OBJECTIVE Admissions to psychiatric emergency services have frequently been cited as a gauge of how well a mental health system manages behavioral disorders. However, few measurements of the longitudinal association between psychiatric emergencies and characteristics of a mental health system have been described. The purpose of this study was to assess whether weekly admissions to psychiatric emergency services would increase when outpatient services were reduced, whether weekly admissions would increase when greater effort was made to identify and treat persons with acute mental illness, and whether weekly admissions would decrease when emergency services were enhanced to include postrelease case management. METHODS Time-series methods were applied to approximately 29,010 admissions to three psychiatric emergency services of the San Francisco Department of Public Health over a 180-week period. RESULTS Reduced outpatient services, efforts to identify acutely ill persons, and changes in emergency services themselves were found to affect admissions to emergency services. However, community events such as extreme weather, holidays, job loss, and the scheduling of receipt of income also affected the workload of the emergency service. CONCLUSIONS The causes and course of mental illness inextricably tie a psychiatric emergency service to the overall mental health system and to events in the community it serves. These connections make it possible for managers to anticipate the use of emergency services and to detect disruptions in the remainder of the mental health services systems.