293 patients took part in a WHO-collaborative study on the efficacy of the dexamethasone suppression test (DST) in depression. The goal was to examine the relationship of psychopathological and psychiatric history to cortisol levels and suppression/nonsuppression status. Differences between centers were large and significant on nearly all measures. Predictor analyses generally suffered from numerically weak correlations with many variables correlating to sex and age. With analyses adjusted for center, sex, and age influences, the best predicting features of cortisol were restless sleep, change of body weight, and affective disorders in blood relatives. The last 2 items together with hypersomnia and ideas of insufficiency were the best predictors of suppression/nonsuppression status.