On May 2, at the World Health Care Congress, held at the Marriott Wardman Park Hotel in Washington, D.C., Frances Martini, R.N., and Katrina Miller, M.D., presented a session entitled, “Consumer-Centric Care: Member Engagement and Retention Strategies for Complex Populations,” as part of the Medicaid & Health Policy Summit and Behavioral Health Summit tracks, at the Congress.
Miller is chief medical information executive at L.A. Care Health Plan, a health plan that serves primarily Medicaid recipients in Los Angeles County, California. Martini is director of integrated population health at BlueCare, the Medicaid division of the Chattanooga-based Blue Cross Blue Shield of Tennessee. BlueCare manages the care of about 600,000 individuals, out of the total of 3 million BCBST members. At their presentation, Martini and Miller discussed the complex and manifold challenges of managing the care of, and attempting to improve the health status of, a population that is beset by problems of poverty, housing instability, food scarcity, behavioral comorbidities, and general instability around the personal and socioeconomic factors of life, while also attempting to engage members of that population around their health. Yet both L.A. Care Health Plan and BlueCare have been making significant strides in managing the care of their Medicaid plan members, and in engaging them around their health concerns.
Recently, Martini spoke with Healthcare Informatics Editor-in-Chief Mark Hagland regarding recent initiatives and developments at BlueCare in Tennessee. Below are excerpts from that interview.
Qualitatively speaking how might you characterize the Medicaid population that you and your colleagues help to manage at BlueCare?
I would tell you that 60 percent are children. Of the adult population, many are pregnant women. We also have developmentally disabled adults, and adults with significant complex illnesses. And 40-50 percent, especially the high-risk population, also has behavioral comorbidities, which adds to the complexity of their care management.
What kinds of elements make it challenging to connect with them around their care?
You can’t look at all 600,000 people at once; it would be totally overwhelming. So we look at sub-populations. We look at age-based cohorts, including 0-4 years of age, and 5-20. And clearly, a four-year-old is very different from a 15-year-old. And clearly also, different things affect newborns from those that affect 55-year-olds. So we divided our population into sub-populations—small children, school age, young adults, and so on. And you know that a 0-4 doesn’t have behavioral or chemical dependency, but they could have disability, etc. And with school-age kids, behavioral or chemical dependency might be high, but cancer won’t be. So we looked at those elements. And for the 5-20 population, behavioral and chemical dependency is number one in terms of risks, and asthma is second. Behavioral is around 15 percent and asthma is 10 percent, and of course, you can have both. Obesity is also very significant; and so are developmental disabilities. So for me, in putting my team together, I have to be familiar with those conditions, know I have a network that can handle it, where those members are, and have a staff to manage them. Because the kind of staff that would manage newborns would be very different from a staff that could manage school-age kids.
And I have to look at what’s happening in the community; and what’s going on in the schools; things like what the school nurses are doing; the contact level with the parents; the educational levels of the schoolchildren. And that’s where the data is so important.
In your presentation that was part of the Medicaid & Health Policy Summit track and the Behavioral Health Summit track, you talked extensively about the value of data, to population health management and care management work of the type that you and your colleagues are engaged in.
Yes, in fact, I actually participated in two panels at the World Health Care Congress; I did a second panel around informatics. And the thing is that, without the data and being able to have someone help you interpret the data and make sure you’re tracking it correctly, it’s useless. So he and I did a presentation together, and he did one of those ‘be careful what you ask for’ kinds of things. Because sometimes clinical people don’t want they’re asking for. We actually call people like him “data clinicians.” He helps us think through what’s happening and what we’re doing, and how to track it. So we might start out saying, for example, we want to set up an asthma program. But what we found out was that the asthma problems we had are with children from zero to four. Clinicians will tell you that most children that age don’t have asthma diagnoses. But we had so many children in the ED with asthma symptoms, that we ended up creating a program to help manage that. So that’s the beauty of informatics, not only helping us to identify a population, but also how to manage that population, and how to intervene.
So once you’ve identified a subgroup, then what are the next steps that you take?
Yes, for example, we had a very small subset of our population—less than 1 percent—with very high medical costs, and who were frequently admitted, but who still weren’t getting better. Our data showed us that they were utilizing services, particularly hospital services, but they weren’t going to the doctor. They weren’t utilizing primary care.
Were they usually just going to EDs with symptoms?
Yes, and often being admitted. So we said, how are we going to address this? So we did was that we decided to look at people with 10 or more admissions a year. We came up with 285 people who were really, really at the top, and we said, wow, that’s not very many. So we have something called Persona. We have not only internal data—our stratification tools and so forth; we actually purchase data from external sources, like Experian. So I looked at a 65-year-old woman with diabetes—she lives in an area with families, and she’s not homeless. That, compared to a 65-year-old woman with a similar diagnosis, but who’s using the ER all the time, but I can’t address her situation the same way. So we did a deep dive, and found that in both cases, we had tried to reach out to them. But we could never find them, could never get to them. So we identified where they lived, and the staff just found them. One of them was in dialysis, so we said, wow, we know where she is, let’s just go there. The other one, we flagged her record, and went to the hospital when she was admitted. And the lady who was in the family area just didn’t want to be a burden. So we talked her and said, don’t you want to be around when your grandson gets married? That kind of thing. The lady who was in dialysis and who was homeless, we worked with her to get housing and disability and things like that. Her life isn’t perfect, but we did what we could. And we helped.
So part of it really was almost like working as private investigators, tracking them down, right?
