With health reform currently in limbo, one of the few areas that command bipartisan support is reforming provider payments to promote integrated-care delivery models such as accountable care organizations, episode-based payments, and medical homes, Paul Ginsburg said as he introduced a February 8 panel on this subject at the National Health Policy Conference. As a result, even if Congress does not pass a comprehensive reform bill, payment reforms could still be enacted, for example as part of a Medicare bill blocking currently mandated physician payment cuts, said Ginsburg, the president and CEO of the Center for Studying Health System Change.
The problem, Ginsburg said, is that “we’re not ready yet … development is needed, piloting is needed, a process to go from piloting to implementation is need.” Ginsburg’s panel therefore addressed the issue of how we get from where we are now to where we would like to be, with an emphasis on three questions: 1) What piloting and development activities are needed?; 2) What kind of governance mechanisms at the Centers for Medicare and Medicaid Services could best move this process forward?; and 3) How can private insurers participate in developing new models of payment reform?
The NHPC is co-sponsored by AcademyHealth and Health Affairs. This year’s version offered those who braved the weather a good overview of hot health policy topics, including the prospects for health reform. For example, at a plenary session featuring congressional staffers, Wendell Primus, top health policy adviser to House Speaker Nancy Pelosi (D-CA) predicted that comprehensive health reform would be almost completed by Congress’ Easter recess. Primus said the House would pass the Senate bill and that both congressional chambers would pass modifications to the Senate bill through the budget reconciliation process, under which Senate passage requires only 51 votes and no filibuster is allowed. Other speakers included Health and Human Services Secretary Kathleen Sebelius.
The limiting factor is not a lack of ongoing experiments in payment reform, said one panelist in Ginsburg’s panel, former Centers for Medicare and Medicaid Services Administrator Mark McClellan. He pointed out that Medicare is currently experimenting with accountable care organization-like models through its Physician Group Practice Demonstration program. He also cited two Medicare Health Quality Demonstration programs in North Carolina and Indianapolis, authorized under section 646 of the Medicare Modernization Act of 2003, involving Medicare and other private and public payers.
McClellan: No Time To Waste
The problem is that the cycle time of learning from and implementing the lessons of these experiments is far too long, said McClellan, who now directs the Engelberg Center for Health Reform at the Brookings Institution. He said the United States has only three or four years to make the health care system more cost-effective before debt concerns will force blunt cuts that will undermine the effort to create “21st century personalized medicine.”
Using the examples of the North Carolina and Indianapolis demonstration projects, McClellan outlined some of the steps necessary to bring payment experiments from conception to reality: turning the legislative language into a clearer conceptual framework; outlining contractual terms and looking for participants through requests for proposals and other means; actuarial analysis to ensure that the project has performance metrics and will be budget neutral; and clearance by the Department of Health and Human Services and the Office of Management and Budget.
The result: despite bipartisan support for the demos and treatment of them as priorities by CMS, the projects are just underway only now, seven years after they were authorized by the MMA in 2003. The projects will take another three years to conduct and another year or two to evaluate, according to McClellan. “That cycle time has to come way, way down,” he said.
How can the cycle be shortened? McClellan outlined several steps. First, he said, we need standardized quality and cost measures and the ability to track those measures on an ongoing basis. “It’s 2010 – it’s time, not just in Medicare but in all of our health insurance programs, to have an ongoing ability to measure important aspects of quality of care and the cost of care. If such measures were collected consistently and on an ongoing basis, that alone would probably shave a couple of years off the time for conducting one of these demonstration programs, because I will tell you from experience, a lot of the negotiating time involves determining what are the appropriate ways of measuring quality of care and cost impacts.”
McClellan cited progress in constructing such measures. “There are a number of National Quality Forum-endorsed performance measures that can be calculated using longitudinal administrative data,” he observed, adding that the Engelberg Center was working on making consistent versions of those measures widely available across Medicare and private plans. “That’s coming very soon, I think, but it should be possible to get even richer data more widely available. Pilot programs around the country are incorporating electronic lab results, imaging results, and in some cases working on more sophisticated clinical registries.”
