Demographics and socioeconomic determinants of health predict continued participation in a CT lung cancer screening program.

PURPOSE: We developed machine learning (ML) models to assess demographic and socioeconomic status (SES) variables' value in predicting continued participation in a low-dose CT lung cancer screening (LCS) program. MATERIALS AND METHODS: 480 LCS subjects were retrospectively examined for the following outcomes: (#1) no follow-up (single LCS scan) vs. multiple follow-ups (220 and 260 subjects respectively) and (#2) absent or delayed (>1 month past the due date) follow-up vs timely follow-up (356 and 124 subjects respectively). We quantified the contributions of 14 socioeconomic, demographic, and clinical predictors to LCS adherence, and validated and compared prediction performances of multivariate logistic regression (MLR), support vector machine (SVM) and shallow neural network (NN) models. RESULTS: For outcome #1, age, sex, race, insurance status, personal cancer history, and median household income were found to be associated with returning for follow-ups. For outcome #2, age, sex, race, and insurance status were significant predictor of absent/delayed LCS follow-up. Across 5-fold cross-validation, the MLR model achieved an average AUC of 0.732 (95% CI, 0.661-0.803) for outcome #1 and 0.633 (95% CI, 0.602-0.664) for outcome #2 and is the model with best predictive performance overall, whereas NN and SVM tended to overfit training data and fell short on testing data performance for either outcome. CONCLUSIONS: We identified significant predictors of LCS adherence, and our ML models can predict which subjects are at higher risk of receiving no or delayed LCS follow-ups. Our results could inform data-driven interventions to engage vulnerable populations and extend the benefits of LCS.

Socio-Economic Factors and Clinical Context Can Predict Adherence to Incidental Pulmonary Nodule Follow-up via Machine Learning Models.

OBJECTIVE: To quantify the relative importance of demographic, contextual, socio-economic, and nodule-related factors that influence patient adherence to incidental pulmonary nodule (IPN) follow-up visits and evaluate the predictive performance of machine learning models utilizing these features. METHODS: We curated a 1,610-subject patient data set from electronic medical records consisting of 13 clinical and socio-economic predictors and IPN follow-up adherence status (timely, delayed, or never) as the outcome. Univariate analysis and multivariate logistic regression were performed to quantify the predictors' contributions to follow-up adherence. Three additional machine learning models (random forests, neural network, and support vector machine) were fitted and cross-validated to examine prediction performance across different model architectures and evaluate intermodel concordance. RESULTS: On univariate basis, all 13 predictors except comorbidity were found to have a significant association with follow-up. In multiple logistic regression, inpatient or emergency clinical context (odds ratio favoring never following up: 7.28 and 8.56 versus outpatient, respectively) and high nodule risk (odds ratio: 0.25 versus low risk) are the most significant predictors of follow-up, and sex, race, and marital status become additionally significant if clinical context is removed from the model. Clinical context itself is associated with sex, race, insurance, employment, marriage, income, nodule risk, and smoking status, suggesting its role in mediating socio-economic inequities. On cross-validation, all four machine learning models demonstrated comparable and good predictive performances, with mean area under the curve ranging from 0.759 to 0.802, with sensitivity 0.641 to 0.660 and specificity 0.768 to 0.840. CONCLUSION: Socio-economic factors and clinical context are predictive of IPN follow-up adherence, with clinical context being the most significant contributor and likely representing uncaptured socio-economic determinants.