Objective:​ To assess the success of recent outreach activities to promote appropriate child restraint in motor vehicles by examining trends in restraint types used by children under age 9 in 3 large regions of the United States.

Methods:​ Cross-sectional study was conducted of children who were under age 9 and in crashes of insured vehicles in 15 states, with data collected via insurance claims records and a telephone survey. A probability sample of 8730 crashes involving 10 195 children, representing 128 291 crashes involving 149 820 children, was collected between December 1, 1998, and November 30, 2002. Parent report was used to determine restraint type used in the crash. Logistic regression models were used to analyze the secular trend of restraint type use.

Results:​ Overall, for children under age 9, seat belt use decreased significantly from 49% to 36% between 1998 and 2002;​ for 7- and 8-year-olds, from 97% to 92%;​ and for 3- to 6-year-olds, from 63% to 34%. Concurrently, gains were achieved in overall child restraint use from 49% to 63%, for 7- and 8-year-olds, from 2% to 5%;​ and for 3- to 6-year-olds, from 35% to 65%. Child restraint use remains stable for children from birth to 2 years of age (from 97% in 1998 to 98% in 2002). Both the use of child safety seat and belt positioning booster seat increased significantly, whereas shield booster seat use decreased significantly.

Conclusion:​ Although considerable achievements have been realized over a short period of time, substantial inappropriate restraint still remains:​ 62% of children aged 4 to 8 remain inappropriately restrained in adult seat belts. Parents hear safety messages when they are relevant to their children. As a result, sustained efforts about appropriate restraint must continue to maintain and improve the gains achieved in appropriate child restraint use. The additional benefits realized by recent changes in child restraint laws remain to be evaluated.