Dialysis centers hit with financial penalties for poor performance don't tend to improve afterward, calling into question a set of U.S. federal programs intended to improve health care nationwide, a new report says.
Dialysis centers face up to a 2% reduction in their annual Medicare reimbursements if they get a low score on a set of quality measures designed by the U.S. Centers for Medicare and Medicaid Services.
The measures aren't just bureaucratic box-checking: Kidney patients treated at low-scoring programs tend to have a higher risk of death in their first year of dialysis, researchers said in background notes.
About 1 in 5 U.S. dialysis centers received such financial penalties in 2017, based on their performance two years earlier, said lead researcher Dr. Kyle Sheetz, a research fellow with the Center for Healthcare Outcomes and Policy at the University of Michigan.
However, those dialysis centers did not consistently improve their performance in either 2017 or 2018, despite the fines, Sheetz said.
"Getting penalized was not associated with future improvement in quality," Sheetz said. "The primary leverage that the program has over centers to improve quality didn't work."
Tying Medicare reimbursements to performance is a relatively recent development, promoted by the Affordable Care Act's focus on improving U.S. health care.
Launched in 2012, the End-State Renal Disease Quality Incentive Program "was the first pay-for-performance program that CMS implemented, where the payment for dialysis was pegged to performance in specific quality measures," said Dr. Paul Palevsky, president of the National Kidney Foundation.
Subsequent CMS "value-based programs" track hospitals and penalize them for poor purchasing practices, too many readmissions, or a surge in infections and illnesses acquired by patients during their hospitalization. Other programs affect the pay of doctors, nursing facilities, and home health care providers.
But the entire movement might have been wrongheaded, based on these findings, Palevsky said.
Pay-for-performance programs may miss the mark
"This article raises serious questions about the methodology that's been used for trying to drive quality improvements through payment policy," Palevsky said. "Essentially, what this study has shown is that it didn't work. It didn't change the performance on the specific measures that facilities were being penalized on."
About a half a million people are on dialysis in the United States, and CMS spends about $100,000 per person every year for kidney patients on dialysis -- more than 6% of the total Medicare budget, researchers said.
Sheetz and his colleagues analyzed CMS data on dialysis centers, including whether the centers had received a financial penalty for their performance.
Centers in 2015 were judged on such things as whether their patients suffered from anemia or bloodstream infections, whether they needed hospitalization, whether their dialysis was adequate to keep them healthy, and whether the dialysis was performed using best practices.
One reason why centers didn't improve might be that the quality measures tend to change from year to year, creating a frustrating moving target for them to hit, Sheetz said.
"If you are being penalized based on your performance two years ago on outcomes that are no longer even part of the program, it may be very difficult for centers to respond to certain quality measures," Sheetz said.
The quality measures themselves might not be an accurate reflection of performance, Palevsky added.
For example, a patient might choose to leave a dialysis session early, something that the center can't control, Palevsky said. Nevertheless, the center would receive a ding on the quality of dialysis it provided that person.
Part of a dialysis center's score also rests on a quality survey of patients.
"Patients complain that the survey is somewhat onerous to complete. It is a lengthy survey and to have to go through this survey completion twice a year every year, the patients complain about it," Palevsky said. "Are we getting the right information from this survey? Is the survey appropriately targeted to what the patients need to be telling us?"
Centers serving minorities penalized the most
CMS might be able to improve the program and get better outcomes if it narrowed its quality measures down to a handful of very important problems, Sheetz said. It also would help if the quality measures remained the same year to year for an extended period, to give centers a chance to address their shortcomings.
Another possibility is that the penalties aren't high enough.
"You could argue that 2% doesn't seem like a lot. The financial incentive isn't significant enough for them to act to change anything," Sheetz said.
However, raising the penalty higher -- to something like 10% of reimbursement -- could cause more problems with struggling centers.
"That's a significant amount of revenue at risk, but if you do that and you penalize centers who serve a difficult patient population -- patients who may have worse outcomes not because of what the center's doing but because they live in a tough area and they have worse health status to begin with -- then you may actually be further disadvantaging centers that are just trying to do the best with what they've got," Sheetz said.
This could compound another problem with the dialysis quality program, which is that centers in poor areas caring for minority patients are more likely to be penalized, Sheetz added.
"There may be unintended consequences if in fact we are penalizing centers that are not able to improve their quality and they're also the centers that are responsible for taking care of some of the most marginalized patients," Sheetz said.
The new study was published recently in the Annals of Internal Medicine.
The U.S. Centers for Medicare and Medicaid Services has more about value-based programs.
SOURCES: Kyle Sheetz, MD, MSc, research fellow, Center for Healthcare Outcomes and Policy, University of Michigan, Ann Arbor; Paul Palevsky, MD, president, National Kidney Foundation; Annals of Internal Medicine, June 1, 2021