Background: Helmets reduce bicycle-related head and facial injuries for bicyclists of all ages in all types of crash. We aimed to identify non-legislative interventions that are effective in promoting helmet use among children, so future campaigns can be designed on a firm evidence base.
Objectives: to assess the effectiveness of non-legislative interventions (compared to a lack of interventions) in increasing helmet use among children; to identify possible reasons for differences in the effectiveness of interventions; to evaluate effectiveness with respect to social group; to identify any adverse consequences of interventions.
Search strategy: We searched 11 electronic databases and manually searched other sources of potentially relevant data.
Selection criteria: We included randomised controlled trials, cluster randomised controlled trials and controlled before and after studies. Studies included participants aged 0 to 18 years and described interventions to promote helmet use not requiring enactment of legislation. Studies had to report at least one of the following outcomes: observed helmet wearing; self-reported helmet ownership; self-reported helmet wearing.
Data collection and analysis: Data were extracted by two reviewers working independently. Study results were adjusted to account for clustering. A random-effects model was used to pool data in meta-analyses. Heterogeneity was explored with sub-group analyses.
Main results: Twenty-two studies were included. The odds of observed helmet wearing were significantly greater amongst those receiving non-legislative interventions (OR 2.30, 95% CI 1.37 to 3.85). Sub-group analysis indicated the effect may be greater for community-based studies (OR 4.30, 95% CI 2.24 to 8.25) and those providing free helmets (OR 4.35, 95% CI 2.13 to 8.89) than for those providing subsidised helmets (OR 2.02, 95% CI 0.98 to 4.17) and for those set in schools (OR 1.82, 95% CI 0.94 to 3.52). We found no significant effect of non-legislative interventions in increasing self-reported helmet ownership, but they were associated with a significant increase in self-reported helmet wearing (OR 3.90, 95% CI 1.42 to 10.69), particularly amongst interventions set in schools (OR 4.73, 95% CI 1.09 to 20.49) but there was significant unexplained heterogeneity between effect sizes for these two outcomes.
Authors' conclusions: Community-based studies that include the provision of free helmets alongside an educational component increase observed helmet wearing in the areas in which they are set. There is some evidence that interventions in schools and those providing subsidised helmets may increase observed helmet wearing, but possibly to a lesser extent than those set in communities and those providing free helmets.