JMIR Publications

ABSTRACT

Background:

Online health communities (OHCs) have emerged as critical platforms for patients with chronic diseases to exchange informational and emotional support. Understanding the mechanisms of influence within these communities is essential for explaining how social support behaviors are organized and how they contribute to long-term health management, particularly for individuals with type 1 diabetes (T1D). Despite growing attention, little is known about how stakeholder roles and disease duration shape social support and influence dynamics in OHCs.

Objective:

This study investigates the mechanisms of network-based social support in a large T1D-focused OHC, with particular emphasis on stakeholder diversity, disease duration, and the determinants of user influence.

Methods:

We analyzed user behavior data from the largest online T1D community in China, covering the period from January 1 to May 20, 2024. The dataset comprised 43,788 posts, 145,423 comments, and metadata from 1,393 users, including identifiers, gender, community role (Patient, Guardian, Peer Supporter, Professional), registration dates, and time since diagnosis. Large language models were used to classify support behaviors, and social network analysis combined with behavioral modeling was applied to construct informational and emotional support networks. Ethical approval was granted by the Clinical Trial Ethics Committee of Anhui Provincial Hospital.

Results:

Peer supporters emerged as the most active providers, primarily delivering informational support, while patients and caregivers mainly sought help and provided emotional reassurance. Disease duration significantly shaped participation: users gradually transitioned from seekers to providers as their experience accumulated. Network analysis demonstrated that both informational and emotional subnetworks exhibited multi-centered, star-like topologies. Informational support followed a cross-stage reciprocity pattern, with long-duration users offering more assistance to newly diagnosed members, whereas emotional support showed homophilic clustering among users with similar disease durations. Influence analysis revealed that peer supporters were the most central actors, with relative centrality far exceeding that of patients and professionals. Female users exhibited significantly greater influence than males (approximately 65% higher). Temporal engagement patterns also mattered, with users active between 14:00–15:00 showing higher influence. Influence followed an inverted U-shaped trajectory over disease duration, peaking around 10 years post-diagnosis before declining, indicating a lifecycle effect in OHC-based influence.

Conclusions:

This study demonstrates that influence in T1D OHCs is shaped by stakeholder diversity, disease duration, and structural interaction patterns. Peer supporters emerged as key influencers, underscoring the critical role of experiential knowledge in long-term disease management. The findings highlight the value of precisely identifying and engaging influential users to enhance community-based interventions. We propose a Collaborative Care model centered on patients, led by peer supporters, and supported by multidisciplinary collaboration to foster resilient, sustainable, and equitable digital health ecosystems.