From 3M Health Information Systems
Podcast Episode Transcript: Empowering primary care with better access to trusted data
Gordon Moore: Welcome to the 3M Inside Angle podcast. This is your host, Dr. Gordon Moore and with me, today, is Beth Bortz who’s named President and CEO of the Virginia Center for Health Innovation (VCHI) in March 2012.
Prior to joining VCHI, she served for nine years as the Executive Director of the Medical Society of Virginia Foundation, where she developed and led programs in healthcare quality improvement, medication assistance, public health awareness, and physician leadership.
Ms. Bortz has served also as the Senior Program Officer and Deputy Director of the Virginia Healthcare Foundation for seven years and as Senior Associate Legislative Analyst for the Virginia General Assembly’s Joint Legislative Audit and Review Commission for three years.
Welcome, Ms. Bortz.
Beth Bortz: Thank you, so much, Dr. Moore. It is a pleasure to be here.
Gordon: The reason that I thought the conversation would be interesting is that there is a lot of literature around what is the basis of high performing healthcare systems, internationally as well as nationally. The consensus is, if you look at the literature, that we have to have a good primary care foundation.
I know that VCHI and you have been working on this agenda for a while, and so, I’d like to first hear a little bit about what VCHI is and what its current portfolio of work is about and then, dive into the governor’s taskforce.
Beth: The Virginia Center for Health Innovation was built as a public private partnership. Going back almost 10 years ago now, the Affordable Care Act had passed nationally and in Virginia, we are a state that believes in fairly limited state government.
And so, a lot of the healthcare entities were looking around saying, “How are we going to move the needle on payment and delivery system reform?” It’s not likely that we’re going to create a new state agency. The group came together and said, “We need a neutral convener that can help us in this effort to bring all voices together to talk about the steps necessary to change how we deliver and pay for care.”
The point that everyone agreed on was that we needed to move the Commonwealth from a system that pays for a volume of services to one that pays for value. That became the mission and the rally cry of all the members. We have around the table health systems, health plans, the state patient advocates, labs, pharmaceutical, medical specialty groups, employers—pretty much everyone that we thought that needed to be part of the conversation. We brought together to start to wrestle with the really hard questions.
The good thing is that we have evolved into this role as trusted convener of the group. I always tell people, “It’s a good day if you spin the wheel and I’ve angered a different member of my constituency.” Because there’s going to be some pain involved, but the goal is really to make sure that we’re listening to all sides, thinking about all perspectives, and then ultimately, doing what’s in the best interest of the patients of the Commonwealth.
Gordon: You guys get together this incredibly broad set of stakeholders in the Commonwealth and you’re working now on the Governor’s Taskforce of Primary Care. Tell me more about that.
Beth: We have worked for years doing a lot of learning collaboratives with the medical community focused more on the primary care community that’s been more of a niche role for us. In doing some of that work, we’ve worked on things like improving vaccines or cardiovascular care at the primary care level; reducing low-value care, but through those partnerships, when the pandemic struck, we started hearing from our frontline primary care providers a heightened level of concern about basic sustainability of their practices.
In listening to that, we are fortunate here in the Commonwealth that Virginia’s Secretary of Health and Human Resources serves on the VCHI board. He’s our vice chair. We also have a physician governor in Virginia. We thought the time was right to take the issue of how do we use this crisis as an opportunity to fix some of the problems that we have in primary care.
How do we develop a longer term payment arrangement for primary care that both helps the provider community, but also helps the patient community, because the last thing that we want is to lose practices in some of our most vulnerable communities?
Gordon: Isn’t there this concept of free markets in the United States, where you say that one of the things that happens when a business is not doing well in some sense, that’s just part of the natural thing—sort of the culling of the herd. Is that not true in primary care?
Beth: That could be true in certain markets. The challenge is I don’t think any of us want to see in some of our rural or high need communities there being no provider, because when we look at our population health statistics and take a step back, it’s those very communities that are struggling the most. When you look at the challenges of providing care in the current market, it’s not surprising that those providers are struggling. They have a heavy Medicaid caseload, heavy Medicare caseload—it can be quite challenging financially to keep the practice afloat.
Gordon: It makes me think what I’ve read from different groups that are looking at how primary care is functioning that there over time has been an increasing administrative workload to push their primary care, for instance, around prior authorization. So, a specialist wants to do a procedure to fix, let’s say, a heart problem. The primary care doc has to go play mother may I with the insurance company. Or another specialist prescribes a medication for a chronic condition. Again, the primary care doc has to go in there and none of that is compensated work.
