Deaths from complications of unsafe abortion: misclassified second trimester deaths

This study measured the contribution of abortion-related deaths to overall maternal mortality and calculated the underestimation of maternal mortality using verbal autopsy and clinical record review where available. We reviewed 807 death certificates of women aged 12–50 who died in 2001 in two sites of about 1.5 million inhabitants each in the state of Morelos (primarily rural) and the municipality of Nezahualcoyotl (primarily urban) in the state of Mexico. Deaths were classified as definite, possible or non-maternal deaths. Finally, we identified abortion-related deaths and calculated the underestimation of maternal mortality. Among 326 possible maternal deaths, we encountered five misclassified cases: one spontaneous abortion and four non-abortion maternal deaths. Among 32 registered maternal deaths, we found four misclassified cases that were actually second trimester, abortion-related deaths. There were no officially registered abortion-related deaths in either Morelos or Nezahualcoyotl, making the overall underestimation of abortion mortality 100%. Abortion contributed 13.5% of all maternal deaths. The overall underestimation of maternal mortality was 13.5%, higher in Morelos (21.7%). There were no unregistered maternal deaths in Nezahualcoyotl. Unsafe abortion continues to be an important cause of maternal mortality, though first trimester deaths appear to be decreasing. We identified domestic violence as an important cause of death among pregnant and post-partum women, and two abortion-related suicides, and believe these should be reconsidered as indirect maternal deaths. The misclassification of second trimester abortion deaths as maternal deaths from other causes is an obstacle to preventing them.