Background: The American Academy of Pediatrics and the National Highway Traffic Safety Administration recommend the use of belt-positioning booster seats in children four years and older until they fit properly in the vehicle seat belt; however, many children graduate prematurely to car seat belts, increasing their risk of serious injury in a crash. Data from 1998 to 2002 showed a notable injury reduction in four- to five-year-old children who used booster seats, but data on older children were not available because most did not use booster seats at the time. Recently, many states have mandated that booster seats be used for children up to six to eight years of age, with a resulting threefold increase in booster seat use. Arbogast and colleagues reviewed more recent safety data to update the estimated effectiveness of belt-positioning booster seats versus seat belts alone in children four to eight years of age.

The Study: The Partners for Child Passenger Safety project collected and analyzed data from December 1998 through November 2007. This large-scale, child-specific crash surveillance system incorporates data reported by insurance claims from the State Farm insurance company, as well as telephone surveys and onsite crash investigations. A driver telephone survey functioned as the primary source of data for this analysis. Inclusion criteria consisted of State Farm–insured vehicles built in 1990 or later that were involved in a crash that included at least one child 15 years or younger. The insurance company initially contacted the policyholders of qualifying crashes to inform them of the study and to get consent for further evaluation of crash data.

A stratified group sample was designed to select and categorize vehicles based on the initial degree of medical treatment required by the child passengers (i.e., no treatment required, physician's office or emergency department only, admitted to the hospital, or death), and whether the vehicle was drivable. All child occupants of a selected vehicle were included in the survey. All selected vehicles with at least one injured child passenger and a 10 percent random sample of vehicles without child injuries were contacted for a full telephone interview. The 30-minute survey included the parents of the involved children (or the driver if he or she was not the parent).

During the study, 97 percent of eligible claims were accurately identified; 80 percent of policyholders consented to participate in the study. Of those who consented, 79 percent were successfully contacted and interviewed, comprising 52 percent of all eligible persons. Injuries were categorized by severity using the Abbreviated Injury Scale (AIS). Only the more severe injuries (those with an AIS score of 2+), including concussions and more serious brain injuries, internal organ and spinal cord injuries, and extremity fractures, were included in the analyses. Only children four to eight years of age who were seated in the backseat and restrained by a booster seat or seat belt were included. Booster seats were further identified by high-back or backless models.

Results: During the nine-year study, the authors interviewed a representative sample of 34,732 children involved in 21,943 crashes, and 7,151 children in 6,591 crashes met inclusion criteria. A total of 70 percent were restrained by seat belts, and 30 percent used booster seats. In both groups the ages were closely divided among the four- to eight-year range, although children in seat belts were more likely to be heavier and older than those in booster seats.

The overall risk of serious injury was 1.15 percent for children in this age group. Children in booster seats had about one half the risk of those in seat belts, which persisted when adjusted for child age, weight, seating position, driver restraint and relationship to the child, crash severity, and car model year and crash year. Regardless of the type of restraint used, head injuries accounted for approximately 65 percent of all injuries; facial injuries accounted for approximately 9 percent. In children in booster seats, lower extremity injuries accounted for 8 percent of injuries, whereas abdominal injuries constituted 12 percent of injuries in those wearing seat belts. Among children with multiple injuries, there was not a significant difference in injury rates between high-back and backless styles of booster seats.

Conclusion: The authors conclude that the use of belt-positioning booster seats reduces the risk of serious injury by 45 percent compared with seat belt use alone in children four to eight years of age. Physicians and health educators should continue to recommend booster seat use for children until at least eight years of age.