Healthcare-associated infection by meticillin-resistant Staphylococcus aureus (MRSA) is still a great concern in an intensive care unit (ICU). Our surveillance data in the ICU revealed that intubated patients were at eight times higher risk of acquiring MRSA than non-intubated patients, so we hypothesised that pre-emptive contact precautions for all intubated patients would prevent healthcare-associated infection by MRSA in the ICU. Patients staying in our ICU for >2 days were included in this study. The study period was divided into two periods. During 2004 (1st period), contact precautions were performed only for patients with MRSA. During 2005–2007 (2nd period), contact precautions were applied to all intubated patients regardless of MRSA infection status. Patients were defined as MRSA-positive on admission when MRSA was detected by surveillance or clinical culture on enrolment. Other MRSA-positive results were defined as healthcare-associated MRSA (HA-MRSA) transmission. HA-MRSA infection was diagnosed according to the National Nosocomial Infections Surveillance Manual. The 1st period comprised 415 patients, and the 2nd period comprised 1280 patients. In intubated patients, HA-MRSA infection rate decreased significantly in the 2nd period (1st period 12.2%, 2nd period 5.6%; P=0.015). HA-MRSA infection of all patients decreased from 3.6 to 2.3 incidents per 1000 patient-days (P<0.05), despite a significant increase in the rate of patients MRSA positive on admission in the 2nd period (1st period 2.9%; 2nd period 6.1%). Pre-emptive contact precautions for intubated patients would be helpful in reducing HA-MRSA infection in ICU.