Outcomes of beta blockers after myocardial infarction in nursing home residents

描述（由申请人提供）：背景：在许多流动人群中，β受体阻滞剂用于心肌梗死（心脏病发作）后的二级预防得到了坚实的证据支持。然而，实际上没有数据表明老年养老院患者使用受体阻滞剂的结果，他们可能特别容易受到受体阻滞剂不良事件的影响，并且从这些药物中获益较少。目的：(1)描述老年心肌梗死后养老院患者使用-受体阻滞剂的模式和预测因素；以及（2和3）确定受体阻滞剂对该人群功能结局、再住院和死亡的影响。方法：采用结合临床和管理数据的观察性研究设计，研究人群将包括2007年5月至2010年3月期间因心肌梗死住院且在此事件之前未使用-受体阻滞剂的大型全国性连锁养老院的长期居民。受试者出院后将随访6至32个月，比较住院后开始和未开始使用受体阻滞剂的患者。本研究的数据将来自(1)最小数据集，这是一个全面的标准化临床评估数据数据库，由临床医生定期为美国几乎所有养老院的居民完成；(2)医疗保险a部分住院患者诊断和服务的索赔数据；(3)来自医疗保险D部分和私人连锁养老院（Manor Care）的药物使用数据，该数据捕获了无法获得医疗保险D部分索赔数据期间的药物使用情况。倾向评分方法和其他因果推理方法将被用于控制-受体阻滞剂使用与感兴趣的结果之间关系的潜在混淆。认识到没有一种方法可以最终解决所有潜在的混杂因素，将使用来自相关方法的一系列估计来收敛于临床有意义的效应大小范围。为了进一步测试和交叉验证结果，将在另外两个队列中重复分析。首先，结果将分析全国范围内所有从医院返回疗养院而没有中间急性后康复期（医疗保险D部分不可见的时期）的医疗保险患者。其次，对退伍军人事务部养老院的病人进行重复分析。退伍军人事务部的数据包含医疗保险无法获得的其他元素，允许对建模假设和潜在偏差进行测试，这些假设和偏差可用于改进主要队列的分析。相关性/公共卫生意义：更好地了解-受体阻滞剂对心肌梗死后虚弱的养老院居民的益处和危害，将大大改善证据基础，并有助于指导这一重要人群的治疗决策。

DESCRIPTION (provided by applicant): Background: The use of beta blockers for secondary prevention after myocardial infarction (heart attack) is supported by a solid evidence base for many ambulatory populations. However, virtually no data exist on outcomes of beta blockers in older nursing home patients, who may be particularly susceptible to adverse events from beta blockers and yield less benefit from these drugs. Aims: (1) To describe patterns and predictors of beta blocker use in older nursing home patients after myocardial infarction; and (2 and 3) to determine the impact of beta blockers on functional outcomes, rehospitalization, and death in this population. Methods: Using an observational study design that combines clinical and administrative data, the study population will comprise long-term residents in a large, national nursing home chain who were hospitalized for myocardial infarction between May 2007 and March 2010 and had not been using beta blockers prior to this event. Subjects will be followed for outcomes for 6 to 32 months after hospital discharge, comparing those who were and were not started on beta blockers after the index hospitalization. Data for this study will come from (1 the Minimum Data Set, a comprehensive database of standardized clinical assessment data that is completed at regular intervals by clinicians for nearly all nursing home residents in the United States; (2) claims data on inpatient diagnoses and services from Medicare Part A; and (3) drug utilization data from Medicare Part D and from a private nursing home chain (Manor Care), which captures drug use during periods where claims data from Part D are unavailable. Propensity score approaches and other causal inference methods will be employed to control for potential confounding in the relationship between beta blocker use and the outcomes of interest. Recognizing that no single approach can conclusively address all potential confounders, a series of estimates from related approaches will be used to converge on a clinically meaningful range of effect sizes. To further test and cross-validate results, the analyses will be repeated in two other cohorts. First, results will be analyzed for all Medicare patients nationwide who returned from hospital to nursing home without an intermediary period of post-acute rehabilitation (the period invisible to Medicare Part D). Second, analyses will be repeated for patients in VA nursing homes. VA data contain additional elements not available through Medicare, allowing testing of modeling assumptions and potential biases that can be used to refine analyses in the primary cohort. Relevance / public health significance: Better understanding the benefits and harms of beta blockers in frail nursing home residents after myocardial infarction will substantially improve the evidence base and help guide treatment decisions for this important population.