• Blog
  • Aug 23 2018

Mark Williams: Improving Hospital Discharge and Targeting Social Needs for Better Health

Dr. Mark V.  Williams doesn’t want people back in the hospital. The former emergency medicine physician and current chief of the Division of Hospital Medicine at UK HealthCare says his goal is for people to understand the medications they need to take and what do when they leave the hospital, so they don’t end up coming back. 

One of his research projects, funded by the Patient Centered Outcomes Research Institute called ACHIEVE, is evaluating nationwide efforts to improve the hospital discharge process and reduce hospital readmissions. The first step was asking patients and caregivers question like these: Did you follow up with your primary care provider? Did you end up having to go back to the emergency department? What matters most to you when you’re being discharged from the hospital?

Williams says, “Right now we’re wrapping up surveying over 9,000 patients as they’ve gone through the hospital discharge process and nearly 3,000 of their caregivers, and we’ll then link that survey data to Medicare data on outcomes.” As part of the project, his research team undertook a massive data analysis effort — five million records from 380 hospitals across the U.S. By linking what hospitals describe as their care transition improvement efforts to the medical records of what happened after patients left the hospital, the researchers will see what works and where hospitals can improve processes. 

Williams wants to impact care nationally, but closer to home in Eastern Kentucky he wants to empower clinics and their patients by developing a database of local resources to address social needs — things like access to transportation, housing and food. Research has shown these factors play a critical role in health. “My personal goal was to be able to set up a platform so we could interact with these clinics, begin to touch the patients in these communities that we see too often when they get extraordinarily ill and have to be transported either by ambulance or helicopter to UK,” Williams says. 

That platform, a website called CARE-KY (www.care-ky.org), is a crowd-sourced database and currently includes more than 1,000 community resources in 27 Kentucky counties. “The worst health outcomes, unfortunately, in the United States are out in Eastern Kentucky, and I'm hopeful we'll be able to tap into all these resources and focus them on undertaking something that tangibly improves health in these communities.”

Williams is director of the UK Center for Health Services Research. “The center started in January 2014 with two people, but now we have 50 and have received more than $20 million dollars in awards, and there's hopefully more to come.”

Last May, he was one of 14 UK faculty selected as a 2018-2019 University Research Professor. Nominated by the College of Medicine, Williams was recognized for his outstanding research achievements. He says his reaction was shock and gratitude. “It's certainly an honor, but it really is recognition of what our team has been able to accomplish. The collaborations have come readily and easily. I think it reflects the University of Kentucky, the willingness of people to work with you. I'd say at other academic medical centers it's more of a competition. You can get so much more accomplished when it's a collaboration instead of a competition.”

VO: Today we meet Dr. Mark Williams, Chief of the Division of Hospital Medicine at UK HealthCare. Williams also serves as director of the Center for Health Services Research and he focuses on care transitions and the role of health literacy in the delivery of health care. 

Mark Williams: Too many decades ago I… my first… major position was actually at Grady Memorial Hospital in Atlanta, Georgia and I was serving as the director of the Medical Emergency Clinic, and I'd come from Mass. General where my very first job was as an attending in the emergency department there. And the experience at Mass. General and the patient's we took care of was pretty different than in inner-city Atlanta at Grady Memorial Hospital. Why I say that is they would come to the emergency department in the early morning with their brown bag lunches in hand, knowing that they're going to have to wait all day because it was so busy and so crowded. Whereas up north, the patients coming and would be tapping their watches after they'd waited 15 to 30 minutes. And it also became apparent that the patients really did not understand a lot of what we the nurses and doctors and so forth were saying to them. I see them back in the emergency department maybe just two, three, days later and the reason was as I began to explore it was because they didn’t understand the instructions, they've not been able to get their prescriptions filled. All these barriers that they were facing out in their environment that individuals with health insurance who have homes and families and support systems in place, don't even think about. And so, as we begin to explore this… you know first I had that realization. The second realization was it was my fault and I needed to figure out, you know, how could I do a better job of helping patients understand what I was saying. And in talking with colleagues, we came up and this concept of literacy skills which everybody knows about, but then as we looked into the literature, there were like two articles on health literacy. We then applied to the Robert Wood Johnson Foundation for the literacy in health care (a proposal) and remarkably they funded it. This was back in around 91 - 92 and that's what got me started in research. We surveyed patients in the emergency departments at Grady and in our urgent care center and elsewhere and found out a high proportion of them had inadequate health literacy skills. The patients with low health literacy had lower quality of life assessments. And so, this ended up getting published as a lead article in JAMA and that at that point I became increasingly interested in just research and more especially around how our health system was set up to deliver care, and how we might do a better job of delivering the care to help the patients and their caregivers.

