Providers who take on downside risk are more likely to put “the pedal to the metal,” acknowledges Mai Pham, MD, Anthem’s new vice president of provider alignment solutions. The country’s second-largest health insurance company is considering different ways of luring providers into downside risk arrangements, including offering a faster, more efficient form of prior authorization—and perhaps no prior authorization in some cases. Pham says she is also looking at providers in upside-only arrangements and the fees they are getting paid. “They are not entitlements.”
Pham, 49, was chief innovation officer at CMS’s Center for Medicare and Medicaid Innovation before joining Anthem last year. At CMMI, she was responsible for implementing an alternative payment model under MACRA and oversaw the Pioneer and Next Generation ACO programs. Pham earned her undergraduate degree from Harvard, her medical degree from Temple, and her MPH from Johns Hopkins.
You’ve moved from the public sector to private sector. How is it different on the private side?
Only slightly less stressful because, I think, the fiscal pressures and the moral imperative to do the right thing with health care payment is the same, regardless of which sector you’re in. I have a much deeper appreciation for the advantages that a public payer has in terms of the efficiency of pricing, not having to negotiate with individual provider organizations in every market, with clarity around universal policies, whereas in the private sector there’s much more variation on a market-by-market basis. In some ways, that can be very constructive and, in other ways, that can be very challenging.
In the public sector, did you have more stakeholders to deal with? At the Leonard Davis Institute symposium I heard you speak at last fall, a gentleman asked about what he saw as a simple, common-sense fix of a health care problem, and you responded that it isn’t that simple in the public sector.
I think each sector has its own set of complexities. I would say what’s different in the public sector sometimes is that your colleagues are regulators—and sometimes law enforcement. While we certainly have to abide by laws and regulations in the private sector, we do not have to wear the enforcement hat. So, the approach to program integrity and all of those related issues are somewhat different in the public sector.
Is it liberating not to have to wear the enforcement hat?
I would say it frees up time, on the one hand. On the other hand, you don’t get to wear the enforcement hat, so you don’t have that same kind of authority always when you’re tackling your partners in the private sector. But it definitely gives you a few more degrees of freedom in terms of the speed in which you can design new policies and implement them. There are fewer stakeholders to get sign-off from as a private payer.
Your title is vice president of provider alignment solutions. Is that a new title?
It’s actually an old title, and it reflects better the portfolio that had existed in the past. What it means in the current day is payment innovation. My team is responsible for the design and implementation of payment innovation for Anthem on an enterprise level.
What does “enterprise level” mean?
It means that we come up with the national payment policy and then work with our teams in the individual markets to apply them. That sometimes means allowing teams within a given market to toggle certain parameters up or down a little bit, give them that flexibility with which to manage their relationships in the marketplace. But the general framework and the broader strategic approach are the things that we provide.
Who do you report to?
I report to Colin Drozdowski, the senior vice president for provider solutions, who oversees what I do and network management and contracting. That includes what we refer to as “provider experience,” which is all the interfacing that Anthem does with its clinical partners—everything from credentialing to claims management and appeals. All of that sits under Colin.
Is the Enhanced Personal Healthcare Program the core program you’re working on?
Historically, EPHC was our largest and flagship program in value-based payment. It now has several different tracks, one of which is called EPHC EssentialsEPHC Essentials has 340,000 attributed members with 1,200 participating practices, according to an Anthem spokesperson. The company’s goal is to have more than 400,000 attributed members in the program by 2019., a relatively new track focused just on primary care.
But my portfolio is broader than EPHC, and we’re growing it all the time. We are definitely evolving our ACO construct and our primary care construct. But we’re also building new programs, too, to complement them, and, in some cases, actually interface with them directly. For example, as we built new payment opportunities for providers in bundled payments targeted at certain specialties, we want to think through how that will interact when that specialist is also in one of our EPHC ACOs. Or when that specialist is competing with an EPHC ACO in the same marketplace.
Is EPHC an upside-only shared savings model?