Yes, I hate to say it, but yes, that’s what it was like. And sometimes, a staff member could figure out where the member was. And sometimes we knew where they were in terms of apartment buildings, etc. We did a lot of what we call “drive-bys,” even though that’s not an ideal term. But we did find every single one of those 285 people; it took about two to three months to do so. But we got to the point where we knew where they were. To give you a couple of examples, one time, someone said, oh, that member was just admitted! And we said, go run over there now! In another case, one of our staff knew the sister of a plan member. He lived with a bunch of other folks in a group home, and we were able to find him.
And once you do track down those members, how do you get member engagement with them?
We call it meeting them where they are. We meet with them and find out what they want. My goal might be, let’s not have you in the ER all the time. Their goal might be, I need food, or to be with my family. And the only way to get results is to help them get what they want. The Case Management Society calls it integrated population health or integrated case management. We don’t start out with, “Have you ever had asthma,” or whatever; we start with, “How are you?” And they might say, “My son is living with me, and I can’t get him out of the house.” So we don’t immediately say, “Have you had a pap smear?” And she might seek ER care because of her living situation. So we have to figure out that situation. And we could say, ‘Why don’t you move?’ But that’s not what she wants. So we have to work out her situation for her. And sometimes it means we have to fix what’s wrong with the family or the situation…
There’s this myth that keeps perpetuating itself that people on Medicaid don’t care about their health, but that’s not true. Could you speak to that myth?
You’re exactly right, it’s a myth. Sometimes, their choice is between buying food or fulfilling their prescription. Here’s an example: the electricity was turned off on a family, and their child had a tracheotomy, and without electricity, they couldn’t suction their child. So I called the electric company, because it was an emergency. And they said, “No, I can’t pay it because I need an account number.” So we went out on a well-check visit with the police. And it turns out, the mother didn’t have enough money to pay the electric or the phone bill—and the other children were suffering, too. So they were able to locate the mother’s father, and the police took the mom and baby over to the grandfather, where there was electricity. And we got her a free cell phone that she was eligible for, because of Medicaid, and got her a grant for the electricity. So then we could focus on the kids getting their immunizations and making sure the baby was able to be suctioned.
And sometimes, the amount of money involved is actually tiny compared to people constantly rotating through the ED or hospital, right?
Yes, that’s exactly right. You think about a $700 ED visit versus a $200 electric bill, which would you rather pay? The struggle is getting them to the PCP. And sometimes, the behavioral problem comes out of needs, like hunger. So we do things like making sure the nurses have snacks, like Lunchables, so that they can better cooperate with the doctor. You know, there’s a quote that talks about population health, especially in dealing with Medicaid. And the quote is about the broken families, the lack of money, the lack of access to any kinds of services. It’s not about solving a purely medical problem, it’s about solving all the problems. It was the population health group at Jefferson Hospital. I use it in almost every presentation, to get people used to what you have to deal with, even with our staff. The quote comes from David B. Nash, M.D., dean of the Jefferson School of Population Health, at Thomas Jefferson University in Philadelphia. He says that, to improve population health “It’s the housing, it’s the lack of access to food, it’s the broken families, it’s all that messy stuff. It’s a complex story, so it can’t have a medical fix to make it work.”
What should CIOs and CMIOs be thinking about, as they consider the advances that you and your colleagues have made, in this initiative?
I would tell them that, first of all, numbers have an important story to tell, and they need to rely on those numbers. Our informatics team is part of our clinical team. We talk about “data journalists,” those who tell stories with their data. I would say their informatics and clinical folks need to be in with their data analysts from the beginning. It’s kind of like what a CIO does, they translate the systems for the users. They’re like that piece in the middle, to make sure the people in the business get what they need.
And often, it’s the clinical informaticists in hospitals and health systems, who bridge the IT and data people and the clinicians, in efforts like this one.
True, but in any case, the more closely they all work together, the more likely it is that you’ll come up with a good approach. And we’re huge on control charts—I was at an IHI workshop on the use of control charts, where you have a mean, and you know what your standard deviation is, and you know when you’re tracking an activity and you’ve got so many data points above the mean, you know you’ve got a significant variance. And when you create a plan, and then you have a process change and you track it, and see positive change as a trend, you know that you’ve made a difference. And if you do something and you don’t see change, you need to stop it and do something else. So you can say, OK, I tried this and it didn’t work, but the variance that occurred wasn’t based on that. And I’m a nurse, and it’s helped me in my career to not waste time and money. And I always look at the example of Florence Nightingale, who tracked deaths on the battlefield, and was able to identify infections and improve sanitation.
Do you have any metrics that you can share, with regard to some of the results you’ve achieved?
We’ve had four major interventions, and have been able to show improvement in inpatient admissions. And we’ve been able to translate that into a savings of over $650,000. We found that, among those 285 people—we ended up with about 250 people as we went through the year, because a few were disenrolled or died—but we saw a statistically significant reduction in admissions, and we saw a reduction in costs, without any additional outlay of expenditures or staff. That’s the beauty of data.
What will happen in the next couple of years, as this initiative evolves forward?
We’ve expanded this approach to the next level, and we’re working to see what lessons we’ve learned—finding Medicaid members is a challenge, but we’ve been using some of the same approaches with the less-acute. And with a lot of the technology out there—a lot of patients and members will self-manage, but you also have to work with the populations you have, and work with the tools you have. It will be interesting to see what happens with healthcare, and with Medicaid expansion or not. We have 6 million people in Tennessee, and 20 percent are on Medicaid. That’s a huge percentage of people who depend on us. And we’re going to have to use data in a good way.
And you’re all about working with providers, correctly?
Absolutely, absolutely! And you know what? Doctors love data. And it’s OK for me to stand up and talk about 600,000 members, but the individual physicians want to know about their 1,000 members each. And we’re paying physicians more the better members get. So that’s where things have been headed.