With the release by the federal Centers for Medicare and Medicaid Services (CMS) of the proposed rule for accountable care organization (ACO) development on March 31, experts having been poring over the details of the rule and analyzing its implications for providers of different types. One industry expert who sees ACO development as an expert-level type of enterprise is Paul Keckley, Ph.D., who leads the Washington, D.C.-based Deloitte Center for Health Solutions, a division of the global Deloitte consulting firm. Keckley spoke this week with HCI Editor-in-Chief Mark Hagland regarding his analysis of the proposed rule. Below are excerpts from that interview.
Healthcare Informatics: Overall, what is your impression of the proposed rule?
Paul Keckley, Ph.D.: I think it’s quite complicated and comprehensive; it seems to me that the delay from when it was first anticipated in January until March 31 was the result of caution [on the part of federal healthcare authorities] in making sure that they could accommodate varied types of markets for those with very well-integrated systems, and also those that are fairly fragmented. Second, they seem to have tried to accommodate antitrust issues; there’s been a question since ACOs were first included in the healthcare reform legislation, that they could potentially create cartels, in which doctors, hospitals, and long-term care, for example, potentially might become so integrated that it would impact costs.
Paul Keckley, Ph.D.
So I think that the delay partly speaks to their addressing those issues. I didn’t see anything in what I’ve now read twice that was necessarily surprising. You definitely see the complexity of this in various sections of the rule. For instance, the way they’ve calibrated quality scores, up to 130 points, with 62 of the points with treatment of various elderly populations; or the methodology they’re using to allocate payments.
This is a rule that’s now being put out as CMS’s best effort. I think it will get a lot of reaction. I don’t think the ACO is necessarily for everyone; but I don’t necessarily think it was intended to be. When you read the Affordable Care Act, you read about accountable care organizations, the medical home, bundled payments, value-based payments, and readmissions work; and the glue that holds all that together is information technology. You must have both clinical and administrative data-sharing. And I think we’ve now seen one piece of that puzzle in this proposed rule. It’s consistent with what else we’ve seen in the Affordable Care Act. And I anticipate they’ll make some changes based on feedback.
HCI: With regard to the creation of two risk models, one involving “one-sided” risk and the other involving “two-sided” risk models, we’ve had experts tell us that they believe very few provider organizations would be able to handle the “two-sided” model, and some are even expressing skepticism about the terms around the “one-sided” model. Your thoughts?
Keckley: I can understand that comment, but we’ve got about 120 integrated systems already taking on risk, many of them already owning their own plans. I believe that the fact that they created the two models means that they believe there are groups ready to take on the risk. And the fact is that under the two-sided model, you get 60 percent of the savings if you achieve 90 percent or higher on the quality scores, whereas you get 50 percent of the savings if you achieve 90 percent or higher on the quality scores under the one-sided model; so the differences aren’t that dramatic.
HCI: So you think some groups will be able to take on the two-sided model?
Keckley: I think what was intended was that they wanted to set the bar low enough to get 75-150 organizations that would pursue this. This is actually a rule, and in this 429-page rule, there is this explicit statement that they anticipate between 75 and 150 organizations to be eligible as either one-sided or two-sided ACOs; and between 1.5 and 5 million Medicare enrollees would be encompassed.
And they said they’re really not sure how many will anticipate, and how effective this will be in reducing costs; and they’ve put this out for comment. So I think they’ve produced a very intellectually honest rule. And they started with a whiteboard, and this is very complicated, and they were pretty intellectually honest here. So I’m impressed with the scope of the rule.
There are a few things that weren’t included that I’m wondering about. One is the start-up costs for developing an ACO. You wonder how they’ve allocated the quality points in these five zones. But those are part of the natural process of rulemaking. And I think again, it’s consistent with this theme of the integration of the healthcare system, that it’s connected, it’s coordinated care, and it’s pay-for-performance. And I think what they’ve done is provided glimpses of what will come out of the final rules around the medical home, and episode-based payments, and some others.
HCI: What are the key gaps you see in terms of IT preparation for ACO development?
Keckley: The two that people will usually begin with are the interoperability issue, and the degree to which you have systems that are meaningfully useful; and the ways in which privacy and security can be addressed. And then if you get past those two, you have two bigger issues. One is clinical integration: you’ve got to rework clinical processes so that you use clinical information in delivering patient care. You can’t buy clinical information technologies and then not change patient care. So the clinical processes have to change, and you have to train and retrain people to better use tools; that’s a huge issue.
And the second issue is the funding for information technology. It does not appear that the $27 billion in the stimulus act will come even close to funding all the implementation needed in clinical IT. It’s clinical data warehouses, and data sets. And I think the industry is being fairly aggressive about solving the interoperability issues. I spend a lot of time with companies—both the big players and startups; and I think there’s an intensity around interoperability and meaningful use, and also around privacy and security. So I don’t think the technical issues are the barriers here; I think the barriers are around the users, and around how we change the funding.
HCI: What would your advice be for CIOs and CMIOs?
Keckley: One, the combination of CIO and CMIO is powerful and is a necessary starting point, because historically, we’ve made clinical IT discussions, and financial and operational IT discussions, and the two have been separate. But I think CIOs and CMIOs have to learn to translate; they have to avoid tech-speak and become very pragmatic about how they roll out solutions. What happens is that we end up spending all our time talking about the needs and not the ends. Everybody wants to deliver higher-quality, more efficient, and effective patient care. But we get buried in the weeds sometimes. I’ve lived in those organizations where the CIO and CMIO can’t speak English. So I think that’s a huge part of the gap.
The second issue is that we’re going to have to reach some consensus with policy-makers around certain issues. It is virtually impossible for us to achieve perfection with meaningful use and with privacy and security. We’re going to have to achieve a level of pragmatism with the use of these tools. We can’t wait for everybody to be completely satisfied with these tools. And if people can crack the code on the most secure data the government holds, healthcare isn’t going to be far behind in terms of being breached. What makes it so emphatically essential to embrace this technology sooner rather than later is that we’re killing people because we don’t ask people the right information about drug-drug interaction, because we don’t provide clinicians with important information at the point of care.
What we’ve not had in this country is a social movement around the lack of information necessary to make critical judgments around care issues. It ought to be the most important social issue of our time; it ought to be unconscionable to be making medical mistakes and not providing the best care, in a system that’s spending $9,217 per capita, per year—that’s the official government figure—$2.417 trillion, plus $360 billion that consumers spend for care out of pocket that the government doesn’t count; that brings it up to $2.83 trillion. And the only way to improve things is to bring clinical and administrative data together. So I believe we need a social movement that requires doctors, hospitals, and health plans to be on the same page in terms of using data information, and we need to require that they do so.