It has a full range of options. So, there is a shared savings, an upside only. There is certainly a downside-only option where we reconcile the financials at the end of the year, and if there are savings or losses, we split them with the provider.
It also has a delegated-risk option where you actually, as a provider, take on not only the risk for your performance—which is what shared savings and losses usually are—but also some insurance risk. The portion that you are taking delegation for—those are the funds that are going to be used for those members, and no other funds. So, it’s a different level of accountability.
Some people argue that the ACO model really won’t have the value-based effect its proponents hope it will until there is two-sided risk, so you have the proverbial stick along with the carrot. Do you see it that way?
First, at a high level, I’m very uncomfortable with a binary question that people often pose: “Do these models work or don’t they work?” To my mind, that’s a very confusing question because the ACO construct is an evolving thing. I also think that every construct works better in some circumstances than others. So, it’s our job to get smarter about how to apply what is in our toolkits over time. ACO constructs are one such tool, but they’re not the only toolkits over time.
So with that aside, in general I would say that we do believe that providers who take on downside risk are more likely to put the pedal to the metal when it comes to smart care management and thinking about their population, thinking about the use of resources, and how to engage members. We are on a track to both encourage and push them more in that direction through a variety of mechanisms.
On the carrot side, we definitely intend to make the business case so downside risk is far more attractive than not going to downside risk.
We’re also looking for ways to reduce the burden on providers who go to downside risk. There are a couple ways you can do that. You can reduce the burden of performance measurement on providers. If you are in a downside-risk arrangement, you are probably tracking over 100 clinical quality metrics yourself within your own organization. I don’t need to pile on more there. I need to focus your attention on maybe a handful of clinical outcome metrics and patient experience, and you’re good to go.
We’re going to start to signal to providers that if you are going to stay an upside-only arrangement, we’re not going to give you free money forever in the form of the care-coordination fees that we’ve been paying out. They’re not an entitlement.”
Another way to reduce burden is to look at what feels burdensome to providers in their daily lives. Often, those are administrative processes that they have to engage with us as a health plan. So, we’re looking at ways to reduce that burden specifically on providers whom we highly value and are taking on downside risk. That may mean certain efficiencies and hand-holding and getting you through prior authorization. Or reducing the burden of prior authorization for you altogether. We’re looking at a range of options in that regard.
On the stick side, we’re going to start to signal to providers that if you are going to stay an upside-only arrangement, we’re not going to give you free money forever in the form of the care-coordination fees that we’ve been paying out. They’re not an entitlement. They actually belong to our customers. And we need do right by our customers. So, if we have given you a reasonable amount of time to demonstrate performance and you haven’t, we’re going to turn off that spigot.
Does that mean that you’ve told certain providers who are in an upside-risk-only model that care-coordination fees are ending this year or next year or some time certain?
Right. Those policies are being refined and rolled out in this coming year. We are definitely on our way to sending those signals to say, “We’re tracking performance, and, you know, if you’re taking funds from our customers and you’re not performing, we need to assess that on a routine basis.”
If I’m in a downside-risk model with Anthem, does this mean that Anthem is saying that there are certain things I will not need prior authorization for because I am in this model?
The details are being worked through, but that is one possibility. We’re also looking at, when you do need prior authorization, how can we make that process easier for you?
How would you make prior authorization easier?
I think there are options in the platform interface—the way you do data exchange. You can, for example, reduce the number of transfers of information from a practice manager to a local medical director to somebody else to somebody else, and then, back again with the decision. And then, oh, you need another round of information from the clinical record.
If you can reduce that traffic—and compress that time frame—that makes life much easier for the provider.
Can you tell me how many of your ACOs are upside only and two-sided right now, and whether you have a ratio in mind for what that might be next year or the year after?
I don’t have that number for you. We can get you that number.In an email after the interview, an Anthem official said the insurer has 160 commercial ACOs, and at least six are in two-sided risk arrangements. The rest are upside-only risk. One reason I don’t have it is because there are ongoing negotiations now. So, it’s a little bit of a moving target.
