The potential of ambulance records for a road traffic safety agenda in low-income cities : Studies from Karachi, Pakistan

Abstract: Road traffic injuries (RTIs) are a major public health problem in low-income countries. The overall aim of the thesis is to increase knowledge about the characteristics of RTI events and victims and their changes over time in an urban, low-income setting. The accuracy of ambulance data for RTI surveillance is also investigated. All studies forming the thesis are register-based and were conducted in Karachi (Pakistan). Data collections took place both in 1996-1997 (reference period 1993-95) and in 2003-2004 (reference period 2003). Common to all studies is the use of Edhi Ambulance Service (EAS) logbooks. In study 1, the epidemiology of childhood RTIs was analysed. Study 2 assessed the magnitude of RTIs for all age groups. EAS logbook data were matched with traffic' police OBS data so as to compile incidence estimates of RTI mortality and morbidity using a two-sample capture-recapture design. Study 3 dealt with the accuracy (validity and quality) of EAS data. A comparison was made of the data points of individual patients with hospital and medico-legal department data. In Study 4 , injury data from ambulance records were compared for two one-year periods (1995 and 2003); significance of differences between years were measured using proportions of injuries. For the period 1993-95, children aged 0-15 incurred 21% of the total number of RTIs recorded. 15% of the fatally injured children died either at the scene of the accident or during transportation to the hospital. Buses, minibuses and trucks were involved in 54% of all cases. For the year 1994, police reported 544 deaths and 793 injuries while ambulance records noted 343 deaths and 2048 injuries due to road traffic crashes. It was estimated that at least 972 (CI=912, 1031) deaths and 18936 (CI=15507, 22342) injuries occurred during the study period. Traffic police records recorded only 56% of deaths and 4 % of serious injuries, while the ambulance data counted only 35% of deaths and 11% of serious injuries. About ten years later, a total of 1,245 adult RTI victims were transported by the EAS during 2003, most of which were males (89.6%). Comparing 1995 and 2003 showed a 41% lower number of RTIs in 2003 compared to 1995, with a 35% increase in the prehos ital mortality. The road users most likely to be injured remained similar. There were missing entries in EAS records in about one in four cases (27%), the majority of which were the unique identifiers like name and age (67%). Data on other variables were reported in 95% or more cases. The agreement rate for the variables available in the three datasets ranged from 61% to 100%. In sum, the burden of RTIs in Karachi is much larger than would be expected if only ambulance or police data alone were used to generate incidence rates. While many victims and events characteristics remain unchanged between the two periods, an augmentation in case mortality rates occurs. Further, while there are limitations to all three major sources of road traffic injury information in the city, ambulance records provide as accurate and more detailed information than either the medico legal reports or the hospital records.