Evidence is accumulating that active surveillance, when combined with appropriate infection control, is a successful measure for controlling hospital-acquired meticillin-resistant Staphylococcus aureus (MRSA). In this study, the impacts of a long-term control strategy of this type, including the use of chlorhexidine baths, on the clinical and molecular epidemiology of MRSA in the Intensive Care Unit of Aberdeen Royal Infirmary were investigated. Characterisation of 85 sequential index MRSA isolates was performed using phenotypic methods (biotyping), antibiotic susceptibility testing and three genotypic methods (pulsed-field gel electrophoresis, spa typing and multilocus sequence typing) over a 4-year period. There was no evidence of loss in effectiveness of the control strategy over the study period. Compliance with screening remained high (>85%) throughout and there was no significant increase in the prevalence of MRSA detected in surveillance (P = 0.43 for trend) or clinical cultures (P = 0.79). There were no significant trends in rates of other index surveillance organisms (P > 0.5). Results of the three typing methods were in general agreement with three prevalent MRSA clones [clonal complex 22 (CC22), CC30 and CC45]. CC22 emerged as the dominant clonal complex alongside a significant decline in CC30 (P = 0.002). CC45 was significantly more likely to be positive in glycopeptide resistance screens (P <0.001). There was no increase in antibiotic or chlorhexidine resistance. Long-term chlorhexidine bathing was not associated with any detectable loss of efficacy or increase in resistance in MRSA or with any increase in infection with other organisms. Changing clonal epidemiology occurred with no overall change in the prevalence of MRSA.