The Learning Healthcare System: from concept to reality

Join Jeff Williamson, MD, MHS, Wake Forest University School of Medicine, and Leah Hanson, PhD, HealthPartners Institute, as they discuss learning healthcare systems (LHS) and their potential value. They define LHS, their importance and also the differences in the cycle of research in the LHS vs the traditional academic research study.

To view a transcript click here then select the transcript tab.

Transcript

Jeff Williamson, MD: Welcome to this edition of Integrating Research into the Learning Healthcare System. We're going to focus specifically on the very interesting problem of integrating concepts and bringing them to reality. 

I'm Jeff Williamson. I'm the head of Geriatric Medicine and Gerontology and Director for the Center for Healthcare Innovation at Wake Forest University School of Medicine, part of the Advocate Healthcare System.

And I'm so pleased today to have our guest, Dr. Leah Hansen. Leah, I'm so pleased to have you here talking with us today. We've known each other for a few years now, and I learn things from you every time I talk to you, but I thought maybe first of all, it would be helpful if you just start off, tell us a little bit about your work, your role at health partners and, um, kind of what brought you to this point in your career.

Leah Hanson, PhD: Sure, thank you, Jeff. I'm a Senior Director and a Senior Investigator at Health Partners Institute. Health Partners is an integrated care delivery system located in the Midwest. [01:00] I'm located in the Twin Cities of Minnesota. My training is as a neuroscientist. I focus on neurologic conditions that occur with aging with a large focus on people living with dementia and caregiving.

Jeff Williamson, MD: So Leah, thank you for that introduction. 

I have a few questions for you. The first one, just basically, describe to us, what is a learning healthcare system and why are learning healthcare systems so increasingly important in the healthcare landscape? 

Leah Hanson, PhD: So learning healthcare systems systematically gather and apply best evidence into practice, and they're committed to innovation. They have a culture of continuous learning and of improvement and in learning healthcare systems, research influences practice and practice influences research. Learning healthcare systems are really important because healthcare systems must continue to adapt to provide the best care at the lowest cost.[02:00] 

Now, this is especially important in the care of people living with multiple chronic conditions who have both suboptimal health outcomes and are higher utilizers of health care. 

Jeff Williamson, MD: I know that many people perhaps listening to this, I know I myself, are familiar with the traditional cycle of research funding.

As a researcher I might have an idea for investigating an important hypothesis, and hopefully that idea matches up with extramural funding like the NIH. And so I apply there and get that, do the research, publish my conclusions in a journal, and move on to the next question. 

But this is a little bit different model, so if you could just talk to me about how the cycle of research in a learning health care system is different from the traditional cycle that so many of us have been familiar with through the years.

Leah Hanson, PhD: Yeah, you're definitely right. 

The, the learning healthcare system research cycle is much more rapid and iterative than the, an academic research study. [03:00] It can be boiled down to kind of six steps. A scan, design, implement, then evaluate, adjust, and disseminate. 

So would you like me to go into a little more detail about the steps?

Jeff Williamson, MD: Sounds perfect. 

Leah Hanson, PhD: So, the first step is a scan to identify and characterize a problem. Now, this is really important. You need to learn what's important to both patients, leaders, and clinicians about the problem and then search for evidence based solutions.

Second, is to design a change in practice. Again, make sure to involve all the key stakeholders, and you're likely going to need to adapt that solution to your environment in your own healthcare setting. And you need to explore what data will be available. It's really important for the evaluation step to know ahead of time what data you have available and create a plan to measure success.

The third step, implement. It's really important to start [04:00] small. So many people start with a pilot before a widespread implementation. You can learn a lot about the barriers and facilitators to implementing your change in practice. In addition, you can build partnerships with clinicians and leaders that will really help you when you go to a full scale study. 

The fourth step is evaluate. Again, going back to your evaluation plan, make sure that you have a comparison or control group versus just a pre/ post design. Oftentimes, you could implement a change in practice in one clinic and then have another clinic that keeps the same standard of care as a comparison. Alternatively, you could use like a historic control or data from previous years or months in the clinic. 

Fifth is the adjust step. So use the evidence collected to make refinements. And this is a step that typically doesn't always happen in these academic research studies. [05:00] It's very similar to the, the method in quality improvement where you plan, do, check, adjust. 

And of course, the last step would be similar to the academic system, which is disseminate. So sharing the results with the stakeholders, going back to the patients, the clinicians, the health care system leaders and operation leaders, doing it in a timely manner, and have discussion. I think this continuing discussion back and forth is just really critical for the culture of continuous learning and improvement. 

