This study investigates patient–peer evaluation inconsistency on online healthcare platforms. We examine how the emotional and informational support in physicians' answers affects this inconsistency, whether the two support types act as substitutes or complements, and how professional titles moderate their effects.
Using a large dataset of 1,933,083 physician–patient interactions from a major Chinese healthcare platform (2017–2023), we analyzed how emotional support and informational support influence evaluation inconsistency. We employed logistic regression models, robustness checks, and heterogeneity analyses to test the hypothesized relationships.
Both emotional support and informational support are positively related to the likelihood of inconsistency in patient–peer evaluation. The two support types act as substitutes for one another. Professional titles buffer the inconsistency effects of both support types by providing a shared quality anchor that reduces both audiences' reliance on textual cues. Emotional support is associated mainly with peer-preferred inconsistency, whereas informational support affects both types of inconsistency.
This study repositions receiver heterogeneity not as a boundary condition on signaling effectiveness but as the generative mechanism of patient–peer evaluation inconsistency, extending signaling theory beyond single-audience contexts. It demonstrates that emotional and informational support operate as substitutes in driving patient–peer evaluation inconsistency. It further identifies an audience-alignment function of professional titles through which credentials narrow cross-audience evaluation divergence.
