Alcohol use, smoking and their co-occurrence during pregnancy among Canadian women, 2003 to 2011/12

Introduction: The co-occurrence of alcohol use and smoking during pregnancy has been shown to have a negative synergistic effect on fetal and perinatal risks. The objectives were to: 1) obtain an estimate of the prevalence of smoking during pregnancy in Canada by province and territory from 2003 to 2011/12; 2) determine if the prevalence of smoking during pregnancy has increased or decreased over time; 3) investigate whether smoking status is differentially associated with alcohol use during pregnancy; and 4) examine the risk factors predictive of alcohol use only, smoking only, and the co-occurrence of alcohol use and smoking during pregnancy. Methods: Secondary data analysis was conducted using five cycles of the Canadian Community Health Survey (CCHS; 2003, 2005, 2007/08, 2009/10 and 2011/12). The prevalence of smoking during pregnancy, and 95% confidence interval (CI) was calculated by province and territory and by year. The likelihood ratio test was used to determine if the prevalence of smoking during pregnancy has increased or decreased over time. The relationship between smoking status and alcohol use during pregnancy was explored using a quasi-Poisson regression model. A mulfinomial logistic regression model was utilized to determine which factors were predictive of alcohol use only, smoking only, and the co-occurrence of alcohol use and smoking during pregnancy. Results: In Canada, between 2003 and 2011/12, the weighted pooled prevalence of smoking during pregnancy was 143% (95% Cl: 13.6%-15.0%). Women who smoked daily during pregnancy, occasionally during pregnancy, or had a lifetime history of smoking (but did not smoke while pregnant) were 2.54 (95% CI: 2.11-3.06, P < 0.0001), 2.71 (95% CI: 2.25-327, P < 0.0001), and 2.09 (95% CI: 1.85-237, P < 0.0001), respectively, times more likely to have consumed alcohol during pregnancy, compared to pregnant women who were lifetime non-smokers when controlling for age, household income, ethnicity and CCHS cycle. Risk factors that predicted alcohol use only, smoking only, and the co-occurrence of alcohol use and smoking during pregnancy differed. Conclusion: It is apparent that smoking in any capacity, whether during pregnancy or not, increases the likelihood that a woman consumed alcohol while pregnant. Ascertaining smoking status among pregnant women and women of childbearing age could be a useful screening method for identifying those at-risk of consuming alcohol during pregnancy, and vice versa. (C) 2015 Elsevier Ltd. All rights reserved.