Development of a Risk Score to Predict Short-term Smoking Relapse Following an Inpatient Smoking Cessation Intervention.

This study aimed to investigate the factors affecting smoking relapse and to develop predictive models among Korean national 5-day smoking cessation program participants. The subjects were 518 smokers and follow-up was continued for 6 months after discharge. A predictive logistic model and risk score were developed from the multivariate logistic models and compared using the area under the receiver operating characteristic curve (area under the curve [AUC]). The smoking relapse rate within 6 months after program participation was 38.4%. The AUCs of the logistic regression model and risk score model were similar (odds ratio [OR] = 0.69; 0.69, respectively) in the development data set, and those of the risk score model were similar between the development and validation data sets (OR = 0.68). The risk score used by the six risk factors could predict smoking relapse among participants who attended a 5-day inpatient smoking cessation program.