Heart Transplant Performance Metrics: Separating Signal From Noise.
Heart Transplant Performance Metrics: Separating Signal From Noise.
复制标题
心脏移植性能指标：从噪声中分离信号。
DOI：
10.1016/j.jchf.2024.02.022
复制
发表时间：
2024
期刊：
影响因子：
--
通讯作者：
Stehlik,Josef
中科院分区：
文献类型：
--
作者：
Hanff,ThomasC;Johnson,MarylR;Stehlik,Josef
心脏移植是一种具有内在风险的巨大后果疗法。多年来，评估移植中心质量的不同工具被用来为移植中心、患者和其他利益相关者提供数据，这些数据可以为他们的决策提供依据。在理想的情况下，对中心质量的公开报道可以创造透明度，促进患者自主，并产生对患者福祉至关重要的信任，包括更广泛的公众对移植过程和器官分配道德的信任。公开报告的数据为确保高质量报告的计划创造了强大的动机，理想情况下，这是提供高质量移植护理的同义词。在现实中，质量指标只有在它们揭示与程序的过程和结果相关的真实潜在属性的能力，以及它们沿着对患者有意义和可理解的质量梯度准确区分程序的能力时才有用。理想情况下，这些数据还应该提供实用信息，表现不佳的项目可以根据这些信息采取行动进行改进。重要的是，如果项目以牺牲质量护理本身为代价而将指标优化放在首位，或者如果质量评估方法的限制导致表现良好的项目被错误地描述为提供次优护理，公共报告也可能产生意想不到的不良后果。2016年，移植受者科学注册中心响应了移植社区的要求，要求进行更细粒度的质量评估，并将移植后移植失败(GF)的报告指标从3级系统扩展到目前的5级系统。1后来，SRTR增加了代表患者等待名单存活率(WS)和接受更快移植的可能性(FT)的指标，从而编纂了等待名单结果的关键重要性。在三级系统中，根据GF的95%可信区间是否包括HR 1.00，将计划绩效分类为“低于预期”、“符合预期”或“高于预期”。尽管很容易理解，但移植社区担心，中间层提供的有价值的信息很少，因为二级计划可能包括从0.5到2的HR点估计，而且考虑到基础贝叶斯模型的限制，较低数量的计划很少能够证明“高于预期”的结果。1这种程度的粒度不能满足患者、提供者或项目的需求，因此SRTR使用5级系统对每个项目的GF、WS和FT进行评级，分别为“比预期差”、“比预期差”、“如预期”、“比预期好”或“好于预期”。
Heart transplantation is an enormously consequential therapy with inherent risks. Different tools that assess transplant center quality have been used over the years to provide transplant centers, patients, and other stakeholders data that can inform their decisions. Public reporting of center quality, in an ideal setting, creates transparency, facilitates patient autonomy, and engenders trust that is critical to a patient’s well-being, inclusive of the trust of the wider public in the transplantation process and the ethics of organ allocation. Publicly reported data create a powerful incentive for programs to secure high-quality reports, which ideally are synonymous with delivering high-quality transplant care. In reality, quality metrics are only as useful as their ability to reveal real underlying attributes related to a program’s processes and outcomes and their ability to differentiate programs accurately along a gradient of quality that is meaningful and comprehensible to patients. The data should ideally also provide practical information that underperforming programs can act on to improve. Importantly, public reporting can also have unintended and undesirable consequences if programs prioritize metric optimization at the expense of quality care per se or if limitations of the quality assessment methods result in circumstances where well-performing programs are mistakenly characterized as delivering suboptimal care.In 2016, the SRTR (Scientific Registry of Transplant Recipients) responded to the requests of the transplantation community for a more granular quality assessment and expanded reporting metrics for posttransplant graft failure (GF) from a 3-tiered system to the present 5-tiered system. 1 Later, the SRTR added metrics representing a patient’s waitlist survival (WS) and the likelihood of receiving a faster transplant (FT), thus codifying the critical importance of waitlist outcomes. In the 3-tiered system, program performance was categorized as “lower than expected,”“as expected,” or “higher than expected” on the basis of whether the 95% credible interval for GF included an HR of 1.00. Although easily comprehensible, the transplantation community was concerned that the middle tier offered little valuable information, given that tier 2 programs could include HR point estimates ranging from 0.5 all the way to 2, and that lower-volume programs were rarely able to demonstrate “higher than expected” outcomes given the constraints on the underlying Bayesian models. 1 This degree of granularity did not meet the needs of patients, providers, or programs, so the SRTR responded with a 5-tier system that ranks each program’s GF, WS, and FT as “worse than expected,”“somewhat worse than expected,”“as expected,”“somewhat better than expected,” or “better than expected.”