Levels of exhaled carbon monoxide measured during an intervention program predict 1-year smoking cessation: a retrospective observational cohort study

Life-long smoking cessation is a critical public health objective, but it is difficult for numerous people. This study aimed to identify the independent predictors of 1-year abstinence in smokers motivated to quit and participating in an intervention program. This 6-year retrospective observational cohort study was conducted in smokers who participated in an intervention program. The exhaled carbon monoxide (CO) was sequentially measured on day 1, 8, 15, and 22 of the intervention program. The primary outcome measure was smoking status at 1 year of follow-up. A total of 162 participants were enrolled and divided into a successful quit group (n = 52) and unsuccessful quit group (n = 110). Using a multivariate logistic regression analysis, we reported that the intention to quit (adjusted odds ratio [AOR] = 1.475, 95% confidence interval [CI] = 1.169–1.862, P-value = 0.001), varenicline use (AOR = 3.199, 95% CI = 1.290–7.934, P -value = 0.012) and the exhaled CO level on day 8 (AOR = 0.937, 95% CI = 0.885–0.992, P-value = 0.025) independently predicted 1-year smoking cessation. Moreover, the level of exhaled CO < 4.5 parts per million on day 8 significantly predict successful 1-year smoking cessation (area under curve 0.761, sensitivity 88.2%, and specificity 57.8%, P-value < 0.001). These independent predictors including intention to quit, varenicline use, and exhaled CO level on day 8, may help primary care physicians rearrange resources and refine the strategies for intervention programs to achieve a higher rate of long-term smoking cessation.