Cigarette smoking during pregnancy presents significant health risks to both the mother and infant. Among pregnant women, the highest rates of smoking occur among teens 15-19 years old, yet pregnant adolescents are likely to underrepresent their smoking. Self-report, exhaled carbon monoxide, and saliva cotinine levels are all indicators used to determine the status of cigarette smoking in adolescents. The best, yet most cost-effective method to evaluate smoking status in pregnant adolescents needs to be identified. The purpose of the current study was to examine the accuracy of self-report and carbon monoxide levels as compared to saliva cotinine as indicators of smoking cessation in pregnant teenagers.
The self-reported smoking behavior, saliva cotinine level, and exhaled breath carbon monoxide of 109 pregnant adolescents who had completed a smoking cessation program were examined. Sensitivity and specificity were calculated to compare the accuracy of self-reported smoking cessation and of the carbon monoxide (CO) measure (using the cut-point of 8 ppm) to the saliva cotinine value of 10 ng/ml or greater at post-intervention. Sensitivity was defined as the proportion of non-smokers detected by self-report or CO measure compared to those detected by the saliva cotinine measures, whereas specificity was the proportion of self-reported smokers and subjects with CO measures of > 8 ppm correctly classified via the cotinine value.
The sensitivity of self-reported smoking cessation when compared to saliva cotinine levels was calculated at 62.9%. Specificity was calculated at 86.5%. The sensitivity of carbon monoxide reading compared to saliva cotinine was 93.5%. Specificity was calculated at 38.0%. These findings indicate that self-report is a better indicator of continued smoking (has better specificity), while carbon monoxide is a better indicator of smoking cessation (has better sensitivity). Racial group (African American versus non-African American) and treatment group comparisons were also conducted.