Objectives : Misuse of antibiotics in surgical prophylaxis is still quite common. The objectives of this study were to reduce the quantity and improve the quality of surgical prophylaxis and to reduce costs. Methods : Prospective multi-site study of elective procedures in 13 Dutch hospitals. The quality of prophylaxis was audited before and after an intervention consisting of performance feedback and implementation of national clinical practice guidelines. Process outcome parameters were antibiotic choice, duration, timing, antibiotic volume and costs. Segmented regression analysis was used to estimate the effect size of the intervention. Patient outcome was documented by the incidence of surgical site infections (SSI). Results : Before the intervention, 1763 procedures were recorded and 2050 thereafter. Antimicrobial use decreased from 121 to 79 DDD (defined daily doses)/100 procedures and costs reduced by 25% per procedure. After the intervention, antibiotic choice was inappropriate in only 37.5% of the cases instead of in 93.5% expected cases had the intervention not occurred. Prolonged prophylaxis was observed in 31.4% instead of 46.8% expected cases and inappropriate timing in 39.4% instead of the expected 51.8%. Time series analysis showed that all improvements were statistically significant ( P < 0.01) and that they could be fully attributed to the intervention. The overall SSI rates before and after intervention were 5.4% (95% CI: 4.3–6.5) and 4.6% (95% CI: 3.6–5.4), respectively. Conclusions : The intervention led to improved quality of surgical prophylaxis and to reduced antibiotic use and costs without impairment of patient outcome.