The utility of the REM latency test in psychiatric diagnosis: a study of 81 depressed outpatients.

To examine the utility of the REM (rapid eye movement) latency test in identifying outpatient primary depressions, 81 consecutive referrals to a sleep disorders center were evaluated in a phenomenologic, sleep polygraphic, and psychometric study. Modified Feighner (St. Louis) diagnoses were definite primary depression (n = 19), probable primary depression (n = 30), depression chronologically secondary to preexisting psychiatric disorders (n = 19), and nonaffective psychiatric disorder (n = 13). There were 18 nonpsychiatric controls. REM latency less than 70 minutes on 2 consecutive nights detected 62% of primary depressions, discriminating them from the other diagnostic groups with 88% specificity. There were no false positives among controls. These data provided a 90% confidence for the diagnosis of primary depression in this outpatient sample. Requiring 2 consecutive nights of shortened REM latency appears to improve significantly the specificity of a test previously considered to have high sensitivity but relatively low specificity for depressive disorders.