Pedestrian injuries to children represent a major urban health problem in the United States. Thousands of children each year are struck by moving motor vehicles; such collisions result in numerous hospitalizations and deaths. At particular risk are young school-age children between the ages of 5 and 9 years. Using a survey methodology, we collected data regarding the method by which children in an urban setting travel to and from school, in addition to the number of streets they cross in a typical school day. This information was compared with data from police records on street intersection locations of pedestrian collisions. There is a wide variation in the number of streets children cross in 1 day, calculated as the number of streets crossed in the entire day, not only those crossed to and from school. Children whose parents own a car and home cross an average of 3.7 streets per day, whereas children whose parents do not own both a car and home cross an average of 5.4 streets per day; this difference is highly significant (P < 0.0001). The largest differences in traffic exposure are between families reporting car- and-home ownership (x = 3.70 streets) versus those who do not own both a car and home (x = 5.39 streets) (Mann-Whitney = -5.5, P < 0.0001). There is a significant correlation between the proportion of children driven home from school and the rate of pedestrian injury in different regions of Baltimore. In areas where children are driven home, rates of pedestrian injury are significantly lower, whereas in areas where children walk home, rates of pedestrian injury are high (r = -0.79, P < 0.01). This study underscores the importance of adapting the child's environment to prevent injury. Interventions that alter the nature of the hazard are indicated. Changing the environment may ultimately prove more useful than attempting to change children's behavior.