Summary of findings for the main comparison. Summary of findings table 1.

Hand washing at child day‐care centres and schools compared to no intervention
Patient or population: Children Settings: Child day‐care centres or schools Intervention: Hand washing promotion (± provision of hand washing materials) Comparison: No intervention
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	No. of participants (trials)	Quality of the evidence (GRADE)
	Assumed risk	Corresponding risk
	No intervention	Hand washing promotion
Episodes of diarrhoea	High income countries		Rate ratio 0.70 (0.58 to 0.85)	4664 (9 trials)	⊕⊕⊕⊕ high1,2,3,4,5
	4 episodes per 100 children per year	2 episodes per 100 children per year (2 to 3)
	Low‐ or middle‐income countries		Rate ratio 0.66 (0.43 to 0.99)	45,380 (2 trials)	⊕⊕⊖⊖ low6,7,8
	22 episodes per 100 children per year	15 episodes per 100 children per year (9 to 22)
Hand washing behaviour	—	—	Not pooled	1845 (3 trials)	⊕⊕⊖⊖ low9,10,11
The basis for the assumed risk is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; RR: risk ratio.
GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.

¹The median incidence of diarrhoea in the control groups was four episodes per 100 children per year.
 ²No serious risk of bias: Most trials are at high or unclear risk of detection or reporting bias due to no description of blinding of outcome assessors. Restriction of the analysis to just the blinded trials finds a slightly smaller effect size but the result remains statistically significant. Not downgraded.
 ³No serious inconsistency: Although statistical heterogeneity was high, this heterogeneity was related to the size of the effect not the direction of effect. The individual effect sizes in trials ranged from an 10% relative reduction in diarrhoea to a 50% reduction.
 ⁴No serious indirectness: These nine trials were conducted in day‐care centres/schools in high income countries (USA, Denmark, Australia, Netherlands and Canada).
 ⁵No serious imprecision: The result is statistically significant and the meta‐analysis adequately powered to detect this result.
 ⁶The incidence of diarrhoea in the control group in the trial from Egypt was 22 per 100 children per year. The incidence in the control group in the Kenya trial was not stated.
 ⁷No serious inconsistency: While both trials found reductions in diarrhoea incidence the reduction was only statistically significant in the trials from Egypt. However, we did not downgrade.
 ⁸Downgraded by 2 for serious indirectness: Only one trial was conducted in a low‐income country (Pickering 2013 KEN). This trial from an urban slum in Nairobi did not find a statistically significant benefit on diarrhoea incidence.
 ⁹Downgraded by 1 for serious risk of bias: In the three trials, the observers themselves could not have been blinded and may have influenced the outcome simply by being present.
 ¹⁰Downgraded by 1 for serious indirectness: These three trials are from day care‐centres in the Netherlands and USA and schools in an urban slum in Nairobi, Kenya. Further trials from different settings are needed to confirm this result can be generalized.
 ¹¹The trials from Netherlands and USA found large and statistically significant improvements in staff hand washing behaviour or hand hygiene compliance. The trial from Kenya found no improvement in hand washing, but large and statistically significant improvements in the use of soap.