The Honorable Kathleen Sebelius The Honorable Donald Berwick, M.D.
Secretary, Health and Human Services Administrator, Centers for Medicare &
200 Independence Avenue, SW Medicaid Services
Washington, DC 20201 Washington, DC 20201
Dear Secretary Sebelius and Administrator Berwick,
We write in response to the Department’s proposed rule for Accountable Care Organizations
(ACOs). We strongly believe that, when properly organized and funded, ACOs have tremendous
potential to improve the quality and reduce the cost of health care in America.
A delivery system focused on the quality of care is crucial to reining in spending, improving
patient outcomes, and truly bending the health care cost curve. While the underlying concept of
the rule is strongly supported, some providers have expressed concerns about their ability to
comply with it and, therefore, believe they may not be able to establish or participate in an
Provider concerns fall into the following four categories:
Many providers are concerned that there is an imbalance between the risk and reward
equation. For an organization to apply to become part of an ACO, there must be some
assurance that such a decision has an opportunity to become financially beneficial. Due to the
retrospective payment of the shared savings payments and the requirement that all ACOs
assume downside risk for some portion of the agreement period, many organizations believe
there is too much risk on their part to assume the upfront investment costs with no certainty for
returns over the long term. In addition, leading states that have already set a high bar for a low-
cost to high-quality ratio have less room for success in achieving the benefits. While it is critical
to help lift up low performers, we should not penalize those who have already taken important
first steps to improve the quality and lower the cost of health care.
In the proposed rule, patients will be assigned to an ACO at the end of the year based on the
primary care physician who cares for the patient during the course of that year. ACOs need to
know who their patients are so that they can manage them effectively. This increased case
management is a key component of the ACO model and the lack of a prospective attribution of
patients may hinder their success.
In the proposed rule, CMS states that it will make certain data available to the ACOs. In order to
successfully implement the types of innovations CMS is requesting, complete and timely data is
The final rule must allow organizations to collaborate, particularly in rural states. While the
proposed rule creates a safe harbor from anti-trust review for organizations with less than 30
percent of the market share, a mandatory review is required for organizations with a 50 percent
or greater market share. In rural states, which frequently have only one major hospital per
service area, there would almost certainly be a mandatory review by the Federal Trade
Commission (FTC) and Department of Justice (DOJ) before the ACO could be established.
This review could present a significant burden to aspiring ACOs, particularly in rural states.
These four practical concerns can be adequately addressed without undermining the objectives
of the proposed rule:
The financial implications can be overcome if the rules for ACOs allow organizations to recover
overall investment costs associated with implementing necessary clinical reforms and provide a
balanced model to consider the financial investment of providers. Additionally, risk-bearing
models which free up resources for improving care that the current fee-for-service models do
not permit could increase financial resources available to support improvement. Allowing all
ACOs to share in first-dollar savings once the “minimum savings rate” is exceeded would help
mitigate concerns about the balance of risk versus reward.
Concerns about patient attribution can be mitigated if CMS provides regularly updated
information about the patients who are likely to be assigned to providers. A policy of
prospective assignment would enable ACOs to better understand the effects of the interventions
and care processes. Ultimately, prospective assignment would allow providers to identify
delivery system improvements that are working and implement successful interventions more
Providers will need timely and complete data for ACOs to succeed. CMS should err on the side
of providing ACOs as much data as necessary to ensure the successful development of the
coordinated care model.
Finally anti-trust concerns could be alleviated if DOJ and FTC provide further guidance on how
collaborative models can be structured to avoid anti-trust concerns, particularly in rural areas
where the proposed rule would trigger a mandatory review. Clarity on how safe harbors will be
implemented for health systems in rural areas would reassure providers interested in
participating in an ACO.
If these four issues are adequately addressed, we believe many concerned providers will
constructively engage in this process to help us achieve our mutual goals higher quality health
care at a lower cost. We look forward to hearing back from you at your earliest convenience.
Member of Congress