• Joseph P Newhouse, Mary Price, John Hsu, Bruce Landon, and J Michael McWilliams.
• Harvard University, 180 Longwood Ave, Boston, MA 02115. Email: newhouse@hcp.med.harvard.edu.
• Am J Manag Care. 2019 Dec 1; 25 (12): e388-e394.

ObjectivesBanner Health, a large delivery system in Maricopa County, Arizona, entered into both Medicare and commercial insurance contracts that varied the amount of financial risk that Banner assumed. Rates of utilization and spending under these various contracts were investigated.Study DesignPrior to 2012, Banner held Medicare Advantage (MA) contracts, and in 2012 it began as a Medicare Pioneer accountable care organization (ACO). Banner also introduced a commercial ACO contract in that year. We compared risk-adjusted healthcare utilization and spending in the MA plan, the ACO, and a local traditional Medicare (TM) comparison group. We also compared risk-adjusted utilization and spending in Banner's commercial ACO with that of a comparison group drawn from the same employment groups who were not attributed to Banner providers.MethodsWe used claims and encounter data to measure utilization and spending. We risk adjusted using CMS and HHS Hierarchical Condition Categories.ResultsWithin Medicare, MA enrollees had lower risk-adjusted utilization and total spending than either the Pioneer ACO participants or a local TM comparison group. Participation in the Pioneer ACO program was associated with a greater reduction in hospitalization rates for ACO patients relative to local TM patients served by non-ACO providers, but the effect on total medical spending was ambiguous. Risk-adjusted differences between the commercial ACO group and the fee-for-service comparison group were generally small.ConclusionsThe results are consistent with CMS' efforts to shift reimbursement away from pure fee-for-service reimbursement.

29 keywords...

hide…