Police and health authorities play an essential role in road traffic injury (RTIs) surveillance, prevention and control. Appropriate reporting of RTI by both authorities can help in adapting local injury prevention policies. The RTI reporting from different sources is seldom compared in low- and middle-income (LMICs). This study aimed to compare police- and health authority-based surveillance systems of RTI in Mongolia. The study setting was Ulaanbaatar, the capital city of Mongolia accounting for 43% of the Mongolian population. This descriptive study combined interviews with key informants (n=30), statistical reports and direct observations. Surveillance data from 2008-2010 was analysed using frequency tables. RTI data collection and reporting were compared to the standards mentioned in the “World Health Organization (WHO) injury surveillance guidelines” (Holder et al., 2001) and “Data systems: a road safety manual for decision-makers: (WHO, 2010a). Police and health authorities independently collected RTI data in Ulaanbaatar. Police records showed that both RTI (n=394 in 2008 and n=564 in 2010) and the fatalities (N=59 in 2008 to N=127 in 2010) doubled from 2008 to 2010. The health authority’s data showed that RTIs increased slightly from 2008 (n=10 174) to 2010 (n=11 157) whereas the fatalities remained almost unchanged over the same period (n=211 in 2008 and n=207 in 2010). Comparisons of RTI and fatality counts between the two sources showed that for every police reported RTI, health authority reported 19.7 RTIs. Similarly, for every RTI fatality reported by police, health authorities reported 2.1 RTI fatalities. At least six of the eight WHO recommended variables were available in both datasets. The individual road user type and intent were unavailable in police dataset. Health authority dataset lacked information about RTI location and activity before crash. In police dataset, the description of RTI depended on police officer as there was no formal definition. This was the case as well in health facilities except hospitals which used ICD codes. We concluded that police underreported RTI and fatalities compared to health authorities, and therefore information about local RTI factors was not adequately collected and used by the current system. Linking both datasets seemed feasible because of identifiers and other comparable information available in both of them. This could increase the utility of RTI information captured by both authorities for making prevention policies. Nevertheless, both surveillance systems might require improvements in reporting of crash circumstances to assist future preventive efforts. (Author/publisher)
Abstract