At the same time, while the foundation of high performing health systems is also receiving maybe about four cents on the dollar for every dollar spent, depending on Medicare, Medicaid or commercial health plans. There’s this disproportionate burden and low relative payment for this highly-valued resource.
I think about work of Barbara Starfield and colleagues that show that for every primary care physician per ten thousand population, the more you have, the better the outcomes are for people because they have better access because they have someone who could follow through on their needs, who has a relationship with them so that they can recognize problems early and get in front of them before they get bad, before they land a person in the hospital.
These factors seem to speak to something where I think the strict business model may not apply as well. We may need to think of it as—not as a public utility, but there’s something essential about having primary care access.
Beth: Absolutely, it’s something essential. One of the things that struck me when you were talking was as we look at some of the movement to capitated payment or different payment arrangements that actually hold primary care even more responsible for system performance. Look at total cost of care measures, for example.
Think of the challenge of being a primary care provider who our numbers suggest maybe five to seven percent of total spend, who’s not getting a feed of data that helps them know what all the other contact points that their patient has with the system, are those good choices and is somehow placed in the position of being responsible for those choices or potentially, they will lose money in the new payment arrangements.
We have significant challenges in terms of the expectations; the data that they’re getting and their ability to analyze what they’re getting; the level of accountability that they can hold others to. These are all part of the challenges that the Governor’s Taskforce on Primary Care is starting to look at.
How do we build a system where we empower our primary caregiver by giving them better access to valid trusted data that they can use, timely data that they can use, what other pieces of infrastructure do they need? We talk a lot these days about social determinants of health, but very few primary care providers are getting good information about how social determinants are impacting their particular patients, so data on that front.
Most primary care docs did not go to school learning how to evaluate a risk-based contract, so how do they establish if it’s fair and can they succeed under the contract terms that have been presented to them. There are supports that are needed there.
Then, there’s the models themselves, looking at what measures they’re going to be held accountable for and are those the ones that they believe are most important. There’s some really good conversations to be had between our medical community, our employer community that’s purchasing these services, and then our health plans. That’s what we’re trying to foster, those conversations.
Gordon: Let’s pursue that a little bit. I’m curious that employers are at the table. As I’ve been talking with colleagues and reading the literature and looking at process improvement and practice transformation for 20 plus years. It doesn’t seem to have shifted a lot. There’s increased interest, but where’s the action? Where’s the sticking point and what do you think is going to be the breakthrough that gets us to this better place?
Beth: I’ll give you an example from Virginia. We have worked at VCHI for the last five years on an initiative to reduce low-value healthcare. As part of that process, we very much believe that you can only change things if you have good data and you can review that data and adjust accordingly.
We invested in using a tool, the MedInsight Milliman Health Waste Calculator, which we were able to pair with our state’s All-Payer Claims Database. I’ll take a moment to say that Virginia does have one of the best APCDs in the country.
We have data in our APCD, all of the Medicare, all of the Medicaid, and about 60 percent of the commercially insured. It’s a pretty rich data source and we started running reports to look at low-value care. When we did that and then, we started sharing that data with our provider community, we can run reports actually by NPI number. We can go all the way down to, “Dr. Moore, here’s your performance on seven measures over the last quarter that were deemed to be unnecessary.” You can start to unpack that and dig into it.
But our concern, truthfully, became, alright, but people are actually getting paid for this in the current system. If we have folks knock it out of the park, the end result may actually be that they lose money. If we’re going to really change the system, we have to change the way care is incentivized. That’s going to require the employer community.
We started conversations and we created an employer taskforce here in Virginia and brought some of our bigger players to the table. One of the things that we found is that they had not seen their own data on low-value care. They were dependent on what the plans were sharing with them and I’m sure the plans share some very good information with them, but reduction of low-value care was not a set of what they were seeing.
It was only when we started showing them, “Here’s what we could give the state employee health plan,” and “Here’s what we could give Medicaid so that they could really start digging into their own data,” that an appreciation for having a neutral source of information generated among this group.
Remember, these employers—health benefit design is just a tiny portion usually of all the things that they’re being asked to do. We have to make it as easy as possible for them to engage; we have to give them good data that’s relevant to their own business; help educate them so that they can be a voice at the table.
I think what we’ve learned is, in the past, we didn’t spend enough time up front on the providing them with data and education.
