Assessment of maternal smoking status during pregnancy and the associations with neonatal outcomes
Introduction: Single assessment of smoking during pregnancy may lead to misclassification due to underreporting or failure of smoking cessation. We examined the percentage of mothers who were misclassified in smoking status based on single assessment, as compared with repeated assessment, and whether this misclassification leads to altered effect estimates for the associations between maternal smoking and neonatal complications. Methods: This study was performed in 5,389 mothers participating in a prospective population-based cohort study in the Netherlands. Smoking status was assessed 3 times during pregnancy using questionnaires. Information on birth weight and neonatal complications was obtained from hospital records. Results: For categorizing mothers per smoking status, Cohen's Kappa coefficient was .86 (p < .001) between single and repeated assessments. Of all mothers who reported nonsmoking or first trimester-only smoking in early pregnancy, 1.7% (70 of 4,141) and 33.7% (217 of 643), respectively, were reclassified to continued smoking based on repeated assessment. Younger, shorter lower educated mothers who had non-European ethnicity experienced more stress, consumed more alcohol, and did not use folic acid supplements had higher risk of underreporting their smoking status or failure of smoking cessation. Marginal differences were found on the associations of maternal smoking with neonatal complications between single or repeated assessment. Conclusions: Our results suggest that single assessment of smoking during pregnancy leads to underestimation of the continued smoking prevalence, especially among mothers who reported quitting smoking in first trimester. However, this underestimation does not materially change the effect estimates for the associations between maternal smoking and neonatal outcomes.