Initial Communication Media, Sustained Patient Engagement, and Health Outcomes: Evidence from an Online Mental Healthcare Platform

Abstract

Online mental healthcare platforms have thrived during the COVID-19 pandemic, yet the choice of media in physician-patient interactions presents a conundrum for stakeholders. Communication media, a key element of service design, can influence patient revisits, which are essential for managing chronic diseases. However, understanding of their impacts remains limited in research and practice. Using data from a leading online mental healthcare platform in China, we assess the relative impacts of synchronous (audio, video) versus asynchronous (text) media on patient revisits and health outcomes. Our findings reveal a significant increase in patient revisits following initial services conducted via synchronous media, an observation consistent across various empirical methods, including matching and regression discontinuity design. More importantly, synchronous media are associated with improved health conditions in both the short and long term. Our mechanism analysis suggests the indispensable role of cue multiplicity (over feedback immediacy) featured in synchronous media, which likely explains these positive effects. We also explore effect heterogeneity across patient attributes, finding that synchronous media benefit most mental health conditions, except schizophrenia. Furthermore, the positive sync-media effects are more pronounced for patients at higher risk of health deterioration, as predicted and ranked through machine learning techniques. These findings advance our understanding of communication media in telehealth engagement and inform the design of sustainable online platforms for healthcare stakeholders.

Note:
Funding Information: None

Conflict of Interests: None

Ethical Approval: This study was approved by the Research Ethics Committee at the School of Management, Harbin Institute of Technology.

Keywords: Digital Platform, Online Healthcare Service, Patient Revisit, Communication Media, Mental Health, Regression Discontinuity Design, Machine Learning

JEL Classification: D83, I12, L86, M15