Despite the prevalence of pay-for-performance compensation in healthcare, a new editorial in the British Journal of Medicine suggests the concept "rests on flawed assumptions about medicine, measurement and motivation" and may encourage providers to game the system.
Physicians and professors from City University of New York School of Public Health and Duke University in Durham, N.C., said they are "worried that pay for performance may not work simply because it changes the mindset needed for good doctoring." Here are a few of the editorial's main takeaways.
Does the model measure a physician's performance or their ability to game a system? Risk adjustment for health outcomes — such as a hospital's mortality rate — is complicated since a key component of the adjustment is left to how aggressively the providers code the condition. For instance, "aspiration pneumonia with acute chronic systolic heart failure" instead of "pneumonia with chronic heart failure" can increase the risk-adjusted quality score and triple reimbursement.
Process-based indicators are "poor proxies for quality of care." Process-based measures, or those measuring tasks instead of outcomes, have "hidden complexity" no matter how seemingly clear, according to the authors. For instance, beginning treatment for pneumonia patients within four hours of their arrival may correlate with quality, but Medicare's incentives for this measure left providers giving antibiotics to nearly any patient in the emergency department who was coughing, according to the article.
Patients' social characteristics present a variable. Physicians at a Harvard teaching hospital who cared for more patients who did not speak English, came from minority backgrounds, were poor and/or were uninsured scored low on pay-for-performance metrics. This was also true for physicians who cared for patients with infrequent visits, according to the article.
Medicare accountable care organizations incentivize 33 metrics — but is that too many? The authors also question ACO's plans to unveil more quality metrics in the years to come and whether parameters established in pay-for-performance programs may be "perceived as controlling and may undermine the intrinsic motivation crucial to maintaining quality when nobody is looking."
How Will Hospital-Physician Alignment Affect Profit, Performance?
Study Finds Majority of Providers Unaware of Performance Incentives
Physicians and professors from City University of New York School of Public Health and Duke University in Durham, N.C., said they are "worried that pay for performance may not work simply because it changes the mindset needed for good doctoring." Here are a few of the editorial's main takeaways.
Does the model measure a physician's performance or their ability to game a system? Risk adjustment for health outcomes — such as a hospital's mortality rate — is complicated since a key component of the adjustment is left to how aggressively the providers code the condition. For instance, "aspiration pneumonia with acute chronic systolic heart failure" instead of "pneumonia with chronic heart failure" can increase the risk-adjusted quality score and triple reimbursement.
Process-based indicators are "poor proxies for quality of care." Process-based measures, or those measuring tasks instead of outcomes, have "hidden complexity" no matter how seemingly clear, according to the authors. For instance, beginning treatment for pneumonia patients within four hours of their arrival may correlate with quality, but Medicare's incentives for this measure left providers giving antibiotics to nearly any patient in the emergency department who was coughing, according to the article.
Patients' social characteristics present a variable. Physicians at a Harvard teaching hospital who cared for more patients who did not speak English, came from minority backgrounds, were poor and/or were uninsured scored low on pay-for-performance metrics. This was also true for physicians who cared for patients with infrequent visits, according to the article.
Medicare accountable care organizations incentivize 33 metrics — but is that too many? The authors also question ACO's plans to unveil more quality metrics in the years to come and whether parameters established in pay-for-performance programs may be "perceived as controlling and may undermine the intrinsic motivation crucial to maintaining quality when nobody is looking."
More Articles on Pay-for-Performance Compensation:
5 Must-Haves for Value-Based Purchasing SuccessHow Will Hospital-Physician Alignment Affect Profit, Performance?
Study Finds Majority of Providers Unaware of Performance Incentives