LAUSR

Introduction Maternal smoking remains a modifiable cause of adverse maternal and child health outcomes. This study investigated smoking transitions across pregnancy. Methods Data from the contemporary child cohort study Growing Up in New Zealand (n = 6822) were used to analyse smoking status across three points across a pregnancy: pre-pregnancy, during pregnancy and after pregnancy. Odds-ratios (OR) were calculated for maternal, socio-economic and pregnancy-related factors associated with each transition using multivariate logistic regression. Results The prevalence of smoking pre-pregnancy was 20.3%. The cessation rate during pregnancy was 48.5%, while the postpartum relapse rate was 36.0%. Heavy smokers were less likely to quit during pregnancy (OR 0.13, 95% CI 0.08–0.20), and more likely to relapse at 9 months (OR 2.63, CI 1.60–4.32), relative to light smokers. Women in households with another smoker were less likely to quit during pregnancy (OR 0.35, CI 0.25–0.48), and more likely to relapse postpartum (OR 2.00, CI 1.14–3.51), relative to women in a smoke-free household. Women without high school qualifications were less likely to quit during pregnancy than women with bachelor degrees (OR 0.21, CI 0.11–0.41) but no more likely to relapse. Maori women were less likely to quit during pregnancy than European women (OR 0.35, CI 0.25–0.49) but no more likely to relapse. Conclusion Heavy smokers and those with another smoker in the household are at high risk of smoking during pregnancy or relapsing after pregnancy. Decreasing smoking across a pregnancy therefore requires a focus on cessation in all households with heavy smokers of child-bearing age. The association between smoking and ethnicity may be confounded as it not consistent across the pregnancy.