The accountable care revolution is being led across multiple venues in the U.S. healthcare industry, and nowhere more vigorously than in multispecialty physician groups. The logic is inexorable: physicians are the providers closest to patients, particularly the high-risk, high-cost patients whose utilization is leading the bulk of expenditures in the current healthcare system nationwide.
One of the true pioneers, in a literal sense, in this area, is the Brighton, Massachusetts-based Mount Auburn Cambridge Independent Practice Association (MACIPA), with over 500 physicians practicing in 10 communities. Founded in the mid-1980s as a contracting agent for its physicians, MACIPA became an accountable care organization (ACO) in effect years before the term was coined, says Barbara Spivak, M.D., MACIPA’s president. Not surprisingly, Spivak and her colleagues leapt at the chance to become a Pioneer ACO under the Medicare Shared Savings Program, and joined that program with the first cadre of Pioneer ACOs, in January 2012.
Dr. Spivak was one of a number of healthcare leaders interviewed by HCI Editor-in-Chief Mark Hagland for the magazine’s July-August cover story. Below are excerpts from her interview with Hagland this spring.
Tell me about the evolution of MACIPA as an organization, and towards accountable care development.
MCIPA was formed in the mid-1980s as a contracting agent for its physicians, when Tufts Health Plan decided it wanted to have a capitation arrangement for its providers. After about 10 years into doing this, in the mid-1990s, with a 19-physician board, the leaders agreed they had to start managing the care of patients. And over time, we have built up what is now considered to be an ACO, even before ACOs were called that.
What has been the timeline around this evolution?
If you look back historically in the 1996-2000 period, we started to build up a very robust program, first of pharmacy management, then of care management, and then of disease management with a quality improvement program that looked both at health screenings and chronic disease. As time went on, we realized that we were very limited by being on paper, and only had the claims data to work with that health plans were sending us. So in 2005-2006, the organization decided it wanted to go electronic. And back in 2006, Blue Cross Blue Shield of Massachusetts gave $50 million to the state of Massachusetts for the e-collaborative to go electronic. It was a competitive bid, and we ended up being number four, so we didn’t get that grant. So we weren’t going to get it paid for by the state; so in essence, we had convinced the physicians in the leadership that going electronic was in fact the right thing to do.
So we worked in a number of different areas, and we ended up picking a vendor, eClinicalWorks, and our first practice went live in 2007; we’re not entirely even today. And not everyone is on eClinicalWorks; our OB/gyn group is on athenahealth, our orthopedic group is on Allscripts, our dermatology-specific group is on a dermatology-specific EMR; our ophthalmologists are on ophthalmology EMRs; and our Cambridge Alliance doctors are on Epic. But the majority of our primary care docs are on eClinicalWorks; and we host eClinicalWorks.
How many physicians are at MACIPA altogether?
We have 500 physicians, and we have over 200 doctors and mid-levels on eClinicalWorks.
Are all of your physicians employed physicians?
No. Of our 500 physicians, about 100 are employed by Cambridge Health Alliance, 150 are employed by Mt. Auburn Hospital, and the rest are private. And we have about 110 primary care doctors, and about 400 specialists. So we have 49 instances of eClinicalWorks at 180 office sites. And even the employed docs are all in different offices, one-to-five-physician offices.
When were the large majority of your practices electronic?
It took us three to four years to get the majority of our practices up, but we’re still implementing some people who had decided to retire but then didn’t. So we still have five or six who aren’t on an electronic medical record, but they’re senior physicians who are going to retire soon.
Which of your organization’s contracts are risk-based contracts?
All of them: Tufts, Harvard Pilgrim Health, Medicare Advantage—all the major commercial plans in Massachusetts, plus Medicare Advantage; and, we’re a Medicare Pioneer ACO.
What made you decide to go that route?
There were several reasons. One was that we as an organization believe that managing the care of patients provides better care, and that in order to effectively manage the care, you need to get data from the health plans. Particularly in an environment like ours where there is a lot of fragmentation of care, even more in the senior population than in the younger population, it’s very important to get information on where they’re going. And if you’re going to enhance their care, somebody has to pay for that. And the only way to get that is to enter into some sort of risk contract with a payer, in this case, Medicare. And you’re really only talking about the sick patients—the healthy patients you may run registries, but managing care is focused on the sick patients. And one of the most frequent complaints we would get is, why is it that you can provide my Medicare Advantage patients with all these enhanced services, but my straight Medicare patients you can’t? And the truth is, we need the data, and someone needs to pay for the additional support. And so we saw the Pioneer program as being an opportunity to invest in care management.
And the other shift is that, particularly for chronic disease patients, people are interested not only in, was the test done, but was the result good? And particularly around cholesterol management, blood pressure management, and blood sugar control, there’s been more of an emphasis in the past ten years around outcomes. And if you look at the alternative quality contract in the Pioneers’ shared savings metrics, that’s a big part of them.
What have been the biggest challenges for you and your colleagues so far, strategically and data-wise, as a Pioneer ACO?
