American Medical Group Association
May 11, 2011
Re: Medicare Shared Savings Program: Accountable Care Organizations
We write today, however, to express our serious concerns over the direction of the Proposed Rule. On its face, it is overly prescriptive, operationally burdensome, and the incentives are too difficult to achieve to make this voluntary program attractive. As you know, most policy experts believe multi-specialty medical groups are best poised to become ACOs in the short term. However, in a survey of AMGA members, 93 percent said they would not enroll as an ACO under the current regulatory framework.
Our membership’s concerns were many and focused on issues such as the risk sharing requirement, static risk adjustment, retrospective attribution, quality measurement requirements, the Minimum Savings requirements and others. Without substantial changes in the Final Rule, we fear that very few providers will enroll as ACOs and that CMS and the provider community will miss the best opportunity to inject value and accountability into the delivery system.
Letter to Donald Berwick, M.D., Administrator, CMS
The Everett Clinic
Univ. of Michigan Med. School
St. John’s Clinic
Novant Medical Group
Middlesex Health System
May 12, 2011
On behalf of the multi-specialty groups participating in the Physician Group Practice (PGP) Demonstration Program, we are writing to you today regarding Section 3022 of the Affordable Care Act’s notice of proposed rule-making (NPRM) for the Medicare Shared Savings Program/Accountable Care Organizations (ACOs).
However, as presently proposed, we ALL have serious reservations about the economics and the complexity of the Medicare Shared Savings Program/ACO NPRM. All of our organizations are planning to individually respond with our own respective concerns, but broadly, the “Shared Savings” model/NPRM has the following aspects that are problematic:
* There is downside risk during the initial 3-year term, unlike the recently concluded PGP Demonstration project. Such downside risk is compounded by significant investment cost on the part of the ACO.
* Savings are measured net of 2% threshold for the one-sided risk model. Additionally, the Minimum Savings Rate (MSR) is set at high levels for ACOs with lower enrollment.
* Limits placed on accounting for beneficiary acuity level that is documented and appropriate will dilute true savings realized by the ACO, and is a disincentive for management of patients with complex care needs.
* There are a large number of quality measures, especially new quality metrics in several domains, that go into effect starting year one. As an example, on average, it costs about $30,000 just to program a single new quality metric. This NPRM has more than 60 new ones, which equates to nearly $2,000,000 for each organization. The excellent results produced by the PGP demonstration are evidence of the benefits of a careful expansion of quality measures.
* Retrospective attribution places limits on the ACO’s ability to bend the cost curve. It impedes optimal patient engagement, timely program planning and course correction, and compounds underlying issues of claims lag and financial settlement.
* The logistics associated with Medicare beneficiaries’ opt-out of the ACO program is simply not practical. We believe this would lead to beneficiary and physician confusion on the terms of engagement.
As currently proposed, ACOs have a greater potential for incurring losses under either track, than for generating savings. This risk-reward imbalance makes it difficult, if not impossible, for internal decision-makers to accept the financial design.
ACO start-up costs higher than estimated, AHA study says
By Rich Daly
ModernHealthcare.com, May 14, 2011
Accountable care organizations will likely face start-up and first-year costs six to 14 times higher than HHS has estimated, according to a study released by the American Hospital Association.
The study, which was based on an analysis of previous research, concluded that the various elements required to successfully manage the care of a specific population will cost between $11.6 million and $26.1 million—depending on the size of the hospital or hospital system involved in the ACO—and far more than the $1.8 million estimated by the CMS in its proposed rule.
The cost findings are based on 23 different capabilities that ACOs will need to develop across four categories: network development and management; care coordination, quality improvement and utilization management; clinical information systems; and data analytics.
By Don McCanne, MD
The accountable care organizations (ACOs) called for in the Accountable Care Act (ACA) don’t seem to be getting off to a good start. Most medical groups do not intend to participate, the hospitals have found that they are too expensive, and the distinguished participants in the Group Practice Demonstration Program all have serious reservations about the proposed rules for ACOs under the ACA. What is going on here?
The goal seems to be to establish integrated health care delivery systems with varying design characteristics based on local needs and logistical considerations. Size and complexity would vary greatly. Such systems should provide a primary care base for entry into the system and for coordination of services. Specialty care should be integrated within the system, providing appropriate access and interaction with primary care. Information technology systems should enable access to clinical records while providing tools that reduce error and duplication of services, while assisting with clinical decisions. Basically, we want health care professionals and hospitals working together to provide more efficient, higher quality care for everyone.
That is the concept behind ACOs. Integrated systems would be rewarded based on quality and efficiency (i.e., lower costs). Because of political considerations, Congress decided that they should apply the ACO concept exclusively to Medicare, while letting the private insurers supervise their own quality and cost programs.
The simplistic solution was to establish the Medicare Shared Savings Program. Reduced costs would be built in by rewarding the ACO systems with a portion of the funds that they saved, and ensuring quality by making the rewards dependent on meeting certain quality measures. The problem with this became evident when it came time to write the rules. Not only were the systems required to enable this very expensive, the rules would create what bureaucracies are so adept at creating – a bureaucratic quagmire. The letter from the participants in the Group Practice Demonstration Program list some of these aspects that proved to be problematic.
Bundling of payments has been another approach, but only certain clinical presentations conform well to bundling. So it would not be an answer to the majority of clinical problems. Also bundling removes risk from the common financing pool and splits it up amongst the provid
ers of health care. Luck or lack thereof plays too great of a role to ask the providers to place bets on it merely by agreeing to accept patients with conditions subject to bundling.
Much of the difficulty has stemmed from the new, widely touted principle that we should no longer pay for volume but instead we should pay for quality and efficiency, but both quality and efficiency have proven to be very difficult to define. The default has been a few parameters of each which in no way can represent the broader spectrum of quality and efficiency.
Health care is a lot of work. No matter what else, the delivery system will have to be paid primarily on volume, but we need to get the volume right. Much attention has been paid to excess volume being driven by the rewards of fee-for-service care. In fact, the largest volume increase has been in imaging services, yet most physicians who order the procedures are not compensated for these services. The Dartmouth variations do remain of concern and further attention in defining these is certainly warranted.
The greater problem with volume has been demonstrated to be underutilization of services. According to a RAND study, most patients receive only about half of the care that they should have. The emphasis on reducing the volume of care has been somewhat misguided since there is a greater need to increase the volume, though we need to get the pricing right when we do that. The SGR dilemma must be addressed, but the basis for it will not go away.
The bottom line? Integrated health systems that are designed for optimal patient service are an admirable goal, but integrated health systems designed predominantly as a business model are not. ACOs designed to comply with the Medicare Shared Savings Program are an abstract concept that will never fly, as today’s quotes indicate. On the other hand, ACOs designed as a business model – the so-called commercial ACOs – are just another label for the intrusive managed care organizations that most of us have been trying to get away from.
The goals of higher quality at reasonable costs would best be achieved by humanitarian integrated health systems financed not by a dysfunctional, fragmented system of private and public plans, but rather by a publicly-financed and publicly-administered universal risk pool – an improved Medicare for everyone.
The ACO misstep is more of a problem of well intentioned but misdirected policy application. The now-tarnished ACO label should be dropped as we move forward with mutual efforts to improve the quality and efficiency of our health care delivery system so it works well for all of us.