- Medicare accountable care organizations (ACOs) operating under two-sided financial risk arrangements save the federal healthcare program more than one-sided risk organizations. But policymakers must address financial incentives, benchmarks, and other issues to ensure two-sided risk ACOs continue to generate savings, the Medicare Payment Advisory Commission (MedPAC) recently reported.
“Medicare ACOs were created to help moderate the growth in Medicare spending and improve quality of care for beneficiaries by giving providers greater responsibility for costs and quality. ACOs have grown rapidly (about a third of Medicare FFS [fee-for-service] beneficiaries are now in ACOs), and several new initiatives have been designed to expand ACOs,” MedPAC wrote to Congress in its June 2018 report.
“Performance to date shows high quality being maintained, some savings relative to benchmarks, and slightly greater savings relative to what Medicare spending would have been without ACOs. However, several issues confront Medicare ACOs—particularly as they transition to models with two-sided risk—that will need to be resolved for the program to be successful in reaching its goals.”
Key issues include the roles of hospitals and specialists in ACOs, benchmarking and rebasing methodologies, and MACRA incentive payments.
The roles of hospitals and specialists in ACOs
Hospitals have the capital and resources to engage with two-sided risk ACO programs compared to their smaller and independent peers. However, the myth that ACO program incentives threaten hospital revenue goals may be stymying hospital participation, MedPAC explained.
Hospitals operating in a fee-for-service environment have an incentive to boost their Medicare admissions, while ACOs have an incentive to limit Medicare spending below a target amount. Reducing costly hospital care by reducing admissions seems like an effective strategy for significantly lowering costs.
But MedPAC found that ACOs have not significantly impacted inpatient admission rates, rather the organizations have achieved savings through post-acute care optimization.
County-level ACO penetration data showed that program participation and whether a hospital was participating in an ACO did not have a statistically significant effect on the change in a hospital’s total admissions or revenue. Instead, inpatient admission rates and revenue were more affected by hospital size and population.
ACOs also have more opportunity to generate shared savings by targeting post-acute care over inpatient utilization, MedPAC reported.
Their analysis of claims data from 2014 found that inpatient use for market areas at the 90th percentile of use was 1.16 times that for market areas in the 10th percentile. In contrast, post-acute care use for market areas in the highest percentile of use was 1.88 times that of market areas in the lowest percentile.
While MedPAC addressed hospital concerns with ACO participation, it also tackled specialist skepticism.
“Another concern is that specialists are not perceived to have a role in ACOs because attribution to ACOs is predominantly dependent on primary care visits, and thus specialists are not required for an ACO to meet the minimum number of attributed beneficiaries,” the commission explained.
“Also, some could be concerned that specialists would attract high-need patients to the ACO, thereby increasing its costs.”
However, MedPAC argued that high-cost patients would be factored into an ACO’s benchmarks to begin with, and those patients may be the ones who could most benefit from ACO care delivery.
Including specialists in an ACO would be mutually beneficial, the commission added. Involving specialists in an ACO would ensure the organizations can control spending and outcomes when attributed patients seek services outside of the ACO’s network.
Specialists in ACOs would be able to access historical claims data and potentially receive emergency department or hospital admission alerts. The doctors could also see additional referrals from providers in the ACO network.
Benchmarking and rebasing methodology challenges
Medicare ACO programs typically use historical spending to set financial benchmarks for organizations. However, MedPAC found that the high rate of churn in attributed patients is jeopardizing ACO success.
Only 66 percent of attributed patients were consistently assigned to an ACO over two years, while about 20 percent of attributed patients left the ACO each year, a cited analysis showed.
High patient churn spells trouble for ACOs because patients who lose alignment with an ACO tended to have lower spending growth compared to those who were consistently attributed and those who joined the ACO. Patients who joined the ACO also had the highest spending growth, MedPAC found.
“This preliminary analysis suggests that, although MSSP ACOs are to some extent controlling the spending growth for beneficiaries who are continuously attributed, there is a tendency for ACOs to have beneficiaries leaving who have lower growth in spending and beneficiaries joining who have higher growth in spending. Attribution is related to service use, which could be a source of concern when setting benchmarks or estimating savings,” the commission explained.
MedPAC suggested that ACO programs use prospective attribution to counter the effects of patient churn.
