Tobacco use is a major modifiable risk factor of adverse maternal and fetal health outcomes. Being a migrant has been associated with lower risk of smoking during pregnancy. However, it remains a matter of debate whether such association could be explained by other sociodemographic characteristics or whether such effect remains or increases according to the length of stay in the host country. This study examined the differences in maternal smoking prevalence between native, long-term and recent migrant women in Portugal.

This study was derived from baMBINO, a national project grounded on 32 public maternity units in mainland Portugal aiming to investigate the differences in perinatal healthcare and outcomes among migrant and native women. Recruitment took place during admission for delivery, inviting both native and foreign-born women. Participants included in this analysis (n=1107) were classified according to their country of birth and length of stay in Portugal into: native (Portuguese-born), recent migrant (foreign-born women who spent 10 years or less in Portugal), and long-term migrants (foreign-born women who spent more than 10 years in Portugal). Logistic regression model was fitted to estimate the association between the aforementioned migration statuses and having or not smoked tobacco during pregnancy, taking into account women's age, parity, marital status, level of education, family income, country of origin of women's parents and gestational age at the first prenatal care visit. Adjusted odds ratio (OR) and respective 95% confidence interval (95% CI) were obtained.

A little over half of the study participants were foreign-born and almost 60% of those were recent migrants. Tobacco smoking during pregnancy was evidently more prevalent among native women than among long-term or recent migrant women (14% vs. 8% and 4% respectively; P<0.001). Compared to native women, both long-term and recent migrant women were more likely to be unmarried, have a family income lower than 1000€, and have their first prenatal visit after 12 weeks of gestation. Long-term migrants were more likely to have Portuguese-born parents than recent migrants (26.0% vs. 2.5%; P<0.001). According to the multivariate regression model, having a higher educational level was significantly associated with not smoking during pregnancy (secondary school: OR=0.38; 95% CI: 0.23–0.62; postgraduate education: OR=0.18; 95% CI: 0.09–0.35), opposed to women who have 9 or less years of education. However no association was observed between smoking during pregnancy and maternal age (OR=1.10; 95% CI: 0.60–2.01 and OR=0.95; 95% CI: 0.57–1.58 for women aged less than 25 and 35 or more years, respectively, in comparison with women aged 25–34 years), family income (OR=1.30; 95% CI: 0.77–2.18 for women with higher income), parity (OR=0.68; 95% CI: 0.43–1.09 for multiparous in comparison with primiparous women), gestational age at first prenatal visit (OR=1.12, 95% CI: 0.59–2.11 for women having their first prenatal visit after 12 weeks), having a foreign-born parent (yes vs. no: OR=0.83; 95% CI: 0.43–1.63) and marital status (single vs. married women: OR=0.76; 95% CI: 0.46–1.24). Smoking during pregnancy was significantly less frequent in recent and long-term migrants (OR=0.24; 95% CI: 0.10–0.57, OR=0.52, 95% CI: 0.25–1.09, respectively) compared with native Portuguese women.

Results of this study demonstrated significant disparities in the prevalence of maternal smoking during pregnancy according to the women's place of birth and duration of stay in Portugal. Native-born women had a higher risk of smoking during pregnancy than foreign-born women. The increase in the duration of stay in the host country seemed to have a negative effect on maternal smoking of migrants in Portugal. Smoking cessation health strategies in pregnant women in Portugal need to be improved by taking into account the mentioned variations in smoking behaviors.