Study designs of evaluations included in the review
Studies using any design were eligible for inclusion. The included trials comprised randomised controlled trials (RCTs) and controlled observational studies. None were blinded.
Specific interventions included in the review
Studies of hand washing or hand hygiene practices were eligible for inclusion. The studies compared different hand washing agents with each other, with the use of protective clothing with or without isolation, and with other unspecified comparators. The interventions in the included studies consisted of non-medicated soap, 60% alcohol, 1% and 4% chlorhexidine, chloroxylenol, 10% povidone-iodine and 1% triclosan, with or without interventions to increase hand washing.
Participants included in the review
Participants eligible for inclusion were patients in ICUs. The studies were set in adult, paediatric and neonatal ICUs.
Outcomes assessed in the review
Studies reporting infection rates in ICUs were eligible for inclusion. The review excluded studies of hand hygiene adherence, hand washing frequency or duration, and studies of the efficacy of different hand washing solutions or soaps in reducing bacterial counts on hands.
All of the included studies reported the rates of patient infection and/or rates of infective episodes. Two also reported the rates of patient carriers. Some studies focused on specific infective organisms such as Klebsiella (1 study) and methycillin-resistant Staphylococcus aureus (1 study). Two studies focused on specific types of infection such as bacterial and fungal pneumonia (1 study) or infections of the bloodstream or cerebrospinal fluid (1 study). The other 5 studies reported all nosocomial infections; in one of these studies the rates were broken down by infection site.
How were decisions on the relevance of primary studies made?
The authors did not state how the papers were selected for the review, or how many reviewers performed the selection.