Analyzing Nurses' Impact on Outcomes Using Detailed Data

描述（由申请人提供）：IOM的报告将全国的注意力集中在医院护理的质量和安全上。越来越多但不一致的证据表明医院护士配备与患者不良预后之间存在关系。由于这些不一致和研究证据的差距，解决当前护理短缺和围绕预计的更大短缺制定战略的努力受到了阻碍。本回顾性历史队列研究通过利用梅奥诊所及其医院5年（2000 - 2004年）的独特数据资源来解决这些需求，在研究期间包括25万入院患者、150万病人日和450万病人轮班的数据。有三个目标。首先，在患者细节层面，我们分析了患者不良结果的频率，并测试了这些结果与目标护理时间和实际护理时间之间的差距之间的关系，以及注册护士、执业护士和无证患者护理助理的组合，并预测哪些患者最有可能经历这些结果。其次，在单位轮班水平的分析中，我们测试了上述确定的护士配备措施与发生在特定班次和单位的不良事件之间的关系。第三，我们提供了在目标1中经历不良患者结果的患者总成本增量增加的估计。重要的是，我们能够冒险调整这些结果。因此，我们相信我们可以确定：a)哪些患者最有可能经历与护士人员配置相关的并发症（每个患者每天的小时数和人员组合），b)不良结局、不良事件和护士人员配置之间的关联是否像报道的那样强，因此，是否应该在监测医院质量的监测系统中依赖它们；C)一种新的用于分析护士敏感患者结果的敏锐度调整指标；d)出院时和随访30天内的结果及其相关费用。随后，本研究可以提供迄今为止缺失的证据，以确定美国医院中与患者安全相关的护士人员配备水平。

DESCRIPTION (Provided by the Applicant): Reports by the IOM have focused national attention on the quality and safety of hospital care. There is increasing but inconsistent evidence of a relationship between hospital nurse staffing and adverse patient outcomes. Efforts to address both the current nursing shortage and to strategize around the much larger shortage projected have been hampered by these inconsistencies, and by gaps in the research evidence. This retrospective historical cohort study addresses these needs by taking advantage of 5-years (2000 - 2004) of the unique data resources of Mayo Clinic and its hospitals, which over the study period include data on 250,000 admissions, 1.5 million patient days, and 4.5 million patient shifts. There are three aims. First, at the patient level of detail, we analyze the frequency of adverse patient outcomes and test the relationship between these outcomes and the gap between targeted and actual hours of nursing care, as well as the mix of RNs, LPNS, and unlicensed patient care assistants and predict which patients are most likely to experience these outcomes. Second, at the unit-shift level of analysis, we test for relationships between the measures of nurse staffing identified above and adverse events that occur on a specific shift and unit. Third, we provide estimates of the incremental increases in total costs for patients who experience the adverse patient outcomes in aim 1. Importantly, we are able to risk adjust these outcomes. As a result, we believe we can determine: a) which patients are most at risk for experiencing a complication related to nurse staffing (both hours per patient day and staff mix), b) whether associations between adverse outcomes, adverse events, and nurse staffing are as strong as reported, and hence, whether they should be relied upon in surveillance systems for monitoring hospital quality; c) a new acuity adjustment metrics for use in analyzing nurse sensitive patient outcomes; and d) outcomes and their associated costs at discharge and within 30 days follow-up. Subsequently, this study can provide heretofore missing evidence needed to define nurse staffing levels associated with patient safety in U.S. hospitals.