The objective of this study was to test provider response to the implementation of recent threshold‐based P4P while controlling for regression to the mean. That is, providers with the lowest performance may be most likely to improve even in the absence of P4P simply because their baseline performance is most extreme in the distribution of performance. While the findings that the lowest performing providers improved the most have been consistent, these studies have not accounted for the possibility of regression to the mean. Similarly, Beaulieu and Horrigan studied physicians enrolled in a threshold‐based P4P program in New York, finding that those with the lowest baseline score had the largest improvements (Beaulieu and Horrigan 2005). examined changes in physician performance after implementation of a P4P program in California and found that the largest improvements were among those physicians with baseline performance farthest below the threshold needed for financial rewards (Rosenthal et al. While threshold‐based incentives have been shown to provide discontinuous incentives in non–health care settings (Grant 2010 Neal and Schanzenbach 2010), early studies of pay‐for‐performance in health care found that contrary to predictions, providers with the lowest baseline performance improved the most. This would result in P4P having limited impact on performance overall and failing to improve performance among those that would most benefit from improvement in terms of measured performance.
Indeed, threshold‐based incentives may result in the most robust performance improvement among providers for whom it is easiest to obtain a financial bonus-those that are just below the threshold. Second, they may give little incentive to low‐performing providers to improve, as their performance may be far from the threshold with little chance of achieving the targeted threshold, making returns on investment in quality improvement unlikely. Thus, a ceiling of quality improvement may be observed at the threshold, blunting the average effect of P4P on quality improvement. First, threshold‐based payments give providers no direct incentive to improve beyond the targeted threshold.
However, the use of thresholds to determine bonus payments may have several downsides, assuming that improving scores requires investment of resources. Performance thresholds are easy for payers to implement and transparent for participating providers, making them common in current P4P programs. Our objective was to investigate empirically the effect of using performance thresholds on provider response to financial incentives, an area of P4P design in which evidence is lacking. Understanding the tradeoffs of these design choices may enable design of more effective P4P programs.
While each of these quality goals has anticipated pros and cons, empirical evidence supporting their use is scarce. Typical P4P incentives give providers financial bonuses or add‐on payments for achieving prespecified quality goals, including achieving a target threshold (e.g., performance above a predetermined level), a relative rank (e.g., performance in the top 10 percent of all providers), or improvement on quality metrics (e.g., improving performance over the prior year's performance). One possible explanation for this mixed evidence is that provider response to P4P has been heterogeneous because of variation in how specific P4P programs are designed. Despite the proliferation of P4P programs, there is mixed evidence to support their use, with the effects of P4P being variable across settings and programs. The use of pay‐for‐performance (P4P) to improve health care quality has become commonplace in the United States.