Abstract

Contact: [email protected] Background: Morbidity and mortality among very preterm births have been the focus of several studies of prenatal determinants obtaining unexpected results. We aimed to compare perinatal death rates among very preterm births (2431 weeks) with those based on all births (≥24 weeks), using birthsbased and fetusesatrisk rate calculations. Methods: We conducted a cohort study of singleton live births and stillbirths ≥24 weeks’ gestation using administrative data from the United States and Canada (20062015). We compared rates of perinatal death based on traditional vs fetusesatrisk calculations, among births to hypertensive vs nonhypertensive women in the United States, and births in Canada vs the United States. Results: Birthsbased perinatal death rates among births to hypertensive women at 2431 weeks were lower than the same rates among nonhypertensive women (rate ratio [RR] 0.65, 95% confidence interval [CI] 0.630.66). However, overall death rates were higher among births to hypertensive vs nonhypertensive women (RR 2.04, 95% CI 2.012.07), as were death rates based on fetuses at risk at 2431 weeks (RR 2.55, 95% CI 2.492.60). Birthbased perinatal death rates among births at 2431 weeks in Canada were higher than the same rates in the United States (RR 1.09, 95% CI 1.061.13), whereas overall perinatal death rates (RR 0.91, 95% CI 0.900.93) and fetusesatrisk perinatal death rates at 2431 weeks(RR 0.79, 95% CI 0.770.82) were lower in Canada. Conclusion: Studies of perinatal risk factors and betweencentre and betweencountry comparisons of perinatal mortality lead to biased inferences when restricted to births at very preterm gestation.