Alicia: So I know in today's conversation there's a lot of… how do we prevent people who come to the ER from coming back or readmission kind of focus, and is health literacy part of that? It sounds like it is.

Mark Williams: No question. When patients and their family caregivers do not understand what's going on, what they're supposed to be doing in follow-up, how urgent it is to get their medications filled, then you end up having situations where the patient's health deteriorates after they leave the hospital and they need to come back to the emergency department. So this is a critical component of this and I've been doing research related to that since 2004, is when I first began to realize the importance of health literacy and all of the information we attempt to impart to patients at the time of hospital discharge, and, frankly, do a poor job.

Alicia: So let's talk a little bit about the first hospitalist program that you were involved in, in 1998. Tell me first what does that term mean, because a lot of people have probably not heard “hospitalist,” and then why is such a program important?

Mark Williams: So hospital medicine evolved in the mid 90’s. There's a famous seminal article that was published in the New England Journal of Medicine authored by Bob Walker and Lee Goldman defining the term hospitalists. And as they described it, it was a physician who took care of patients exclusively in the hospital. The old model had been if one of their patients showed up in the emergency department and needed to be admitted to the hospital. The internist would try to get in to see them in the hospital and it was disruptive if you will to their day, because that admission might occur at 2:00am in the morning or it might occur on a Saturday or Sunday, or it might occur at 2:00pm in the afternoon when they had a whole waiting room crowded with patients. And so, they came up with this concept of having physicians based predominantly in the hospital taking care of patients and it wasn't novel. This is how it's been done in Europe for decades. And so from that, when that article was published, then Bob came to realize there were a lot of people already doing this in the U.S. and we think at that time there were probably about a thousand hospitalists in the U.S. I read that article and had an epiphany of this is what I really enjoyed doing the most and I began to shift my career then out of emergency medicine umm… presented a proposal to two Grady Memorial Hospital to set up a hospitalist program at Grady, and that ended up becoming the first hospitalist program at a public hospital in the United States. The whole… specialty, if you will, of hospital medicine has been the fastest-growing medical specialty in the history of American medicine. We've gone from that 1,000 to 2,000 or so hospitalists in the mid 90’s, there are now over 55,000 hospitalists in the United States. We outnumber emergency medicine, we outnumber cardiology. It's a large group of physicians. And they've been successful I think because one: there's now somebody always available in the hospital. Our program here and at the University of Kentucky Healthcare has over 80 clinicians in it, physicians who are hospitalists as well as advanced practice providers who we work with, taking care of patients. And we are here 7 days a week 24 hours a day, 365 days a year, able to respond to patient's if something happens. Two: we're able to communicate with the primary care providers and then primary care providers are no longer pulled out of their clinics to have to go see the patient in the hospital. A critical aspect though that must occur for this to be successful is good communication between the hospital-based physician and the primary care clinic-based physician or specialty-based physician.

Alicia: I imagine the challenge when people show up in the hospital, they don't bring their medications with them. So how do you deal with those kind of issues?

