Temporal trends in maternal mortality in Canada II: estimates based on hospitalization data

Abstract

Objectives: World Health Organization reports based on Canadian Vital Statistics data suggest a recent increase in maternal mortality rates in Canada. Since Vital Statistics data typically provide inaccurate estimates of maternal mortality, we examined temporal trends in Canada using hospitalization data.

Methods: We identified in-hospital deaths among women aged 15 to 54 years from the Canadian Institute for Health Information's hospitalization database from 1996-1997 to 2007-2008. Maternal deaths during delivery were identified, and other in-hospital deaths were linked with prior pregnancy/delivery hospitalization records. Maternal mortality rates, 95% confidence intervals, and risk ratios (RRs) were estimated.

Results: The maternal mortality rate in Canada was 9.2 per 100 000 deliveries (95% CI 7.6 to 11.2) in 1996 to 1999 and 9.0 per 100 000 deliveries (95% CI 7.4 to 10.9) in 2005 to 2007 (P for trend = 0.22). Older maternal age (RR 9.9 and 3.1 for ≥ 45 years and 40 to 44 years vs. 20 to 24 years), delivery by Caesarean section (RR 4.5 vs. vaginal delivery), and early gestation delivery (RR 20.1 and 5.9 for 20 to 27 weeks and 28 to 36 weeks vs. ≥ 37 weeks) were associated with higher maternal mortality. Cardiovascular diseases (rate 4.7 per 100 000 deliveries, 95% CI 3.9 to 5.5) were the most common diagnoses associated with maternal death. The rate of late maternal death (from 43 to 365 days after delivery) was 5.4 per 100 000 deliveries.

Conclusion: There was no increase in maternal mortality in Canada from 1996 to 2007. Increases observed in Canadian Vital Statistics data likely reflect improvements in the ascertainment of maternal death. Hospitalization data can serve as a timely and comprehensive source for monitoring trends in maternal mortality in Canada.