Prepartum and postpartum predictors of smoking

Abstract

Introduction: Reduction of smoking during pregnancy and in the postpartum period remains a public health priority. This study tested whether variables (demographic, pregnancy related, emotional, and smoking related) assessed in abstinent women during their pregnancy predicted resumed smoking at 1 month and 1 year postpartum. Additionally, pregnancy-related and smoking-related variables obtained in the immediate postpartum period were examined as predictors of smoking at 1 year postpartum.

Methods: Participants were pregnant women in their second and third trimester (N = 504) who enrolled in a randomized controlled trial of a smoking relapse-prevention intervention. Multivariate regression analyses were conducted with baseline data collected during pregnancy and follow-up assessments completed at 1 month and 12 months postpartum.

Results: Independent predictors of smoking at 1 month postpartum included not intending to remain abstinent, lower quitting confidence, the presence of other household smokers, and not planning to breast feed. Smoking at 12 months postpartum was predicted by only not intending to remain abstinent and having a partner who maintained his or her smoking rate. In contrast, when assessed at 1 month postpartum, smoking at 12 months was predicted by current smoking status and lower quitting confidence.

Conclusions: Baseline variables measured during pregnancy differentially predicted early versus late smoking status. After delivery, the best predictor of later smoking status was current smoking status, reinforcing the need to focus on preventing early postpartum relapse. The importance of quitting confidence, or self-efficacy, was also reinforced. Findings may be useful for screening women at risk of relapse and targeting interventions to key variables.

Figure 1.

Change in smoking status over time. Values based on the first of 10 datasets created by multiple imputations.