Mark Williams: So it's getting better. It used to be a lot harder. Before, you would maybe call their family caregiver and I myself asked somebody to go back to the house and go get the pill bottles. These days you can often call a pharmacy and they'll have this data on their database, but this is difficult. It's harder than people would think. I've been involved actually in medication reconciliation efforts is the term people used to reconcile medicines to make sure they're always accurate at the right time, right place, and when we first got involved in this we had a national meeting and Microsoft and Google actually came to the meeting. I remember talking to the Microsoft umm representative about the difficulties they were encountering because our medical record system is not connected and that was the biggest problem. You can get some medicines at one pharmacy, you can get other medicines at a different pharmacy, or maybe you end up getting some of the same medicines at the other pharmacy. But you know, one is Rite Aid, one is CVS, maybe a third is Walgreen, nobody's reconciling across those and the medical records - what we have here is different than is what the hospital up the street has, and that's the biggest issue. And that said, we're starting to get closer and closer to be able to have a unified electronic health record that I think will end up being owned by patients instead of by the institution, and that'll make things a lot easier.

Alicia: So tell me a little bit about how you ended up here in UK Healthcare and what your current positions entail.

Mark Williams: Right now, I serve as our director of our Center for Health Services Research as well as on directing our division of hospital medicine, and I became interested in this position basically because of the job description. When got an email in March of 2013 and looked at the job description, it was basically what I've been wanting to do ever since the late 90’s, and saw this is as a huge opportunity. I thought well let me just go talk to them and when I came up here was thrilled to see the level of collaboration on this campus, and this Center was set up to be collaborative. The mandate to me was to work with all of the health sciences colleges to foster collaboration and we use these three words: connect, catalyze, create. And our very first major project… project ACHIEVE, did just that. It involved people from the College of Public Health, the College of Communication and Information, the College of Arts and Sciences, and the College of Medicine. And those colleges, some weren’t even present where I was working before, where I previously worked up north there was no College of Pharmacy. There was no College of Nursing to complement the College of Medicine and that's why I saw this as a huge opportunity and was excited to come down here and was able to thankfully recruit Dr. Jing Li, who's our associate director for the Center. Now we have 50 people in our Center and have gotten over $20 million dollars in awards and there's hopefully more to come.

Alicia: So tell me a little bit about Project ACHIEVE. What is the focus of that project?

Mark Williams: So Project ACHIEVE is a fascinating initiative. The Patient Centered Outcomes Research Institute, which was established through the Affordable Care Act, aimed to start funding patient-centered research versus disease-centered research. And so, they were very interested in evaluating all these natural experiments that were going on across the United States looking at how can we improve the hospital discharge process or other health care transitions and reduce hospital readmissions. One of those I was involved in, something called Project BOOST (Better Outcomes by Optimizing Safe Transitions) that we had rolled out to about 200 hospitals across the U.S. And so, PCORI thought that if we could evaluate all these natural experiments that were going on, we might be able to find out what really worked, and also find out what was most important to patients. And we're right now wrapping up surveying over 10,000 patients as they've gone through the hospital discharge transition process and over 3,000 of their caregivers, and we’ll then link that survey data to Medicare data on outcomes. So these patients will have been discharged, we'll ask them how the experience was and then we'll find out umm did they follow up with their primary care provider? Did they end up having to go back to the emergency department? Did they end up having to be re-hospitalized? Did they potentially even have worse outcomes? And what was related to those outcomes? And then just equally important, the first thing we did is part of the project was to sit down with patients and their family caregivers and ask them: what matters most to you when you're being discharged the hospital? Simultaneously, we undertook a massive data analytic effort where we analyzed an eight terabyte file from CMS. It literally took us days to unpack it once they send it to us on a hard drive, it has over five million records. We've looked at results from 380 hospitals across the U.S., who we surveyed regarding what care transition initiatives they implemented and then comparing that to their outcomes.

