The crowdsourced online mental health services: Quality identification, influencing mechanisms, and intervention strategies

Crowdsourced online mental health services address the limitations of traditional counseling; however, low barriers to entry for providers result in inconsistent quality. This study develops an "identification-influence-intervention" framework from the provider perspective: it proposes a five-dimensional quality assessment with multi-task learning for automated prediction; examines the dual mechanisms through which the initial provider's expertise affects subsequent participation, along with the moderating roles of rewards and patient distress; and investigates the diminishing marginal and complementary effects of self-experience and peer experience on dynamic quality improvement. The findings support platforms in automated assessment, incentive optimization, and service quality enhancement.