John D Kraemer1. 1. Department of Health Systems Administration and O'Neill Institute for National & Global Health Law, Georgetown University, Washington, DC. Electronic address: jdk32@georgetown.edu.
Abstract
PURPOSE: To assess bicycle helmet laws' effect on helmet and bicycle use among U.S. high school students in urban jurisdictions. METHODS: Log-binomial models were fit to Youth Risk Behavior Survey data from five jurisdictions. Adjusted helmet and bicycle use proportions were calculated with post-regression marginal effects. Difference-in-differences were estimated, comparing intervention to concurrent controls. A placebo outcome was used to falsify possible confounding or selection effects. RESULTS: In San Diego and Dallas, helmet use increase increased 10.6 (95% confidence interval [CI] 6.5 to 14.7, p < .001) and 8.1 (95% CI 4.3 to 12.0, p < .001) percentage points more than out-of-jurisdiction controls. Increases in Florida counties were 5.0 (95% CI 1.8 to 8.2, p = .003) and 4.0 (95% CI -.7 to 8.8, p = .098) points against age-based and out-of-jurisdiction controls, respectively. Bicycle use fell 5.5 points in both San Diego (95% CI -9.8 to -1.1, p = .015) and the Florida counties (95% CI -11.5 to .5, p = .075) against out-of-jurisdiction controls, but other comparisons had no significant changes. The placebo outcome never changed significantly. CONCLUSIONS: Laws increased helmet use in all jurisdictions, with limited evidence of reduced cycling. Although sound health policy, laws should be coupled with physical activity promotion.
PURPOSE: To assess bicycle helmet laws' effect on helmet and bicycle use among U.S. high school students in urban jurisdictions. METHODS: Log-binomial models were fit to Youth Risk Behavior Survey data from five jurisdictions. Adjusted helmet and bicycle use proportions were calculated with post-regression marginal effects. Difference-in-differences were estimated, comparing intervention to concurrent controls. A placebo outcome was used to falsify possible confounding or selection effects. RESULTS: In San Diego and Dallas, helmet use increase increased 10.6 (95% confidence interval [CI] 6.5 to 14.7, p < .001) and 8.1 (95% CI 4.3 to 12.0, p < .001) percentage points more than out-of-jurisdiction controls. Increases in Florida counties were 5.0 (95% CI 1.8 to 8.2, p = .003) and 4.0 (95% CI -.7 to 8.8, p = .098) points against age-based and out-of-jurisdiction controls, respectively. Bicycle use fell 5.5 points in both San Diego (95% CI -9.8 to -1.1, p = .015) and the Florida counties (95% CI -11.5 to .5, p = .075) against out-of-jurisdiction controls, but other comparisons had no significant changes. The placebo outcome never changed significantly. CONCLUSIONS: Laws increased helmet use in all jurisdictions, with limited evidence of reduced cycling. Although sound health policy, laws should be coupled with physical activity promotion.
Authors: Emily E White; Jordan Downey; Vidiya Sathananthan; Zahir Kanjee; Avi Kenny; Ami Waters; Jenny Rabinowich; Mallika Raghavan; Lorenzo Dorr; Amal Halder; Joseph Nyumah; Derry Duokie; Tamba Boima; Raj Panjabi; Mark J Siedner; John D Kraemer Journal: Am J Public Health Date: 2018-07-19 Impact factor: 9.308