Gordon: Have you gotten any sense that this is helping them shift from fee-for-service to value?
Beth: What I’ve seen so far is that several of our large employers—right now, with our state APCD, large self-insured employers are exempt from submitting data. That’s a federal exemption through ERISA. Truthfully, it’s been a little frustrating to us that that’s a missing—when I say that we have 60 percent of the commercially insured, I’d really like to tell you that I have 100 percent in there. We were missing these large self-insured.
What I’ve seen is, that since we started sharing our low-value care data from state and regional level, we have had several of those large employers go through the hoops with their legal teams so that they are now submitting their data to the state APCD voluntarily.
Gordon: It’s got to start somewhere.
Beth: It’s got to start there, exactly.
Gordon: I start to think, “Oh, wait a second. Maybe there’s something I can work on here.”
Beth: Right, and what else am I not seeing? What else should I be made aware of that might be an area that we can target? That group, when they started to look at data, actually helped us identify the next set of low-value care measures that we’re going to tackle. It’s interesting. They chose different ones than our clinician community chose for some really good reasons and their voice is just as important in that conversation.
Gordon: Are those measures something for primetime discussion?
Beth: Sure. I think we’ve just put them up on our website. If not, I’ll make sure they’re up after today. They just adopted them and I will tell you that they focused more on, when we look at low-value care, these are—I should probably define for folks. Tests and procedures that, in some circumstances, are highly valuable, but in others, due to clinical nuance, may provide no benefit for the patient that we’re looking at.
Think about one of the biggest buckets for us that we’ve been working on has been imaging and labs for low-risk patients before low-risk procedures, kind of the unnecessary preoperative testing.
We started our work because we were working with clinicians with measures that clinicians were the ones that were 100 percent driving the ordering. They were things that most patients didn’t come in and ask for. We felt like the clinicians had a wonderful opportunity to, just through their own behavior, change performance.
Our employer community chose measures that we believe are a little more consumer driven. Things that a patient may come in and request antibiotics that would be unnecessary or may request imaging for low back pain very early in the process without any red flags being present.
They were willing to tackle the piece that logically makes sense, working with their own employees.
Gordon: As I think about these, they make sense and I’ve seen the Choosing Wisely campaigns, like consumer reports and others came together to say, “Can we pick a handful of things that we all agree just don’t make good sense for people? They’re not in their best interest, but we’ve been doing them by habit in the past.”
Is that also going to translate to the lower cost, which would allow for lower premiums or lower deductibles and things like that?
Beth: We hope so. When we ran our first reports in Virginia looking only at—Choosing Wisely has now identified more than 500 tests and procedures across the different medical specialties. The Milliman Healthways Calculator focuses on 48 of those, right now, that are ones that can best be assessed solely using claims data.
As you can imagine, some require more clinical information to be more clinically nuanced. We chose a small subset that you could make determinations using claims data. When we looked at those and, again, we have data on about 5.5 of the eight million Virginians, so not the full set. But we saw, on any given year, between 500 and 700 million in unnecessary care in Virginia.
Gordon: That would clearly make an impact.
Beth: Exactly. The initiative that we have underway—the Smarter Care initiative, we actually have more than a thousand practice sites—about 7,000 clinicians throughout the Commonwealth that are working to reduce seven of those measures. Our goal is, over a three-year period, to achieve a 25 percent relative reduction.
Gordon: How do you get to that problem of the perverse incentive where doing the wrong thing puts finances into the practice; doing the right thing starves me?
Beth: That’s a real challenge and I will tell you that’s why, when we designed the initiative, the very first meeting that I had was not with the chief medical officers, but with the CFOs of some of the large health systems that were going to participate, because my fear was that, by the time the information reached the CFOs and they saw this, they were going to shake their heads and say, “No way. We can’t participate in this.”
What I have learned is that those systems are so eager to get good quality data, which is what we’re giving them. We give them, from the state APCD, every quarter a report on every single clinician in their network on each of the seven measures. They felt that, even though we’re talking real money, we’re talking about a small enough bucket of money that it would help them up their skills in entering into contracts where they would bear a risk.
The whole learning process of learning to change their practice based on that data, they were willing to sacrifice the financial loss with the hopes—remember, I shared—we have a whole employer taskforce on the other side of that work that, by the time that we got to the end of this three-year period, the employer side would be helping to change benefit design.
