A conversation with Kevin Barnett, Grace Damio, and Carl Rush
Kevin Barnett is a Senior Investigator at the Public Health Institute. Grace Damio is a public health nutritionist and Director of Research and Training at the Hispanic Health Council. Carl Rush serves as a core team member of a policy center on CHWs at the University of Texas – Houston School of Public Health. (For the bios, click here.) All three have been engaged in Community Health Workers issues for many years.
Interview by Molly Miller and Tom Peterson
Stakeholder: What are some of the best or most innovative ways hospitals are using community health workers to improve the health of the communities surrounding them? And what are some barriers and misconceptions that might keep hospitals from using them?
Rush: There doesn’t seem to be any systematic data gathering about what hospitals specifically have been doing. But I’ve been hearing a number of examples – particularly under community benefits funding. Partnerships are forming between hospitals and organizations in the community to engage in large-scale population health initiatives to improve health. It depends on what you mean by community health because there are some interesting approaches with safety net hospitals that still have large percentages of uncompensated care involving community health work in care coordination or redirecting individuals into more appropriate sources of care, from the emergency room in particular.
Barnett: Other than a few exceptions, hospitals have not engaged community health workers or promotoras in any substantive manner. Where they have engaged them, they’ve come to the issue in looking at the kinds of challenges most relevant to them. So a good number of hospitals engage with community health workers to help with the enrollment process as part of the Medicaid expansion.
As they have looked at the kinds of issues people face related to readmission – as CMS has established penalties around readmission – it has brought to light that there are issues people confront post-discharge that now have financial implications for the hospitals.
Then there is the broader issue – where we have a good amount of evidence relating to chronic disease, pregnancy, and other related issues that are easily defined and which involve substantial additional costs if these conditions are not well managed. There are numerous examples where hospitals have reduced admission and utilization for things such as asthma, cardiovascular disease, and other conditions.
So, they come to the issue with the focus on how can we solve this immediate problem? They don’t come with the perspective, at least automatically, of how community health workers will help us begin to solve some of the larger drivers of poor health in the community. That, in many instances, is down the road.
Damio: That description resonates with our experience as a local agency. We’ve partnered for decades with one local hospital, and they’ve done every combination of referring to our programs, funding our programs, or hiring their own staff to be part of our programs. So we have active with that hospital some partnership in cancer work with community health workers, a decades-long partnership in prenatal case management, a decades-long partnership in breastfeeding peer counseling, and then a very rigorous randomized-control trial on diabetes peer counseling.
Several of these have a lot of evidence behind them. The breastfeeding work was impressive to a second city hospital. They got a Kellogg grant to build in the expansion of the program to work with their population and their hope is that they’ll convince their hospital to actually pick-up support as the grant goes away in a few years. That said, it is issue-based, some around health promotion and some around chronic disease management. The background issue is that funding is in jeopardy as hospital funding gets cut.
A close colleague and ally from a local hospital recently said to me, “We’re so interested in this topic but what we’re really trying to do is find the right model.” And I said, “We have models that we’re working with you on and we’ve actually done research on them.” I’ve heard that more than once from people who either are well-intentioned but don’t see what’s right in front of them or who are looking to create the groovy new thing and want to be the one who found and brought in the right model.
So, slipping into the barrier question, it’s interesting how the rigor within the community that has some evidence-base behind it isn’t necessarily seen for what it is. And the idea is that we’ll find something Georgetown did and adapt it.
Stakeholder: So you all have been working with this for a long time, and there are evidence-based models where community health workers have been working well with hospital systems. Why isn’t this catching on all over the country?
Rush: Part of the uncertainty many hospitals feel is related to the fact that the community health worker has such a broad scope of practice. It does not overlap a lot with clinical occupations. They do so many different things at an individual patient level as well as a community population level. Folks have difficulty pigeon-holing them. Also, many feel a sort of common sense constraint – they’re looking for a source of support, particularly a sustainable one through a third party payer to support these positions.
I’m seeing, to some extent, a growing interest, not just among hospitals but also among health plans. For example, when looking for a dedicated funding stream to fund these positions they finance them out of their operating budget on the basis of internal return. There still aren’t a lot of examples of that, but there is a growing number. Some fairly prominent health systems are jumping in saying, “Let’s look at how our system needs improving and see where the community health workers fit in that.” This is a bit of a departure from what Kevin was pointing out, looking at an immediate problem and then casting about for a solution to that.
We see another barrier in some systems: I’ve heard direct quotes from folks who say, “Community health workers simply aren’t part of our model.” They don’t want to hear any more about it. Thankfully, that’s relatively few. But a lot of folks in these systems are so steeped in the medical model and the modus operandi for the community health worker is simply not based on the medical model, nor is their expertise in the domain of clinical knowledge and clinical training. They don’t quite know how to wrap their heads around that. So that takes a little bit of cultural shift on the part of a lot of these organizations.
