The predictive power of the salivary cortisol dexamethasone suppression test for three-year outcome in major depressive illness

Seventy patients satisfying DSM-III and Research Diagnostic Criteria for major depression were given a salivary cortisol dexamethasone suppression test, with samples collected at 0700h, 1500h and 2300h after dexamethasone. The patients were classifird as nonsuppressors (mean post-dexamethasone salivary cortisol concentration ≥ 2.0 ng/ml, N = 27) and suppressors (mean post-dexamethasone salivary cortisol concentration < 2.0 ng/ml, N = 43). At 3-yr follow-up there was no difference in illness outcome as assessed by the life table method or by the length of rehospitalization for several periods after the index episode. In multiple regression and discriminant function analyses, with outcome as the dependent variable (readmitted within 1 yr, readmitted between 1 and 3 yr, not readmitted), the mean post-dexamethasone salivary cortisol concentration was not significantly predictive of outcome.