Variation in risk for falls by clinical department: implications for prevention.

To examine rates of falls and injuries by clinical department, a retrospective analysis was performed of 874 inpatient falls during a one-year period in a 1,000-bed acute-care hospital. Incident reports were reviewed, and for those patients injured, medical records were abstracted. Fall rates were calculated by age, sex and service using total patient days as the denominator. During 1983, 874 falls occurred during 282,713 adult patient days (31/10(4) days). Fall rates were significantly higher (odds ratio = 2.12; 95% confidence interval [CI]: 1.84, 2.45) in the neuroscience and psychiatry departments (53/10(4) days) compared with other adult services combined (25/10(4) days); this difference persisted after direct adjustment for age. Differences by clinical department persisted throughout four additional years of surveillance, although fall rates for the oncology service increased. Injuries following falls occurred among 200 (22.8%) patients. These data indicate that falls are not random events, and that falls and injury prevention programs can be more appropriately focused by consideration of clinical department.