Stakeholder: What are some things you all have seen as systems have been able to overcome that bias towards just a clinical model?
Rush: Some systems are simply experiencing, especially with low-income patients, some serious barriers with communication. I’m talking about the barriers that result from power differentials between patient and provider, mistrust of institutions, and other things like that.
A second major current is recognizing a need for better communication, more continuous communication, with patients to deal with issues like care transitions, readmissions, adherence to treatment for chronic disease and that sort of thing. But also coming to respect much more the importance of social determinants and recognizing that most of their workforce is not well equipped to deal directly with them.
Damio: The diabetes peer counseling model we developed in Hartford has 17 home visits and the peer counselors were completely community based. They were integrated into a clinical health care team and conferred in face-to-face meetings once a week, but from the field, from the client’s homes. From that, it became obvious to the provider that there were major gaps in health care going on during the visit, that they needed internal care coordination. They needed cultural competence training within their own institution. And the relevance of what the community health workers produced was pretty stellar in terms of the reduction of HbA1c.
It really was a full, rigorous, community-based model – well integrated, but on the outside dealing with social determinants. Everything from lack of access to pharmaceuticals, to depression not being well-managed, to not knowing how to manage glucose in the home, to educational issues. Many things are going on within the home setting and creating an effective interface. Unfortunately, it was a study that ended and there wasn’t funding to continue. But the hospital saw not only the benefit of what the peer counselors did but also the bigger picture within their institution of needed changes.
Barnett: Clearly, one of the biggest obstacles to the substantive engagement of community health workers is our fee-for-service system. But with the growing volume of managed care, of capitated contracts for Medicaid, we’re still largely operating in a fee-for-service system. The only real entry-point that makes any financial sense for hospitals operating in that arena is through looking at ways to manage uninsured and underinsured patients coming into their emergency room.
That was some of the earliest work – around Emergency Department diversion. Now, with the notion of beginning to be at-risk for the populations that are insured and that a growing number of people who are being insured are under the Medicaid expansion or are lower-income people with Silver and Bronze plans, it now is becoming important to look at creative models and look at the concept of how team-based care needs to be expanded significantly beyond the clinical arena as we see that the drivers of poor health are within that larger community context.
A number of communities in Ohio and Michigan are testing of the concept of the hub model of community health workers. In essence, a separate 501(c)(3) organization contracts with multiple providers and payers. Because they are based in a specific geographic community and are independent and able to contract with those entities they have more of a capability to look at strategies to address some of the drivers of poor health. These may be poor indoor conditions in housing that contribute to the exacerbation of asthma conditions, or broader issues around food access and food insecurity, safety, or other issues that the sector needs to begin to address.
This hub model offers the potential for hospitals to partner with entities that are doing this. Some hospitals are contracting with an external entity to do this for them. I’m most interested in communities where they’re thinking about community health workers serving multiple providers and payers.
Rush: Picking up on that, a number of states have engaged community health workers as part of their design for their SIM strategy. There seem to be two main approaches, and sometimes they are taking both. The SIM plan may address community health workers as a part of a redesign strategy, but it also may be separately considered as part of a workforce strategy under the SIM plan. So the plan may not address the role of community health workers directly in the redesign of health systems, but they may be an important element of workforce strategy.
One of the strengths behind the community hub model is that it allows the hub to draw funding from, or to bill, a number of different entities, such as the Housing Authority, depending on the services being rendered.
Back to the uninsured and uncompensated care, early on, and more or less on their own, several hospitals in the Houston area started applying this approach, mainly emergency room diversion. Almost immediately they began to see dramatic financial return in terms of reducing uncompensated care and the net return over the cost of the program. Employing community health workers has had a return on investment of around 3-to-1. The CHRISTUS system, in one of the regions in Southwest Louisiana, had a return of something like 15 to 1. So they’re certainly convinced of the financial benefit.
Stakeholder: So is emergency room diversion by using community health workers the low-hanging fruit for many health systems?
Rush: I think that’s appropriate. A couple of more sophisticated models have grown up as well, not explicitly hospital-centric, but out of New Mexico is really driven by the health plan that sees similar returns. They engaged a couple of provider networks with supplemental care coordination fees and employing community health workers to get to high utilizers. This is similar to the hotspotting model used in New Jersey, which is in multiple hospitals and partnerships. And they’re high on the concept out of Camden.
Now, they’ve gone even farther as a community partnership in terms of being able to pull information from schools and law enforcement and a variety of things in terms of targeting interventions in the highest priority census tracts to address some social determinants directly and identify people at risk, rather than waiting for them to show up.
