Identifying emerging predictors for adolescent electronic nicotine delivery systems use: a machine learning analysis of the Population assessment of Tobacco and health Study.

Introduction:
This text analyzes a study that aimed to develop a predictive model for electronic nicotine delivery systems (ENDS) use behaviors among a nationally representative longitudinal cohort sample of adolescents using machine learning techniques. The study also identified population-level predictors for ENDS use behaviors as an exploratory investigation.

Key Points:

* The study used data from the Population assessment of Tobacco and health (PaTh) Study, Waves 1-4, and extracted a sample of adolescents who never used any tobacco products (n=10,246).
* The study included five distinct clusters of predictors, which were comprehensively selected from all measured PaTh Study variables that presented associations with ENDS use in the literature.
* The study used a supervised machine learning (ML) model to predict the probability of adolescent ENDS use behaviors. The penalized logistic regression method outperformed three other ML methods based on the area under the precision-recall curve.
* The study found that social media use emerged as an important variable in predicting adolescent ENDS use.
* The study also identified other population-level predictors for ENDS use behaviors, including psychosocial factors, experience with substances other than tobacco, and marketing, packaging, and accessibility to tobacco products.
* The study found that ML models appear to be a promising method to identify unique population-level predictors for U.S. adolescent ENDS use behaviors.
* The study suggests that more research is warranted to investigate emerging predictors of ENDS use and experimentally examine the mechanism by which these emerging predictors affect ENDS use behavior across different spectrums of populations.

Main Message:
The study highlights the potential of using machine learning techniques to develop a predictive model for ENDS use behaviors among adolescents. The study's findings suggest that social media use is an important predictor of adolescent ENDS use, and other population-level predictors were also identified. The study's results can inform effective ENDS use prevention efforts for adolescents, and it underscores the need for more research to better understand the emerging predictors of ENDS use.

han Dh, Lee Sh, Lee S, Seo DC. Identifying emerging predictors for adolescent electronic nicotine delivery systems use: a machine learning analysis of the Population assessment of Tobacco and health Study. Preventive medicine. 2021;145:106418. doi:10.1016/j.ypmed.2021.106418