There are several big challenges in the Pioneer program; one is just the sheer volume of adding that [work] to our primary care doctors’ lives. It just means a lot more intensive management. The example I use is a patient of mine who I believe is being managed by the Joslyn Center for Diabetics, but when we look at the data, hasn’t been to the Joslyn for four years, so all of a sudden I have to take a more active role, even though she’s told me she’s been going to the Joslyn Center. So this adds a lot of work, and some of the quality metrics that we’re asked to measure don’t add value.
Could you provide an example of that?
I would argue that it’s really important that seniors get flu shots. But in this environment today, most seniors get their flu shots at work or at the pharmacy. And so we’re having to track that, even though we’re not doing it. So it’s no added value involved to have to document when and where the patient got a flu shot; not that it’s not important, but it doesn’t add value to what we do. And the medication reconciliation—there was a much easier way for them to measure that by using transition of care codes; but instead, we have to put something in the record. So the process has to be—the electronic health records don’t have anything that satisfies the measure, so you have to find the language that the government will accept for covering it. You have to get the language, put it into 49 instances of eClinicalWorks; then you have to go out and train the physicians and staff in what to do, and then monitor them to make sure they’re doing it right. And that whole process takes at least eight months, and it’s difficult to do it quickly. When you have the number of physicians in private practices that we have—we don’t have people in the practices employed by MCAIPA. We have 68 staff people.
How difficult has it been to combine and analyze claims and EHR/clinical data?
Well, I think it’s more complicated than people realize. The things that I think about: one, when you’re trying to meld multiple insurance plans’ data into one physician’s record—not every plan calls something the same thing, so you have to have a business intelligence person meld all that data. And a diabetic—a patient may have seen someone who put them down as a diabetic for some reason, and the provider may have done that because the service they’re providing may only be covered for diabetics, or they had high blood pressure.
And the example I use is that patients on anti-psychotic meds are required to have their blood sugar checked for the drug, but the lab coded them as being diabetic. So not every diabetic is a diabetic. And there are lots of examples of this that complicate things. You don’t want to ask someone who’s not a diabetic to take a diabetic eye check, that doesn’t make sense. And in terms of blood pressure, you have to make sure that everyone enters the blood pressure numbers in the same way. Those numbers have to be in reportable fields, because you can’t do chart reviews on thousands of charts.
So it requires the standardization of data entry?
Yes.
Does your organization make use of performance dashboards for its practicing physicians?
We’ve found that physicians don’t actually look at these things, so we provide electronic registries to their staffs, or even on paper, per their preference. So we provide registries and information about their patients—such as when a patient’s last mammogram, PAP, or colonoscopy was. And for our diabetics, we might give the provider a spreadsheet that says, you know, Joe’s last blood pressure was normal, his last LDL was two months ago, was 105, and the blood sugar was perfectly fine; and we might put something about what they need. We normally recommend two hemoglobin a1c tests a year for a diabetic.
And we also make use of a metabolic team—a couple of docs in their specialty who meet with the pharmacist and look at patients who are really out of control or whatever, and make recommendations to the primary care doctor on control. We do that outside the EHR—it’s a team meeting, and electronically, someone will send a message to the primary care doctor, your patient’s not in control, and the team recommends that. We don’t send it in the record, we send it through our secure e-mail system, so it’s not part of the patient’s record unless the doctor feels it’s called for.
How have the physicians at MACIPA adapted to the culture around care management, and the culture of using electronic tools?
If you look at the different measures that our primary care doctors are held accountable for, across all the health plans and measures, it’s over 100 measures. And even in diabetes, the measures are not all the same. So it’s very overwhelming for the docs. The number of patients has increased, and the number of metrics has increased.
There’s a cultural consensus in the organization, though?
Yes, because our doctors are used to being in capitated contracts, and we’ve made a big effort to help them to understand that the current contracts are capitated, and the cash we bring home varies depending on the quality, and therefore, that it’s important to keep the quality up.
You and your colleagues have been learning from and sharing with leaders at the other Pioneer ACOs, of course?
Absolutely; not only in the Boston area, where there are five; but also nationally, where there are a lot of meetings, phone-a-thons where people are sharing. So a lot of sharing is taking place. I don’t think that any of the Pioneers feel the information is proprietary; I think everyone understands the experiment is to share.
In that regard, what have the biggest lessons learned been so far in your organization, both in the Pioneer program, and more generally, around accountable care and care management?
I think that one is, I’m impressed that care does improve with management, and that if you look at the data as we look at it from today versus 2011, before we managed it, there are some things that we really see a big improvement in, and I’m pleased with that; and I think that’s true of all the Pioneers, that we are actually improving the care of our patients, and are keeping the costs down. I would say that it’s interesting that most of the money saved has come in post-acute care rather than in inpatient changes or preventing readmissions, which is where people have always thought there was the biggest opportunity; but it’s really in post-acute care. And when you look at Medicare Advantage, if you look at what we can do there in post-acute care, there’s really an enormous opportunity in Medicare fee-for-service to improve communication, improve care, improve the transitions, and keep costs down and get quality much better.
