When Cutting Healthcare Spending, MA Pulls Ahead of Medicare ACOs

Average spending on Medicare beneficiaries enrolled in Medicare Advantage (MA) plans was much lower compared to the same metric of spending for beneficiaries attributed to a Medicare ACO, according to a recent study in JAMA Network Open.

The study of over 15,700 Medicare beneficiaries treated by a health system in the southern US found that per-member per-year spending was 22 to 26 percent higher for beneficiaries attributed to a Medicare Shared Savings Program (MSSP) ACO compared to beneficiaries enrolled in MA plans even after controlling for clinical risk and other factors.

Practice patterns and quality of care were similar for MA and MSSP ACO beneficiaries at the health system. However, ACO beneficiaries had greater outpatient hospital and inpatient spending compared to beneficiaries enrolled in MA plans, the study showed.

Additionally, the study highlighted that coding intensity did not explain higher Medicare ACO spending.

The findings have two important policy implications for policymakers and health systems, according to researchers.

First, greater Medicare ACO spending is unlikely attributable to clinician, health system, or practice pattern factors. Differences in spending also persisted when researchers standardized prices, suggesting that lower negotiated reimbursement rates for MA beneficiaries did not explain the findings.

“What drives spending differences is likely multifactorial. While controlling for area-level socioeconomic factors did not affect spending differences, it is likely that unmeasured selection factors relating to more adverse socioeconomic factors for MSSP beneficiaries may contribute to higher MSSP spending. These may include unstable housing or transportation that predispose to higher acute care utilization,” they wrote in the study.

Utilization management measures used by MA payers, not health systems, could explain differences in healthcare spending on Medicare beneficiaries at the large health system. Program design variability may be behind lower outpatient hospital spending among MA beneficiaries, researchers stated.

Second, researchers indicated that “the effort and cost of adding EHR clinical data to large scale risk-adjustment models such as the CMS-HCC score may not be justified.”

 “Over the past 20 years, CMS has undertaken several efforts to improve risk adjustment in the MA program to account for clinical and socioeconomic differences between MA and FFS populations,” they explained. “These have included expanding diagnostic information to include both inpatient and outpatient information under the CMS-HCC and better accounting for specific factors such as substance use, mental health, and CKD.”

“Our study expands on these improvements in risk adjustment by incorporating rich EHR-based clinical information to estimate risk using clinically validated risk measures derived from clinical practice. Nonclinical risk factors and factors related to plan design likely play an outsized role in unexplained spending differences between the MA and FFS populations. Collecting individual social determinants of health or other data on health behaviors and individual characteristics (eg, health literacy, engagement in personal health) may be important for accurate risk adjustment,” they continued.

Overall, health systems may have trouble achieving spending targets through the MSSP program compared to the MA program. Program design differences between the two programs could put health systems at a disadvantage with each MSSP beneficiary they treat.

The study’s findings also indicate a problem for CMS, which aims to have all traditional medicare beneficiaries tied to a value-based care model like the MSSP by 2030.

Aligning MA and MSSP programs may help both health systems and policymakers, researchers concluded.