Inpatient Safety and Costs in U.S. Hospitals

描述(由申请人提供)：近年来，人们对美国医院医疗差错广泛性的担忧有所增加。虽然改善患者安全的举措已经开始，但到目前为止，人们对各种不良住院事件的代价了解有限。该项目通过医院层面的实证分析来解决这一认识上的差距，该分析检查了相对的经济 在同时影响医院成本的其他因素的背景下，一系列不良患者安全事件的成本。主要研究目标是： 1.建立潜在可预防的住院不良事件和并发症与住院费用之间的经验关系。 2.评估与减少美国医院的不良事件和并发症相关的成本节约的潜力。 该项目采用医院成本函数法在医院层面确定成本和患者安全之间的关系，使用AHRQ患者安全指标(PSI)作为医院质量的控制。应用随机前沿成本回归模型对医院效率进行调整。应用面板数据方法，从2001年至2006年的纽约、密歇根、佛罗里达和亚利桑那州的医疗保险成本报告、美国医院协会年度调查数据库和AHRQ HCUP州住院患者数据库中提取数据。除了PSI，成本功能还控制住院出院、门诊就诊、平均住院时间、投入价格、住院和门诊病例组合、医院规模、当地医院市场竞争、教学任务、所有权状况和城市位置。根据回归估计，预测的医院成本水平被用来模拟不良安全事件的成本，以及不同类型事件和不同类型医院的成本分布。

DESCRIPTION (provided by applicant): Concern over the breadth of medical errors in U.S. hospitals has increased in recent years. While initiatives to improve patient safety have mounted, to date there is limited understanding of the costliness of various adverse inpatient events. This project addresses this gap in awareness through a hospital level empirical analysis, which examines the relative economic costs of a broad range of adverse patient safety events in the context of other factors that simultaneously affect a hospital's costs. The major research objectives are to: 1. Establish the empirical relationship between potentially preventable adverse inpatient events and complications and hospital costs. 2. Assess the potential for cost savings associated with reduction of adverse events and complications in U.S. hospitals. The project takes a hospital cost function approach to determining the relationship between costs and patient safety at the level of the hospital, using AHRQ patient safety indicators (PSIs) as controls for hospital quality. It adjusts for hospital efficiency through application of the stochastic frontier cost regression model. Panel data methods are applied, drawing data from the Medicare Cost Reports, the American Hospital Association Annual Survey Database, and the AHRQ HCUP state inpatient databases for New York, Michigan, Florida, and Arizona for the years 2001 through 2006. In addition to the PSIs, the cost function controls for inpatient discharges, outpatient visits, average length of stay, input prices, inpatient and outpatient case mix, hospital size, competition within local hospital markets, teaching mission, ownership status, and urban location. From regression estimates, predicted levels of hospital costs are used to simulate the costliness of adverse safety events, as well as the cost distribution across different types of events and different types of hospitals.