Gordon: Let me pivot to the change practice. That is hard work. I think about the average primary care clinician practice and they are running—even pre-COVID. They are running to try to keep up with all this work. Of course, we need to decompress some of the non-value added administrative stuff that’s rolling to them.
I like that, for instance, you’re talking about using claims-based measures so we don’t have yet another measure or reporting requirement, which adds to the work without changing the care. That’s a good thing. I like that.
When I think about that average clinician, it is very hard for them to pick up and start acting on new [books 21:43] of work—especially ones that aren’t funded, like let’s start thinking about the non-medical factors that are getting between this person and optimal health. The fact that they don’t have access to good housing or transportation or for good food or if because of their socioeconomic status, they’re severely disadvantaged for access to healthcare, in general.
These are monumentally impactful on outcomes and yet, completely outside what a typical clinical practice can do. When I think about the models that work, they start adding community health workers and social workers and health coaches and all these other salaried line items that, if not funded, are just impossible and then, the practices have to figure out, “How do I work with this person? How do we fold these data into our dataflow? How do I analyze this information to make use of it?”
I wonder if there’s a need for some kind of cooperative extension almost that would support clinicians in doing this transformative work.
Beth: Absolutely, I think that’s one of the things that our new Governor’s Taskforce on Primary Care is looking at. Do we need, potentially, to create some kind of a center for primary care, amping up some of the work that VCHI and others have already been doing to try to support primary care?
I think the other piece that we can do, is to work hard to better align the different places where physicians are having to plug into and report. Some of the work, when we talk about performance measures on the Governor’s Taskforce, it’s really important that, if we want folks to move into more paying for value-type contracts, that they not have to participate in 10 different arrangements where they’re reporting on 10 different sets of things.
Let’s see what we can do to better align that side of the house and align it with other things that they’re already doing. One of the other hats that I wear is I serve as public member on the American Board of Family Medicine, which is involved in certification for physicians. Shouldn’t that certification process be a piece of the benefit that plans can look at and say, “This physician is certified. They’re already met the first set of criteria for us.” Trying to get some of these different systems that truthfully are overwhelming primary care to work better together.
Gordon: I think about what again what I hear from clinicians in practice right now is that they say—or, to be honest—the ones who are head in the wheel of just working hard, they even don’t have the time to step up and look. They just say, “The electronic medical record is getting in my way. I’m being bombarded with administrative trivia.”
The ones who are the practice leaders or the medical directors over groups have a little bit of breathing time to step back and look at the big picture. They’re saying, “We need to report externally a whole bunch of quality metrics, which in and of itself is a good thing. That’s fine, if we could automate that through administrative data, that’s terrific, but it’s often not the case and we, therefore, have to dive in and painfully extract information out of electronic records that are not facile in doing that work.”
And there’s a difference for each health plan and, sometimes, within the health plan for each of the programs or, sometimes, because a self-funded employer wants a particular set of data about their employees. The complexity of that is just overwhelming.
I’ve heard, in Colorado, hundreds of measures.
Beth: Hundreds of measures. We looked at it about seven or eight years ago. We had some funding from the feds as part of their State Innovation Model program and we looked at measures in Virginia. We found that, at that point in time, there were more than 500 measures if you were participating in all the contracts.
No one’s moving the needle on 500 measures. It’s why we actually, after doing that work, created the Virginia Health Value Dashboard. One of the things that we’re probably proudest of at VCHI is that we have brought the whole group together. We have more than 50 different organizations represented on our board and leadership counsel and have agreed to a handful of measures for the Commonwealth.
That doesn’t mean we have the ability to tell the plans, “This is all you can use,” but we’re trying to get folks to say, “We can only move the needle on a handful of things. Let’s be very strategic about what we’re choosing.” We chose a handful that are looking at reducing low-value care. Then, we chose a handful that are increasing high-value care.
Finally, we closed it out with a few measures that make sure we are building, as a commonwealth, the infrastructure we need to be able to measure and reward value in care. Trying to keep it somewhat simple and report out frequently on Virginia’s performance across regions of the state.
That’s one of the other things I think is really important. We don’t need to be doing all the same initiatives in all regions of the Commonwealth. Things that are significant issues for northern Virginia may not be issues at all in southwest Virginia. I’ll tell you, if you look at the low-value care data, for example—the provision of low-value care is a much bigger problem in northern Virginia than it is in southern or southwest Virginia.
