Department of Biostatistics, University of Kansas Medical Center, Kansas City, KS 66160, USA.
Med Care. 2012 Sep;50(9):801-7. doi: 10.1097/MLR.0b013e31825a8b88.
Little is known about the recent development of the quality of nursing care.
To examine trends in the rate of total inpatient falls, one of the nursing-sensitive quality indicators, in US hospitals.
A longitudinal study of unit-level data collected during 2004-2009 by the National Database of Nursing Quality Indicators. Hierarchical Poisson regression models were used for the analysis of the unit-level fall rate.
Approximately 37,000 observations from 8915 nursing units (1994 critical care, 1328 step-down, 1663 medical, 1279 surgical, 2217 medical-surgical, and 434 rehabilitation units) in 1171 hospitals were examined.
The outcome measure was the annual count of unit-level inpatient falls with the annual count of unit-level patient days taken as the exposure variable. Independent variables included hospital size (≥300 or <300 beds), teaching status, and Magnet status and unit-level total nursing hours per patient day and proportion of total nursing hours supplied by RNs (skill-mix) at baseline.
The mean fall rates for most unit types remained stable or decreased, whereas those for surgical units increased over time. A higher register nurses skill-mix and the total nursing hours per patient day were both associated with lower fall rates (P<0.001); hospitals with more beds tended to have lower fall rates (P=0.001). Hospital Magnet and teaching status were not associated with the fall rate.
Overall, the fall rate in the United States hospitals decreased over time, but the large variation in the fall rate at both the hospital and the unit level indicates much room for improvement in the quality of nursing care related to fall prevention.
关于护理质量的最新发展情况，人们知之甚少。
考察美国医院住院患者总跌倒率（护理敏感质量指标之一）的变化趋势。
这是一项对国家护理质量指标数据库在 2004 年至 2009 年期间收集的单位水平数据进行的纵向研究。使用分层泊松回归模型分析单位跌倒率。
共分析了来自 1171 家医院 8915 个护理单元（1994 个重症监护病房、1328 个特护病房、1663 个内科病房、1279 个外科病房、2217 个内科-外科病房和 434 个康复病房）的大约 37000 次观察结果。
因变量为单位层面的年度住院患者跌倒次数，自变量为单位层面的患者住院日数。其他自变量包括医院规模（≥300 床或<300 床）、教学状态和马格内特地位以及单位层面的每位患者每天的总护理时间和注册护士提供的总护理时间比例（技能组合）。
大多数类型的护理单元的平均跌倒率保持稳定或下降，而外科护理单元的跌倒率则呈上升趋势。更高的注册护士技能组合和每位患者每天的总护理时间与较低的跌倒率相关（P<0.001）；拥有更多床位的医院往往具有较低的跌倒率（P=0.001）。医院马格内特地位和教学状态与跌倒率无关。
总体而言，美国医院的跌倒率随时间呈下降趋势，但医院和护理单元层面的跌倒率存在较大差异，表明在与预防跌倒相关的护理质量方面仍有很大的改进空间。
