Philip H. Kamp is CEO of Valence Health.
Tell me about yourself and the company.
We’ve been around since 1996. The vision of the company is that providers should be in charge of how healthcare is delivered, and for them to be in charge, they have to be at risk financially. Our job is to help provider organizations decide on the level of risk, help them figure out how to organize around the risk, and then manage the risk.
We have products around analytics. We operate all the way to a provider-sponsored health plan. We manage those things, so we pay claims, we do care management, customer service … all the functions you would do to run a health plan. It’s moving groups through the value-based care spectrum to any level of risk that they want to be.
How do you think HHS’s seemingly ambitious goals in moving quickly toward value-based payment will go?
I think it’s good. The general issue for us has been that their approach to doing training wheels was a problem. The old version of the shared savings models — we don’t think they were good enough or strong enough. They tried getting into risk, but what’s happened is those organizations – from funding it and actually operating — are finding they are not being successful, most of them, and they just need to assume more risk. We would like to see more risk in those rather than less risk. We think the moves that they’re making now make a lot more sense.
Insurance companies and doctors have always been blamed for healthcare costs, but people are recognizing that hospitals and their increasing clout in driving favorable contract terms are a lot of the problem. Do you see HHS or CMS addressing the cost role of hospitals?
It’s interesting because the hospitals can play a major role in this if they do it right. You’re right that if they get market share, they certainly can drive pricing from that perspective. But I think if you get into more risk, where you’re actually getting a PM/PM amount, the hospitals are a large part of the cost and they’re going to be the organizers.
Hospitals should play a major role, but it’s not on a fee-for-service basis. They’re the ones who can organize the doctors and pull together an organization that can assume the risk, so I see hospitals having a major role. The market share issue is an important one, but if you’re getting into competitive PM/PM insurance-type rates, that’s where you need to go with this.
Back in the HMO days and attempts at moving to a capitated model, assembling risk pools that made actuarial sense was also politically awkward because you had to decide who you could afford to cover. If you’re a hospital and you’re trying to figure out the steps toward accepting risk, how do you define and measure your risk for the population that you have?
We still have conversations with some insurance companies that are saying, I get a PM/PM and I want the healthy people, I don’t want the sick people. I think that’s what you’re getting at. From a risk entity perspective, as provider groups get into risk, the population that really needs to be taken care of are those more expensive people.
What we need to do is make sure that the PM/PM that we’re getting for that population is the right PM/PM. That was a problem in the old models. You had in a market where say the commercial PM/PM was $200 and you had an academic medical center taking in a population and got adverse selection, their cost might have been $300 PM/PM. We need to make sure that for the population that you’re responsible for, we’re looking at the dollars that are being spent on that population, then managing for that as compared to the old insurance model. That’s an important element of this.
Nobody wants the high-cost patients, but it’s not usually acceptable to charge those patients more. How many ways can the buck be passed?
Somebody is paying for it now, right? Those people are getting coverage, or they’re not getting coverage and they’re just using the ERs and hospitals to get their care. It’s recognition of that population and making sure that they have coverage. There’s great opportunity to manage that population better than it’s being managed now and the overall dollars will go down.
You just need to make sure you tie the actual expense for an individual, so if X-person costs $1,000 PM/PM, that should be the basis for the risk of that particular patient or that particular population. You need to tie the cost of the person to the actual risk that you take, or the cost of the population to the risk that you take, as compared to saying, “The overall population is $200 PM/PM, so let me try to get all the people that are $100 and leave the $300 people out.”
Somewhere, that $300 person is getting healthcare. We just have to do a much better job of making sure that the dollars that we’re spending per person is where you take on the risk for that population at that dollar base.
As health systems and insurance companies start looking more alike, what technologies and information do each have that the other needs?
From a health system perspective, very few have the technology and the information to actually manage the population. A hospital has its information and each individual physician may have his or her information. The hard part has been aggregating the data for a population.
