Religious Commitment Predicts Substance Use in Pregnant Women

Document Type


Publication Date



Introduction: Substance use, including cigarette smoking, while pregnant can lead to a plethora of health concerns for both the mother and unborn child including premature birth, low birth weight, and stillbirth. Compared with women nationally, pregnant women in Tennessee are more than three times as likely to smoke during pregnancy. Preliminary findings suggest high levels of religious commitment may be reliable predictors of negative health behaviors. However, the association between religious commitment and substance use has not been thoroughly investigated in pregnant populations. Using a brief measure of religious commitment, it was hypothesized that pregnant women with higher levels of religious commitment would be significantly less likely to engage in cigarette smoking and other substance use. Methods: Participants included 654 pregnant women involved in the Tennessee Intervention for Pregnant Smokers program who completed multiple interviews during pregnancy. Of interest in the current investigation, participants’ religious commitment was measured using two items from the 12-item Surrender Scale, and a 1-item church attendance measure from the Brief Multidimensional Measure of Religiousness/Spirituality. Participants also completed a background information form assessing demographic characteristics, smoking habits, and drug use, with final substance use variables composites of both self-report and urine drug screen results. Results: Direct logistic regression was performed to assess associations between religious commitment and both smoking status (at conception and delivery) and other substance use. All models included level of education, age, marital status, and insurance status. The full direct model predicting smoking status at conception was statistically significant, χ2 (5, n = 654) = 178.76, p < .001, indicating the model could distinguish between participants who did and did not report smoking early in pregnancy. The model as a whole explained between 24% and 32% of the variance in smoking status, and correctly classified 71% of cases. All variables made statistically significant and unique contributions to the model, including religious commitment (OR=.857). A similar pattern was found in the model predicting smoking status at delivery χ2 = 157.01, p < .001. A third regression, using the same predictors, examining the impact of religious commitment on any illicit drug use prior to or during pregnancy, was also statistically significant, χ2 = 58.46, p < .001. Conclusions and Implications: In this sample, religious commitment predicted smoking status and other drug use during and prior to pregnancy. Inquiry into religious commitment as an additional gauge of health behaviors may be beneficial to healthcare professionals. Future research should investigate the possible mechanism of how religious commitment influences health behaviors in pregnancy.


Johnson City, TN

This document is currently not available here.