Participation in Accountable Care Organization (ACO) Investment Model (AIM) during the first performance year by providers serving rural or underserved areas was associated with lower Medicare spending compared with non-ACO providers serving the same areas, according to a study published in The New England Journal of Medicine.
Researchers analyzed Medicare claims and enrollment data for a group of fee-for-service beneficiaries attributed to providers participating in AIM in 2016 (n=387,017). They then compared their findings with fee-for-service beneficiaries who resided in the same ACO markets but who had not been attributed to another Medicare ACO (Medicare Shared Savings Program [MSSP], Next Generation, or Pioneer; n=2,313,339).
A difference-in-differences study design was used to compare changes in outcomes from the baseline period (2013 through 2015) with the performance period (2016). Comparison beneficiaries were selected if they were eligible for attribution in terms of having ≥1 visit to a primary care provider but who received more services from non-ACO providers than from AIM-participating providers.
The primary outcome of interest was the total Medicare Part A and B spending. Researchers examined 4 secondary utilization outcomes: inpatient admissions, emergency department (ED) visits not resulting in admission, days of skilled nursing facility care, and inpatient readmissions.
For total Medicare spending, an estimated spending reduction of $28.21 per beneficiary per month translated into an aggregate reduction of $131 million. In sensitivity analyses, the smallest estimate in aggregate spending reduction was $92 million. CMS provided $76.2 million in up-front and ongoing payments to AIM ACOs in the first performance year, and paid an additional $6.2 million in shared saving bonuses for a total of $82.4 million in disbursements after AIM repayments were recouped. Primary estimates suggest that AIM was associated with net savings to CMS of $48.6 million ($10.46 per beneficiary per month) after accounting for AIM prepayments and shared saving bonuses. The researchers’ most conservative point estimate ($2.06 per beneficiary per month) suggests that AIM was associated with $9.6 million in net savings to CMS.
This study was limited by including only the first AIM performance year, so researchers cannot assess whether their findings can be sustained in the second AIM performance year, after AIM funding ends. This will be an important area for further investigation.
The researchers suggested that MSSP ACO participation in AIM was associated with lower Medicare spending per beneficiary per month in its first year compared with spending for beneficiaries residing in the same markets. The researchers also noted that “differential changes in spending were accompanied by differential reductions in the probability of inpatient admissions and emergency department visits that did not lead to admissions, as well as the number of days spent in skilled nursing facilities.” Providers receiving prepayments can successfully establish ACOs in rural and underserved areas.
This study was supported by the Centers for Medicare and Medicaid Services (contract number HHSM500201400261; task order number, HHSM500T0004).
Trombley MJ, Fout B, Brodsky S, McWilliams JM, Nyweide DJ, Morefield B. Early effects of an accountable care organization model for underserved areas [published online July 10, 2019]. N Engl J Med. doi: 10.1056/NEJMsa1816660