If you take any given market, you probably have some physicians that are employed by a hospital, you have independent physicians … they all have different EMRs and different practice management systems. One of the keys is data aggregation. Maybe 10 years from now it will be different, but right now the key is tying all those systems together into one platform. It sounds like an HIE, but it’s the analytics behind that data that becomes important. There’s got to be ways of collecting that data from everybody and then doing the analytics around that. It’s the pharmacy data, it’s the lab data. There may be 300 different practices that may have 70 different practice management systems and EMRs.
You’ve got to be able to tie that data together to do the analytics. To me, that’s the biggest gap that exists today — pulling that data together, figuring out what it’s actually costing for that population, and managing that data to manage the care better for that population.
What health system metrics will be important to monitor for long-term success?
Today they’re focused on a patient entering their system and how they manage that patient who’s sick. They have to move to a higher level of managing the population as a whole. They’ve got to get a whole different level of data.
Once you’re within a system and you’ve determined how care should be provided for that population, you now have to determine if the services that are needed are actually getting done within that village. If you have multiple health systems in a marketplace, which is usually the case, how do you make sure that the services are all getting done, especially if people leave the system? The physicians in that particular organization have agreed on how care should be provided. You need to push those services to within that organization and make sure that you’re tracking it and making sure things are getting done.
If a diabetic is supposed to get an eye exam, we need to make sure that the eye exam is done by the ophthalmologist that’s in the network and not outside the network, because if it’s done outside the network, we probably don’t have that information. It gets back to the data aggregation piece and managing the population as a whole.
Explain what “narrow network” means.
I think of it as a village of hospitals and physicians that have come together to agree on how care should be provided and agree on the level of risk that they’re going to take. Do they become a health plan, do they just do risk contracting with health plans, do they do a combination of things? It’s that organization that’s made that decision and then it’s got it technology that it needs to manage that in its operation.
What’s really important is that you don’t separate the technology from the operations. There are a lot of smaller technology firms or single-source technology firms that are doing one piece of this thing. But there really needs to be an overarching perspective on how the technology relates to the actual performance on these risk contracts. There’s operational pieces, there’s technology pieces, and then there’s just network development pieces.
For people on the IT side, the biggest thing is around data aggregation, the management of that data overall, and how that helps the operational people succeed in making sure that you’re caring for that population as best you can so you’re managing that population and the best quality of care is given at the lowest possible cost. Getting that information that’s not just from the hospital, but from these other independent sources, and getting it on a daily basis so that you can track what’s actually going on and manage the population going forward and helping your physicians and their practices figure out what needs to happen in the next six months with a population of people.
Do you have any final thoughts?
The main thing to me, and one of the things that I hear in the marketplace, is that as a health system doing this, you have to be really big. I don’t think that’s the case at all. There is a certain life threshold that you need to manage and there’s no question about that, but that life threshold doesn’t require you to be a mammoth system at all.
Groups, provider organizations coming together to cover a state is certainly an approach to do this, but again, I don’t think you need to be a 5,000-bed system in a particular marketplace. You don’t want to be a 50-bed hospital. There’s a certain size that you need to be to do this, but it’s not huge.
The level of risk for a provider assuming risk is very different than the level of risk for an insurance company to assume risk. If you’re UnitedHealthcare and you have a patient in a hospital and that patient cost $50,000, UnitedHealthcare has a direct expense of $50,000 for that patient. Now you take a health system that has that same patient in their hospital, what does it actually cost that health system to provide care for that patient? Eighty percent of the costs in a hospital are fixed. It’s much less costly for them to assume that risk as long as the care stays within that network. If it leaves the network, they have the same situation as United has. But as long as they stay within the network, it is much less risk for a health system to assume risk.
Another way to say that would be, if you’re a billion-dollar health system and I told the CFO tomorrow they were going to get a billion dollars in revenues next year, would they assume that’s more risk or less risk? I would say that everyone would say that it’s less risk. That’s full capitation from that standpoint. The concept of risk on the provider side is much less risky.