OBJECTIVE: The objective of this study is to determine the optimal timing for surgical antimicrobial prophylaxis (AMP). SUMMARY BACKGROUND DATA: National AMP guidelines should be supported by evidence from large contemporary data sets. METHODS: Twenty-nine hospitals prospectively obtained information on AMP from 4472 randomly selected cardiac, hip/knee arthroplasty, and hysterectomy cases. Surgical site infections (SSIs) were ascertained through routine surveillance, using National Nosocomial Infections Surveillance system methodology. The association between the prophylaxis timing and the occurrence of SSI was assessed using conditional logistic regression (conditioning on hospital). RESULTS: One-hundred thirteen SSI were detected in 109 patients. SSI risk increased incrementally as the interval of time between antibiotic infusion and the incision increased (overall association between timing and infection risk P = 0.04). When antibiotics requiring long infusion times (vancomycin and fluoroquinolones) were excluded, the infection risk following administration of antibiotic within 30 minutes prior to incision was 1.6% compared with 2.4% associated with administration of antibiotic between 31 to 60 minutes prior to surgery (OR: 1.74; 95% confidence interval, 0.98-3.04). The infection risk increased as the time interval between preoperative antibiotic and incision increased or if the antibiotic was first infused after incision. Intraoperative redosing (performed in only 21% of long operations) appeared to reduce SSI risk in operations lasting more than 4 hours (OR of 3.08 with no redosing; 95% confidence interval 0.74-12.90), but only when the preoperative dose was given correctly. CONCLUSIONS: These data from a large multicenter collaborative study confirm and extend previous observations and show a consistent relationship between the timing of AMP and SSI risk with a trend toward lower risk occurring when AMP with cephalosporins and other antibiotics with short infusion times were given within 30 minutes prior to incision.