Kentucky: Focus on emergency room “super-utilizers” could lower health care costs and improve care

One percent of Medicaid beneficiaries now account for more than 10 percent of the state's annual emergency room Medicaid costs.

Emergency rooms are a costly source of primary care that too many patients rely on too often. Image credit: iStockphoto by Getty

Research by the Robert Wood Johnson Foundation finds that more than half of all Medicaid spending nationwide is driven by just five percent of Medicaid beneficiaries. These “super-utilizers” often suffer from multiple chronic conditions, such as diabetes, as well as other challenges such as homelessness or substance abuse. They often rely on emergency rooms, not primary care doctors, for all or part of their care.

Finding these super-utilizers and helping them get better care outside emergency rooms can dramatically reduce health care costs. In Camden, New Jersey, Dr. Jeffrey Brenner’s pioneering technique to map super-utilizer “hot spots” helped lower hospital visits and health care costs by as much as 40 to 50 percent.

Under a directive from Governor Steve Beshear, Kentucky is one of seven states participating in an innovative initiative launched by the National Governors Association, the Robert Wood Johnson Foundation and Atlantic Philanthropies that builds on Dr. Brenner’s work to develop new strategies aimed at providing super-utilizers with better, and more cost-effective, care. 

We spoke to Dr. John Langefeld, who is Chief Medical Officer for Kentucky’s Department of Medicaid Services and one of the principal architects of Kentucky’s super-utilizer initiative:

Dr. John Langefeld, Chief Medical Officer for Kentucky Medicaid Services

Dr. John Langefeld, Chief Medical Officer for Kentucky Medicaid Services

R3.0: Why is emergency room “super-utilization” a concern in Kentucky?

Langefeld:  To put it in context, since Kentucky is a Medicaid expansion state, approximately 25 percent of our population in Kentucky this year will be Medicaid recipients. In a 12-month period, we had over 356,000 Medicaid recipients utilizing the emergency room at least once. Over 45,000 used it four or more times, and over 4,500 used it ten or more times. And we had at least two that went over 120 times each; we had one recipient who went to 30 different emergency departments (“ERs”). When we looked at the cost, [it] was in excess of $341 million just for coverage of those ER visits.

R3.0: What is that in terms of the total Medicaid budget?

Langefeld: Our total Medicaid expenditures in that year was just under $6 billion. If we look at the high utilizers – those with 10 or more visits in a 12-month period – that 1 percent of Medicaid patients consumed almost 10 percent of the dollars spent for emergency room use.

R3.0: What common characteristics do these super-utilizers have?

Langefeld: We looked at the almost 4,500 [people who] used the ER ten or more times in a 12-month period, and when we really tried to get at some of the specific questions around that subpopulation, we saw that many of them – almost 80 percent – had a behavioral health [mental health] diagnosis, and 45 percent of them also had a substance abuse disorder diagnosis. We know from other studies of our population that access to a continuum of care, particularly around behavioral health, has been lacking. It’s one of the things we’re focusing on.

“The whole idea is to take a system that is fragmented and dysfunctional and restructure it in a way that supports [patients] in a holistic way and addresses issues that have traditionally not been addressed through our medical system.”
R3.0: Why are these super-utilizers coming to the ER?

There are multiple reasons people utilize the emergency room and not all of them have to do with medical issues. We know that a lot of the utilization is just a matter of people needing to be evaluated but not necessarily having emergency needs.

For example, many of them have difficulty with transportation.  Some of them are homeless, and their easiest access are to facilities where they can get emergency transport [because] they can’t get to a primary care office.

When we look at diagnoses across all emergency room use, the most common would be what we characterize as “non-specific” – things like abdominal pain and headache and non-specific chest pain. This really reinforced the concern that access to resources, period, is an issue.

R3.0: Describe Kentucky’s initiative to tackle this problem.  

Langefeld: First, we developed criteria to identify who we would characterize as super utilizers. One of the criteria was what I mentioned – 10 or more visits in a 12-month period of time. The second criteria was the methodology developed by Dr. Jeff Brenner and the Camden Coalition – what they call “hotspotting” or “geomapping.”

Our primary resource for initial analysis included Medicaid claims data, which included medical claims and pharmacy claims, behavioral claims, [and information about] who the providers are.  As we developed the program, we developed a strategy to include our acute care hospital partners, including our university hospitals – University of Louisville, University of Kentucky and University of Pikeville – but also several hospitals that are dispersed geographically around the state.

From a technology standpoint, one of the fundamental resources that Kentucky has that is a differentiator from other states is our information exchange. Our Kentucky Health Information Exchange is our technological connector – it has the ability to flow data and information between multiple entities, externally and internally.

High-utilizing ER Medicaid recipients is not a new thing. What is new is the technical capability to identify and understand these populations from an analytics viewpoint.

We can also use the data technology infrastructure to transfer information to a care coordination team at the local level.

For example, if someone shows up at a facility, an alert can be immediately sent to all the members of a care coordination team at the local level. The technological infrastructure can deliver secure messaging to that care coordination team as well. It creates the connectivity and the resources that can be focused in a concerted way.

R3.0: Your work has broader community implications. This isn’t just about the super-utilizers per se; the super-utilizers are in fact symptomatic of broader needs within the community.

Langefeld: Absolutely. The discussion around ER use and high ER utilization is actually the same discussion we have about the needs of any residents in any community. You’ve heard terms such as “patient-centered medical home” and “accountable care” – all of those discussions really revolve around the central issue of how we move from a very fragmented, dysfunctional system to one that is much more connected and truly patient-centric.

How do we support that from a patient standpoint, from an analytic standpoint, from a connected community of caregivers standpoint?  How do we pool the resources that have traditionally not even been part of that discussion – like the local health department, the department of transportation, advocacy groups at the local level, housing for the homeless – and also address all of the other social determinants that have impact?

For example, one of the traditionally underutilized resources are local health departments. We have 120 counties in our state, and every county has a local health department. We asked health departments to be key convenors of discussions to identify all the resources that could be available to local residents and Medicaid recipients who were using the emergency room at high rates.

R3.0: What targets have you set? How will you know you’re succeeding?

Langefeld: The initial metrics will be focused on more traditional measures of utilization and cost, but as we drill into that, we’ll want to understand how patients have transitioned from accessing acute resources to primary resources, and how their utilization has been redistributed. If they need access to services, are they getting it at the primary care office versus at the emergency room?

R3.0: If another state wanted to do this, how much would it cost?

Langefeld: A lot of this is redirecting existing resources and personnel from what they’ve been doing traditionally.

We have five managed care plans that help supply services to 85 percent of our Medicaid population. We have to work with those managed care plans and [develop] economic models that may be different from just paying straight fee-for-service.

How do you pay for community care workers? How do you pay for support services? What about housing and transportation?

The whole idea is to take a system that is fragmented and dysfunctional and restructure it in a way that supports [patients] in a holistic way and addresses issues that have traditionally not been addressed through our medical system. I think that’s a challenge for us not just in Kentucky but nationally.