Cluster-Randomized Trial of Personalized Site Performance Feedback in Get with the Guidelines-Heart Failure

Background - There is significant variation in the delivery of evidence-based care for patients with heart failure (HF), but there is limited evidence defining the best methods to improve the quality of care. Methods and Results - We performed a cluster-randomized trial of personalized site performance feedback at 147 hospitals participating in the Get With The Guidelines-Heart Failure quality improvement program from October 2009 to March 2011. The intervention provided sites with specific data on their heart failure achievement and quality measures in addition to the usual Get With The Guidelines-Heart Failure tools. The primary outcome for our trial was improvement in site composite quality of care score. Overall, 73 hospitals (n=33 886 patients) received the intervention, whereas 74 hospitals (n=37 943 patients) did not. One year after the intervention, both the intervention and control arms had a similar mean change in percentage points in their composite quality score (absolute change, +0.31 [SE, 1.51] versus +3.18 [SE, 1.68] in control; P=0.21). Similarly, none of the individual achievement measures or quality measures improved more at intervention versus control hospitals. Conclusions - Our site-based intervention, which included personalized site feedback on adherence to quality metrics, was not able to elicit more quality improvement beyond that already associated with participation in the Get With The Guidelines-Heart Failure program.