The Cost-Effectiveness of Pay-for-Performance

Abstract

T here is considerable dialogue in the clinical and health policy community concerning the implementation of value-based care, 1,2 defined by the ratio of health outcomes to dollar spent and operationalized by payment models that link quality metrics to reimbursement, such as bundled payment for episodes of care or pay for performance, the focus of this issue's article by Hsieh et al. 3 For those of us engaged in medical practice or its scientific study, these are familiar but concerning ideas, because this set of top-down theories does not always translate well to the real-world clinical environments and patient populations that they are purported to support. Indeed, the deceptively simple ratio that underpins value health involves a multidimensionality that makes measuring its true impact on cost-effectiveness far from transparent.Consider, for example, studies focused on the cost-effectiveness of pay for performance. Such studies are of increasing importance as the incentives are now linked to provider performance on quality benchmarks deployed by the Centers for Medicare and Medicaid Services (CMS) by requirements of the 2010 Patient Protection and Affordable Care Act (PPACA). 4 Achievement of the quality metrics associated with processes of care and health outcomes drives reimbursement; therefore, critical operating margins for practices and systems are at risk. Because understanding the association between incentives and the cost and quality of care delivery is essential to current medical practice survival, it provides an opportunity to think through how we might creatively investigate this question to provide a realistic underpinning for policy implementation going forward.To date, pay-for-performance studies demonstrate inconsistency in their findings whether relating to the association between pay-for-performance and process and intermediate outcomes or to the linkage between incentive strategies and cost-effectiveness. [5][6][7][8][9][10][11][12] This disagreement in evidence has been attributed to varying levels of methodological rigor. 13 Despite the lack of a firm understanding of impact, health systems are now engaged in the rollout of payment strategies proposed to support value-based care in the absence of definitive research, indicating that providers can both maximize the utilization of health resources as well as improve the quality of health outcomes.Hsieh and colleagues examine these questions by pay-for-performance reimbursement strategies implemented in Taiwan's National Health Insurance Program. In their article, the authors describe the care context of the policy implementation and its evolution as it relates to process measures, and then present evidence on the intermediate and long-term health outcomes. Of particular note, pay-for-performance is analyzed in terms of its effects on quality-adjusted life years, a linkage few US-based studies have included. The authors provide a 2-stage analysis of the evolution of Taiwan's incentive program: a 2001 diabetes pay-for-performance design th...