In 1992, the Agency of Health Care Administration in Tallahassee, Florida, started releasing, as part of the patient discharge information, the names of the treating physicians, in addition to demographic and diagnostic data. This information is available to the general public for a small price and is being used by health planners, hospital administrators, finance departments, third-party payers, and other agencies involved in health care. Patient discharge information was used to assess the effects of volume on the short-term outcome of primary and revision hip and knee arthroplasty as a function of surgeon and hospital in the State of Florida, during 1992. A total of 19,925 primary and 2,536 revision arthroplasties of the hip and knee were performed during 1992 in Florida and were available for study. After the doctors and hospitals were arbitrarily divided into three case volume groups (low, medium, high), results showed that in primary arthroplasty, surgeons with a low volume of primary cases (< 10) have a significantly higher mortality rate (24%), higher average charges ($25,000), and increased average length of hospital stay (9.3 days). In revision surgery, physicians with a low volume of cases (< 10) have a higher mortality rate (13%) and increased average length of hospital stay (9.8 days). Patients discharge information has many potential uses for investigators interested in the short-term outcome of arthroplasty. In their present form, these databases should not be released to the general public or the media. Lastly, the volume-outcome relation for a specific surgical procedure should, in addition to case severity, account for characteristics affecting the degree of technical difficulty.