Crowdsourced professional signals: How physician-group congruence shapes patient choice on online healthcare platforms

On online healthcare platforms, patients struggle to distinguish collective professional patterns from outlier opinions when multiple physicians respond to the same question. We introduce physician-group congruence, which captures the alignment between a focal physician's communication patterns and the collective patterns of co-posted physicians, as a group-level signal that enables patients to assess professional alignment without medical expertise. Analyzing 500,371 physician answers, we find that physician-group congruence significantly increases the likelihood of answer selection, with linguistic style congruence exhibiting the strongest effect, followed by terminology and emotional intensity congruence. At the cross-level, question-answer congruence significantly amplifies the effect of physician-group congruence. Within the group-congruence dimensions, linguistic style and emotional intensity congruence combine synergistically, while terminology congruence serves as an independent credibility filter. A validation study further shows that linguistic style congruence enhances perceived professionalism, which mediates 48.5% of its effect on answer selection, confirming that patients rely on linguistic convergence as a perceptual signal rather than an indicator of medical quality. These findings extend signaling theory to multi-party contexts, suggesting that collective patterns serve as a normative framework for decision-making. Practically, platforms can visualize physician-group congruence alongside credentials to help patients distinguish between mainstream and outlier perspectives, thereby reducing information asymmetry while preserving clinical diversity.