Last autumn, Earl Anderson, executive director of Tennessee Orthopaedic Clinics, a 23-orthopedic-surgeon group based in Knoxville, Tennessee, presented at an event held in Nashville, and sponsored by the Nashville-based Southwind Company, a division of The Advisory Board Company, based in Washington, D.C. Anderson presented on the very important emergent topic of bundled-payment contracts. He served as a panelist for a discussion entitled, “Inside the Medicare Bundled Payment Initiative.”
Anderson, who has more than 25 years’ experience working in the field of orthopedics, both as a clinician and as an administrator, has helped Tennessee Orthopaedic Clinics (TOC) more than double in size. He spoke late last fall with HCI Editor-in-Chief Mark Hagland regarding his organization’s path forward in the direction of bundled-payments contracting, which is expected to go live sometime in 2014. Below are excerpts from that interview.
Tell me a bit more about Tennessee Orthopaedic Clinics as an organization?
We’re an independent, single-specialty practice. We have 27 doctors, mostly orthopedists. We also have a full staff of mid-level practitioners and others.
How many are orthopedists?
In our group of doctors, 23 are orthopedic surgeons, and the rest include a few podiatrists, a family practice physician specializing in workers’ compensation, and a physiatrist. We have over 200 employees altogether.
What level of patient volume does your group have?
We have about 10,000 patient visits per month, and we do about 800-900 surgical procedures per month, so over 10,000 procedures per year.
With regard to the topic of the Advisory Board Company event, is your organization a participant in the Medicare bundled payments program?
We are not. We had applied with our hospital partner, but backed out at the last minute, primarily because we didn’t feel we had the data to make it work. That said, we’re involved in structuring some other projects now.
Do you have any live contracts yet?
Not yet; we probably will sign our first contract sometime in the first or second quarter of 2014. And that’s actually going to be through an initiative driven by the state of Tennessee, for their Medicaid population and some state employees. We’re signed on whether we wanted to be signed on or not. But we felt this would be a good project to start on. In a bundling project, as with any other change in structure in payment, you want to keep your volumes at a minimum just to get your processes in place at first, and then get your data aligned. In a bundling project, data is the key. If you don’t have your data across the continuum, obviously, you’re at significant risk.
In general, with fairly few exceptions, by and large, we just don’t have the data corralled enough to look at the total cost structure of a bundle and what our risk is, if we take a bundle at a particular cost point. We’re working with a hospital partner to get all those things in line. And like many others, we’re working with consultants, and working with payers, to get the data. I think if you start a bundling project, there’s some low-hanging fruit. We’re looking at everything from the cost from pre-admit through discharge, for total joints, meaning total knees and total hips.
Is putting claims and clinical data together a core challenge?
A lot of the challenge is taking the data from an independent practice like ours, marrying it up with the hospital data, and then incorporating things like post-acute care, which is very much a wild card in a bundle, and trying to put all that together. There are some hard points along the way there; for example, on the physician practice side, that data is pretty easy to get, because we have our claims data. But then you have to look at the implant cost; and then the more difficult thing is look at the total hospital costs, including the anesthesia care costs, the supply costs, and some of the overhead costs in the hospital; and that’s not even accounting for post-acute care. That’s something that is still a little bit of wild card, going forward.
Do you need a data warehouse to do this?
Will we need a data warehouse to develop a bundle for something as simple as joints? Probably not. But in moving forward towards more complex things? Yes. And let’s face it: the end point on all this is capitation. And so ultimately, I would think we would need a data warehouse; and that’s probably something you’ll hear from other people when you talk about a shared savings approach.
Our main hospital partner in Knoxville is Covenant Health. And we’re trying to form a really tight clinical integration with them. And if you’re talking about any organization that doesn’t already have a multispecialty model, like Cleveland Clinic, that’s a paradigm shift. And the guys in San Antonio have done a pretty good job of that.
Are your doctors on board with the strategic and operational aspects of this?
My talk in D.C. focused a lot on this paradigm shift with physicians. And the short answer is, not yet. Because I think physicians are still in the mindset of, the more RVUs I produce, the more collections I produce, the better I’ll do. And in any physician practice, you have about a third of the docs who get it and are trying to move forward, and then you still have two-thirds of the doctors who are still in the other world. And so we’re trying to establish internal report cards to help them start looking at data. Right now, it’s performance-based, but soon, we’re going to start including cost and quality metrics in there. But I think we’re in a time now where physician income is coming down to the extent that they’re realizing the direction in which things are going.
We cannot sustain the healthcare system we have right now, in a purely fee-for-service-based system, right?
Yes. There’s a level of cynicism about what happened in the 1990s, where they’ll say, well, we never got to full capitation. But now we have the data and tools to make this happen, and the needle is definitely going to move, and I think we’re going towards value-based payment.
Do you have someone who fills the CIO role, in your group?
Our group has one full-time health IT person, and several other support people.
What would your advice be for other specialist group leaders, around how to pursue IT strategically?
This is going to be the crossroads for groups like ours. The question is, do I have enough resources to put into IT and data management, or is it going to be too much for us to handle? And my advice is, if you want to stay in business, you’re going to have to look at some virtual and outsourced resources. At the same time, you have to have someone inside who will be able to help our physicians measure their performance.
Is it a potential solution to hire a mid-level practitioner who also has some informatics skills and knowledge?
It’s a potential, but that is still a rare creature who is a mid-level practitioner and who also has informatics skills. So ideally, yes, those are the kinds of people you have to have, but they’re tough to find. So what we’re doing is looking at physician leadership from the inside. So those physicians who understand it, who get it, are the ones we’re pushing towards things like Advisory Board sessions, to learn this.
In other words, out of necessity, medical groups will need to focus on training super-user physicians?
Do you have anything to else add?
I think that bundling is kind of a value-based model in its simplest form. So in one sense for us, it’s an opportunity to get our arms around shared savings, and not get into really complicated ACO-type models. So in a sense, I see it as an opportunity to get into it.
Do you believe you can be successful in achieving meaningful shared savings?
I do; but only if we can get through this paradigm shift with providers [clinicians]. But the data is there, so this can be done.
And you’ll need performance dashboards to help physicians at the individual practice and point-of-care levels, right?
Yes, and we’ll have to change our compensation and incentive models. And as I say, anytime you start talking about changing models, the temperature goes up in the room. But this has to happen, and if you don’t have your compensation model to support that, you’re going to be in a lot of trouble.