To determine whether any of several quality improvement interventions with none specifically targeting methicillin-resistant Staphylococcus aureus (MRSA) were associated with a decline in endemic MRSA prevalence in an intensive care unit (ICU) where active screening and contact isolation precautions for known MRSA colonised patients are not practised.

Medical-surgical ICU with 2,000 admissions/year.

8.5-year retrospective time-series analysis.

ICU re-location, antibiotic stewardship utilising computerised decision-support and infectious-diseases physician rounds, dedicated ICU infection control practitioners, alcohol-based hand rub solution (ABHRS).

Regression modelling was used to evaluate trends in S. aureus prevalence density (monthly clinical isolates per 1,000 patient-days), antibiotic consumption, infection control consumables, ABHRS and their temporal relationship with MRSA prevalence.

Methicillin-resistant S. aureus prevalence density decreased by 83% [95% confidence interval (CI) -68% to -91%, p < 0.001]. Rates of MRSA bacteraemia decreased 89% (95% CI -79% to -94%, p = 0.001) with no statistically significant change in methicillin-sensitive S. aureus bacteraemia. Hospital MRSA prevalence density decreased 17% (95% CI -5% to -27%, p = 0.005), suggesting that ICU was not shifting MRSA elsewhere. In ICU, broad-spectrum antibiotic use decreased by 26% (95% CI -12% to -38%, p = 0.008), coinciding with a decrease in MRSA, but time-series analysis did not show a significant association. On multivariate analysis, only ABHRS was significantly associated with a decrease in MRSA, but it was formally introduced late in the study period when MRSA was already in decline.

General quality improvement measures were associated with a decrease in endemic MRSA in a high-risk setting without use of resource-intensive active surveillance and isolation practices.