About This Episode

People with kidney disease receiving dialysis have poor health-related quality of life. By considering biomedical and patient-reported outcomes when conducting research and measuring care quality, we can enrich our understanding of patient experiences and identify opportunities to improve patient health status by aligning treatment goals with issues important to patients. In this episode of GMO Dialogues, Dr. Jennifer Flythe, Associate Professor of Medicine at UNC School of Medicine, discusses Patient Related Outcome Measures and how they can improve the quality of patient care.

Featured Guest: Dr. Jennifer Flythe, MD, MPH 

Dr. Flythe is a nephrologist and clinical investigator at the University of North Carolina (UNC) Kidney Center, Associate Professor and Vice Chief of Nephrology and Hypertension at the UNC School of Medicine, and Director of Dialysis Services at UNC Hospitals in Chapel Hill, NC. She completed medical school at the UNC School of Medicine and Nephrology fellowship training at the Brigham and Women's Hospital- Massachusetts General Hospital Combined Program. She earned her Master's in Public Health from the Harvard School of Public Health. Dr. Flythe conducts patient-oriented qualitative, epidemiologic, and prospective research aimed at improving outcomes and experiences among individuals with kidney disease. 


Listen to This Episode


Episode Transcript:

Dr. Maddux Voice Over: People with kidney disease receiving dialysis can have poor health-related quality of life. By considering biomedical and patient reported outcomes when conducting research in measuring care quality, we can enrich our understanding of patient experiences, and identify opportunities to improve patient health status by aligning treatment goals with issues important to patients. Dr. Jennifer Flythe, Associate Professor of Medicine at the University of North Carolina School of Medicine, joins us today to discuss patient related outcome measures and how they can improve the quality of patient care. Welcome, Jenny.

Dr Jenny Flythe: Thanks for having me, Frank, happy to be here.

Dr. Maddux: Describe to me the evolution of patient reported outcome measures and qualitative sort of information that we're now studying in clinical trials.

Dr Jenny Flythe: Over the years, particularly over the last decade, there's been increased appreciation of how using patient reported outcomes can enrich our understanding of the patient experience. A patient reported outcome is a report of the status of a patient's health that comes directly from the patient, without any interpretation by a clinician. We tend to think of things like health-related quality of life, symptoms, functional status, and we measure patient reported outcomes with what we call PROMs or patient reported outcome measures, which are the questionnaires that help us capture this information. In dialysis, we know those is the kidney disease, quality of life, KTQOL 36, ICH caps. There's also PROMs that measure depression and pain.

Dr. Maddux: In many of these cases, these tools and surveys are validated in ways that allow us to make sure that they're as unbiased as they can be and useful. Many times, that doesn't always allow for sort of free text or just conversational information from patients. How does that get integrated?

Dr Jenny Flythe: It is really important. These measures be reliable, and they be valid, because we want to make sure that the information that we're collecting is actually reflecting the construct, or the domain, or the concept that we think that it is. However, to your point, patients also have their own thoughts, they have their own experiences that are not always very well captured with a Likert Scale or a "Yes/No" response, which is how many of these items are constructed. And so as we advance our use of patient reported outcome measures, there are different approaches to collecting them that enable free response. If you're doing it on a computer, for example, sometimes you can do voice activated typing, or if it's a written survey, you can do write-in responses. But ultimately, you're going to be somewhat limited in the type of data that a survey-based patient report outcome measure collects.

Dr. Maddux: Many of the survey-based measures that we've learned of over the years have come in written format, some of them are quite long. Are the techniques for actually collecting this data changing as they're sort of electronic world is expanded?

Dr Jenny Flythe: I think that we're really in our rudimentary stages in nephrology. It varies clinic to clinic and practice to practice, but most patient reported outcome measures are still collected on paper. To your point, there's a lot of advantages to doing electronic based data collection. For one, it allows you to tailor the questions to the patient, so you can do what's called Computerized Adaptive Testing, such that if a symptom, for example, is not relevant to the patient, they don't need to proceed and answer five more questions about that symptom that's not relevant to them. That's the case that you have with static paper based surveys. Other opportunities that come with electronic based PROMs is it allows integration with medical record. We can integrate it with biomedical data, with laboratory results, with hemodynamics, and start to put together a very full picture of the patient.

Dr. Maddux: E-PROMs offer this opportunity to do... what I don't know what the right term is, but let's call it asynchronous assessments of patients so that you could instead of have a series of a dozen questions right now there to answer, they may get a dozen questions scattered over several weeks that come through some electronic tool that they are using or that they would use on a regular basis. Is there any advantage or disadvantage and sort of study design and the interpretation of that?

Dr Jenny Flythe: I think the advantage is that you get much more of real time assessment of what's happening with the patients as most of the patient reported outcome measures that we currently collect in clinical care, we tend to administer annually or with ICH-caps twice a year. You're capturing these very small windows. We know particularly with dialysis patients are experiencing things all the time, and they change on a week-to-week basis.

