Forty-six orthopedic patients were studied to determine the incidence, natural history, and risk factors associated with post-operative delirium. Pre-operatively, patients were given a neuropsychological screening evaluation, the Mood Adjective Checklist (MACL), the Zung Depression Scale, the Anxiety Inventory Scale, and the Health Assessment Questionnaire (HAQ). A psychiatrist interviewed each patient on post-op day four for evidence of delirium as defined by DSM III criteria. Of the patients studied, thirteen (26%) were possibly or definitely delirious following surgery. Treatment with propranolol, scopolamine, or flurazepam (Dalmane) conferred a relative risk for delirium of 11.7 (p = 0.0028). Delirium was associated with increased post-operative complications (p = 0.01), poorer post-operative mood (p = 0.06), and an increase of about 1.5 days in length of stay (not significant). Delirious patients were significantly less likely than matched controls to improve in function at six months compared with a pre-operative baseline HAQ (t = 6.43, p less than 0.001).