Variation in performance measure criteria significantly affects cardiology practice rankings: Insights from the National Cardiovascular Data Registry's Practice Innovation and Clinical Excellence Registry

Abstract

Million Hearts is a national initiative to prevent 1 million heart attacks and strokes over 5 years by improving cardiovascular prevention. An important tool in the success of programs like Million Hearts is public ranking on the quality of practices, yet different measures may provide different rankings, so the true quality of practices is difficult to discern. We evaluated the quality of ambulatory cardiology care using performance measure metrics. We compared rankings of practices participating in the National Cardiovascular Data Registry's Practice Innovation and Clinical Excellence Registry using measures from (1) the physician quality reporting system and (2) the American College of Cardiology/American Heart Association/Physician Consortium for Performance Improvement. We compared achievement rates for measures between the 2 frameworks and determined correlations in rankings using Spearman correlation coefficients. From January 1, 2008 to December 31, 2012, there were 1,711,326 patients enrolled from 111 US practices. Among eligible patients, the physician quality reporting system and American College of Cardiology/American Heart Association/Physician Consortium for Performance Improvement measures were achieved in 76.1% versus 77.4% for antiplatelet prescription (P < .001), 68.3% versus 90.8% for blood pressure control (P < .001), 26.9% versus 43.4% for cholesterol control (P < .001), and 37.4% versus 40.6% for smoking cessation (P = .383). Practice rankings were strongly correlated for antiplatelet prescription (correlation coefficient 0.98) and cholesterol control (0.92) but poorly correlated for blood pressure control (0.39) and smoking cessation (0.22). Evaluation of preventive care and individual practice rankings vary significantly depending on how measures are defined. Publicly reported measures need to be validly associated with outcomes to avoid incorrectly evaluating practice performance and failing to achieve public health goals.