Barnett: The inclination is to see community health workers as doing something mysterious, that we don’t understand, and somehow they have connections to people because they come from those communities that today will help doctors and clinical teams get more patient compliance. It’s a simplistic way of interpreting it, but it’s often the way that it’s viewed.
The truth is, there are complex dimensions of knowledge, experience, and understanding of leadership qualities that are needed in order to be effective. Understanding the interactions between the social and physical environment and family and culture and ways in which that plays out in communities, you don’t just give that job to anybody. Going forward, we need to look at community health workers as providing the off-ramp out of the body shops we have created for just providing acute medical treatment to illnesses – many, if not most, of which are preventable. So they provide that off-ramp of the medical expenditure super-highway into the communities where health is actually created and, in that sense, are in a position to provide leadership and to help us begin to impact health.
So the low-hanging fruit is all of these people who are ending up in our emergency rooms who offer the potential to bankrupt not only our health delivery system, but our larger economy, with more and more people acquiring preventable diseases like diabetes. It’s important to understand that there are some strong feelings within the community health worker community to say, “Don’t undermine what is most important about our potential contributions by simply having us be medical service navigators.”
Damio: I would also mention that it was also interesting within the SIM discussions because the original thinking was that the community health worker would be based within the clinical setting and largely play a navigator role. An enormous amount of, not just pushback, but strong input that was given shifted the thinking with the complexity of the role and why it needs to be in more of a community setting.
When the requirements were rolled out, the discussion came up: does the community health worker have to be in a community setting? Many thought, no, they can be in a clinical setting and still be effective; we’re not going to force that. I said that if they’re in a clinical setting they are much more likely to be bound to a navigator role and not get beyond that. We didn’t have enough people saying that for it to be fully built into the language of the plan. When people learn what a community health worker is, there is a mindset to first think of that navigator role, and that’s so limited. They think, really, dump the person into an advanced practice network or an FQHC, and within the FQHC the idea is that they’re in the community. A lot gets lost with that kind of thinking.
Rush: Grace brings up the important point. It’s not so much where their base is – as in where they’re supposed to be spending most of their time – as it’s whether they have the flexibility to work where they’re needed. Sometimes that’s intervening and being present during discharge discussions. But it’s also doing home visits and so on. This can be uncomfortable for those used to managing people in a clinical setting – that all of a sudden they’re not there a lot of the time. This requires a certain organizational maturity and a supplemental set of supervisory skills to be able to oversee a workforce that may not be physically present in the institutional setting a lot of the time.
Another trend, there’s a lot of discussion in the trade press recently about integration of behavioral health in both the primary care setting and the hospital setting. This is where community health workers show tremendous potential. They seem to gravitate naturally towards certain techniques that are helpful in dealing with folks with behavioral health issues.
And to sort of put a ribbon on this point about the safety net hospitals, there seems to be an appreciation for some intangible things that community health workers bring, including anecdotally a fair amount of impact on patient satisfaction and loyalty and things like that. There was a short demonstration with community health workers, mostly in pediatric departments. When the state funding ran out, three of the four hospitals participating kept those community health workers on the payroll without a distinct means of support. When the one here in San Antonio, Christa Santa Rosa Hospital, experienced an across the board staff cut not long after that, the head of one of the departments that employed the community health workers put one of those positions on the block and the CEO came down and said, “You will not eliminate that position.”
Stakeholder: Is it your experience that if hospitals will try them, they’ll like them?
Rush: It can be that, but with a certain proviso of making sure they do it right, with clear guidance that they are making sure that they hire the right people. In many ways that’s more important than the training. And making sure the pieces are in place to integrate them into the organization, which can be jarring for an organization with an established culture.
Damio: The use of the community health worker services needs to be somewhat seamless for the other providers so that it doesn’t feel like a burden or something that’s too peripheral. Otherwise they just forget about it or think it’s a nuisance. And that means communication systems and sharing of information and ways to get updates and even the referral process need to be carefully and elegantly designed to fit their needs.
When we did a care coordination process with the Children’s Hospital, the pediatric leader that envisioned it said at one of our planning meeting, “This might feel like heresy, but I’m more concerned about provider satisfaction that patient satisfaction.” And he only meant it in the sense that it needed to be well-integrated into the practice or it wasn’t going to work, so that’s one.
Barnett: We can’t just drop community health workers into a team context where there’s not a clear and in-depth understanding of their contributions. We hear all too often from community health workers that they don’t have a ready ear and a perspective among other members of the team to integrate the value of what they have to offer as part of that process. So as we look at engagement and community health, it’s important to think about what kind of training is needed for other members of the team to optimize contributions.
Photo: Hubertl, Creative Commons.