This review assessed the effect of handwashing on the risk of respiratory infection. The authors stated that there is some evidence to suggest that handwashing reduces the incidence of respiratory infection, but it is of poor quality, limited to developed countries, and does not consider severe disease. The data reported in the review appear to support these findings.

To determine the effect of handwashing on the risk of respiratory infection in the general population.

PubMed, CAB Abstracts, EMBASE, Web of Science and the Cochrane library were searched from inception to June 2004. The search terms were reported and language restrictions were not applied. The reference lists of retrieved articles and the authors' own reference collections were handsearched for additional studies.

Study designs of evaluations included in the review

Any interventional study was eligible. The study designs included in the review varied: randomised controlled trials (RCTs), crossover trials, before-and-after studies, cluster randomised controlled trials and controlled trials.

Specific interventions included in the review

Any method of handwashing or cleansing was eligible. The studies included in the review assessed soap or hand sanitisers, with or without additional supervision, training or education on how to effectively wash your hands. Two studies assessed handwashing but did not report the agents used.

Participants included in the review

Healthy individuals in the general population were included. Studies of individuals in hospitals, health care or geriatric care settings were excluded from the review, as were studies with immunocompromised participants or individuals suffering from genetic disorders. All of the included studies were conducted in developed countries and focused on children, young adults or students who were based in child care facilities, schools or university halls of residence.

Outcomes assessed in the review

All studies had to calculate the risk of any respiratory outcome related to hand cleansing. To be included in the meta-analysis, the studies had to provide risk or rate ratio estimates with 95% confidence intervals (CI), or information with which to calculate them. The included studies reported outcomes such as days of respiratory illness, incidence of colds, and the incidence of parental- or self-reported respiratory-related symptoms.

How were decisions on the relevance of primary studies made?

Two authors independently determined the relevance of both abstracts and full papers. Consensus was reached on all of those studies included in the review.

The authors stated that they extracted data on the methodological shortcomings of the studies. Such shortcomings included: use of randomisation; baseline incidences reported; use of concurrent control group; use of satisfactory case definition; use of placebo; assessment of impact on behaviour; loss to follow-up; use of correction for repeated episodes; and use of control for clustering.

Two authors independently extracted and tabulated data on the methodological shortcomings of the included studies.

Two authors independently extracted and tabulated the data from the included studies. Data on the study design, sample size, measures of effect (in order to calculate risk or rate ratios and 95% CIs), nature of the intervention, location, study population and methodological shortcoming of the studies were abstracted.

How were the studies combined?

Effect sizes (risks and rate ratio estimates) were pooled using a random-effects model. The risk of publication bias was assessed using Begg's-adjusted rank correlation test and a funnel plot.

How were differences between studies investigated?

Heterogeneity was assessed using a chi-squared test. Differences between the studies were also described in the data tables and discussed in the text. Sensitivity analyses were carried out to test the effects of differing study methodology and quality.

Eight studies (n=47,692) were included.

Only two studies were RCTs and all of the studies had methodological flaws. Two studies were of particularly poor quality. No evidence of publication bias was detected.

All eight studies reported that handwashing lowered risks of respiratory infection, with risk reductions ranging from 6 to 44% (pooled value 24%, 95% CI: 6, 40). However, significant heterogeneity was detected (P>0.0001), so the authors excluded one uncontrolled study. Pooling the remaining seven homogeneous studies gave a pooled risk reduction of 16% (95% CI: 11, 21). Other sensitivity analyses, which excluded poor-quality studies and the one study with a crossover design, provided similar results and no evidence of heterogeneity.

Handwashing is associated with lowered respiratory infection. However, the studies were of a poor quality, none related to developing countries, and only one related to severe disease.

This review was based on clear inclusion criteria for the intervention, population and outcome measures; the criteria for study design were more loosely defined. The review process was reported in detail and appropriate steps were taken to reduce the risk of bias throughout the review process. The authors carried out appropriate assessments of study validity, heterogeneity and publication bias, although given the small numbers of included studies the validity of these tests, in particular for publication bias, is questionable. The literature searches would therefore have benefited from additional attempts to locate unpublished data.

There were differences in study design, settings, participants, outcome measures, types of effect measures (rate and risk ratios); statistical heterogeneity was considerable. The authors' decision to pool effect sizes is therefore of concern, although they rightly pointed out that their pooled findings can only be considered indicative. Despite the limitations and the paucity of good-quality studies, the data generally appear to suggest a positive benefit of handwashing. However, better quality research is required to confirm this.

Practice: The authors did not state any implications for practice.

Research: The authors stated that rigorous trials of the impact of handwashing on acute respiratory tract infection morbidity and mortality are urgently needed, especially in developing countries.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.