The Impact of AI-Driven Recommendation Systems on Patients’ Value Co-Creation Behavior in Online Health Communities: A Moderated Mediation Model

This study explores how AI-driven recommendation systems (ADRS) influence patients' participation in value cocreation behavior (VCB) within online health communities (OHCs), aiming to explore the mechanisms behind the fostering of sustainable OHCs. Using a moderated mediation model grounded in social assistance theories and value cocreation, survey data from 450 respondents were analysed via partial least squares structural equation modelling (PLS-SEM) and a multilayer perceptron (MLP) neural network model. The findings reveal that social assistance perceived (SAP) from ADRS positively influences VCB, both indirectly and directly through patient knowledge acquisition (PKA), and that the indirect effect is more robust when patient ability/readiness (PAR) is high. The MLP analysis further confirms the robustness of the model and the substantial moderating role of PAR. This research introduces a novel framework that sheds light on the role of the ADRS in enhancing SAP and PKA, ultimately driving VCB in OHCs. The findings provide practical insights for developing user-centric interventions to foster effective learning and collaboration in AI-driven health platforms, ensuring long-term sustainability and engagement in OHCs.