This study evaluated the utility of morning pretreatment systolic orthostatic blood pressure (PSOP) in predicting clinical response to treatment with nortriptyline (N = 11) or electroconvulsive therapy (N = 6) in 17 depressed geriatric inpatients (mean age, 70.4 +/- 5.1). Morning PSOP showed a significant inverse correlation with percent change in Hamilton depression ratings (rho = -0.59, p less than 0.01; r = -0.52, p less than 0.02). In nortriptyline-treated patients (N = 10, excluding one outlier), PSOP was significantly correlated with percent change in Hamilton ratings (rho = -0.55, p less than 0.05); a similar association was also found in the subsample of electroconvulsive therapy-treated patients (N = 6, rho = -0.77, p less than 0.05). Patients with PSOP less than or equal to 10 mm Hg averaged 83% improvement in Hamilton depression ratings versus 64% improvement in patients with PSOP less than or equal to 10 mm Hg (p less than 0.05). In an age-equated contrast group of 15 inpatients with mixed clinical pictures of depression and cognitive impairment (11 with primary degenerative dementia with depressive features and four with major depressive disorder with cognitive impairment), no relation between PSOP and treatment response (as measured by Hamilton ratings) was found. The current findings extend earlier work in medically healthy, nonsuicidal geriatric depressed outpatients and suggest that PSOP may also be useful in predicting treatment response in older, cognitively intact depressed inpatients (many with concurrent medical illness and/or suicidal) but not in mixed depression-dementia.