• Maternal mortality;
  • nationality;
  • quality of care

Objectives  To test the hypothesis that the risk of postpartum maternal death in France remains significantly higher for women of foreign nationality after individual characteristics are taken into account and to examine whether the quality of care received by the women who died differs according to nationality.

Design  A national case–control study.

Setting  Metropolitan France.

Population  A total of 267 women who died of maternal death from 1996 to 2001 as cases and a representative sample (n = 13 186) of women who gave birth in 1998 as controls.

Methods  Crude and adjusted odd ratios were calculated with multivariate logistic regression, and the quality of care for cases was compared according to nationality with chi-square tests or Fisher’s exact tests.

Main outcome measures  Odd ratio for postpartum maternal death associated with nationality and quality of care.

Results  After taking individual characteristics into account, the risk of postpartum maternal death was twice as high for foreign women. The odds ratio was 5.5 (95% CI: 3.3–9.0) for women from sub-Saharan Africa and 3.3 (95% CI: 1.7–6.5) for those from Asia, North and South America. There was no significant excess risk of postpartum maternal death for the other European and North Africa women. The risk of dying from hypertensive disorder or infection was four times higher for foreign women. Among women who died, care was more often considered not optimal for foreign women (78 versus 57%).

Conclusions  The excess risk of postpartum maternal death persisted for foreign women after individual characteristics were taken into account and was especially important for some nationalities and for some causes of death, primarily hypertensive disorders. These results point to an immediate need to pay special attention to early enrolment in prenatal care, screening and prenatal management of hypertension, especially in women of sub-Saharan African nationality.