Jeff Williamson, MD: One thing I wanted to maybe get you to emphasize or highlight a little bit more is the institutional role and the resources that an institution would bring to an endeavor like this in a learning health care system. Again in traditional biomedical research, institutions do provide resources, laboratory space or startup funds, or things like that for traditional researchers. [06:00] But it seems to be a little bit different, the kinds of resources that an institution might need to bring forth for an effective integration of research into health care. And I'm just wondering if you have thoughts about that. It seems like the role might even be a little bit larger than it is for. Traditional biomedical research. 

Leah Hanson, PhD: Yeah, big picture resources are definitely needed to create this environment. Resources are needed, you know, first for just creating that culture of learning, like having opportunities for staff to come together in communities of shared interest to discuss what are the problems they're encountering and how might they solve them. 

You also need staff and the ability to both adapt and then implement these evidence based solutions into practice. There needs to be access to data and ability to evaluate that data in order to measure success. And really engagement all along the way of clinicians, patients, [07:00] operations leaders, health system leaders is very important, and it does require intentional work on part of the organization. 

Jeff Williamson, MD: How would an investigator whose interest is in this area of research begin to build that sort of coalition in a healthcare system? Do you have thoughts about that? Have you seen successful models of that or unsuccessful ones that we can learn from? You probably have but I'm just curious to that. I think our listeners would be very interested in that too. 

Leah Hanson, PhD: Yeah, so my perspective is from being an investigator that's embedded already in a health care system. So I, I have an advantage of having those connections already. I know if someone is coming in, we often have people approach us from outside institutions who want to partner on projects and, you know, we have the same goal of improving care, but they're really coming in with a different cultural mindset. 

And [08:00] they can come in and they can tell me, I have the solution to your problem, instead of asking me, what are your problems? What have you tested in order to fix those problems? And really, I think that's the wrong approach of, if you're trying to partner with the healthcare system, is going in and trying to understand and build that relationship.

Jeff Williamson, MD: Sure, I want to come back to this in a minute. Because I think this is a key place where people often come off the rails, I know in my own experience, just trying to transition from traditional research into actually integrating the healthcare system. This is where I've personally made some mistakes and learned from those.

But before we come back to that, just talk to me a little bit about how do you measure success in these kinds of endeavors? And it's a little bit different, I would suspect, than success outcomes in a traditional, you know, biomedical research endeavor.

So what type of outcomes are important in An academic learning health system [09:00] cycle.

Leah Hanson, PhD: Yeah, of course, it's going to be different depending on each project, but I think in all cases, it's really important to focus on identifying outcomes that are important to all the stakeholders. So, um, instead of thinking about you know, a measurement on a scale, really focus on both patient experience, clinician experience, population, health on also health care costs. So thinking of all these perspectives together to measure sort of a value of- value care. 

One of the outcomes in a project I've worked on, we're looking to increase the diagnosis of cognitive impairment through our primary care clinics and oftentimes in an academic research project, you might say, let's look for that diagnosis code. And that shows someone has been diagnosed, but in a learning healthcare system, we focus on process. So you might [10:00] look for a referral to a clinician or a test that was ordered. So focused, um, on a process outcome more than sort of the final end of the road outcome. 

Jeff Williamson, MD: I'm gathering from you, but also from my own experience, that this is not only important work, but it's huge work.

Those, those of us who'd like team science, where we're- this is probably ultimate team sport, I would suppose. And so, you know, I think we've all seen scientists to through the sheer force of their own will get their own project done. And it probably goes no further than them. But this seems to be a little bit different.

So, talk to us a little bit about you know, sometimes when I think about this work, I think of the work of community organizers. You know, so, but in healthcare, you're organizing a community as well. So, talk to us about some strategies. I, I, I see young people who really want to get into implementation science or other aspects and they, they're still trying to do it using the [11:00] old model. And so, not thinking about, okay, what strategies do I, do I need to build to effectively bring together a community and a coalition, so to speak. 

Can you talk about that? How you've seen that happen, Leah? 

Leah Hanson, PhD: Yeah, sure. I have probably three general suggestions about how to make successful partnerships. The first would be start building your relationships really early in the process before you have a solution and you're ready to start something.

Even better, you know, find opportunities to make connections within the health system before you have a quote unquote "ask" so that you have sort of a lot of probes out there and partners that you can connect with. 

