Background: Studies have shown that safety seats are more effective than lap-shoulder seat belts in reducing the risk of injuries and death in children. However, one study by researchers at the National Bureau of Economic Research found that lap-shoulder seat belts were as protective as child safety seats in children two to six years of age, and that they were significantly less expensive. This study had limitations though, such as including children up to six years of age even though the appropriate restraint at that age would be lap-shoulder seat belt, fatal crash data that lacked a good control variable for crash severity, and fatality data with nonfatal injury information collected from police reports. Zaloshnja and associates evaluated the effect of child safety seats compared with lap-shoulder seat belts on nonfatal injuries in children two to three years of age.

The Study: Data were collected from the Crashworthiness Data System of the National Highway Traffic Safety Administration from January 1, 1998, to December 31, 2004. The sample included crashes that had at least one vehicle towed away and had only children two to three years of age riding in the backseat of the car, either sitting in a child safety seat or wearing a lap-shoulder seat belt. Injury data, including injuries that posed only a minor threat to life, were extracted from the database. The major outcome measure was presence of any injury after a crash.

Results: A total of 463 children were included in the analysis, with 409 children being restrained by child safety seat and 54 by lap-shoulder seat belts. There were 178 reported injuries in the child safety seat group and 28 in the lap-shoulder seat belt group. Unadjusted injury probability for children in severe crashes was 46 percent lower in the safety seat group compared with the lap-shoulder seat belt group. After controlling for vehicle type and weight, crash severity, child's seating position with regard to driver, and crash impact angle, the safety seat group had an 81.8 percent lower chance of injury compared with the lap-shoulder seat belt group.

Conclusion: Child safety seats appear to be more effective than lap-shoulder seat belts in reducing the risk of nonfatal injuries in children two to three years of age. The authors add that laws requiring child safety seats in children younger than four years are good, and that efforts should be made to retain these laws.