Introduction - A number of suicides are not captured in the national death statistics. In particular, some are recorded as deaths from unknown causes or violent deaths where intent is undetermined. The underreporting of suicide deaths was estimated at 20% in the late 1990s. Given changes in certification and coding methods, this study presents an analysis of the quality and completeness of suicide mortality data in metropolitan France, updated with more recent data, and offers recommendations for improving the national statistics on suicide deaths. Method - The study is based on all deaths that occurred in metropolitan France during 2006. Using a random sample of 500 deaths involving causes that could “mask” suicides, a retrospective survey of certifying physicians made it possible to assess the proportion of suicides among these causes and to propose a corrected estimate of the number and rate of deaths in the national statistics. Results—In 2006, a total of 10,423 deaths by suicide were recorded in metropolitan France. An investigation into groups of causes of death that could “mask” a suicide identified a number of suicide deaths and concluded that the official death toll was underestimated by 9.4%. After correction, the number of suicides rose from 7,593 to 8,250 (+8.7%) among men and from 2,830 to 3,154 (+11.5%) among women. The regional situation regarding suicide has changed little after taking into account the corrected standardized death rates. Conclusion—The study highlights the importance of obtaining comprehensive information on the medical causes of death, particularly following a forensic investigation. Furthermore, information on the circumstances of violent deaths is also essential for obtaining reliable statistics, especially for an indicator as sensitive as suicide. Recommendations involving collaboration between the Ministries of Health and Justice and the CépiDc, as well as changes to the death certificate, are proposed. (R.A.)