Predicting Dropouts From an Electronic Health Platform for Lifestyle Interventions: Analysis of Methods and Predictors

Abstract

BackgroundThe increasing prevalence and economic impact of chronic diseases challenge health care systems globally. Digital solutions can potentially improve efficiency and quality of care, but these initiatives struggle with nonusage attrition. Machine learning methods have been proven to predict dropouts in other settings but lack implementation in health care.ObjectiveThis study aimed to gain insight into the causes of attrition for patients in an electronic health (eHealth) intervention for chronic lifestyle diseases and evaluate if attrition can be predicted and consequently prevented. We aimed to build predictive models that can identify patients in a digital lifestyle intervention at high risk of dropout by analyzing several predictor variables applied in different models and to further assess the possibilities and impact of implementing such models into an eHealth platform.MethodsData from 2684 patients using an eHealth platform were iteratively analyzed using logistic regression, decision trees, and random forest models. The dataset was split into a 79.99% (2147/2684) training and cross-validation set and a 20.0% (537/2684) holdout test set. Trends in activity patterns were analyzed to assess engagement over time. Development and implementation were performed iteratively with health coaches.ResultsPatients in the test dataset were classified as dropouts with an 89% precision using a random forest model and 11 predictor variables. The most significant predictors were the provider of the intervention, 2 weeks inactivity, and the number of advices received from the health coach. Engagement in the platform dropped significantly leading up to the time of dropout.ConclusionsDropouts from eHealth lifestyle interventions can be predicted using various data mining methods. This can support health coaches in preventing attrition by receiving proactive warnings. The best performing predictive model was found to be the random forest.

Challenges in Acceptance and Compliance in Digital Health Assessments During Pregnancy: Prospective Cohort Study

“…In any digital health program, compliance is a key challenge and an essential factor for a successful outcome, as low compliance can threaten the validity of a study [ 18 ]. A substantial number of patients stop using apps before the completion of a program [ 19 ]. Several studies from different fields reported that compliance declined through the course of their study [ 20 , 21 ].…”

Critical Predictors for the Early Detection of Conversion From Unipolar Major Depressive Disorder to Bipolar Disorder: Nationwide Population-Based Retrospective Cohort Study (Preprint)

“…In our study, the results indicated that RF has the highest average AUC in the process of 10-fold cross-validation, and RF use in the testing set showed performance consistent with that in the training/validation set. Many previous studies also found that RF performs better than many standard supervised learning techniques [54][55][56][57]. The main advantages of RF are as follows:…”

Digital Mental Health Interventions for Depression

“…Continued research that helps to identify those at baseline and throughout treatment who are most likely to eventually drop out of DHMI is critical to address central questions about who needs clinician support, when it should offered, and in what form it should be delivered. Recent research suggests that women and older users are particularly vulnerable to dropout when using digital lifestyle interventions and that dropout risk is highest in early sessions (Pedersen et al, 2019). Finally, further work is needed to improve DMHIs to address issues of patient safety given that clinical deterioration and suicide risk are substantial concerns among individuals with depression.…”