Attempts to address the growing problem of healthcare-associated infections (HAIs) and their impact on healthcare systems have historically relied on infection control policies that recommend good hygiene through standard and enhanced precautions (e.g. barrier precautions and patient isolation). In order for infection control strategies to be effective, however, healthcare workers' behaviour must be congruent with these policies. The purposes of this systematic review were to evaluate studies testing the effectiveness of interventions aimed at changing healthcare workers' behaviour (in reducing HAIs) and to summarize the findings of the studies with the highest quality scores. A total of 33 published studies met the inclusion criteria and were evaluated. Four of these earned a study quality score of > or =80%. In all four significant reductions in HAI or colonization rates were reported. Behavioural interventions used in these high quality studies included an educational programme (in four), the formation of a multi-disciplinary quality improvement team (three), compliance monitoring and feedback (two), and a mandate to sign a hand hygiene requirement statement (one). In all 33 studies, bundles of two to five interventions were employed, making it difficult to determine the effectiveness of individual interventions. The usefulness of "care bundling" has recently been recognized and recommended by the Institute for Healthcare Improvement. Considering the multi-factorial nature of the HAI problem and the logistical and ethical difficulties of applying the randomized clinical trial approach to infection control research, it may be necessary to study interventions as sets of practices.