Estimation of the completeness of road traffic mortality data in Zambia using a three source capture recapture method.

BACKGROUND: Accurate and reliable data are essential for tracking progress and evaluating the effectiveness of road safety intervention measures. However, in many low- and medium-income countries, good quality data on road traffic crashes are often difficult to obtain. This situation has led to an underestimation of the severity of the problem and distortions in trends when the reporting changes over time. This study estimates the completeness of road traffic crash fatality data in Zambia. METHODS: Data from the police, hospitals, and the civil registration and vital statistics (CRVS) databases was collected for the period 1st January to 31st December 2020 and analyzed using a three-source capture-recapture technique. RESULTS: A total of 666 unique records on mortalities as a result of road traffic crashes were collected from the three data sources during the period under review. The capture-recapture technique estimated the completeness of police, hospital, and CRVS databases to be 19%, 11% and 14% respectively. The combination of the three data sets was found to increase completeness to 37%. Based on this completion rate, we estimate that the actual number of people who died as a result of road traffic crashes in Lusaka Province in the year 2020 was approximately 1,786 (95% CI [1,448-2,274]). This corresponds to an estimated mortality rate of around 53 deaths per 100,000 population. CONCLUSIONS: There is no single database contains complete data to provide a comprehensive picture of Lusaka province and by extension the country's road traffic injury burden. This study has shown how capture and recapture method can address this problem. It shows the need for the continuous review of the data collection processes and procedures in order to identify gaps and bottlenecks, improve efficiency, and increase the quality and completeness of road traffic data on injuries and fatalities. Based on the findings of this study, it is recommended that the city of Lusaka province and Zambia as a whole utilize more than one database for official reporting of road traffic fatalities to increase completeness.

Bold actions needed to improve civil registration data for road traffic deaths.

Persisting problems of underreporting and poor quality of road traffic mortality data exist. Bold and sustained actions need to be undertaken by countries to improve civil registration data collection and reporting. Countries need to implement three specific actions to improve civil registration and vital statistics data systems and/or road traffic injury data collection:Ensure that civil registration and vital statistics systems produce high-quality cause of death data by working with the health sector to improve medical certification of cause of death and statistical coding in line with the International Classification of Diseases.Adopt a consistent definition of a road traffic death for use in police databases, particularly in countries with incomplete civil registration and vital statistics data, where police data are the most reliable source of informationLink data sources - including civil registration records, police data, health records, insurance data - to improve official road traffic fatality estimates while ensuring the privacy of individual identification.