Vaccination rates, however, are more of a problem in southwest Virginia and southern Virginia than they are in northern Virginia. I think that’s the other piece of what we’re trying to learn with good data is how to target our work.
Gordon: That’s interesting. That makes a lot of sense to me. One size does not fit all. Let’s assess the—
Beth: Absolutely. One other thought—I just want to come back to, when you were talking about the supports and what we need. I think that—and earlier, you asked me about the market effect. I think we do have challenges with some of the small independent primary care practices in that, truthfully, that model is almost impossible to keep up with all the things that are currently required. While we don’t want those practices to go away, we do have to figure out how to do something different.
Gordon: It occurs to me that—I’ve seen some data. I’ve got to go back and look at the sites here, but I’ve seen some data that, when independent practices are bought out by healthcare systems, we sometimes add a lot of cost. For instance, adding a bunch of facility fees on top of what you suggest be provision of care fees. There’s no improvement in quality outcomes and so, it just adds costs to systems.
That may not be true in all cases. I’ve certainly seen health systems that are brilliant and, when clinicians get folded into them, they have access to resources that previously they couldn’t attach.
I’m thinking, isn’t there an opportunity then, to think about models where, through some primary care cooperative extension or some other managed service organization could provide that external support? It could provide, you know, “We have folks in your region who have health coaching experience or community health worker experience and they’ve been trained externally and they’re supported through these other funding streams so that they can work on behalf of Virginians who are coming to your practice and everybody wins when we do that.”
Beth: Absolutely, and you mention community health workers. One of the pieces that we did with our SIM grant here in Virginia several years ago and the work is ongoing in the Commonwealth was a recognition that, when we talk about community health workers, we kind of talk as though they’re all the same.
I’ll tell you, one of the things that I’ve learned is that, if you’ve met one community health worker program, you’ve met one. Because the training that they provide, the focus, the targeted populations, they can be very different. If we want—and I think we do—if we want there to be reimbursement for those services; if we want to move beyond them constantly being in search of a grant to support their work, then there has to be some kind of credentialing and standardization for community health workers.
We can’t expect the mom and pop private practice to be able to be making those distinctions right now. That’s another piece that we have worked on, trying to add a credentialing process for community health workers.
Gordon: That makes a lot of sense. Let an external entity figure out what the criteria are, train people up, and verify their authenticity. It makes me think—I’ve heard that within the Affordable Care Act, there is a provision in legislation for creating a primary care extension service that was not funded, initially. It’s there; it’s all written in and we just need to get that funded and I think we can breathe life into that for any state that wants to do it.
Beth: We can. I’ll tell you, though, we got funding a few years ago from ARC here in Virginia to do a modified version of some of that. We had about 220 primary care practices working with us. While we learned a lot, and I would say that overall, the project was very successful, we did have some unanticipated challenges. The biggest challenge—I feel like I’m harping on this, but was the data piece.
What we found was that the overwhelming majority of our partners could not produce a report on the measures that we were looking at from their EHR. These were not complicated reports; these were all things that should have been done through meaningful use. Our partners were all using compliant EHR systems. I’m just going to tell you that they don’t function the way, perhaps, the vendors might promise that they function.
Gordon: We still have a hill to climb with that. I’m glad you have, at least, the idea of combining different payor data into that all-payor claims database, which gives you insight into five plus million Virginians.
Beth: Absolutely. I do think we’re making improvements on the EHR front, but one of the outcomes was we actually had to go testify at ONC. I think sometimes we assume that maybe the providers just don’t know how to use the software. We learned the hard way that was not the case. It was really challenging to get the reports out.
I keep coming back to, I feel like we have to solve the data problem because the data problem will help us solve the other problems.
Gordon: Yeah, and then, we need that infrastructure where I don’t think we need to train up the primary care work force in how to understand risk adjustment methodologies and other group, or what have you, but we need them to trust that the folks working on their behalf know that well and can help them address these factors.
Beth: Absolutely, that the data is reliable because—and I will say, having looked at a lot of it, it takes a lot of digging, sometimes, for it to be reliable. They’re not whining unfairly. They have a legitimate concern, at times, with the quality of the data that they’re getting.
Gordon: Well, Beth Bortz, thank you so much for your time, today.
Beth: I’m happy to talk with you. I hope you can tell this is a passion for us. We really do believe that our primary care providers are an essential piece—the frontline for Virginia’s healthcare system. It’s really important to us, to all of us, that they are able to thrive.