Objective: To prospectively investigate the predictive value of diagnosis, suicidal behavior, and subjectively experienced depressed mood for imminent risk of suicide death.Methods: This prospective study included a representative and diagnostically mixed sample of 7,000 acutely hospitalized psychiatric patients between May 2005 and July 2014 in a Norwegian catchment area of 400,000 inhabitants. Suicide deaths were registered at 1 and 2 weeks and at 1, 6, and 12 months following admission. Survival and hazard functions were estimated, and Cox regression was used to estimate the predictive values of suicidal ideation, suicide attempts, a diagnosis of depressive disorder, and severely depressed mood. Assessments were conducted at admission and included ICD-10 diagnosis, clinical interview in the form of the Health of the Nation Outcome Scales, and qualitative assessments of suicidal ideation and suicide attempts during the past week.Results: During 1-year follow-up, 101 patients (1.4%) died by suicide, of whom almost 70% were men. Only severely depressed mood, including inappropriate self-blame and guilt, predicted suicide within the first week after admission (hazard ratio [HR] = 7.3; 95% confidence interval [CI], 1.4-37.1; P = .01). Suicidal ideation predicted death by suicide at 2 weeks (HR = 3.8; 95% CI, 1.2-12.8; P = .02) and all follow-up time points after, whereas a recent suicide attempt predicted suicide from the 1-month followup (HR = 7.3; 95% CI, 2.2-23.7; P < .001) onward. Conclusions:We recommend thoroughly examining severity of depressed mood during assessment of imminent suicide risk.