USe of cAptURe-RecAptURe Method to evAlUAte deAthS of peRSonS AdMitted At hoSpitAl AfteR A fiRe

Abstract

introduction: In France, the epidemiological surveillance of fire related mortality is carried out mainly by using three data sources: mortality data of the deaths certificates produced by the Centre d’épidémiologie sur les causes médicales de décès (CépiDc) (476 deaths in 2008), hospitalisation data of the hospital discharges collected by the Programme de médicalisation des systèmes d’information (PMSI) (194 deaths) and firemen notifications (378 deaths). The objective was to estimate the number of fire related deaths at hospital using the capture-recapture method. Method: The capture-recapture method was applied to two sources: CépiDc and PMSI, as firemen data, that were not comprehensive, could not be crossed with the others. Therefore, the study was reduced to deaths at hospital. Principles and conditions of capture-recapture method were considered. Results: Using the capture-recapture method, the number of fire-related deaths at hospital in mainland France in 2007 was estimated to 252 (CI = [134-370]). The two joint sources showed a 97 % (244/252) exhaustivity rate, 85 % (168/198) for the CépiDc and 79 % (168/214) for the PMSI. discussion: Six criteria should be met when using the capture-recapture method: true cases, independent sources, homogenous capture, closed population, same time period and geographical area. The automated process that was settled for this occasion can be repeated every year. Its results will be useful for the epidemiological surveillance.