Abstract
INTRODUCTION: Little is known about the effectiveness of cigarette health warning labels (HWLs) among U.S. pregnant and non-pregnant women of reproductive age. Our intent was to examine HWL responses by pregnancy status and their associations with knowledge and perceived harm. METHODS: We conducted cross-sectional analyses of first (2013-2014) and second (2014-2015) Population Assessment of Tobacco and Health waves. The sample (N = 19,095) was representative of U.S. women of reproductive age (18-45 years). We examined three HWL responses (seeing HWLs, forgoing cigarettes because of HWLs, likely-to-quit because of HWLs), perceived harm from smoking, and knowledge of health effects to fetuses and others. Weighted logistic regression compared HWL responses by pregnancy status and their associations with perceived harm and knowledge, adjusting for demographics and tobacco use. Results were presented as marginal predicted probabilities. RESULTS: Pregnant smokers were less likely than never-pregnant smokers to report likelihood of quitting (57% vs 67%, p = 0.020). Forgoing cigarettes and likely-to-quit due to HWLs-but not seeing HWLs-were positively associated with perceived harm and knowledge. Pregnant smokers were less likely to have knowledge of fetal harm (83%) than ever-pregnant (91%, p = 0.006) and never-pregnant (92%, p = 0.003) smokers. However, pregnant smokers who reported likelihood of quitting were much more likely to have knowledge of fetal harm than those who did not (93% vs. 67%, p = 0.028). CONCLUSIONS: Implementation of HWLs that elicit stronger reactions-such as pictorial HWLs-and increase knowledge of fetal risks is one action to help reduce tobacco use in pregnant and non-pregnant U.S. women of reproductive age.