Lalloué Benoît, Jiang Shu, Girault Anne, Ferrua Marie, Loirat Philippe, Minvielle Etienne
EA 7348 MOS-EHESP.
Gustave Roussy 114, rue Édouard-Vaillant, 94805 Villejuif, France.
Int J Qual Health Care. 2017 Oct 1;29(6):833-837. doi: 10.1093/intqhc/mzx111.
Most studies showed no or little effect of pay-for-performance (P4P) programs on different outcomes. In France, the P4P program IFAQ was generalized to all acute care hospitals in 2016. A pilot study was launched in 2012 to design, implement and assess this program. This article aims to assess the immediate impact of the 2012-14 pilot study.
From nine process quality indicators (QIs), an aggregated score was constructed as the weighted average, taking into account both achievement and improvement. Among 426 eligible volunteer hospitals, 222 were selected to participate. Eligibility depended on documentation of QIs and results of hospital accreditation. Hospitals with scores above the median received a financial reward based on their ranking and budget. Several characteristics known to have an influence on P4P results (patient age, socioeconomic status, hospital activity, casemix and location) were used to adjust the models.
To assess the effect of the program, comparison between the 185 eligible selected hospitals and the 192 eligible not selected volunteers were done using the difference-in-differences method.
Whereas all hospitals improved from 2012 to 2014, the difference-in-differences effect was positive but not significant both in the crude (2.89, P = 0.29) and adjusted models (4.07, P = 0.12).
These results could be explained by several reasons: low level of financial incentives, unattainable goals, too short study period. However, the lack of impact for the first year should not undermine the implementation of other P4P programs. Indeed, the pilot study helped to improve the final model used for generalization.
大多数研究表明，绩效薪酬（P4P）计划对不同结果没有或几乎没有影响。在法国，IFAQ绩效薪酬计划于2016年推广至所有急症护理医院。2012年启动了一项试点研究，以设计、实施和评估该计划。本文旨在评估2012 - 14年试点研究的即时影响。
从九个过程质量指标（QIs）构建了一个综合得分，作为加权平均值，同时考虑了达成情况和改进情况。在426家符合条件的志愿医院中，选择了222家参与。入选资格取决于质量指标的记录和医院认证结果。得分高于中位数的医院根据其排名和预算获得经济奖励。使用几个已知会影响P4P结果的特征（患者年龄、社会经济地位、医院活动、病例组合和地理位置）来调整模型。
为评估该计划的效果，采用差异中的差异方法对185家入选的符合条件医院和192家未入选的符合条件志愿医院进行了比较。
尽管所有医院在2012年至2014年期间都有所改善，但差异中的差异效应在粗模型（2.89，P = 0.29）和调整模型（4.07，P = 0.12）中均为正向但不显著。
这些结果可以由几个原因来解释：经济激励水平低、目标无法实现、研究期过短。然而，第一年缺乏影响不应削弱其他P4P计划的实施。事实上，试点研究有助于改进用于推广的最终模型。
