For decades, Medicare’s approach to improving the care a patient receives has remained largely unchanged: Reward success and penalize failure.

But clinicians and policy experts have long questioned whether giving bonuses to hospitals that achieve quality standards and docking so-called underperforming doctors and facilities is the best way to achieve the goal. Skeptics now have a growing body of convincing evidence showing the approach is not working.

“At this point, it’s clear that Medicare’s pay-for-performance programs have cost the system more than they have benefited patients. The programs have largely been a failure,” said J. Michael McWilliams MD, PhD, a professor of medicine and health care policy at Harvard Medical School in Boston and a senior advisor to the Center for Medicare & Medicaid Services (CMS) Innovation. “It’s time to pull back and rethink.”

Health systems with fewer resources, less money, and a higher proportion of disadvantaged patients often remain at the bottom. Others say some clinicians and hospitals game the system, choosing measures that are easy to meet to become a “top performer.”

As Medicare heads into its 60th year, some experts say policymakers need to take a completely different tack, such as rewarding incremental improvement by health systems. Other analysts believe more clinician input and better use of technology are all that is needed to improve what the program already does. One aspect on which everyone agrees? It’s time for change.

Digging Into Metrics

Medicare’s quality improvement (QI) initiatives include the Hospital Value-Based Purchasing program, which rewards acute care hospitals with payments based on metrics such as mortality, healthcare-associated infections, and patient experience. CMS withholds 2% of hospitals’ Medicare payments each year and redistributes the money as a bonus or penalty of 2%, depending on how they perform.

The Merit-based Incentive Payment System (MIPS), meanwhile, gives rewards or penalties to physicians who care for patients with Medicare Part B. Clinicians must select measures from one of four domains: quality, cost, improvement activities, or promoting interoperability. Payment adjustments in MIPS generally range from -9% to +9%.

Multiple studies and reports have found the metrics CMS uses are not valid in proving if the care provided was high quality or not. Rebecca Andrews, MD, a professor of medicine and director of primary care for Patient Centered Medical Home and Comprehensive Pain Center, UConn Health, in Farmington, Connecticut, said many of the measures do not take into account the usual workflows that exist in hospitals or outpatient environments.

“So then what the hospitals have to do is craft all these new workflows just trying to meet these performance measures and quality improvement initiatives, rather than focusing on the whole goal of them, which is improving patient care,” said Andrews, who chairs the American College of Physicians (ACP) Board of Regents, which manages the business and affairs of the college.

Other studies show low-performing hospitals frequently care for needier patient populations with less access to care to begin with. These patients present to hospitals sicker and prone to readmissions, mortality, or complications because of social determinants of health.

“Patients may have more social determinants of care playing into some of those outcomes, and there’s been efforts over the years to try to equalize that, but with anything that you try to scale this large, the devil’s always in the details, and some people get hurt,” said Jeff Pothof, MD, MMM, the chief quality officer at UW Health in Madison, Wisconsin.

“There’s also a belief that some hospitals may avoid treating sicker patients to maintain lower readmission or complication rates,” he said.

What Is Working?

But like any massive government program, outcomes can be nuanced and in some areas, Medicare’s quality strategy has worked, said David Levine, MD, chief medical officer and senior vice president for Vizient, a company that helps health systems improve performance.

The gains have been greatest in areas such as readmissions for heart failure, pneumonia, and myocardial infarctions, with hospitals reporting fewer readmissions overall, Levine said.

In addition, when Medicare created connected penalties to the prevalence of catheter-associated urinary tract and central line-associated bloodstream infections, health systems prioritized those conditions and rates fell, Pothof said.

Some health systems likely had robust departments prior to the advent of Medicare’s quality programs. But when penalties became involved, hospitals with less investment in QI built more robust systems to prevent preventable issues like hospital-acquired infections, Pothof said.

What Could Work

Policy experts said a shift from punishing low-performing hospitals to helping them would go a long way in achieving positive change and enhancing quality.

Strategies could incorporate measures of a hospital’s individual progress alongside fixed benchmarks and provide tiered incentives that allow hospitals at all performance levels to earn rewards if they show meaningful gains, Levine said. Another idea: allow a “ramp-up” period for low-performing hospitals to avoid punitive consequences while working toward benchmarks.

“An effective program would reward progress and not just reward a certain performance level,” he said. “It takes time to improve, and that should be recognized and rewarded.”

Pothof said more focus on patient-reported outcomes would be helpful to advance QI.

“This is, if I replace your knee, how do you feel we did with that knee replacement?,” Pothof said. “Are you doing the things you love to do? Are you mobile? Things like that.”

The US could also learn from international models, Levine said. The Care Quality Commission (CQC) in England is an independent regulator that inspects, rates, and supports clinicians and hospitals. When problems are identified, the CQC provides action plans, improvement coaches, and follow-up support rather than penalizing, he said.

Australia, meanwhile, focuses on minimum standards but emphasizes continuous QI, Levine said. Australian regulators conduct regular audits, provide education and capability-building, and accreditation is not punitive, he said. Instead, underperforming hospitals are required to improve with external support.

“These strategies, if adapted thoughtfully, could help create a more equitable and effective approach to quality improvement in American healthcare,” he said.

The Next 60 Years of QI

Pothof said the future of QI will likely include some form of artificial intelligence to improve patient outcomes.

For example, at UW Health, staff are already using a computer model that examines hundreds of patient variables and creates associations to alert clinicians to which patients may develop conditions like sepsis, Pothof said.

“Ideally, we would get a day’s heads-up that Bob in Room 5 is actually going to get pretty sick with sepsis about 24 hours from now,” Pothof said. “These models are not necessarily fully flushed out yet, but I think that’s going to be a big deal for healthcare.”

While no major legislative reforms are currently in the works, physician organizations are trying to address how QI programs can be improved. For instance, the ACP meets regularly with quality reporting and regulatory agencies like CMS to discuss changes and find common ground to improve current systems, she said.

“We have been able to push back a little bit and say, ‘Let us give you some perspectives about why there’s no evidence for this,’ or ‘you’re penalizing an individual hospital or an individual physician here, when it’s a much bigger issue with elements that actually are out of our control,’” said Andrews, who chairs ACP’s Performance Measures Committee.

The American Medical Association (AMA) has proposed an incentive payment system aimed at replacing MIPS. The new payment system would reinvest excess penalties in QI by helping under-resourced practices. It would also eliminate the win-lose model and reduce physicians’ maximum penalty.

In 2025, nearly 50% of solo practitioners, nearly 30% of small practices, and 18% of rural practices received a MIPS penalty of up to -9%, according to the AMA.

These percentages are too high, said Bobby Mukkamala, MD, president of the association. A new model, he said, would “help practices avoid steep penalties, improve patient outcomes, and use resources more efficiently.”

McWilliams said he hopes in the future, QI becomes more local. For example, accrediting bodies might ask physicians and nurses at individual facilities about safety and quality issues that could be improved. The health system would then be required to address those problems to maintain accreditation.

Creating more meaningful metrics would likely help health systems retain clinicians who want to provide the best patient care, said Andrews.

Clinicians “feel like they have to check boxes that are not doing things that improve either the quality of health or the quality of life for their patients or that add to safety,” Andrews said. “And that’s very frustrating because they went into medicine to provide good care. It’s part of the reason that a lot of physicians leave healthcare.”

Alicia Gallegos is a freelance healthcare reporter based in the Midwest.