I don’t expect to cover your full portfolio in this interview. But you mentioned bundled payments. I’m not sure that’s a new thing you’re working on, or if there is some other value-based initiative that you’re getting off the ground now.
We’re working on a range of things. We have a number of longstanding pay-for-performance programs, as well as designation programs—programs where we identified centers of excellence in certain categories based on the value proposition of both cost and quality. In terms of pay for performance, we have a number of programs in Medicaid for both facilities and providers. We also have them in behavioral health for our commercial population, in maternity care, and in nursing homes. And we’re building out more all the time. Bundled payments are not a new concept at all, but we are on track to enlarge our offering of bundled payments for both chronic and acute episodes.
In addition, as we think about the future payment portfolio, there are exciting new projects, but we’re not quite ready to announce them. We are doing a lot of building and work around the concept of what does it take to bring together a system—a whole team and system—that serves our members and customers. We are thinking through the marriage of payment innovation with network and product design, but also, what does it mean to really offer an enhanced member experience?
The way we’re thinking that through has real potential to differentiate Anthem, but we’re not quite ready to announce the details of that yet.
We are also very interested in digging deeper into payment arrangements with specialists and with ancillary care providers.
Do you think the payment arrangements with specialists have been neglected in the value-based discussion so far?
I do think so. Bundled payments are a familiar construct, but for given specialists, they usually don’t account for that large a portion of your book of business. And ACOs, for the most part, don’t do a great job of engaging the specialists even within their four walls.
So we, as a payer, believe that we need to more deeply engage specialists. And if we focus on specific, high-priority specialty areas, we think there’s a real opportunity to do that. You can work with them on a range of things—how they can better interface with primary care, how they can better support chronic care management. Working with them to rationalize service, so doing fewer of the things that aren’t really necessary. I think there are some very specific ways that we can engage.
But if there’s one group that’s done well under fee for service, it’s specialists. So, you can use euphemistic words like engage, but aren’t you going to get some fight from them because they’ve done very well with fee for service and volume?
Yes. I think when you’re trying to change the status quo, you need to paint a picture of the future for people in which they can see themselves in some positive, constructive light. And then, yes, you do need to apply pressure to get them there.
But you need to do both things. That’s going to be our approach. We’re going to say to specialists, “Here are the ways that we think you could bring spectacular additional value to the system, and here are other ways we’d like you to back off and remove some of the negative influence that you have.”
What did you think when CMS canceled mandatory bundled payment in the orthopedic and cardiovascular programs?
I wasn’t surprised. I think Secretary Price made his opinions about that pretty well known very early on. And I think there were perfectly credible reasons for wanting to go that direction. If you want to just go with voluntary payment models, that’s fine. Just make sure that you have a way of generating credible evaluations from it. That was supposed to be one of the benefits of having a mandatory model.
My understanding of the mandatory model is that it was a way of running a true experiment.
Right. Not doing it in a true randomized fashion—it’s not insurmountable. It’s just harder. You just have to think about how to do that.
I also think it’s important to maintain the momentum for value-based payment. So, if we’re not going to have mandatory models, we need to find other ways to consistently signal that this is the direction we want the system to move in.
When you were at CMS, it seemed like CMS was the prime driver of efforts to move the health care system to value-based payment. Do you think that’s going to change under the Trump administration? Is it going to be the private sector that’s going to be the main influence rather than CMS?
I think what’s needed to drive change happens on multiple levels. I think the private sector has a tremendous opportunity right now to really push the envelope and do things that even CMS had not planned to do.
The sort of passing back and forth of intellectual leadership or innovation between public and private sectors—that’s not new. That’s happened in both directions multiple times. I think everyone just needs to be nimble and prepared to step in when necessary.”
I think there is an important role for government to play, both as a payer and as a regulator. And we’ll see what happens in the Trump administration.
