To compare an interventional protocol with a standard protocol for preventing the acquisition of methicillin-resistant Staphylococcus aureus (MRSA) in the intensive care unit (ICU).

Prospective, randomized, controlled, parallel-group, nonblinded clinical trial.

Medical ICUs of 2 French university hospitals.

Five hundred adults with an expected length of stay in the ICU greater than 48 hours.

For the intervention group, the protocol required repeated MRSA screening, contact and droplet isolation precautions for patients at risk for MRSA at ICU admission and for MRSA-positive patients, and decontamination with nasal mupirocin and chlorhexidine body wash for MRSA-positive patients. For the standard group, the standard precautions protocol was used, and the results of repeated MRSA screening in the standard group were not communicated to investigators.

MRSA acquisition rate in the ICU. An audit was conducted to assess compliance with hygiene and isolation precautions.

In the intent-to-treat analysis (n = 488), the MRSA acquisition rate in the ICU was similar in the standard (13 [5.3%] of 243) and intervention (16 [6.5%] of 245) groups (P =.58). The audit showed that the overall compliance rate was 85.5% in the standard group and 84.1% in the intervention group (P =.63), although compliance was higher when isolation precautions were absent than when they were in place (88.2% vs 79.1%; P<.001). MRSA incidence rates were higher without isolation precautions (7.57‰) than with isolation precautions (2.36‰; P =.01).

Individual allocation to MRSA screening, isolation precautions, and decontamination do not provide individual benefit in reducing MRSA acquisition, compared with standard precautions, although the collective risk was lower during the periods of isolation.

Clinicaltrials.gov identifier: NCT00151606.