An evaluation of the National Highway Traffic Safety Sdministration's 'Safe Communities' model of traffic injury prevention: A case study of the Northwest Oak Cliff Community of Dallas, Texas.

摘要

The purpose of this study was to determine the effect of the Northwest Oak Cliff Safe Communities Demonstration Program on restraint use and traffic-related morbidity and mortality in the target community. The target population of the study consisted of residents of the Northwest Oak Cliff Community while the comparison population was residents of the City of Dallas, in the years 1996 and 1998.; This study utilized a simple before-and-after design with a yoked comparison to determine statistically significant changes in restraint use, driving under the influence (DUI) of alcohol and traffic-related morbidity and mortality in the target community. A logistic regression analysis was utilized to determine the effect of age, race, restraint use, DUI and gender, on severe injury in the target community before and after implementation of the Program. A second purpose of the study was to examine the efficacy of three accident and injury data sources. Secondary data from four different accident/injury databases (Emergency Medical Services, Medical Examiner, Trauma Registry and a Linked database) were analyzed for the study.; The major findings were: (1) A statistically significant increase in restraint use by the target population. (2) A non-statistically significant decrease in total injuries in the target community. (3) A non-statistically significant decrease of ‘'moderate’ injuries in the target community. (4) The EMS database indicated a non-statistically significant increase in severe injuries in the target community, while the Linked database indicated a non-significant decrease and the Trauma Registry indicated no change. (5) Comparative analysis of the three injury databases revealed a wide variation in the total frequency of injuries captured by each. The Linked database captured the greatest frequency while the Trauma Registry captured the smallest. All three databases indicated a non-statistically significant decrease in the number of injuries in the target community. (6) Results of DUI frequency analyses were inconclusive because they were suspect to non-response error introduced by the large frequency of missing cases (blood alcohol test was not conducted). (7) Logistic regression analyses revealed that before implementation of the program, age group and restraint use were the strongest predictors of severe injury in the target community. After program implementation, age group was still a strong predictor while DUI was the other, indicating that the program was successful in increasing restraint use.