One nice aspect of Project ACHIEVE is our ability to set up a platform within the Center for Health Services Research to begin to undertake other projects. As we've been successful with obtaining grant funding and getting very talented individuals in our research center in establishing terrific collaborations across UK HealthCare, we've been able to look at other opportunities that were available. And so, one of these that came up was CMS decided to look at how social factors impact care delivery, and can we identify patient’s social determinants of health? And if they have issues that are impacting them such as inadequate access to housing or food or they have personal safety issues, but if we can help patients get connected to resources, will they have better outcomes? We're the only project here in the Commonwealth of Kentucky, and I think more importantly, there are no other projects like us in the United States. Almost all of the other projects funded by CMS are in a single health system or city or urban area. Our project is looking at care delivery in 27 different counties in Eastern Kentucky. We're really focused on Appalachia and we have great collaborators in the Kentucky primary care association and the Appalachian Regional Health System. We have called all these counties and found out what are their resources to help people with transportation? What are their resources to help people with access to food, access to utilities, access to housing, help with personal safety, as well as family and social support? And we then collated those into a large database. We set up a website and also then turned it into an app. And so, these clinics out in Eastern Kentucky, which did not have access to 2-1-1, as an example, can now look online. Say this person has issues with food and housing, well, here are the resources they can access in their local communities and this is a crowd-sourced database that we're constantly improving. And in fact, I think in one week we had 29 different adjustments and improvements and enhancements to it from feedback from these communities, and we're sustaining that and our goal is to then eventually spread that beyond the 27 counties to all of Eastern Appalachian. We've already connected with the United Way to have this be the source for 2-1-1 expanding out into Eastern Kentucky. So this is a huge benefit to patients out in these areas that goes beyond just doing research.

Alicia: Absolutely, that’s the kind of community impact you hope for, but this is a very tangible example.

Mark Williams: Exactly, and in fact applying for this and getting this funding for the Kentucky Consortium for Accountable Health Communities, my personal goal was not to get a grant. My personal goal was to be able to set up a platform so we could interact with these clinics, begin to touch the patients in these communities that we see too often when they get extraordinarily ill and have to be transported either by ambulance or helicopter to UK, and if you will I wanted to have the opportunity to begin to touch them in their clinics upstream if you will, so that they never have to get transported here. You know, the worst outcomes, unfortunately, in the United States are out in Eastern Kentucky and I'm hopeful we'll be able to tap into all these resources here and focus them on undertaking something that tangibly improves outcomes in these communities. It's been interesting because we went out in a lot of these communities, the initial reaction is you know, “What do you want?” You know, “What are you trying to do to us?” And we spend a lot of time talking about, “I don't want anything, and I want to try and figure out how can we support you locally to deliver better health care and provide better access to resources.” And so then, taking that a little bit different approach instead of, “We've got this research project we want to collect these subjects and get the data, so we can publish papers” has really made a difference. It's been wonderful and gratifying to have these clinics excited about the Accountable Health Communities Project, wanting to get involved, and they're investing a lot of their own resources and time to see that this succeeds.

Alicia: So transitioning just a bit, this past May you were selected as 1 of 14 faculty, selected by your own college, for your outstanding research and scholarship. You were named a University Research Professor. So what was your reaction to that recognition?

Mark Williams: Shock… (laughter) and I think my main reaction is recognition we've got a terrific team. I think in all these situations it's not one person. And so gratitude, I mean it has been… the collaborations have come readily and easily. I think it reflects the University of Kentucky willingness of people to work with you. I'd say at other academic medical centers it's more of a competition and you can get so much more accomplished when it's a collaboration instead of a competition. And so, I think that and I'm getting old (laughter). It's certainly an honor, but it really is just I think recognition of what our team has been able to accomplish and I look forward to be able to use that recognition to do more here. As people point out we’re one of eight land-grant institutions that have all these colleges near each other. I mean we're working with the College of Business, the College of Engineering, we're now even starting to work at the College of Agriculture related to Accountable Health Communities because the College of Agriculture has these regional extension centers and they're out in these areas. And so you know, we're working with the College of Pharmacy, we're working with the College of Nursing, we're working with the College of Health Sciences, and we work with the College of Dentistry, umm and of course the College of Medicine. And so, I think that is what's really unique about this place and it's all here. It's within a two-block walk pretty much.