Background: Asthma is associated with several adverse pregnancy and perinatal outcomes. Familial factors may confound these associations.
Objective: To examine the role of measured and unmeasured confounding by conducting a study that compared differentially exposed cousins and siblings from the same families.
Methods: We retrieved data on adverse pregnancy outcomes, prescribed drugs, and physician-diagnosed asthma from nationwide registers for all women in Sweden with singleton births between 2001 and 2013. Logistic and linear regression estimated the association between maternal asthma and several outcomes in the whole population and within differently exposed pregnant relatives.
Results: In total, 1,075,153 eligible pregnancies were included and 10.1% of the study population had asthma. We identified 475,200 cousin and 341,205 sister pregnancies. Women with asthma had increased risks for preeclampsia (adjusted odds ratio [aOR], 1.17; 95% CI, 1.13-1.21), emergency cesarean section (aOR, 1.24; 95% CI, 1.22-1.27), and having a child small for gestational age (aOR, 1.18; 95% CI, 1.12-1.23). In the conditional regression analyses, after adjustment for familial factors, the associations remained: preeclampsia in cousins (aOR, 1.16; 95% CI, 1.07-1.25) and siblings (aOR, 1.23; 95% CI, 1.08-1.38), emergency cesarean section in cousins (aOR, 1.28) and siblings (aOR, 1.21), and small for gestational age in cousins (aOR, 1.17) and siblings (aOR, 1.13).
Conclusions: Factors shared by siblings and cousins do not seem to explain the observed association between maternal asthma and adverse pregnancy outcomes. This implies that targeting the asthma disease will continue to be important in reducing risks for adverse outcomes in pregnancy.
Keywords: Asthma; Epidemiology; Family design; Pregnancy; Pregnancy outcomes.
Copyright © 2017 The Authors. Published by Elsevier Inc. All rights reserved.