RAND Corporation, 1776 Main Street, Santa Monica, CA, 91403, USA.
UCLA Fielding School of Public Health, Los Angeles, CA, USA.
BMC Health Serv Res. 2019 Jul 18;19(1):500. doi: 10.1186/s12913-019-4293-9.
Payers and policy makers across the international healthcare market are increasingly using publicly available summary measures to designate providers as "high-performing", but no consistently-applied approach exists to identifying high performers. This paper uses publicly available data to examine how different classification approaches influence which providers are designated as "high-performers".
We conducted a quantitative analysis of cross-sectional publicly-available performance data in the U.S. We used 2014 Minnesota Community Measurement data from 58 medical groups to classify performance across 4 domains: quality (two process measures of cancer screening and 2 composite measures of chronic disease management), total cost of care, access (a composite CAHPS measure), and patient experience (3 CAHPS measures). We classified medical groups based on performance using either relative thresholds or absolute values of performance on all included measures.
Using relative thresholds, none of the 58 medical groups achieved performance in the top 25% or 35% in all 4 performance domains. A relative threshold of 40% was needed before one group was classified as high-performing in all 4 domains. Using absolute threshold values, two medical groups were classified as high-performing across all 4 domains. In both approaches, designating "high performance" using fewer domains led to more groups designated as high-performers, though there was little to moderate concordance across identified "high-performing" groups.
Classification of medical groups as high performing is sensitive to the domains of performance included, the classification approach, and choice of threshold. With increasing focus on achieving high performance in healthcare delivery, the absence of a consistently-applied approach to identify high performers impedes efforts to reliably compare, select and reward high-performing providers.
国际医疗保健市场的支付方和政策制定者越来越多地使用公开可用的综合指标来指定表现出色的供应商,但目前还没有一种一致适用的方法来确定优秀的供应商。本文使用公开数据来研究不同的分类方法如何影响被指定为“高绩效”的供应商。
我们对美国公开的绩效数据进行了定量分析。我们使用了来自 58 个医疗集团的 2014 年明尼苏达州社区测量数据,对四个领域的绩效进行分类:质量(两种癌症筛查过程指标和两种慢性病管理综合指标)、总医疗成本、可及性(一个 CAHPS 综合指标)和患者体验(三个 CAHPS 指标)。我们根据所有纳入指标的绩效使用相对阈值或绝对值对医疗集团进行分类。
使用相对阈值,没有一个医疗集团在所有四个绩效领域的前 25%或 35%达到了绩效水平。只有当一个医疗集团在所有四个领域的绩效达到 40%时,才能被归类为高绩效。使用绝对阈值,有两个医疗集团在所有四个领域都被归类为高绩效。在这两种方法中,使用较少的领域来指定“高绩效”会导致更多的医疗集团被指定为高绩效,尽管在确定的“高绩效”医疗集团之间存在一些一致性或中度一致性。
将医疗集团归类为高绩效对纳入的绩效领域、分类方法和阈值选择非常敏感。随着医疗服务提供方面对实现高绩效的关注度不断提高,缺乏一种一致适用的方法来确定高绩效的供应商,这阻碍了可靠地比较、选择和奖励高绩效供应商的努力。