Machine learning to identify and understand key factors for provider-patient discussions about smoking

Abstract

We sought to identify key determinants of the likelihood of provider-patient discussions about smoking and to understand the effects of these determinants. We used data on 3666 self-reported current smokers who talked to a health professional within a year of the time the survey was conducted using the 2017 National Health Interview Survey. We included wide-ranging information on 43 potential covariates across four domains, demographic and socio-economic status, behavior, health status and healthcare utilization. We exploited a principled nonparametric permutation based approach using Bayesian machine learning to identify and rank important determinants of discussions about smoking between health providers and patients. In the order of importance, frequency of doctor office visits, intensity of cigarette use, length of smoking history, chronic obstructive pulmonary disease, emphysema, marital status were major determinants of disparities in provider-patient discussions about smoking. There was a distinct interaction between intensity of cigarette use and length of smoking history. Our analysis may provide some insights into strategies for promoting discussions on smoking and facilitating smoking cessation. Health care resource usage, smoking intensity and duration and smoking-related conditions were key drivers. The "usual suspects", age, gender, race and ethnicity were less important, and gender, in particular, had little effect.

Keywords: Bayesian machine learning; Smoking cessation; Survey data; Variable selection.

Fig. 1

Data selection procedures.

Fig. 2

Variable selection algorithm using BART-Machine.

Fig. 3

Visualization of BART-Machine variable selection. The lines are threshold levels for the variable selection algorithm described in Fig. 1 corresponding to $\alpha =0.1$. Variables passing this threshold are displayed as solid dots and asterisks. Solid dots represent variables selected with more stringent rule $\alpha =0.05$ and asterisks correspond to those with less stringent rule $\alpha =0.1$. Open dots correspond to variables that are not selected.

Fig. 4

Effect estimates and 95% confidence intervals (CI) for six key determinants and one most important interaction. For continuous variables, effect estimates represent changes in odds ratio per 10 years increase in years since first smoking and per five more cigarettes smoked per day. For factor variables, effect estimates compare odds for different levels to the reference level. The dashed red line corresponds to an odds ratio of one. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)