Dr. Maddux: Tell me about whether in the world of patient reported outcome measures, the opportunity to use more discrete choice surveys, and that kind of technique is being used or has the opportunity to be used.

Dr Jenny Flythe: It depends on what kind of information you're trying to elicit from the patient. Designs, like discrete choice experiments, are very useful when you're collecting patient preferences. The idea of presenting treatment A and treatment B, which would you prefer? And why? Or commonly with discrete choice, comparing risk. How much risk are you willing to accept for a given benefit of something?  I think those can be very informative at a population level, also very useful in the regulatory environment in terms of thinking about approval of devices from the patient perspective. For shared decision making, they may be useful as well.

Dr. Maddux: When you're preparing a clinical trial, and you're designing the study and so forth, to what degree do you proof-of-concept the tools that you're using? Not so much the technology of the tools, but the kinds of questions. Would you typically have focus groups or other patient groups that get involved in that study design?

Dr Jenny Flythe: If you're looking to use a patient reported outcome measures an endpoint and a clinical trial, you need to make sure that you're using a validated and relevant and again, a reliable patient reported outcome measure. The way that you establish that validity, and particularly, I think one of the things that's most important to consider is the content validity.  What that means is, are the concepts that are being measured by this instrument relevant to my patient population? Among dialysis patients, is this measuring what we think that it's measuring? You can establish content validity through interviews with patients, focus groups with patients, but it's an absolutely critical step that you return to your target population.

Dr. Maddux: I've always wondered, in all of these surveys that I used clinically and participating in a number of trials, where are we missing potential areas of bias because of the way we ask questions, the language. The form of the question may be different than the cultural world that a given patient population may live in? How do you address those sorts of potential bias areas that are really tied into disparities we see in care in general?

Dr. Jenny Flythe: Ideally, your measures should be culturally validated. They should be validated from a language perspective. Then also considerations of cultural differences.  I think that that's an incredibly important component and something that's often skipped because it is really labor intensive to go back and validate, and confirm these things about measures. Some critics would say, well, if it's been validated in the cancer population, it must be true in the dialysis population. I think there is some truth in that statement sometimes. But again, it's really important to include your target population.

Dr. Maddux: It's like a sort of a 1 to 10 scale of maturity of using these kinds of tools and in kidney disease care, and nephrology, where are we today?

Dr. Jenny Flythe: We're pretty low on the totem pole. We do collect them, and we're mandated to collect them by different payment programs. So, that progresses this down the spectrum a little bit. But I think we're pretty early.

Dr. Maddux: I wonder if there are disciplines, we need to bring closer into the field, anthropology in some of these other areas. What are your thoughts on what would be useful as you've been a PI and RPI on some major studies related to these things? What kind of people do we not always put in our community that might be useful in understanding how to do this?

Dr. Jenny Flythe: We don't even need to go that far. I don't not sure that we even need to leave medicine in some respects. You know, there are other disciplines within internal medicine that have advanced the field of PROMs. I think oncology is definitely a leader in that regard, in terms of incorporating it, not only into routine clinical care, but also into the assessment of regulatory safety around drugs, chemotherapeutic agents,

Dr. Maddux: How quickly are we going to get to where all of these tools that we use are not on paper? Tell me about the maturity and development and the political push to move towards electronic forms.

Dr. Jenny Flythe: We're gonna get there pretty quickly. I don't know if it will be mandated or not, but I think the benefits of doing it electronically, there's also a fair amount of reticence in the community around whether all patients would be able to complete them electronically. Our group did some work developing an electronic patient reported outcome measure, and what we found in a very diverse set of people treated with haemodialysis, they were able to complete it. If they weren't, a small amount of assistance usually enabled them to do it. But it takes time. We actually designed an instrument that has these large radio buttons that someone just has to get their finger in the vicinity of. You can't hand someone something that's in 12 point font and ask them to click a box with a mouse. We have to be creative. Those things can be done.

Dr. Maddux: Following that work that you did that was very disciplined in the way you've approached it. What advocacy is needed – our federal government, CMS, the FDA, whoever it may be – that is looking to incorporate these things. How much do they understand the state of the art really would be electronic tools for this? And not a 60-question paper thing sent anonymously to your home?

Dr. Jenny Flythe: They're starting to understand it. One of the challenges is there's not a lot of funding for actual measure development. People are interested in trials of patient reported outcome measures. They want you to use patient reported outcomes for endpoints and in clinical trials. But they will often say: we don't want to fund the development of an instrument. Ultimately, there's got to be investment made. Some of that, and again, I think it's important to know that we don't need to start from ground zero. There's been tremendous work done in other disciplines, the PROMIS measures which are available through the NIH, it is truly just validating them in our population and using them. It's not that hard.

Dr. Maddux: One of the things that I've watched over the years is the National Quality Forum, look at the science of measured development and not being a measured development expert myself. I'm wondering, in patient reported outcomes and less quantitative measures, are there different ways to assess reliability and validity and these characteristic things within the science of measure development, is it different?