And oftentimes as an academic researcher, that might feel like not valuable use of time to spending that investment of engagement, but it's really key and it really pays off when working together.

Jeff Williamson, MD: Street cred. That's what I hear you say.

Leah Hanson, PhD: Develop the street [12:00] cred and show how you can add value. I can't count the number of times I've offered to give a talk about dementia to a community center or a long term care center, sort of to get my foot in the door and gain that, you know, "I care about you, you care about me. We're working on this together."

Second, when you start. In the initial stages of working together, it's really important to set expectations that you want to work collaboratively. You know, talk about having that bi directional communication and really intentional, we call it co- design of all parts of the process. 

So it's understanding current workflows. You can talk about the evidence based solution that you might have that might not fit into the workflow and then figure out how to adjust that so that it's adding the least amount of work possible. Something that has a very high burden is not going to be successful. You won't be able to implement that into practice.

The third thing I would emphasize is having [13:00] that focus on shared priorities. So we've talked, you and I, Jeff, about what does a win look like? And it's the win-win or the win-win-win. So it needs to be a important and a priority for the researcher, the healthcare system and clinic, and then also for patients and families.

Jeff Williamson, MD: And these skills are so important. As Leah, I think you know, I've had five teenagers come through my house, and I learned, maybe on the last one finally, that I needed to first understand before I sought to be understood. And so I think that's a rule that helps us in this endeavor, too, is me going with my agenda, usually the discussion doesn't last very long. But if I listen to where their pain points are, and then try to modify my agenda for health system change to their pain points and figure out a way they intersect, then I really start to get somewhere. 

Is that a reasonable approach, just spending some of that time just listening? I'm not pontificating, so to speak, but listening as to what, where the health system [14:00] decision makers are really concerned.

You think that's a good start? 

Leah Hanson, PhD: That is definitely a good start. And I love that analogy. I think, you know, seek first to understand is, is so important. 

Jeff Williamson, MD: Now, it also comes down to sometimes getting money in the healthcare system or, or maybe not money, but in kind support. Say, you know, we had a pragmatic trial to change our healthcare system where they, healthcare leaders said, "Okay, I'll tell you what, I'll make up the difference for what your nurse practitioners bill as we're exploring this care model. Um, they can bill, but there may be some loss. I'll make that up." 

So how do you even raise that possibility with people? How do you start assembling beyond the coalition to actually get some legs behind the support?

Leah Hanson, PhD: Yeah, there's different models depending on the organization. Quite a few learning health care systems have quality improvement departments that the leaders of the health care system come together and each quarter [15:00] or each year identify priorities for doing projects. And they have a team of people that can be dedicated to help bring together a community of people with shared interests, explore the problem, find a solution and actually help to implement it. And in that way, there's some internal funding for priorities of healthcare systems. 

Thinking about a pragmatic trial, like you mentioned, it's more difficult because you're relying on getting grant funding and you need to have that partner kind of engaged as you're putting in the grant. Oftentimes, you know, you've submitted grants, you need to wait six or nine months before you can even know if you'll be funded and start that work. 

Jeff Williamson, MD: Sure. You know, I think that advice about, and I hadn't thought about that: become integrated into the discussions of your quality improvement team, will help you begin to leverage ideas that you have that might mesh with their ideas. That's really, really helpful. 

One thing I've found, almost [16:00] all health systems now have a population health team as well. They're also working with quality improvement many times. And often the problems they're thinking of do intersect with the problems we're thinking of, especially in this case, aging and things like that. I mean, that's, that's a high, a group of high health system utilizers. 

So patient continuance, that's a word my wife the lawyer, taught me. Lawyers have a lot of patient continuance, she said. So being patient and just continuing the discussion is sometimes a success. 

Well, I see we've only got a couple more minutes, but I don't know if there are thoughts that have entered your mind as we've been talking. Anything that you feel that we need to cover that I haven't asked you about? 

Leah Hanson, PhD: No, I think we got through the main points I wanted to, to share, but I do look forward to talking with you again about your case study in multiple chronic conditions when we do this again. 

Jeff Williamson, MD: Thank you so much. I really, I look forward to that too. And it's a big, big area. There's so many opportunities we have [17:00] in older people. If we just pick a doable question and then enter it into the cycle of a healthcare system, we can make. A lot of progress, so.

Leah, thank you again. Really appreciate this opportunity to talk about this aspect of integrating our research into learning healthcare systems.