Dynamic, data-driven typologies of long-term smoking, cessation, and their correlates: Findings from the International Tobacco Control (ITC) Netherlands Survey

Identification of smoking cessation phenotypes as a basis for individualized counseling: An explorative real-world cohort study

INTRODUCTION

The rate of relapse in smokers attempting to quit is generally high. In order to maximize the chances of success, it is of interest to better understand the dynamic of lapse and relapse during smoking cessation. We hypothesized that specific behavioral patterns in tobacco consumption could predict the probability of quitting success and could open the possibility for a more targeted approach. The aim of the current study was to characterize clusters of quitting trajectories among participants involved in a smoking cessation program.

METHODS

In a retrospective real-world cohort study, data from 843 consecutive participants between March 2012 and December 2014 were collected. Data consisted of baseline information on demographics, smoking history and dependence level, as well as longitudinal data about tobacco consumption. The correlations among time series were characterized using principal coordinates analysis. Clusters were identified using k-means clustering and the average profile associated with each cluster was computed. The association between the participant’s baseline characteristics and clusters of tobacco consumption was assessed.

RESULTS

Four distinct clusters of transition phenotypes were identified based on tobacco consumption during the cessation phase: the long-term quitters (30%), the persistent smokers/reducers (44%), the short-term returners (16%) and the repeated try and failers (10%). Significant between-cluster differences were found in terms of baseline characteristics and smoking behavior during follow-up.

CONCLUSIONS

Meaningful clusters of quitting trajectories could be identified. Such specific behavioral patterns were useful for the application of personalized assistance needed to achieve successful and long-term cessation.

Are smoking restrictions at public venues and psychosocial beliefs associated with intentions to quit smoking among smokers in Malaysia?

The Malaysian government reinforced smoking restrictions at public venues to protect nonsmokers from secondhand smoke (SHS) exposure. This study examined whether smokers' reports about smoking restrictions and psychosocial beliefs were associated with quit intentions among Malaysian smokers. Data from 1047 cigarette smokers (103 females and 944 males) aged 18 and older from the 2020 International Tobacco Control (ITC) Malaysia Wave 1 Survey were analyzed with bivariate and multivariable logistic regression models. Most Malaysian smokers (85.2%) reported having quit intentions. Smoking was completely restricted in 34.8% of the nighttime venues, 85.3% of air-conditioned (AC) food and beverage (F&B) venues (restaurants, food courts, coffee shops), 87.3% of non-AC F&B, and 69.4% of indoor workplaces. Smokers who visited nighttime entertainment venues where smoking was fully restricted were less likely to have quit intentions. There was no significant association found with quit intentions for smokers who visited AC and non-AC F&B venues and indoor workplaces where smoking was fully restricted. All five psychosocial beliefs assessed, age, and education were positively associated with quit intentions. Malaysian smokers are interested in quitting and psychosocial beliefs were positively associated with quit intentions. There is a need for the Malaysian government to implement and reinforce comprehensive smoking restrictions in all public venues and indoor workplaces to protect nonsmokers from SHS exposure and to encourage smokers to think about quitting, which may influence their quit intentions.

Classification of drug use patterns

Characterizing long‐term prescription data is challenging due to the time‐varying nature of drug use. Conventional approaches summarize time‐varying data into categorical variables based on simple measures, such as cumulative dose, while ignoring patterns of use. The loss of information can lead to misclassification and biased estimates of the exposure‐outcome association. We introduce a classification method to characterize longitudinal prescription data with an unsupervised machine learning algorithm. We used administrative databases covering virtually all 1.3 million residents of Manitoba and explicitly designed features to describe the average dose, proportion of days covered (PDC), dose change, and dose variability, and clustered the resulting feature space using K‐means clustering. We applied this method to metformin use in diabetes patients. We identified 27,786 metformin users and showed that the feature distributions of their metformin use are stable for varying the lengths of follow‐up and that these distributions have clear interpretations. We found six distinct metformin user groups: patients with intermittent use, decreasing dose, increasing dose, high dose, and two medium dose groups (one with stable dose and one with highly variable use). Patients in the varying and decreasing dose groups had a higher chance of progression of diabetes than other patients. The method presented in this paper allows for characterization of drug use into distinct and clinically relevant groups in a way that cannot be obtained from merely classifying use by quantiles of overall use.