Overactivity of the HPA axis as manifested by an increase in cortisol secretion is now a well-established phenomenon in depression (Carroll et al. 1976; Sachar et al. 1973). The first studies (Sachar et al. 1973) showed that up to 50% of depressed patients have higher mean plasma cortisol concentrations and an increased number and duration of cortisol secretory episodes, suggesting increased cortisol secretory activity. Numerous studies have subsequently validated these findings (Carroll et al. 1976; Halbreich et al. 1985; Krishnan et al. 1990a; Pfohl et al. 1985; Rubin et al. 1987). As many as two-thirds of endogenously depressed patients fail to suppress cortisol, or show an early escape of cortisol, following overnight administration of 1 mg of dexamethasone (using a cortisol cutoff of 5 g/dL to define "escape") (Carroll et al. 1981). While nonsuppression of cortisol in response to dexamethasone is strongly associated with endogenous depression, this finding is less robust in outpatients with depression. Although both hypercortisolemia and feedback abnormalities in response to dexamethasone are present in depressed patients, they do not necessarily occur in the same individuals (Carroll et al. 1981; Halbreich et al. 1985). Other abnormalities, such as reduced glucocorticoid fast feedback (Young et al. 1991) and a blunted ACTH response to exogenous CRH, have also been reported in depressed patients (Gold et al. 1986; Holsboer et al. 1984; Young et al. 1990).