CMS issued the final rules for accountable care organizations (ACO) on Thursday and, in so doing, sparked excitement throughout healthcare with what initially appears to be a more achievable set of qualifying parameters. But it will take some time for the healthcare industry to absorb and understand the changes.
George Roman, senior director of health policy at American Medical Group Association (AMGA), spoke with Government Health IT Editor Tom Sullivan Thursday afternoon, following the CMS conference call announcing the new rules.
Stipulating that he'd read just the first 60 pages of the 696-page document, and clear that his answers are based on that, Roman shared his first impressions of the final ACO rules.
Q: What was the biggest surprise in the final rules?
A: The biggest surprise at first blush is the fact that CMS seems to have listened more than it typically does to the commenters and it appears to have made changes of some consequence. That’s a pleasant surprise. Now, I cannot say that applies to everything because I have not read everything yet, but on several matters there are clear improvements. Whether that’s enough to make the program attractive remains to be seen. The devil is in the details.
I can give you some examples. In the proposed framework there was a two-sided model that involved shared savings and shared losses. We and many others commented that this was a methodology that didn’t render the ACO model as very attractive. Why should people engage in risk if they can be guaranteed Medicare payment if they don’t participate? This is still fee-for-service, after all. CMS heeded that and has offered one of the participation tracks to be one-sided where there is just shared savings risk and no downside.
[Reporter's notebook: Biggest surprises within the final ACO regs.]
Another thing I read that I liked was the reduction of the required quality measures from 65 to 33. We commented that 65 is way too much – it’s onerous, it’s burdensome, doesn’t necessarily advance anything and it may not be possible to attain. So it’s a serious disincentive. Now, I don’t know what the new measures are yet, but having said that the fact they reduced the number and the burden is a very favorable sign.
Q: During the comments period, a lot of health experts said there was a chasm between the regs and reality. Has CMS successfully addressed that with the final rules?
A: I hate to be trite, but let me characterize the final rule as a mosaic, and each component is a tile of that mosaic. The full assessment and value of participation is going to have to be taken into account by individual entities evaluating them within their own context. So, even though some of the various tiles have changed for the better, the full picture is what counts. In addition to that, there are various external documents which will play a very significant role in whether or not, methodology aside, this will be attractive. Those come from HHS OIG, IRS, FTC, DOJ, etc. And I believe the documents were almost concomitantly released with this rule.
The rule itself as a methodological document is critical but without some greater assurance that they’re not going to put the handcuffs on me as an executive of a large entity seeking to form such an ACO, that would dampen my interest considerably. The assurances offered in the preliminary documents were not enough.
Q: AMGA this summer cited a member survey that found 93 percent of respondents "indicated that they would not participate in the ACO program unless the requirements in the final rule reflect major modifications to the proposals.” What would you put that percentage at now that the final rules are out?
A: I wouldn’t venture into that realm now – not because I wouldn’t want to give you an assessment, but because I am not able to, and I’ll you why. Go back to my rather trite but reasonably illustrative mosaic example. It depends what all the pieces are together and how they’re viewed by individual participants. And it has to be viewed in the context of these other documents I mentioned a moment ago. So I don’t know, but based on a preliminary reading, if the improvements and changes and changes as reflected in the limited thing I’ve talked about are carried forward through the entire rule and the attendant anti-trust anti-kick back documents then it certainly will produce something more attractive, but the devil remains in the details. I haven’t read the whole document yet; I’d be foolish to opine beyond what I’ve said.
Q: During the public comments period there were these two sort of opposing forces, one being patient-centric organizations, notably the Campaign for Better Care, which maintained the ACO bar had to be high enough to effect change, and the other being associations such as AMGA, AHA, CHIME, which said the rules and associated expenses were prohibitively stringent. Do the final rules close that gap?
A: I’m hoping. We believe that the bar should be set right, so that it is not attainable by anyone who wants to participate without doing something, but not so high as to discourage participation by those who might be able to contribute something. The main promise of the ACO model over time is to turn the very fragmented system into one that is more integrated. And that will produce not only benefits for the Medicare program but I’m hoping for greater care quality delivered and for societal benefits. I have to take my white hat off now, but it’s true. We really do believe and that’s why we were so enthusiastic when this thing was put into law and so deeply disappointed when the proposed framework was issued. This appears, at first blush, to be better.
Q: The final rule is more realistically achievable, so it’s not a huge leap to think that more ACOs will emerge because of that.
A: I don’t know enough to characterize that, but I hope it’s the case.
Q: CMS administrator Donald Berwick said during the announcement that the final rule makes a stronger ACO framework. But does making it easier to qualify as an ACO really strengthen that framework?
A: I’m not sure that qualifying is all that much easier. Look, we’ve got to do something. Many healthcare policy folks agree the current system is unsustainable. Fragmented, un-coordinated care is a large contributor to the state of affairs now. There are many others, as well. This thing has great promise; we want that promise to be realized.
Q: What did CMS not change that you would have liked them to?
A: We characterized early on 10 or 12 deal breakers. There are a bunch of them but I’ll give you 5 or 6 of the key ones.
1. The way patients are assigned or aligned. We believe it should be prospective and there should be data sharing with identified patient information available. You can’t manage a population if you don’t know who’s in it.
2. Risk adjustment. We think there should be a commonly used methodology designed to take into account the disease severity of the population being served. Healthier patients cost less than sicker patients cost. And this is an adjustment for purposes of recognizing costs, computing savings, costs, etc. And so CMS proposed a method by which risk- adjustment would be done. It would be a look-back three years and a number would be arrived at based on a complex methodology and that the same number would be applied each and every year of the performance period without change. That’s fundamentally flawed and unfair. Because people can choose the doctor they seek care from, they can wander. And as much as 25 percent of the people go elsewhere. So we called for dynamic risk-adjustment. Doing exactly what has been suggested but doing over each year to reflect the population more accurately. That makes a big difference in terms of dollars and sense.
3. The two-sided model. They have changed that, it’s great.
4 . Minimum savings rate. This is a mechanism designed to ensure that whatever savings are realized from operations are the result of clinical intervention and quality care, not random fluctuations of statistical results based on the population size. Let’s say you save $100 and the minimum savings rate is 2 percent. Two percent comes off the top in the NPRM and goes to CMS, the program keeps it. And your potential shared savings are then calculated on the remaining 98 dollars. There are a lot of other machinations but that’s the essence. If you demonstrate that it’s not through fluctuations then you get first-dollar savings, meaning your shared savings split is based on the full $100. That’s huge.
5. Data reporting.
6. Too many measures; they changed that, too.
7. Anti-trust, anti-kickback, civil money penalty issues. I haven’t read that yet, but that’s easily a deal-breaker.
There’s a whole bunch of others, but those are the key ones we’ll be reading for. Some of those I mentioned may be changed, too. I haven’t read the whole document yet, I’d be foolish to opine beyond what I’ve said.
Previous ACO coverage: