Learning to be Proficient? A Structural Model of User Dynamic Engagement in E-Health Interventions

Abstract

User attrition has been a major challenge confronted by eHealth interventions. In this study, we aim to examine how individuals dynamically engage in eHealth interventions to shed light on users’ continued participation. In the participation process, users’ intervention perceptions play an important role in affecting their subsequent participation decisions. To capture the updates in individuals’ intervention perceptions and the associated behavior dynamics, we establish a hierarchical Bayesian learning framework to structurally characterize individuals’ decision-making processes. Through analysis of a 4-month dataset collected for users’ intervention participation in an online weight-loss platform, our study provides empirical evidence for individuals’ learning behaviors in online health management. That is, we show that although individuals tend to have inaccurate perceptions of intervention effectiveness initially, they can significantly improve their intervention perceptions during continuous participation. Despite this learning, we find that users’ learning efficiency can vary across intervention types, depending on the level of noise contained in the intervention experiences. When users are unable to accurately infer the effectiveness of eHealth interventions, they are more likely to discontinue their usage in subsequent periods. Therefore, to help users improve their learning efficiency, we further propose several learning schemes to help individuals “de-noise” their intervention experiences. Our empirical estimation also reveals individual heterogeneity in experienced intervention effectiveness and users’ preference structure for descriptive intervention features. These findings provide various implications for user engagement and healthcare platform design.

Keywords: personal health management, lifestyle change, eHealth interventions, Bayesian learning, perceptions of interventions, dynamic participation