Dr. Jenny Flythe: There are certainly different ways to do it and kind of the field of study that you're referring to are psychometrics. There are psychometricians that do this for a living. In fact, some of the patient reported outcome measures that are used in kidney disease, in recent years, people have gone back and done some of this validity and reliability testing, and they've actually had pretty good results. Even KDQOL, there's been psychometric testing around that published in the last few years.

Dr. Maddux: I've always thought the process of asking a patient what they think, is actually a therapeutic intervention. Regardless of what the answer is, just the fact that you're willing to listen to them. What are your thoughts about the benefits of incorporating this routinely? Whenever we interact with patients for study reasons, or non-study reasons?

Dr. Jenny Flythe: That's an excellent point. There are really two pathways that I would say that routinely collecting patient reported outcome measures, and not just collecting them, it's important to follow up. It's not just about the collection, but kind of two pathways that I think that they can improve outcomes. One is the most obvious is it brings to light issues that clinicians may have underappreciated, allows them to be treated, and then allows them to be reassessed to see if there's response to whatever therapeutic intervention there was. But the second is really more about the process of collecting that information. It's about the communication that's created between the clinician and the patient. They really can support shared decision making in a way that, phosphorus value just does not have the ability to do.

Dr. Maddux: What's your assessment of our peers in the nephrology community being ready to embrace a responsibility for things that go beyond the core pathophysiology we are all taught to address? Because they elicit issues that are important to the patients that have nothing to do with the traditional medicine.

Dr. Jenny Flythe: Even if you're willing to take an expanded view and say, as a nephrologist, "I'm responsible for my whole patient." One of the challenges people face is that we don't always know what to do with the responses. Even if we considered it in our domain, we don't always know what therapy, and then you're collecting information from a patient and you're having to tell them that you don't know what's the best treatment. It gets at that issue of: we've got to collect these in order to be able to do the research and do the science to figure out how we can really give targeted interventions.

Dr. Maddux: I've been a proponent of, much like oncology, all patients should be offered access to clinical research. Given that your life is around clinical research, what's your perspective? How do we get nephrology and nephrologists to make that part of the daily life of our interaction with patients?

Dr. Jenny Flythe: A lot of it is bringing the research to the patient. Nothing more than dialysis is better positioned to do that. Because our patients, unfortunately, come to us as much as they do, having opportunities for research available is really important. But in doing so, we also have to educate patients about what it means to participate in research, and then make sure that we feed the results back to them so they understand the time and effort that they gave, something was actually done with that information. That's something we've fallen down on often.

Dr. Maddux: Before we finish, I know recently, you and Dr. Laura Denver have been involved with a study that you're looking to get forward. You want to give us a little briefing on that particular study. I think it's quite interesting.

Dr. Jenny Flythe: We started working on the foundation for close to seven years ago. This is an example of developing a patient reported outcome measure, we started the review of the literature, we were interested in capturing patient's symptoms. We went on with input from people from Fresenius other dialysis organizations. After interviewing well over 100 patients, we ultimately developed an electronic patient reported outcome measure system that captures dialysis related symptoms. The system is composed of a 12 item, simple symptom PROM that's done on a tablet computer. It also includes some supports for clinicians. Because one thing that we haven't talked as much about is we can collect patient reported outcome measures, that actually it's not that difficult, but we have to follow up on the responses. I would argue that not following up on the responses can mean that collecting patient reported outcomes can do more harm than good. This system includes supports to help physicians with follow up, and recently, we've been funded by PCORI to do a pragmatic randomized trial which we're doing in partnership with Fresenius where we'll randomize 30, clinics 15 to usual care, and 15 to smart HD. We'll compare the effects on symptom burden, hospitalizations, other outcomes.  About 50% of the study is looking at the implementation of the electronic patient reported outcome measure system, and hopefully, we'll yield learnings that are applicable to other PROMs outside of symptoms. But we're just excited to have the opportunity to partner.

Dr. Maddux: What's the timing of that study? When do you expect it to get started? And how long do you think it will take to begin to get some results?

Dr. Jenny Flythe: That's an excellent question.  We've just been notified in the last month about the funding. We are aiming for a late fall startup. It's a five-year award. So, we anticipate that we will have results in close to 2028.

Dr. Maddux: Congratulations for that. Later this spring, I think we're going to see one of the first translational activities with a medicine for managing a symptom. And that's pruritis. It doesn't get at the root cause. It's just managing a symptom, so it seems to me really topical right now that this is becoming a big part of what our world is going to look like in kidney care. I've been here today with Dr. Jenny Flythe from the University of North Carolina. We've been talking about patient reported outcomes and think that these elements of trying to understand the patient's perspective on living a life with advanced kidney disease is going to be very important for all of us to expand and mature our understanding of this.  Jenny, thanks so much for being here.

Dr. Jenny Flythe: Thank you for having me.