Policy Levers for First-Contact Healthcare Provider Choice: A Discrete Choice Experiment in Shanghai’s Hierarchical Medical System

BACKGROUND: China's hierarchical medical system seeks to route first-contact care for mild, nonurgent conditions to primary healthcare (PHC), yet hospital-centrism persists. Evidence is needed on how financing and visible access/quality signals jointly shape initial provider choice, and which groups respond most. This study quantifies those trade-offs and simulates policy impacts in Shanghai. METHODOLOGY: A discrete choice experiment (12 tasks; two labeled provider alternatives plus opt-out) was administered to adult residents and long-term migrants in Shanghai (N = 441). Attributes included reimbursement rate, deductible, travel time, provider title (ref. General physician), hospital tier (ref. PHC), and online booking. A respondent-panel mixed logit with random coefficients for reimbursement, deductible, and travel time captured unobserved heterogeneity; effects were summarized as marginal rates of substitution (MRS) in reimbursement percentage points. RESULTS: Higher reimbursement pulled patients toward PHC. Beyond price, visible access and quality signals exerted strong influence on first-contact choice. Price responsiveness varied meaningfully across people, with those who better understand insurance reacting more strongly to reimbursement. Policy simulations raised PHC's first-contact share from 23.4% (UEBMI baseline) to 29.7% with a PHC-tilted reimbursement schedule, and to 33.3% when pairing the same schedule with assured booking and senior staffing at PHC. CONCLUSIONS: Price gradients are necessary but insufficient; coupling PHC-tilted financing with visible access/quality signals delivers the largest shift of mild first-contact care away from tertiary hospitals.