Severe Maternal Morbidity in Canada: Temporal Trends and Regional Variations, 2003-2016

OBJECTIVE: This study sought to quantify temporal trends and provincial and territorial variations in severe maternal morbidity (SMM) in Canada. METHODS: The study used data on all hospital deliveries in Canada (excluding Québec) from 2003 to 2016 to examine temporal trends and from 2012 to 2016 to study regional variations. SMM was identified using diagnosis and intervention codes. Contrasts among periods and regions were quantified using rate ratios (RRs) and 95% confidence intervals (CIs). Temporal changes were also assessed using chi-square tests for trend (Canadian Task Force Classification II-1). RESULTS: The study population included 3 882 790 deliveries between 2003 and 2016 and 1 418 545 deliveries between 2012 and 2016. Severe hemorrhage rates increased from 44.8 in 2003 to 62.4 per 10 000 deliveries in 2012 (P for trend <0.0001) and then declined to 41.8 per 10 000 deliveries in 2016 (P for trend <0.0001). Maternal intensive care unit admission and sepsis rates decreased between 2003 and 2016, whereas rates of stroke, severe uterine rupture, hysterectomy, obstetric embolism, shock, and assisted ventilation increased. Rates of composite SMM in 2012-2016 were higher in Newfoundland and Labrador (RR 1.15; 95% CI 1.04-1.26), Nova Scotia (RR 1.11; 95% CI 1.03-1.19), New Brunswick (RR1.22; 95% CI 1.13-1.32), Manitoba (RR 1.09; 95% CI 1.03-1.15), Saskatchewan (RR 1.15; 95% CI 1.09-1.22), the Yukon (RR 1.74; 95% CI 1.35-2.25), and Nunavut (RR 1.76; 95% CI 1.46-2.11) compared with the rest of Canada, whereas rates were lower in Alberta and British Columbia. CONCLUSION: This surveillance report helps inform clinical practice and public health policy for improving maternal health in Canada.