More broadly, McClellan advocated the use of standard templates for approving, conducting, and evaluating experiments, such as CMS used in connection with the Medicare Health Support program, another program authorized by the MMA. Similarly, CMS has created a model waiver form for the “money follows the person” reform in the Medicaid long-term care arena.
Pham: Considering Reactions By Providers And Politicians
Hoangmai Pham spoke about the need to consider provider reactions to different payment methods. “You can have a great model, but if you don’t have an audience that is willing to accept it, it’s not going to get very far,” said Pham, senior health researcher and co-director of quantitative research at the Center for Studying Health System Change. She said provider reactions would vary according to factors such as how much financial risk particular payment structures imposed on providers, and how novel the payment structure being considered was. The more risk imposed on providers, and the more novel the payment model, the more experimentation would likely be needed before providers bought into the model.
One “gaping hole” where more experimentation is needed, Pham added, is risk adjustment. “When we talk to providers about care integration and risk-sharing, bundled payments, and ACOs, they get their backs up over the issue of patient case mix,” she said. “Payers have to be honest: We don’t know how to case-mix adjust for episodes of care. We can’t even agree on the definition of an episode of care. All of our case-mix experience has been on years of whole-patient data, and we are just translating that in a very improvised way to other units of payment.”
Pham applauded the proposal for a Center for Payment Innovation at CMS, and she urged that any such Center be open to ideas for innovations from the field, not just innovations developed at CMS. “There are likely to be a number of communities and pockets of providers and private payers out there who have some really novel ideas that maybe haven’t received the exposure that medical homes and ACOs have.”
But, she warned, “I don’t think much of it matters unless the resources are there, not just for the Center but for the agency as a whole, which is understaffed and working on a shoe string all the time.” As an illustration of CMS’ ailing infrastructure, she spoke of her experience as part of the agency’s medical home demonstration design team: “We got stuck on the issue of how to turn around a capitated payment to an individual physician on a monthly basis. We couldn’t do it with the way that the computers and the machinery and the software at CMS are set up.”
Pham also emphasized the importance of political protection: “The agency needs protection so that, for example, the Physician Quality Reporting Initiative does not need to need to cater to every single physician specialty, in order that every specialty be eligible to earn bonuses. It should cater to the national quality priorities … rather than political concerns.”
Pham said she was open to various ways of providing political protection to CMS. She and Ginsburg have written about the idea of an independent medical commission, which was included in the Senate bill. Ginsburg said that he and Pham “were initially excited about all of the attention it’s gotten in the reform debates.” However, many people in the provider community have come to “think of the independent board idea as a mechanism to ratchet down their payment rates.” Ginsburg expressed concern that this perception would eclipse the original concept of the board as a protection for CMS against political interference.
Sandy: What Private Insurers Bring To The Table
Private insurers have both advantages and disadvantages in testing payment reforms, said Lewis Sandy, Senior Vice President, Clinical Advancement, UnitedHealth Group. On the positive side, private insurers can try innovations out more quickly, and they can revise experimental designs more quickly; in government programs, it is very hard to be flexible, because the design tends to be congressionally mandated and locked in stone,” he said. Sandy also agreed with Pham that private insurers know their markets and the providers in those markets much better than Medicare does.
On the negative side, with a few exceptions, private insurers have relatively small market shares and lack good mechanisms to coordinate across particular markets. “If health plan A tries innovation one and health plan B tries innovation two in the same market, you’re creating noise and churn and confusion.”
To deal with this problem, Sandy called for “a mechanism to deal with ongoing collaborative activities within a market, so the delivery system is looking at consistent signals regarding payment reform, and the delivery system can evolve in a coherent way.” Elaborating on this, he urged “very explicit multi-stakeholder collaborations” to move from fee-for-service to more integrated payment models. The pace of this movement would vary from market to market depending on the unique characteristics in each market.