But the sort of passing back and forth of intellectual leadership or innovation between public and private sectors—that’s not new. That’s happened in both directions multiple times. I think everyone just needs to be nimble and prepared to step in when necessary. Anthem is certainly committed to driving this momentum, and we have not let up with our provider partners. We think many of them remain very excited about this work. It is true, of course, that it would be easier if CMS would continue to also bring that pressure to bear. We’ll see.
As I recall, during the panel that you were part of at the Leonard Davis Institute symposium, there was some discussion that for value-based payment to work, there had to be some heavy lifting on benefit design—that the beneficiary or the member has to be motivated and have incentives to use the health care system in a different way. Are you working on benefit design?
We are. We are thinking through benefit design, and we think it is a very important part of the puzzle. We think there are several pieces that need to come together in an interlocking fashion. We want providers to go to downside risk. We want members to have rational incentives in front of them and to respond to those incentives. Employers want members to have an enhanced experience, and we as a payer want to hold providers accountable for care outcomes. All those pieces—we have to find a way to bring them all together as one. There is no one piece that’s going to be the magic bullet without the other.
It’s not realistic, we have learned, to expect providers to manage care for a population when that population feels like it can go wherever it wants, whenever it wants, and get whatever care it wants.
First of all, the patients don’t necessarily feel satisfied in that context. Second of all, it may be an insurmountable challenge for providers. It may not be possible to expect them to go to downside in that context. It’s a question of what each party—provider, health plan, customer, patient—is willing to give up in order to make this work.
Are the members willing to give up a little bit of freedom of choice in exchange for an enhanced member experience and a lower price point? Are employers willing to have those difficult conversations with their employees—in exchange for what? Are providers willing to go to downside risk—in exchange for what? And what are health plans willing to do to reduce administrative burden to offer a better business case to make it work?
We’re trying to bring all those pieces together.
Can you talk about anything concrete that you’re doing now with benefit design that speaks to what you were talking about? In between the lines, you seem to be saying that maybe a member would agree to some sort of narrow network or one provider system, in exchange for either a lower premium or lower co-insurance or copays. Is that the sort of thing you’re talking about?
That is certainly on the table. That’s a very familiar construct in the industry—exchange a narrow network for a lower price. We think there’s more that can be brought to the table. It’s that “more” that we’re thinking through now. We’re not quite ready to discuss it now.
Can you discuss it conceptually?
We’re thinking through what it means to provide an enhanced member experience. It’s not just about dollars and cents.
As I’m sure you’re aware, Uwe Reinhardt died recently. A lot of people have re-read his Health Affairs piece, “It’s the Price, Stupid.” Have you thought on occasion that we’re doing all this design—incentives this and incentives that—but if the prices are so high, is all you can have just a marginal effect?
I never said that prices were outside our purview.
Oh. So, you agree, it is the prices, stupid.
I think prices are a huge contributing factor to the ills in the system. However, I don’t know that Uwe would have contradicted me if he heard me say that culture is also really important.
So, the reason to do value-based payment design on top of restructuring prices, to the degree that you can, is because prices alone do not a culture make.
It’s the nature of the payments, combined with the prices, that produce the behavior. And where we would like the system to go is a healthier culture around care delivery and the use of health care resources in the long run.
But if prices were reduced, costs would be lower, and we wouldn’t be worried about the culture if what you mean by “culture” is utilization.
Only for some length of time, right? Prices have a remarkable ability—a remarkable and self-reinforcing ability—to go up over time. Very rarely in health care do prices come down over time. You can count on two hands the instances of that happening.
I thought that the Reinhardt and other critiques found that utilization is not that high in this country, and that the reason why the cost of health care is roughly approaching 20% of GDP is just prices. Therefore, all this work on utilization just won’t have that much effect because the prices are so high.
I’m not a black-and-white person. Uwe was an economist who tended to think in terms of broad aggregate average numbers, where I see a lot of subpopulations and a lot of subgroups of providers and a lot of subcategories of services. And I see a lot of waste in the system, in addition to the high prices.