CONTEXT: Experts and policy makers have repeatedly called for universal screening at hospital admission to reduce nosocomial methicillin-resistant Staphylococcus aureus (MRSA) infection. OBJECTIVE: To determine the effect of an early MRSA detection strategy on nosocomial MRSA infection rates in surgical patients. DESIGN, SETTING, AND PATIENTS: Prospective, interventional cohort study conducted between July 2004 and May 2006 among 21 754 surgical patients at a Swiss teaching hospital using a crossover design to compare 2 MRSA control strategies (rapid screening on admission plus standard infection control measures vs standard infection control alone). Twelve surgical wards including different surgical specialties were enrolled according to a prespecified agenda, assigned to either the control or intervention group for a 9-month period, then switched over to the other group for a further 9 months. INTERVENTIONS: During the rapid screening intervention periods, patients admitted to the intervention wards for more than 24 hours were screened before or on admission by rapid, multiplex polymerase chain reaction. For both intervention (n=10 844) and control (n=10 910) periods, standard infection control measures were used for patients with MRSA in all wards and consisted of contact isolation of MRSA carriers, use of dedicated material (eg, gown, gloves, mask if indicated), adjustment of perioperative antibiotic prophylaxis of MRSA carriers, computerized MRSA alert system, and topical decolonization (nasal mupirocin ointment and chlorhexidine body washing) for 5 days. MAIN OUTCOME MEASURES: Incidence of nosocomial MRSA infection, MRSA surgical site infection, and rates of nosocomial acquisition of MRSA. RESULTS: Overall, 10 193 of 10 844 patients (94%) were screened during the intervention periods. Screening identified 515 MRSA-positive patients (5.1%), including 337 previously unknown MRSA carriers. Median time from screening to notification of test results was 22.5 hours (interquartile range, 12.2-28.2 hours). In the intervention periods, 93 patients (1.11 per 1000 patient-days) developed nosocomial MRSA infection compared with 76 in the control periods (0.91 per 1000 patient-days; adjusted incidence rate ratio, 1.20; 95% confidence interval, 0.85-1.69; P =.29). The rate of MRSA surgical site infection and nosocomial MRSA acquisition did not change significantly. Fifty-three of 93 infected patients (57%) in the intervention wards were MRSA-free on admission and developed MRSA infection during hospitalization. CONCLUSION: A universal, rapid MRSA admission screening strategy did not reduce nosocomial MRSA infection in a surgical department with endemic MRSA prevalence but relatively low rates of MRSA infection. TRIAL REGISTRATION: isrctn.org Identifier: ISRCTN06603006.