Every year, road traffic crashes claim the lives of hundreds of thousands of children worldwide, and for families who own cars, the single most effective shield is a properly used child restraint system, or CRS. Yet a new cross-sectional study from China reveals a striking divide between two neighboring cities: in Shenzhen, where a local law has required CRS use for children under four since 2015, 75.88 percent of car-owning families with children aged 0 to 10 reported owning a CRS, compared with just 55.80 percent in Shantou, a city governed only by non-compulsory national recommendations. The gap in actual use was even sharper, with 63.75 percent of Shenzhen families reporting CRS use versus 43.34 percent in Shantou. The findings, published in BMC Public Health, offer some of the clearest policy-relevant evidence yet on how local legislation shapes child passenger safety in China.
The research team, led by Bowen Zhu and Yingjie Luo as co-first authors, with corresponding authors Dan Zhao of the Shenzhen Nanshan Center for Chronic Disease Control and Liping Li of Shantou University, surveyed 1,963 car-owning families recruited from selected urban districts of the two cities in 2023. Participants answered detailed questions about CRS ownership, day-to-day use behavior, caregiver safety knowledge, attitudes toward mandatory CRS legislation, awareness of the national CRS-related provision, and whether they had ever received reminders about CRS from traffic police. Because the survey captured both behavioral and policy-perception variables, the investigators could statistically disentangle how much of the city difference might be attributable to demographics versus the broader policy environment.
The statistical architecture of the study was deliberately robust. The researchers applied multivariable logistic regression to adjust for demographic characteristics and measured behavioral and policy-perception factors, then ran sensitivity analyses using stricter definitions of what counts as CRS use, and finally employed propensity score matching, or PSM, a technique that pairs families from the two cities with similar characteristics to isolate the effect of city itself. Across all three approaches, the city difference held firm. In the fully adjusted model, families in Shantou had 58 percent lower odds of CRS use than those in Shenzhen, with an adjusted odds ratio of 0.42 and a 95 percent confidence interval of 0.33 to 0.53. The baseline demographic model showed a similar association, and the PSM and sensitivity analyses confirmed the result was not an artifact of how use was defined or how families were matched.
Beyond the city contrast, the study uncovered a pattern that should alarm parents everywhere: CRS use declines steadily as children get older. Children aged 7 to 10 years had significantly lower odds of being restrained than those aged 0 to 3 years, with an adjusted odds ratio of 0.57 and a 95 percent confidence interval of 0.43 to 0.76. This age-related falloff appeared in both cities, suggesting that even in Shenzhen, where the local law targets children under four, families tend to abandon restraints once children outgrow the legally covered age range. The authors note that improving CRS use may require practical support for sustained, age-appropriate restraint use, not just initial adoption, because the transition from infant seats to booster seats for older children is where compliance quietly erodes.
Several factors emerged as powerful predictors of whether families actually used their restraints. Awareness of the national CRS-related provision, support for mandatory CRS use, reported exposure to traffic-police CRS reminders, and higher CRS safety knowledge were all positively associated with CRS use. In other words, families who knew the rules, believed in them, and had been personally reminded by enforcement officers were significantly more likely to buckle their children in. This cluster of associations points to a behavioral mechanism: legislation does not work through the threat of penalty alone, but through a feedback loop in which clear rules, public communication, and enforcement reminders reinforce one another to normalize restraint use as standard parenting practice.
The policy backdrop is what makes this comparison so compelling. Since 2015, Shenzhen has required CRS use for children under four, making it one of the first Chinese cities to translate national recommendations into local law. Most other cities in China, including Shantou, still rely only on non-compulsory national guidelines, which carry no penalty for non-compliance and no systematic enforcement communication. The roughly 20-percentage-point gap in both ownership and use between the two cities, observed even among children aged 0 to 3 years who are directly covered by the Shenzhen law, suggests that compulsory local requirements can move the needle on child passenger safety in ways that voluntary recommendations simply cannot.
The authors are careful about what the data can and cannot prove. They acknowledge that the city difference may partly reflect differences in local policy contexts, but they also caution that city-level socioeconomic, transportation-related, and other contextual factors cannot be fully separated in a cross-sectional design. Shenzhen and Shantou differ not only in legislation but in income levels, urban form, traffic density, and enforcement capacity, and any of these could contribute to the observed gap. The propensity score matching was designed to reduce confounding by measured characteristics, but unmeasured city-level differences remain a possibility. Still, the consistency of the association across multiple statistical approaches, and its alignment with the known behavioral predictors, strengthens the case that the policy environment plays a meaningful role.
What makes this study resonate beyond China is its implication for the global debate on child passenger safety legislation. Road traffic injuries are a leading cause of death for children worldwide, and the effectiveness of CRS in preventing fatalities and serious injuries is well established in the crash-safety literature. Yet many countries and regions still lack compulsory requirements or enforce them weakly. The Shenzhen-Shantou comparison functions as a kind of natural experiment: two cities in the same country, surveyed in the same year, with the same national recommendations, differing primarily in whether a local law exists. The result, a 58 percent reduction in the odds of use in the non-legislating city, is a vivid demonstration of how the presence or absence of a single legal requirement can shape everyday parenting behavior at population scale.
The study also carries a warning for cities that have passed CRS laws but assume the job is done. The decline in use among older children in both cities, including Shenzhen, reveals a blind spot in current policy design. Shenzhen’s law covers children under four, and once children cross that threshold, families appear to relax, even though booster seats remain critical for protecting older children in crashes. The authors suggest that integrated strategies combining clearer policy requirements, enforcement communication, caregiver education, and practical support for sustained age-appropriate restraint use are needed. In practice, this could mean extending legal coverage to older age groups, training pediatricians and caregivers on when and how to transition between restraint types, and using traffic-police reminders not just for young children but across the full age range.
For public health officials, the message is clear and actionable. Legislation works, but only when paired with the communication and enforcement infrastructure that makes families aware of the rules and reminds them to follow them. The positive associations between traffic-police reminders, safety knowledge, and CRS use show that enforcement is not merely punitive but educational, creating touchpoints that keep restraint use top of mind. As China continues to urbanize and car ownership spreads, the number of children riding in vehicles will only grow, and with it the stakes of getting this policy right. The Shenzhen-Shantou divide offers a roadmap: pass the law, communicate it clearly, remind families persistently, and support them practically as their children grow. The alternative, relying on voluntary recommendations, leaves more than half of children in the non-legislating city unprotected on every car ride.
Subject of Research: Child restraint system ownership and use in selected urban districts of Shenzhen and Shantou: a cross-sectional comparison
Article Title: Child restraint system ownership and use in selected urban districts of Shenzhen and Shantou: a cross-sectional comparison
Article References: Zhu, B., Luo, Y., Wu, F., Yang, J., Chen, Y., Zhao, D., & Li, L. (2026). Child restraint system ownership and use in selected urban districts of Shenzhen and Shantou: a cross-sectional comparison. BMC Public Health. https://doi.org/10.1186/s12889-026-29892-8
Image Credits: AI Generated
DOI: 10.1186/s12889-026-29892-8
Keywords: Child, restraint, system, ownership, selected, urban, districts, Shenzhen, Shantou, cross-sectional, comparison, scientific research
News Source: Ophelia Keating. (October 18, 2026). Mandatory Law, Safer Kids: Shenzhen Outpaces Shantou in Child Car Seat Use. Scienmag.
