Digital Healthcare: Individual Engagement and IT-Enabled Design

Abstract

In this dissertation, I study the transformational effect of information technology on healthcare. I take two main perspectives to investigate the topic: individual engagement and IT-enabled design. Along the first perspective, I study individuals’ engagement process in online healthcare communities. Individuals’ engagement in online healthcare communities provides opportunities for healthcare providers to deliver services in a timely and cost-effective manner. In an aim to better understand and facilitate individuals’ online engagement, I apply econometric structural approaches to disentangle individuals’ online behavior dynamics in my first two essays. Specifically, the first essay studies individuals’ participation states that drive their self-regulatory behaviors online, and the second essay investigates individuals’ learning behaviors during their health-management process. Along the second perspective, I follow the design-science paradigm to study personalized healthcare recommendation systems in my third essay. To address several key challenges in healthcare recommendations, I propose a bandit-based recommendation framework that is enhanced by deep-learning feature engineering and diversity promotion constraint. I evaluate the proposed framework through extensive simulation experiments, and the results justify the superiority of the model as compared to a wide range of state-of-the-art recommendation systems.