医保支付方式改革是推进医疗健康高质量发展的核心驱动力,综合价值评价是衡量改革成效的关键工具。本研究立足真实世界研究（RWR）范式,以按病种分值付费（DIP）为切入点,采用文献计量初筛与两轮德尔菲专家咨询相结合的方法,从候选指标库中遴选出23项指标,构建“效率—质量—公平—可持续性”四维医保综合价值评价体系,并运用层次分析法（AHP）确定各指标权重（一致性比率CR<0.1）。以深圳市11家妇幼专科医院2018—2024年10万余份真实病例数据为实证样本,综合运用倾向性评分匹配（PSM）与中断时间序列分析（ITS）量化DIP政策的净效应,采用熵权-TOPSIS法对各机构医保综合价值进行动态排序评价。研究发现,DIP实施后,次均住院费用下降15.6%,高危孕产妇救治成功率提升至97.1%,医保基金结余率由2.3%升至5.8%,医保综合价值评分平均提高25.9%,但专科病种适配性不足、公平维度改善滞后、跨机构数据共享受阻等问题仍制约评价精度。据此,本文提出完善专科病组权重体系、建立动态指标调整机制、搭建“数据沙箱”共享平台、健全四维评价反馈闭环等政策建议。

The reform of medical insurance payment is a core measure to promote the high-quality development of healthcare, and comprehensive value evaluation is a key tool to test the effectiveness of the reform. Based on the paradigm of real-world research, this paper takes the Diagnosis-Intervention Packet (DIP) payment model as the starting point, employs a combination of bibliometric initial screening and two-round Delphi expert consultation to select 23 indicators from a candidate indicator pool, and systematically constructs a four-dimensional comprehensive value evaluation system for medical insurance, including efficiency, quality, equity and sustainability. The weights of the indicators were determined by the Analytic Hierarchy Process (AHP), with all consistency ratios (CR) below 0.1. Combined with more than 100000 real case data in the field of maternal and child health care in Shenzhen City, this paper comprehensively employs Propensity Score Matching (PSM) and Interrupted Time Series (ITS) analysis to quantify the net effect of the DIP policy, and adopts the Entropy Weight-TOPSIS method to conduct a dynamic ranking evaluation of the comprehensive medical insurance value across various institutions. Empirical results indicate that after DIP payment, average inpatient costs fell by 15.6%, success rate of critical maternal care reached 97.1%, the fund surplus rate increased from 2.3% to 5.8%, and overall value scores rose by 25.9%. Yet, the evaluation accuracy is still constrained by insufficient adaptability to specialty disease groups, delayed equity improvement, and barriers to cross-institutional data exchange. Accordingly, this paper recommends enhancing the specialty group weighting system, introducing a dynamic indicator adjustment mechanism, establishing a “data sandbox” for secure sharing, and reinforcing a closed-loop feedback process for the four-dimensional value assessment.