Road traffic injuries remain a leading cause of death and serious harm among children, and a substantial proportion of this burden occurs among child passengers. Despite clear safety recommendations, incorrect installation and suboptimal child restraint system (CRS) selection remain widespread, representing a persistent public health implementation gap. Against this backdrop, the present study examined determinants of CRS selection decisions and self-reported installation and use practices, with a particular focus on how selection and retail-support experiences relate to intention to choose the same CRS again. A cross-sectional online survey (Google Forms) was conducted from February to December 2024, yielding 3,750 questionnaires, of which 1,150 complete cases were included in analysis. Descriptive statistics were supplemented with chi-square tests (Cramér's V) and multivariable logistic regression (odds ratios, 95% CI) for top-box intention to choose the same CRS again (“yes/probably yes”). Price (58.1%), seat weight (48.2%), and installation method (46.1%) were the most frequently reported selection determinants, while safety tests (28.3%) and comfort (26.4%) were less commonly cited. ISOFIX solutions predominated (74.6%), with installation most often behind the driver (45.0%) and a notable share on the front seat (25.4%). Intention to choose the same CRS again was low (27.4% top-box; 62.3% negative). Seller installation demonstration was associated with intention to choose the same CRS again (χ 2 = 22.82, p = 0.0036; V = 0.100), whereas reported child discomfort was not ( p = 0.9378). In logistic regression, only age 26–35 years (vs. 18–25) predicted top-box intention to choose the same CRS again (OR = 1.42, 95% CI 1.05–1.91), with overall low explanatory power (McFadden pseudo-R 2 = 0.0167). The findings suggest that economic and usability considerations outweigh explicitly stated safety criteria, and that closing the implementation gap requires stronger, scalable support for correct CRS selection and installation.
Kopańska et al. (Wed,) studied this question.