Six studies were included in the review: one placebo-controlled cluster RCT, two cluster non-RCTs, two crossover studies, and one cluster trial that was published only in abstract form and in which randomisation was unclear.
Inter-observer agreement was 92% for study selection, 89% for data extraction and 80% for quality assessment.
The quality of the included studies was poor. Of the five studies that were reported in full, only one was a double-blind RCT and this trial did not describe an appropriate method of randomisation. In addition, allocation concealment was unclear. Drop-outs were generally inadequately reported and one study reported that over 50% participants dropped out, with no explanation for such a high rate. Four of the studies received industrial sponsorship. None of the studies reported a sample size calculation and the statistical methods used were vague.
All studies found a statistically significant effect of the antimicrobial rinse-free hand sanitiser interventions in reducing absenteeism due to communicable illness. In the four studies where relative effect data were available, the relative effect ranged from 20% (95 CI: 19, 21) to 56% (95% CI: 31, 72). In the two studies where relative effect data were not available, one reported an overall reduction of 28% in absenteeism due to illness for children in the intervention group. The other reported a reduction in the odds of being ill of 43% in the intervention group; this was significant using McNemar's test (chi-squared 7.787; P=0.0053).