Risk-Based Contracts May Not Cut Low-Value Care in Medicare Advantage

In this investigation, the authors evaluated the impact of a voluntary transition to risk-based contracts under Medicare Advantage on health care use.

Despite making a voluntary switch to risk-based payment contracts, either upside only or 2-sided, there was no measurable impact on low-value service use or health care utilization overall, according to Medicare Advantage data published online today in JAMA Internal Medicine.¹ An upside-only plan has potential for financial bonuses, and 2-sided plans include possible bonuses and penalties, the authors wrote.

“The growth of risk-based payments raises key questions for government institutions, private insurers, and health care organizations,” they added.

Human administrative claims and enrollment data provided the data for this retrospective cohort investigation, which covered the period of January 1, 2015, through December 31, 2021, for beneficiaries enrolled in health maintenance organization plans. There were 658 organizations making the move from a fee-for-service arrangement to upside-only risk, 114 organizations making the move from upside-only plans to 2-sided plans, and 3385 organizations that made up the control cohort. They accounted for 1,042,272; 706,303; and 2,491,985 beneficiary-years, respectively. The overall patient age was 73 years, and at least 55% of each study group comprised female patients. Most patients were White and lived in an urban location.

The investigators used a difference-in-differences analysis to examine changes to 9 domains:

• Utilization measures: inpatient encounters, outpatient visits, and testing
• Low-value service use: cancer screening, diagnostic and preventive testing, preoperative testing, imaging, cardiovascular testing and procedures, and other surgeries

“Because adoption of risk-based contracts was staggered over time for different health care organizations,” the authors explained, “we employed the Callaway-Sant’Anna (C-S) estimator to calculate group-time average treatment effects.”

Per year of contract switch, the early years (2016-2018) saw a higher percentage of upside-only risk contract adopters compared with the later years (2019-2021), which saw more 2-sided risk contract adopters:

• 2016: 16% (upside-only) vs 8% (2-sided)
• 2017: 12% vs 3%, respectively
• 2018: 17% vs 7%
• 2019: 13% vs 21%
• 2020: 14% vs 31%
• 2021: 28% vs 31%

Overall, most plan switches took place in January: 50% of beneficiary-year observations among those transitioning to an upside-only contract and 91% changing to a 2-sided risk contract. There was also a longer mean (SD) observation period for risk-based contract adopters: for upside-only contract adopters, this time was 5.16 (1.87) years vs 3.21 (2.15) years in the control group, while for 2-sided risk contract adopters, this time was 5.14 (2.06) years vs 4.24 (2.30) years. Patient panel sizes were also larger for risk contract adopters, with a mean 322.36 (512.91) beneficiaries per year for upside-only adopters and 1578.94 (1911.40) beneficiaries per year, 2-sided contract adopters vs 128.12 (158.72) and 923.11 (1256.03), respectively, for control organizations.

Differential reductions in the following health care services were seen from baseline:

• Adoption of upside-only contracts:
  • Emergency department (ED) visits: −4.6% (95% CI, −8.7% to −0.4%)
  • Primary care visits: −5.0% (95% CI, −7.9% to −2.1%)
  • Advanced imaging: (−8.3% of baseline level; 95% CI, −13.1% to −3.5%)
  • Specialty visits: (−8.7% of baseline level; 95% CI, −16.4% to −1.0%)
• Two-sided risk contracts:
  • Cardiovascular stress testing: –11.1% (95% CI, –21.4% to –1.0%)
  • Advanced imaging: −4.2% (95% CI, −8.1% to −0.2%)

Despite these improvements, however, just 2 remained statistically significant after Rambachan-Roth adjustment, and only in connection with upside-only contract adoption:

• ED visits: −8.4% (95% CI, −15.5% to −1.3%)
• Cardiovascular stress testing: −12.1% (95% CI, −23.4% to −0.7%)

No differential changes were seen in total count of low-value services received per beneficiary for either type of contract adoption.

Overall, the authors observed that the most consistent changes in both service utilization and use of low-value services were seen for those adopting 2-sided risk contracts in the later years of the study. There were reductions seen for inpatient hospital stays, ED visits, nursing facility stays, primary care visits, specialty visits, and total low-value services.

Trying to clarify why the results were inconsistent, an accompanying editorial noted, “These efforts require time, money, and having a large enough share of patients in value-based models for them to make financial sense. Clinicians in the organization (who would likely prefer to be payer agnostic) must know which of their patients are eligible for which programs based on their insurance plans.”²

In addition, the authors themselves said that their limited study time frame may not have permitted enough time for changes to be noticed and that it can be challenging to sell clinicians on risk-based contracts when not all patients stand to benefit.

“As with any voluntary payment model, organizations already achieving the model’s goals may be more inclined to participate,” they concluded. “However, changes to utilization among participating organizations may be limited.”

References

1. Schwartz AL, Kim S, Chhatre S, et al. Changes in health care utilization and low-value service use after risk-based contract adoption in Medicare Advantage. JAMA Intern Med. Published online November 10, 2025. doi:10.1001/jamainternmed.2025.5917
2. Ganguli I. Full risk yet little reward?—when clinicians take on risk-based contracts in Medicare Advantage. JAMA Intern Med. Published online November 10, 2025. doi:10.1001/jamainternmed.2025.5924