Highlights • In certain collisions, booster seats seem effective at reducing odds of injury. • 62% of children aged 4 to 8 years were reported in booster seats at time of collision. • The findings support legislation promoting booster seat use in young children.

Abstract Background Alberta remains the only province in Canada without booster seat legislation. To date, analyses of booster seat effectiveness compared with seatbelt only use have demonstrated mixed findings using observational data. Methods This study uses Alberta police collision report data for the years 2010–2016, inclusive. Using a case-control study design, children aged four to eight years, who were reported by police to be injured (cases), were compared with uninjured controls for restraint use (seatbelt, booster seat or no restraint). Logistic regression was used to estimate the relation between booster seat use and injury with adjusted odds ratios (aORs) and 95% confidence intervals (CI), stratified by collision types. Results There were 12,922 children involved in collisions, of whom 570 were injured. Approximately 62% of all children included in the analysis were in a booster seat or safety seat at the time of collision. Crude analysis indicates higher odds of injury for seatbelt wearers compared with booster seat use (OR = 1.21; 95% CI: 1.02–1.44). Front-end vehicle-vehicle collisions, demonstrated higher odds of injury for seatbelt wearers relative to those in booster seats (aOR = 1.46; 95% CI: 1.03–2.07). Conclusion This analysis indicates a protective effect of booster seats that varies depending on the type of collision and impact location. These regionally-specific injury data may encourage and inform policy on the use of booster seats. Moreover, stratification by collision type may be necessary to inform analyses on booster seat effectiveness.