Pedestrian fatalities outnumber passenger fatalities in U.S. school-aged children. The problem is particularly severe in cities. As yet, there are no known effective prevention approaches to complement variably effective education efforts. The proposed study will (1) analyze the interactions of biology, development/behavior and the traffic environment leading to pedestrian injury (PI), and (2) seek to identify developmental, behavioral, and family characteristics that place children at risk for PI. The study expands on work done in a pilot study. 200-250 PI victims 6-11 years old (cases) will be identified among inpatients at one Chicago pediatric hospital trauma center, or in the offices of practicing pediatricians affiliated with the hospital. With physician and family consent, date will be collected from medical records concerning injury severity and preexisting medical problems. Families will be interviewed concerning family constellation and details of the injury event, and will complete standardized surveys of development, temperament (Alpern-Boll, Thurstone), behavior (Achenbach), and family environment (Moos, Coddington). Teacher assessments of behavior and development will also be obtained correlation analysis will be used to identify types of children (in terms of demography, development, behavior and family) and sties (in terms of traffic/environment), and to link types of sites and children. Logit analysis will be used to compare mild and severe cases. Cases will be qualitatively analyzed at monthly staffings by a multidisciplinary term (pediatrics, social work, psychology, traffic engineering) to clarify the dynamic interaction of medical, developmental/behavioral, and traffic factors and to suggest potential mechanisms, conceptual frameworks and countermeasures. Uninjured children will be recruited to provide controls equal in number to cases and matched to cases on age, sex, social status and geographic origin. Controls will be recruited from the hospital and from pediatric practices. Control families will fill out the same surveys as cases. Cases and controls will be compared using discrete choice models, chisquare, ANOVA and t statistics. They study will (1) characterize children with PI and environments in which injuries occurred, (2) identify risk factors for PI, and (3) develop conceptual models and directions for future research.
| Wills, K E; Tanz, R R; Christoffel, K et al. (1997) Supervision in childhood injury cases: a reliable taxonomy. Accid Anal Prev 29:133-7 |
| Christoffel, K K; Donovan, M; Schofer, J et al. (1996) Psychosocial factors in childhood pedestrian injury: a matched case-control study. Kid's'n'Cars Team. Pediatrics 97:33-42 |
