[Transmission of death certificates to CepiDc-Inserm related to suspicious deaths, in France, since 2000].

Background: The purpose of this study is to assess variability in death certificates' transmission related to suspicious deaths.

Methods: The medical part of death certificates of suspicious deaths (violent or sudden deaths, unknown or ill-defined causes of death) were examined. We studied the frequency of suspicious deaths, in France, for deaths aged under 65, from 2000 to 2013, searching for temporal (2000-2013) and spatial correlations between unknown causes of death and other suspicious causes, and report of an autopsy. These results were compared with external data.

Results: Standardized proportion of deaths by unknown cause more than doubled during the study period (from 3.4 to 7.5%). The spatial correlation was strongly negative between standardized proportions of unknown causes of death and violent deaths (ICC=-0.80). Report of autopsy varied with cause of suspicious death and estimated zone of the forensic institute. The distribution was consistently distributed with external data, except for homicides.

Conclusion: Information transmission to CépiDc-Inserm needs to be improved in case of suspicious death, in particularly from forensic institutes. This study emphasizes the urgent need for a complementary form to the death certificate.