oa Data linkage: An untapped resource for reducing serious traffic injuries in fast developing countries.

Reliable high quality traffic safety data has been essential for monitoring, developing and evaluating strategies to reduce the burden of injuries from traffic injuries. Information is often available in many different sources including police reports, ambulance/emergency medical service (EMS) run data, emergency department(ED) and hospital data, and mortality data from death certificates or medical examiner/coroner systems. However while each of these data systems has their own strengths and weakness, none can provide a comprehensive picture of traffic related injuries. Hospital trauma center records for example have extensive information on injuries and related medical problems but lack details on the mechanism of injury and crash and roadway characteristics. Often medical data cannot even distinguish drivers from pedestrians. Police or insurance crash reports on the other hand have considerable detail on the crash and circumstances surrounding it but have very poor data on the actual injuries sustained or their severity. This impedes targeting of prevention strategies towards serious injuries and fatalities. This presentation will use examples from current work on data linkage to demonstrate the value of linking traffic record data and explore how this approach can greatly advance traffic medicine and road safety in fast developing countries. In the United States linkage of crash data and medical data has been important for many traffic safety programs. One example is Maryland’s Crash Outcome Data Evaluation System (CODES) that uses probabilistic methodology to link police crash records to injury outcome records. CODES was established in an effort to improve highway safety through the use of linked motor vehicle crash data collected from police, EMS, hospitals (emergency department or inpatient admissions), and death certificates. This linked data have been used for a variety of statewide studies such as of injured motorcyclists, effectiveness of seat belts at preventing injuries, studies of older drivers, an analysis of mismatch collisions, geographic analysis of pedestrian injuries, patterns of injury in frontal collisions, costs and consequences of lower extremity injuries, the safety of newer vehicles, and the impact of casino gambling on alcohol-related crashes. A new project seeks to overcome the serious lack of reliable routine data on drug involvement in fatal crashes by linking detailed toxicology data from medical examiners with the Fatality Analysis Reporting System (FARS) in a sentinel group of states. Similar data linkage is also used in other countries such as a New Zealand study estimating underreporting of serious injuries by police reports. An Australian study noted that workers compensation datasets can identify most serious cases of occupational traffic injuries but lack valuable information on the circumstances and risk factors contributing to work-related crashes. Through linkage to police crash reports the study was able to identify high risk groups such as heavy vehicle drivers and the role played by fatigue and speeding. Many fast developing countries also have useful crash reports either from police or insurance reports. However only through linkage with medical data can their full potential for reducing the burden of serous traffic injuries be realized?