Key points are not available for this paper at this time.
Background: Most studies of disparities in birth and postnatal outcomes by parental birthplace combine all immigrants into a single group. We sought to evaluate heterogeneity among immigrants in Canada by comparing birth and postnatal outcomes across different immigration categories. Methods: We conducted a population-based retrospective study using Statistics Canada data on live births and stillbirths (1993–2017) and infant deaths (1993–2018), linked to parental immigration data (1960–2017). We classified birthing parents as born in Canada, economic-class immigrants, family-class immigrants, or refugees, and evaluated differences in preterm births, small-for-gestational-age (SGA) and large-for-gestational-age (LGA) births, stillbirths, and infant deaths among singleton births by group. Results: Among 7 980 650 births, 1 715 050 (21.5%) were to immigrants, including 632 760 (36.9%) in the economic class, 853 540 (49.8%) in the family class, and 228 740 (13.4%) refugees. Compared with infants of Canadian-born birthing parents, infants of each of the 3 immigrant groups had higher risk of preterm birth, SGA birth, and stillbirth, but lower risk of LGA birth and neonatal death. Compared with infants of economic-class immigrants, infants of refugees had higher risk of early preterm birth (0.9% v. 0.8%, adjusted risk ratio RR 1.08, 95% confidence interval CI 1.01–1.15) and LGA birth (9.2% v. 7.5%, adjusted RR 1.12, 95% CI 1.10–1.15), but lower risk of SGA birth (10.2% v. 11.0%, adjusted RR 0.92, 95% CI 0.90–0.94), while infants of family-class immigrants had higher risk of SGA birth (12.2% v. 11.0%, adjusted RR 1.01, 95% CI 1.00–1.02). Risk of stillbirth, neonatal death, and overall infant death did not differ significantly among immigrant groups. Interpretation: Heterogeneity exists in outcomes of infants born to immigrants to Canada across immigration categories. These results highlight the importance of disaggregating immigrant populations in studies of health disparities.
Yang et al. (Mon,) studied this question.