“Under prospective attribution, the year of data used to attribute an individual differs from the performance year data used to evaluate spending relative to the benchmark,” the report stated. “Therefore, an episode of illness that results in a beneficiary being attributed to an ACO will be in a previous year and thus in the benchmark.”
While determining benchmarks was a challenge, rebasing financial targets also presented an issue for two-sided risk ACO sustainability.
Redetermining financial benchmarks strictly based on historical spending creates diminishing returns for consistently efficient, cost-saving ACOs. But adjusting for patient severity and local spending (as in Medicare Advantage plans) allows organizations to know their benchmarks prior to participation, causing only ACOs below their regional benchmark to join two-sided risk programs.
MedPAC suggested that policymakers blend historical spending and a regional average when rebasing benchmarks.
The MSSP has recently implemented the blended rebasing approach by comparing the ACO’s average risk-adjusted spending over the past agreement period with the average risk-adjusted fee-for-service spending in the region.
If the ACO’s per capita risk-adjusted expenditures are greater than the regional average, then the organization’s benchmark is lowered toward the regional average. But if spending is lower, then the benchmark is increased toward the average.
Should CMS distribute the Advanced APM bonus differently?
Two-sided risk ACOs qualify for the five percent bonus under MACRA’s Advanced APM option, but patient and revenue thresholds for participation could stifle two-sided risk ACO sustainability, the report stated.
To qualify for the Advanced APM bonus, clinicians must have at least 25 percent of their Medicare Part B revenue stem from the Advanced APM or 20 percent of their patients attributed to that model in 2019 and 2020. These thresholds are also slated to increase in later performance years.
However, the revenue and patient thresholds create “payment cliffs,” meaning an ACO provider who has 24.9 percent of Part B revenue under the Advanced APM would not qualify for the bonus. This creates a disincentive to participate in two-sided risk ACO programs.
Instead, MedPAC recommended that CMS apply the five percent bonus to just the clinician’s Part B revenue derived from the Advanced APM, rather than his entire Physician Fee Schedule revenue.
“The proposed system would eliminate such payment ‘cliffs’ or discontinuities,” the report stated. “Instead, under our proposed refinement, the bonus would be certain because the incentive would depend solely on the clinician’s revenue coming through the A–APM, whatever that level may be.”
The modified Advanced APM bonus would also incentivize more eligible clinicians to participate in the advanced models, which is a goal of MACRA.
Other considerations for sustaining two-sided risk ACO programs
While MedPAC addressed several challenges with two-sided risk ACO participation, the commission also promised to continue studying how asymmetric models and Medicare Advantage impact two-sided risk ACO sustainability.
Policymakers recently developed an asymmetric ACO model, which has a greater shared savings rate than the shared lose rate, to encourage two-sided risk ACO participation.
Participation in the new MSSP track is currently limited to one three-year agreement period. But MedPAC questioned if the MSSP Track 1+ and other potential asymmetric models should be a temporary step to two-sided financial risk adoption or a permanent program in Medicare ACOs.
“It appears that Track 1+ could put the Medicare program at risk of financial loss if Track 1+ ACOs’ losses relative to the benchmark are greater than ACOs’ relative savings because of the model’s asymmetries. If Track 1+ were incorporated into permanent Medicare law, the costs may need to be offset if performance is essentially random,” the report stated.
MedPAC promised to monitor MSSP Track 1+ progress and results to determine whether the asymmetric model saves or costs Medicare relative to MSSP Track 1 and fee-for-service Medicare.
The commission also plans to address how Medicare Advantage and ACO programs compete.
“The ACO program is large, continues to expand, and continues to evolve. However, some suggest that MA plans are the more efficient model and that, eventually, ACOs should evolve into MA plans. As a matter of policy, the question is whether all ACOs should be encouraged to become MA plans or whether there are circumstances in which it is better for ACOs to remain ACOs,” the report stated.
Whether an ACO should take the step into Medicare Advantage depends on the local market. Therefore, MedPAC stated that “it is not clear a priori whether ACOs are in all circumstances a stepping stone to MA or should remain as ACOs.”
These unanswered questions may threaten two-sided risk ACO sustainability as these organizations mature. Policymakers should tackle these questions to ensure Medicare continues to save as care quality improves.
“ACOs in Medicare have proven to be a popular choice for providers, but whether they remain that way in the long run may depend on the choices policymakers make going forward,” MedPAC concluded.