ADVANCING MEASURES FOR RISK ADJUSTMENT AND PERFORMANCE ASSESSMENT

项目总结（见说明）： 美国公众的健康和美国医疗保健的可负担性都受到美国医疗保健系统效率低下的威胁。一系列广泛的公共和私营部门支付和交付系统改革正在进行中，其目的是改善护理，改善健康和减缓支出增长。然而，这些改革的成功需要可靠的方法来衡量健康质量和成本，同时考虑到患者风险的潜在差异。虽然已经达成共识，目前的风险调整和性能监测方法存在严重的局限性，但即将大规模采用电子健康记录预计将能够常规收集广泛的个体生物标志物和患者报告的风险，症状和功能状态指标。目前的提案旨在促进健康风险、健康结果和卫生系统绩效衡量的科学，同时帮助解决紧迫的政策需求。该项目将利用核心C中收集的重要数据：来自健康和退休调查（HRS）的详细个人风险和健康状况数据，来自三个不同卫生系统的新收集的患者水平数据，以及全面的医疗保险索赔数据。第一个目标侧重于风险调整，我们将开发两个新的风险调整器：（a）一个基于通过HRS收集的详细生物特征和患者报告数据，（B）第二个基于整合邮政编码和县级健康状况，死亡率和疾病发病率的措施。这些将与医疗保险目前用于计划支付的基于索赔的措施进行比较，目的是为改善风险调整的长期和临时方法提供信息。在第二个目标下，我们将使用在常规临床实践中收集的患者风险和生物特征数据来探索使用这些数据进行风险调整和性能测量的潜在限制，包括评估无应答偏倚可能存在问题的程度，临床医生如何看待它们在初级保健实践中的常规使用，以及它们的使用是否有助于改善医疗保险患者的护理。

PROJECT SUMMARY (See instructions): The health of the American public and the affordability of U.S. healthcare are both threatened by inefficiency in the U.S. healthcare system. A broad array of public and private sector payment and delivery system reforms is underway, the aim of which is to improve care, improve health and slow spending growth. The success of these reforms, however, requires reliable approaches to measuring health quality and costs while accounting for potential differences in patient risk. And while consensus has emerged that current approaches to risk adjustment and performance monitoring have serious limitations, the impending broadscale adoption of electronic health records is expected to enable the routine collection of extensive individual biomarker and patient-reported measures of risk, symptoms and functional status. The current proposal is intended to advance the science of health risk, health outcome and health system performance measurement while helping to address pressing policy needs. The project will draw on important data assembled in Core C: detailed individual risk and health status data from the Health and Retirement Survey (HRS), newly collected patient-level data from three diverse health systems, and comprehensive Medicare claims data. Under the first aim, which focuses on risk adjustment, we will develop two new risk-adjusters: (a) one based on the detailed biometric and patient-reported data collected through the HRS, and (b) a second based on integrating ZIP code and county level measures of health status, mortality and disease incidence. These will be compared to the claims-based measures currently used by Medicare for program payment, with the aim of informing both long-term and interim approaches to improving risk adjustment. Under the second aim, we will use the patient risk and biometric data collected in routine clinical practice to explore potential limitations to using these data for risk adjustment and performance measurement, including assessing the degree to which non-response bias is likely to be problematic, how their routine use in primary care practices is perceived by clinicians, and whether their use contributes to improved care for Medicare patients.