On December 22, the Centers for Medicare & Medicaid Services (CMS) announced that 89 new organizations will join the Medicare Shared Savings Program (MSSP) in January 2015. These accountable care organizations (ACOs) will bring approximately 23,000 additional physicians and other providers into the ACO program, and will likely be the last group of program entrants before new rules – released for public comment on December 1 – will go into effect.
As the last “class” of ACOs enter the program under regulations finalized in November 2011, I can’t help but reflect on what we have learned so far. Earlier this year, CMS released information on the “early” results achieved by the 2012 and 2013 classes of MSSP ACOs. At a high-level, these results showed that the ACOs improved on 30 of the 33 quality measures in the first two years, and that they outperformed group practices on 17 of 22 measures. CMS also reported a combined total program savings of $417 million to Medicare across the MSSP and Pioneer ACO programs.
While these results focused on the top-line takeaways, there are many program insights that have not been captured in a headline or fact sheet. Some of this information was included in the recently released proposed rule, and would only have been noticed by those who dug into the nitty gritty regulatory language.
- More patients than anticipated are sharing claims data with their ACO. In the proposed rule, CMS noted that it has shared claims data on over 5 million beneficiaries with over 200 MSSP ACOs since program implementation began in 2012. Of the approximately 5 million beneficiaries, only about 100,000 – or 2% – have opted not to share their claims data with an ACO. Although CMS indicated that this is consistent with other initiatives that have included data sharing, until now, the conventional wisdom was that at least 5% of patients were opting to withhold data from their ACO. Although the difference between 2% and 5% may not seem that significant, the delta is an additional 150,000 Medicare patients for whom data is available to the ACO.
- Almost all of the patients assigned to an MSSP ACO are assigned based on the primary care services they receive from their primary care physician. One area of significant stakeholder concern has been the two step process CMS currently uses to assign beneficiaries to a particular ACO. Explained in very simplistic terms, the first step of the process assigns patients to an ACO by evaluating whether they received a plurality of their primary care services from a primary care physician participating in the ACO. If the first test is not satisfied, a beneficiary could be assigned to the ACO based on primary care services provided by any physician and other providers (e.g., nurse practitioners, physician assistants, clinical nurse specialists) participating in the ACO. Although many stakeholders were concerned that a large number of patients would be assigned in step two, this has not turned out to be the case. In the proposed rule, CMS noted that for the first 220 ACOs participating in the program, on average, about 92% of the beneficiaries were assigned to an ACO in step one – based on the primary care provided to them by their primary care physician. CMS had to look to primary care services rendered by other physicians and providers for only 8% of beneficiaries participating in the program.
- A significant number of patients “churn” out of ACOs over the course of a year. Long suspected to be the case, CMS confirmed in the proposed rule that wide variation was seen in the list of patients assigned to an ACO from one performance year to another. On average, 76% of beneficiaries assigned to an MSSP ACO at the end of the first year were still assigned to the same ACO at the end of the second year – reflecting a “churn” rate of 24%. Of this 24%, only 7% dropped off the ACO’s list of patients because they were no longer eligible. The remaining 17% of patients dropped off the list because they no longer received a plurality of their care from a provider participating in the ACO. In most instances, these patients received some, but not a “plurality,” of care from a provider participating in the ACO (e.g., 7% had at least one physician or non-physician primary care visit, but none with ACO professionals; 7% had at least one primary care visit with a physician in the ACO, but did not receive the plurality of their primary care services from ACO professionals; 6% had at least one primary care physician visit with a physician who is an ACO professionals, but the plurality of their primary care services were rendered outside the ACO).
- ACOs are constantly adding new providers/suppliers. CMS indicated that 17,000 new providers were added to existing MSSP ACOs in 2014 alone. In the proposed rule, CMS frequently noted that it devotes tremendous resources to making these changes and that it was consequently suggesting a number of proposals to streamline the process for an ACO to make changes to its provider/supplier list. Although the reasons for the frequency with which these changes are being made likely vary, one could hypothesize a variety of reasons. For example, providers may drop out because they realize they are unable to meet program requirements. On the flip side, providers may want to join the program because they find the potential “upside” attractive or because they have an existing partnership with an ACO participant. CMS and stakeholders will need to continue to work together to ensure efficiency and promote transparency regarding how and why providers are being added to/removed from the list of ACO providers/suppliers.
- Pioneer ACOs are serving as incubators for innovations that may eventually be implemented in the MSSP. In the proposed rule, CMS noted that it is conducting several pilots in the Pioneer ACO program that may be extended to the MSSP in the future, especially to two-sided or risk-bearing MSSP ACOs. The first is a waiver of the requirement that beneficiaries must have a prior inpatient hospital stay of no fewer than three consecutive days to be eligible for Medicare coverage of inpatient skilled nursing facility (SNF) care (the so-called “SNF 3-day rule”). As long as safeguards are in place to ensure program transparency and prevent abuse, CMS noted that it believes that such a waiver could “allow ACOs to realize cost savings and improve care coordination.” The second pilot is a test of beneficiary “opt in” attestation for the 2015 performance year. Participating Pioneer ACOs will be allowed to mail certain beneficiaries a cover letter, asking them to confirm that an ACO participating provider/supplier is their “main doctor.” If the patient confirms the relationship, he/she will be aligned to the Pioneer ACO for the following performance year, regardless of whether or not the practitioners participating in the Pioneer ACO rendered the plurality of the beneficiary’s primary care services during the performance year.
- Additional regulatory changes may be needed to integrate other delivery system and payment reforms into Medicare ACOs. As Sean Cavanaugh, Deputy CMS Administrator and Director of CMS’ Office of Medicare, noted in his December 22 blog post, ACOs are “one part of this Administration’s vision for improving the coordination and integration of care received by Medicare beneficiaries.” However, for the most part, there has been little discussion on how – or if – MSSP ACOs should begin to incorporate other delivery system and payment reforms currently being tested (for example, what would happen if bundled payments were incorporated into the MSSP ACO model?). A technical correction included in the proposed rule may, however, have given us a brief glimpse into the potential difficulties associated with combining or overlapping models. In the recently released rule, CMS proposed a change to the definition of “hospital” to allow Maryland’s acute care hospitals to continue to participate in the MSSP. Under the existing definition, Maryland’s hospitals would not be considered “hospitals” because they are subject to a waiver from the Medicare payment methodologies under which they would otherwise be paid. Although relatively minor as a standalone proposal, this correction may foreshadow additional regulatory changes that may be needed when organizations start to layer reform efforts on top of other reform efforts such as accountable care. As models continue to develop and mature, this will no doubt be an area of ongoing dialogue between CMS and its stakeholders.
With 405 MSSP ACOs serving 7.2 million beneficiaries, there are many stakeholders across the country invested in the success of this program – one of the largest Medicare demonstration programs to date. We must all continue to look beyond the headlines and connect the dots to facilitate continuous improvement of the program and our health care system.
