Several weeks ago CMS announced its newest accountable care organization (ACO) program, the Next Generation ACO. A significant departure from past Medicare ACO programs, the Next Generation ACOs will again challenge providers to decide what type of risk is best suited for their organizations.
Why not become a Medicare ACO? The administrative and organizational cost of developing a high-performing ACO is great, measured in dollars, effort and pain. In addition to the application and ongoing compliance costs, it is necessary to transform many clinical and operational workflows and systems within the organization. The return needs to justify this investment.
Beyond cost, the real question for provider organizations should be about both the level and nature of the risk assumed. Although Pioneer had higher limits than the Shared Savings programs, both programs have very limited upside opportunity, greatly limiting how much risk the provider would take. In practice, only five percent (24 of 424) of providers under current Medicare ACOs have taken symmetrical (two-sided) risk.
More importantly, the nature of the risk under Pioneer and Shared Savings is hard to manage, as the retrospective attribution model is a design of CMS administrative convenience and their outdated dedication to “any willing provider” thinking. While I applaud CMS’s desire to retain open access, payment reform in healthcare must result in winners and losers, where poor performing providers lose their business to providers offering more value. Given the limits CMS has created on member engagement, Medicare ACOs are being asked to take accountability without the needed ability to influence or manage patients.
Finally, another flaw in the nature of the risk under existing Medicare ACO programs is the future performance measurement based on a provider’s recent historical performance. A provider that has spent millions over the years to improve the quality and efficiency of their care is penalized because their benchmark for improvement (and financial success) is far more difficult to attain. Moreover, the baseline moves each year, making successive improvement even more difficult.
This is why I challenged providers to consider truly taking on full risk and becoming a Medicare Advantage plan. Any organization considering the launch of a completely new and wholly-owned insurance company should stop and take a deep breath; it is a difficult and serious decision. But it comes with advantages: The chance to further monetize a provider’s investment in their brand (patients sign up for the insurance from the provider they love), to control and understand the risk they are assuming, and to fully retain 100 percent of the savings possible.
Launching a Medicare Advantage plan is the most extreme form of a provider taking risk. There is a wide gulf between it and the limited risk levels in Pioneer and Medicare Shared Savings Program (MSSP). The Next Generation ACO programs not only fall well in-between these risk levels but also offer a better nature of risk.
New level of risk
Perhaps the most important new feature of the Next Generation ACO program is the change made to the level of risk. The program allows providers to earn a higher share of savings (80-100 percent). It also adds two new ways to limit unpredictable losses, with both a limit on aggregated savings and losses and the exclusion of individual level expenses beyond the 99th percentile.
These limits allow CMS to essentially serve as reinsurance for a provider, protecting them from catastrophic outcomes that cannot be planned or prevented.
Another major advancement in the program is the promise of capitation (instead of fee-for-service) in later years. The details here (to be determined in the future) are important as capitation models can simply result in underutilization and shifted risk, but stepping away from fee-for-service as a payment method is a critical step for CMS.
Providers will be glad to see their benchmark prospectively set at the start of the year. This will greatly improve predictability and the ability of the provider to manage against a known budget. While the program will include better benchmarking opportunities in the fourth and fifth years of the program, it does not sufficiently address the existing problems of being penalized today for already delivering high value care.
Finally, there are better but still inadequate measures to engage patients. The Next Generation ACO offers the option for a patient to “opt in” to an ACO, thus potentially replacing the disastrous retrospective attribution model. Further, providers will have new options to engage members, to include certain incentives of value to help the providers influence and manage patients.
Neither of these last measures go far enough. Unless 100 percent of patients opt in and that is the only attribution process, there is still retrospective attribution occurring. Also, CMS has not detailed the limits on patient engagement.
Taken as a whole, these changes do a good job of changing the limits on the risk and significantly improve the nature of the risk. CMS intends to constrain this program to a small number of organizations that have the capability today to be successful under these terms, with its greater operational requirements and the significant downside risk that comes with the better upside risk.
Many of the providers who fit this new model are also candidates for launching Medicare Advantage plans. The decision between Next Generation ACO and Medicare Advantage becomes more difficult for the small number of provider organizations that can now be successful under either.
However, as an industry, we still need better programs for the large number of providers that are not candidates for the Next Generation ACO yet need better than the Shared Savings program.
Jay Sultan is principal strategy advisor at Edifecs.