Substantial evidence now exists that ongoing surveillance of surgical wound infections can contribute to reduced infection rates. What is not yet determined is whether surveillance should be limited to the postoperative hospital stay or should be continued after patient discharge. To determine the number of infections occurring after discharge, the authors contacted a random sample of their patients who did not have wound infections during their hospitalization after orthopedic surgery. This was done 30 days after the procedure. The authors selected 273 patients of 1375 who underwent orthopedic surgery over a 7-month period and were able to contact 199 (73%). At the 30-day follow-up 23 patients (11.6%) had wound infections, as judged by wound discharge and physician prescription of antibiotics in 20 and the patient's description of pus issuing from the wound in 3. During the same period postoperative wound infections were found in only 19 (1.5%) of 1278 patients who were subjected to in-hospital surveillance. The authors conclude that, in patients who undergo orthopedic procedures, the majority of wound infections occur after discharge from the hospital and that infection rates based only on in-hospital surveillance greatly under represent true surgical wound infection rates for orthopedic procedures.