Medicare is moving from fee-for-service payments to risk-based contracts designed to minimize costs and make insurers, health care organizations, and providers financially responsible for patients’ health. 

A new study from the Leonard Davis Institute of Health Economics (LDI) finds these contracts do not meaningfully change how care is delivered. 

“We did not see consistent, dramatic reductions in health care utilization or low-value care,” said LDI Senior Fellow Aaron Schwartz, who led the study, with coauthors including LDI Senior Fellows Sumedha Chhatre and Amol Navathe. “Risk-based contracts need to be redesigned if they’re going to rapidly transform patient care.”

One Quality and One Utilization Measure Changed With Risk-Based Contracts

The study advances our previous understanding of risk-based contracts by using national data on more than 6,100 health care organizations from 2015 through 2021 from a leading Medicare Advantage insurer. The researchers studied the effects of the contracts on patient outcomes using nine utilization measures (e.g., hospital stays, 30-day readmissions, nursing facility stays, and outpatient and emergency care) and 26 measures of low-value services (e.g., nonguideline-based cancer screening and unnecessary imaging, cardiovascular testing, and surgery). 

The researchers examined contracted risk using claims from 658 organizations that moved from fee-for-service to partial-risk contracts with bonuses only, and 114 organizations that transitioned from partial-risk to full-risk contracts with bonuses plus penalties. The analysis also included 3,385 control health care organizations whose contracts remained unchanged. Rigorous analytic methods ensured that any observed changes in care could be attributed to shifts in the organizations’ risk levels.

No utilization or quality measures consistently changed when health care organizations transitioned to risk-based contracts or increased their contracted risk. Significant differences were seen only for measures of emergency department visits (8% fewer compared to no contract change) and unnecessary cardiovascular stress testing (12% reduction), and only when organizations moved to partial-risk contracts. 

“Those reductions are not trivial,” Schwartz said. “They could affect costs and quality of care, but the changes were not consistently observed across measures. We saw no significant differences overall in counts of low-value services.”

Contract Redesign Could Encourage More Transformative Improvements

The researchers considered several explanations for the unexpectedly low impact of risk-based contracts on utilization and low-value care. Most full-risk contracts were adopted in 2019 or later, preventing analysis of long-term effects. 

Medicare Advantage plans already have policies and incentives to decrease unnecessary services, so additional bonuses and penalties may have little impact. 

In addition, health care organizations volunteer for risk-based arrangements, encouraging selective participation by organizations that already meet a contract’s cost-effectiveness goals. “Once you’ve squeezed the orange,” Schwartz said, “there’s not much juice left when you squeeze it again with a different tool.”

Despite the uninspiring results, Schwartz said, “This isn’t the nail in the coffin for Medicare Advantage risk-based contracts.” 

Contract redesign could motivate health care organizations toward more transformative achievements in high-value care. Options include mandatory contracts, tying bonuses to reaching specific cost-effectiveness goals, or reducing targeted low-value services. “These mechanisms might induce different organizations to participate, and all of them to try harder to transform care,” Schwartz said.

---

The study, “Changes in Health Care Utilization and Low-Value Service Use After Risk-Based Contract Adoption in Medicare Advantage”  was published in JAMA Internal Medicine on November 10, 2025. Authors include Aaron L. Schwartz, Soohyun Kim, Sumedha Chhatre, Amanda Sutherland, Aina Katsikas, Sara Riaz, Gosia Sylwestrzak, Emily Boudreau, and Amol S. Navathe. 

---

Author

---