Restraint Use by Children Involved in Crashes in Hawaii, 1986–1991

The use of child safety seats for crash-involved children in Hawaii from 1986 through 1991 is described. Child safety seat use increased over this period, whereas unrestrained children and belt use for children decreased. Male drivers are more likely to have unrestrained infants; female drivers are more likely to have unrestrained toddlers. Toddlers are more likely to be unrestrained in automobiles and trucks, and infants are more likely to be unrestrained in vans, on the freeway, during nighttime hours, and in urban areas. Child-restraint use for infants is twice the rate for toddlers, and infants are less likely to suffer nonincapacitating, incapacitating, and fatal injuries. A logistic regression model shows that children riding in automobiles are less likely to be restrained; belted drivers are far more likely to restrain children; and one- and two-year-olds are less likely to be restrained. Separate logistic regressions for crash types reveals that restrained children are less likely to suffer a non-incapacitating, incapacitating, or fatal injury in head-on and rear-end crashes and more likely to sustain injuries in broadside crashes.