Earlier this week, the Centers for Medicare & Medicaid Services announced that it will be expanding its existing portfolio of Accountable Care Organization (ACO) programs to include the “Next Generation ACO Model,” designed for providers experienced in coordinating care for patient populations. Oliver Wyman’s Niyum Gandhi draws from his work designing value-based systems around the country to highlight key updates:
The Next Generation ACO Model initiative builds on experience from the Pioneer ACO Model and the Medicare Shared Savings Program (MSSP). The new model addresses a lot of the issues that arose with the implementation of the MSSP option, essentially carving out a new ACO category rather than significantly revising MSSP. While this is welcome news, especially for leading providers, we would ultimately like to see the enhancements apply across-the-board in all ACO programs, as we’ve discussed here. In the meantime, here’s what earns Next Generation its name:
- Prospective setting of the benchmark. Thus far, participating in a CMS ACO model has been a little like playing in a baseball game where you don’t know the score you need to beat—much less your own score—until the game is over. This change will provide target performance savings numbers upfront. The model for setting the benchmark also includes adjustments for ACOs that have historically performed well and for regions that are historically less expensive—up to a 1.5% addition to the benchmark for those factors. While this tweak addresses concerns of low-cost ACOs, it’s actually offset by a 2-3% discount that varies based on the quality of the ACO. Even if a provider was previously considered highly efficient in a given region, with good quality, the organization will still need to beat its benchmark by at least 0.5%. This discount factor effectively replaces the minimum savings rate that frustrated many ACOs that were beating their benchmarks, but not by enough to qualify for savings.
- Risk adjustment is no longer a one-way ratchet. It’s capped at a 3% increase per year to prevent gaming the system. But at least the new program acknowledges that the average risk score for members can go up year over year, and the policy is consistent with CMS’s intent to not reward for documentation improvement, only for actually taking on higher risk patients. As the model transitions to a region-based benchmark in future years, this will presumably align even more with Medicare Advantage (MA).
It makes sense that CMS is moving into this model as a small program for leading ACOs, but ultimately, we’d like to see all of these enhancements be standard in the ACO programs, regardless of the level of risk assumed. – Oliver Wyman’s Niyum Gandhi
- Move to a region-based benchmark. Perhaps most importantly, instead of the benchmark resetting after the first three years and ACOs having to outperform prior performance, their benchmark will now likely move to a region-based model for years 4 and 5. This shift to “valuing attainment more heavily than improvement” creates long-term sustainability of the program, rather than a target that becomes harder and harder to hit over time. CMS sought comment on this as a benchmarking option in its latest MSSP revised rule, which we addressed here. It’s encouraging CMS took the only truly viable path towards program sustainability.
- More risk and more payment options. Options for taking on risk: either 80% sharing moving to 85% sharing (in gains and losses) or all the way to 100% right off the bat. Payment options include normal fee-for-service (FFS) payments with a retrospective true-up to either of the risk arrangements; taking some payment upfront; population-based payments; or moving all the way to capitation in the latter years. And in the capitation model, ACOs can pay participants rates other than Medicare FFS rates, which really does make it more like MA.
- Voluntary alignment. Beneficiaries can choose their ACOs and will be rewarded by CMS for keeping their care within an ACO. This is the first step CMS has taken into beneficiary incentives in any of their ACO programs. Many commenters have been urging this direction for awhile. CMS can now stay committed (within Medicare FFS) to allowing beneficiaries to use any Medicare provider, while at the same time it offers an incentive for staying “in network.” The minimum bar here—of just 50% of Part A and B services—is probably too low, as many ACOs have in-network utilization in the 40-60% range already. But that is just a minimum for now. CMS plans to set the exact threshold in the participation agreement, and we recommend a significantly higher range (70%+) for ACOs meeting CMS network adequacy requirements.
- Benefit enhancements. CMS included all of the potential benefit enhancements they discussed in the proposed revised MSSP rules in this program. This includes the 3-day SNF waiver, telehealth expansion, and post-discharge home visits.
The Next Generation program addresses many of the most common concerns with ACOs, but it is a lot of risk for some organizations. It makes sense that CMS is moving into this model as a small program for leading ACOs, but ultimately, we’d like to see all of these enhancements be standard in the ACO programs, regardless of the level of risk assumed.
CMS has indicated that they want to use program design to reward ACOs willing to move from the farm leagues of shared savings into the major leagues of greater risk. The upside should indeed be greater for those taking downside, but they should at least be allowed to play by the same rules.