Rehospitalization following surgery is widely recognized as an important outcome measure. The purpose of this study was to identify rehospitalizations for adverse events following 8 procedures, using diagnosis and procedure codes contained in Medicare claims files. Adverse events were broadly defined as: 1) complications; 2) failure of the procedure to achieve its therapeutic goal; and 3) untoward events associated with the natural history of the disease being treated with the procedure. Expert panels identified specific diagnosis and procedure codes that might indicate an adverse event if they appeared on the Medicare record of a rehospitalization. Among patients undergoing percutaneous transluminal coronary angioplasty, almost 36% were rehospitalized for an adverse event within a year of surgery; among patients undergoing coronary artery bypass graft surgery, 20% were rehospitalized for an adverse event. Following the other 6 procedures (cholecystectomy, partial excision of the large intestine, total knee replacement, total hip replacement, replacement of the head of the femur, and reduction of fracture of the femur) between 4% and 9% of patients were rehospitalized for an adverse event. Findings from this exploratory study indicate that rehospitalizations for adverse events appear to be a useful outcome measure for the cardiac procedures; they appear to be less useful for the other procedures, at least at the individual hospital or small area level, because of their relative rarity. Future studies should investigate procedures associated with more frequent rehospitalizations, and medical admissions, which often tend to be associated with higher rehospitalization levels.