A Machine-Learning Approach to Predicting Smoking Cessation Treatment Outcomes

Abstract

Aims Most cigarette smokers want to quit smoking and more than half make an attempt every year, but less than 10% remain abstinent for at least 6 months. Evidence-based tobacco use treatment improves the likelihood of quitting, but more than two-thirds of individuals relapse when provided even the most robust treatments. Identifying for whom treatment is effective will improve the success of our treatments and perhaps identify strategies for improving current approaches. Methods Two cohorts (training: N = 90, validation: N = 71) of cigarette smokers enrolled in group cognitive-behavioral therapy (CBT). Generalized estimating equations were used to identify baseline predictors of outcome, as defined by breath carbon monoxide and urine cotinine. Significant measures were entered as candidate variables to predict quit status. The resulting decision trees were used to predict cessation outcomes in a validation cohort. Results In the training cohort, the decision trees significantly improved on chance classification of smoking status following treatment and at 6-month follow-up. The first split of all decision trees, which was delay discounting, significantly improved on chance classification rates in both the training and validation cohort. Delay discounting emerged as the single best predictor of group CBT treatment response with an average baseline discount rate of ln(k) = −7.1, correctly predicting smoking status of 80% of participants at posttreatment and 81% of participants at follow-up. Conclusions This study provides a first step toward personalized care for smoking cessation though future work is needed to identify individuals that are likely to be successful in treatments beyond group CBT. Implications This study provides a first step toward personalized care for smoking cessation. Using a novel machine-learning approach, baseline measures of clinical and executive functioning are used to predict smoking cessation outcomes following group CBT. A decision point is recommended for the single best predictor of treatment outcomes, delay discounting, to inform future research or clinical practice in an effort to better allocate patients to treatments that are likely to work.

Development of Machine Learning Models for Prediction of Smoking Cessation Outcome

“…Some examples included prediction of postoperative in-hospital mortality [ 16 ], complications in patients with diabetes mellitus [ 17 ], and occurrence of cardiovascular diseases in patients on dialysis [ 18 ]. For smoking cessation, a decision tree model developed with machine learning to predict smoking cessation treatment outcome was proposed by Coughlin et al in 2018 [ 19 ]. The study included 161 participants, with 90 in the training dataset and 71 in the validation dataset, yielding an average correct classification rate of about 64%.…”

Predicting Response to Group Cognitive Behavioral Therapy in Asthma by a Small Number of Abnormal Resting-State Functional Connections

“…In the study on treatment response to CBT, Reggente et al (2018) observed an accuracy of 67% for predicting response in OCD patients. In a similar vein, Coughlin et al (2018) found that fMRI characteristic could predict the treatment response to GCBT in cigarette smokers. Our finding with an accuracy rate of 67.8% is in the range of previous reports.…”

A Comparative Analysis of Machine Learning Methods for Class Imbalance in a Smoking Cessation Intervention

“…In recent years, few experts have contended that machine learning techniques have been successfully employed to build a prediction of the success of smoking cessation. However, model interpretability tends to be disregarded in those existing studies [29][30][31] in terms of their complex black-box system. Accordingly, common decision-making responses suffer from the class imbalance problem due to overwhelming skewed distribution.…”

“…Nowadays, a limited amount of studies [29][30][31] demonstrated that machine learning classifiers, such as decision tree, SVM, and MLP have the potential to substitute statistical methods in constructing the prediction models of smoking cessation. Coughlin et al [29] presented the prediction model of smoking quit outcomes to improve the success of evidence-based tobacco use treatment and current methods. The significant features were selected by generalized estimating equations followed by using decision tree classifier to identify the smoking quit prediction.…”