Validation of Definitions of Endogenous Depression

To the Editor.—  In the March 1986 issue of the Archives, Zimmerman et al¹ attempted to "validate" four definitions of endogenous depression (DSM-III melancholia and Newcastle, Feinberg-Carroll, and Research Diagnostic Criteria endogenous depression) by examining which of several predicted correlates of endogenous depression were related to each of the four definitions in 152 patients with major depressive disorder. Included were demographic, family-history, social, psychological, biological, and treatment-response variables. By simply counting up the number of predicted relationships that occurred for each definition, the authors found "some support for the validity of each of the four definitions... the Newcastle scale was the most frequently supported."Methodologic questions that were both evident in the report and inferrable render the "edge" given to the Newcastle scale quite tenuous. An example of an evident methodologic question is the authors' use of preexisting personality disorder as one of the 13 variables. Not surprisingly, they