Erythropoiesis-stimulating agent use among chronic kidney disease patients

描述（由申请人提供）：背景。在过去的十年中，慢性肾脏病（CKD）增加了30%。贫血几乎总是在进行性慢性肾功能不全患者中发生，并与许多不良结局相关。红细胞生成刺激剂（ESA疗法）-于1989年批准用于肾病人群-用于纠正贫血。何时开始ESA治疗-在什么样的血细胞比容水平-以及在整个治疗过程中维持血细胞比容的阈值水平是未知的。例如，当患者严重贫血（即，血细胞比容低于30%）或仅中度贫血（即，红细胞压积30 - 33%）？更重要的是，供应商似乎不确定从贫血CKD患者中获得的依泊苷治疗的临床益处。Objective.我们将研究三种可能的治疗策略来解决以下研究问题：CKD人群中不同的贫血管理治疗策略是否会导致CKD进展、心血管疾病和死亡率方面的不同患者结局？本项目将研究的策略包括：1）贫血患者不使用ESA; 2）重度贫血患者（hct < 30%）使用ESA; 3）中度贫血患者（hct在30 - 33%之间）使用ESA。方法.我们建议使用一项观察性研究，该研究采用创新的因果建模技术，通过适当处理治疗和中间临床结局之间的时间依赖性混杂，“模仿”RCT。具体而言，当使用常规统计技术时，在整个研究期间不遵循上述三种策略之一的患者将产生时间依赖性选择偏倚。相反，边际结构模型（MSM）可以处理这种偏差，但需要注意的是，在建模过程中必须使用治疗间数据（包括ESA给药信息和血细胞比容值）。我们建议使用Kaiser Permanente Mid-Atlantic数据，其中包含治疗信息以及详细的实验室数据（GFR，肌酐，铁）和药物使用。意义值得注意的是，尽管对ESA的使用缺乏共识，并且大量CKD患者从未接受过治疗，但在过去十年中，透析前患者使用ESA的数量增加了5倍。我们的应用程序-检查不同的策略，启动和维持ESA治疗-有重要的意义，贫血管理和其成本之间的新兴CKD人群。

DESCRIPTION (provided by applicant): Background. There has been a 30% increase in chronic kidney disease (CKD) over the past decade. Anemia almost invariably develops in patients with progressive chronic renal insufficiency, and is associated with a number of adverse outcomes. Erythropoiesis-stimulating agents (ESA therapy) - approved in 1989 for use among the renal disease population - is used to correct anemia. When to start ESA therapy - at what hematocrit level - and at what threshold level to maintain hematocrit throughout treatment is unknown. For example, should ESA therapy be initiated when patients are severely anemic (i.e., hematocrit below 30%) or only moderately anemic (i.e., hematocrit between 30 - 33%)? Even more fundamental, providers appear to be uncertain about the clinical benefits to be derived from epoetin therapy for their anemic CKD patients. Objective. We will examine three possible treatment strategies to address the following research question: Do different anemia management treatment strategies among the CKD population result in different patient outcomes with regard to progression of CKD, cardiovascular disease, and mortality? The strategies that will be examined in this project include: 1) ESAs are not prescribed for patients with anemia; 2) ESAs are prescribed for severely anemic patients (hct < 30%); and 3) ESAs are prescribed for moderately anemic patients (hct between 30 - 33%). Methods. We propose to use an observational study using innovative causal modeling techniques that 'mimics' RCTs by appropriately handling time-dependent confounding between therapy and intermediate clinical outcomes. Specifically, patients who do not follow one of the three strategies listed above throughout the study will generate time- dependent selection bias when conventional statistical techniques are used. Instead, Marginal Structural Models (MSMs) can handle such biases with the caveat that treatment-to-treatment data (including ESA dosing information and hematocrit values) must be used in the modeling process. We propose to use Kaiser Permanente Mid-Atlantic data containing treatment to treatment information as well as detailed laboratory data (GFR, creatinine, iron) and medication use. Significance. Notably, despite the lack of consensus for the use of ESAs, and the large number of CKD patients who never receive therapy, the volume of ESAs administered to predialysis patients has increased 5-fold in the last decade. Our application - to examine different strategies for initiating and maintaining ESA therapy - has important implications for both anemia management and its costs among the burgeoning CKD population.