Pediatric road traffic injuries in Ghana: a cross-sectional study of factors associated with injury severity within the Haddon Matrix framework

Objectives To apply the Haddon Matrix framework (HM) to examine factors associated with the severity of paediatric road traffic injuries (RTIs) in Ghana. Design Prospective hospital-based cross-sectional study. Variables were organised according to the three domains of the HM framework. Three negative binomial regression models were developed to identify the best-fitting model for predicting paediatric RTI severity Settings Three teaching hospitals in the Northern, Ashanti and Central Regions of Ghana. Participants A total of 484 consecutive children and adolescents aged 0–18 years presenting to the emergency department with RTIs between 18 September 2023 and 30 September 2024 were included. Data were collected using interviewer-administered questionnaires and medical record reviews. Main outcome measure The primary outcome was injury severity measured using the Injury Severity Score, an anatomical measure based on the sum of the squared Abbreviated Injury Scale scores for the three most severely injured body regions. Results The third model (variable selection threshold p≤0.1) demonstrated the best predictive model for paediatric RTI severity, with the lowest Akaike Information Criterion and the highest log-likelihood. Increased injury severity was independently associated with self-reported frequent alcohol use (β=1.54, p=0.016), Harmattan weather conditions (β=0.21, p=0.024), afternoon crashes (β=0.17, p=0.041) and referral cases (β=0.28, p=0.003). Whereas single bicycle crashes (β=−1.58, p=0.011) and walk-in hospital visits (β=−1.09, p=0.026) were associated with lower injury severity. Conclusion The HM identified modifiable environmental, behavioural and health system factors associated with paediatric RTI severity. Interventions targeting alcohol use, Harmattan-related road safety, timely referral and prehospital care may reduce injury severity among children in Ghana.