Background: The aim of this study was to determine whether depressive symptoms affect pituitary-adrenal function in adolescents, as they do in adults. Methods: Salivary cortisol was measured before and after physical exercise in 23 hospitalized adolescent psychiatric patients and 13 age- and sex-matched healthy controls in a placebo-controlled design. In patients, cortisol profiles were assessed from 08:00 to 20:00 h before and after administration of low doses of dexamethasone or the natural steroid hydrocortisone. Patients were classified according to DSM III-R criteria and assigned to either a depressed group (n = 10) or a non-depressed group (n = 13). Subjective depressive symptoms were rated with the 90-item symptom checklist. Results: Physical exercise increased cortisol output significantly in all subjects, but there were no differences between groups. In patients, no differences in basal cortisol levels were found between the depressed and non-depressed groups. Dexamethasone, but not hydrocortisone, was able to suppress cortisol levels in both groups. Differences between groups were only found in total cortisol output over the 3 days when data were analyzed according to the patients 'subjective' feelings of depression, with the highest cortisol levels in the 'subjectively more depressed' patients. Conclusions: The results obtained only partially support the hypothesis that depressive symptoms in adolescent psychiatric patients determine pituitary- adrenal (mal)function, and appear to depend on the strategy used. (C) 2000 Elsevier Science B.V.