of psychiatric disorders in children and adolescents."4(p1228) However, whether one agrees with this conclusion appears to be a case of whether one sees the glass as being half full or half empty. Specifically, it is difficult to agree with the authors' conclusions that the interrater reliability of DSM-III, as well as the raters' agreement with the expected diagnosis, are acceptable. The reliability figures presented are modest, at best, and for a variety of reasons are probably inflated estimates of the true reliabilities. The reasons for this conclusion will be briefly discussed.The experimental procedure used by the authors involved presenting