Abstract：
BACKGROUND: We investigated whether an anonymous automated electronic monitoring system (EMS) could be used to compare hand hygiene (HH) performance of individual nurses.
METHODS: Prospective observational cohort study. HH performance in 10 acute-care patient rooms was estimated through an EMS that anonymously measured HH events, and room entries and exits. The association between patient room\'s HH compliance and the nurse in charge of each room was investigated by comparing percentile rank distributions, and through a negative binomial model.
RESULTS: Over 99 days, there were 38,596 HH events and 135,546 room entries and exits (global HH performance, 28%). For 10 of 54 (19%) nurses, the median HH percentile rank of the rooms to which they were assigned was higher than the group average (P < .001; range of percentiles, 64th to 85th). A lower median percentile was seen in 9/54 (17%) participants (P < .001; range of percentiles, 22nd to 39th). The negative binomial model confirmed this association and identified 15 of 54 high performers (range of adjusted incidence rate ratios [aIRR], 1.17-1.83) and 16 of 54 low performers (range of aIRR, 0.37-0.77).
CONCLUSIONS: An association exists between a room\'s HH rate and its assigned nurse. This association could hold potential value for an individualized feedback strategy.
METHODS: Prospective observational cohort study. HH performance in 10 acute-care patient rooms was estimated through an EMS that anonymously measured HH events, and room entries and exits. The association between patient room\'s HH compliance and the nurse in charge of each room was investigated by comparing percentile rank distributions, and through a negative binomial model.
RESULTS: Over 99 days, there were 38,596 HH events and 135,546 room entries and exits (global HH performance, 28%). For 10 of 54 (19%) nurses, the median HH percentile rank of the rooms to which they were assigned was higher than the group average (P < .001; range of percentiles, 64th to 85th). A lower median percentile was seen in 9/54 (17%) participants (P < .001; range of percentiles, 22nd to 39th). The negative binomial model confirmed this association and identified 15 of 54 high performers (range of adjusted incidence rate ratios [aIRR], 1.17-1.83) and 16 of 54 low performers (range of aIRR, 0.37-0.77).
CONCLUSIONS: An association exists between a room\'s HH rate and its assigned nurse. This association could hold potential value for an individualized feedback strategy.
摘要：
背景：我们调查了匿名自动电子监测系统（EMS）是否可用于比较个别护士的手卫生（HH）表现。
方法：前瞻性观察性队列研究。通过匿名测量HH事件的EMS评估了10个急性护理病房的HH表现，和房间入口和出口。通过比较百分位等级分布，调查了病房的HH依从性与每个病房的护士之间的关联。并通过负二项式模型。
结果：超过99天，有38,596个HH事件和135,546个房间进出(全球HH性能，28%）。对于54名护士中的10名(19%)，他们被分配的房间的HH百分位数中位数高于组平均值(P<.001;百分位数范围,第64至85）。在9/54（17％）参与者中发现了较低的中位数百分位数（P<0.001；百分位数范围，22日至39日）。负二项模型证实了这种关联，并确定了54个高表演者中的15个(调整后的发病率比率范围[aIRR]，1.17-1.83)和54个低绩效企业中的16个(aIRR的范围，0.37-0.77）。
结论：房间的HH率与其指定的护士之间存在关联。这种关联可以为个性化反馈策略保持潜在价值。
方法：前瞻性观察性队列研究。通过匿名测量HH事件的EMS评估了10个急性护理病房的HH表现，和房间入口和出口。通过比较百分位等级分布，调查了病房的HH依从性与每个病房的护士之间的关联。并通过负二项式模型。
结果：超过99天，有38,596个HH事件和135,546个房间进出(全球HH性能，28%）。对于54名护士中的10名(19%)，他们被分配的房间的HH百分位数中位数高于组平均值(P<.001;百分位数范围,第64至85）。在9/54（17％）参与者中发现了较低的中位数百分位数（P<0.001；百分位数范围，22日至39日）。负二项模型证实了这种关联，并确定了54个高表演者中的15个(调整后的发病率比率范围[aIRR]，1.17-1.83)和54个低绩效企业中的16个(aIRR的范围，0.37-0.77）。
结论：房间的HH率与其指定的护士之间存在关联。这种关联可以为个性化反馈策略保持潜在价值。
