A conversation with Dr. Ethan Basch on the PRO revolution

A conversation with Dr. Ethan Basch on the PRO revolution

I had the great pleasure to chat with Dr. Ethan Basch yesterday about Patient Reported Outcomes (PROs) and how they are changing care and clinical trials. Dr. Basch is well-known for his JAMA studies recently showing that cancer patients live longer when they self report, and symptoms can be better managed. Good discussion about what’s really happening on patient experience and patient reported outcomes.


Leonard:  Hi there, this is Leonard Kish with Real World Evidence or RWE today and I'm here in our podcast with Dr. Ethan Basch, who is a Professor of Medicine Division of Hematology and Oncology at UNC. He also holds an appointment as Professor of Health Policy and Management at the School of Public Health there. In addition, he's Director of UNC Cancer Outcomes Research Program and associate editor at JAMA and other many numerous leadership positions within oncology and healthcare including the methodology committee PCORI, the Patient-Centered Outcomes Research Institute, and and advisor to Self Care Catalysts which is driving some changes in our Real World Evidence as collected. So welcome Dr. Basch and thank you for joining us.

Dr. Basch:  Nice to be here.

Leonard:  So you've made a name for yourself in patient-reported outcome measures. Those have changed quite a bit from the standard kind of surveys that have been done in the past or I should say maybe they are changing. So what are these outcome measures to you? How they've been used and how are they changing in the context of the 21st Century Cures Act?

Dr. Basch:  So a patient-reported outcome is essentially a piece of information about how a person feels for functions, for example, their symptoms or information about physical functioning, symptomatic adverse event, overall quality of life.  Patient-reported outcomes (PROs) are generally captured through questionnaires or items that are administered in various contexts. Classically, they have been given out to patients in questionnaires in drug development during clinical trials and in clinical care. Increasingly they're being used to try to figure out how people are feeling during their cancer treatment. There've been some substantial changes over time. I'd say over the past 10 to 15 years that has taken us from the days of long static quality of life questionnaires to the ability to pick and choose individual items that can be given in very short questionnaires fairly frequently and the other major change that we've seen is in the technology which of course we've seen across all of health care and these days it is much more common for these questionnaires or items to be administered electronically via a hand-held device app or web application or an automated telephone system or a text messaging system rather than by paper and pencil, which is the way that they used to be administered. I'd say the other change that's been remarkable is how patient-reported outcomes have come to the forefront of thinking in both drug development and clinical care. Ten or 15 years ago, it was really a small community of people working methodologically on these questionnaires who were most involved with PROs. But these days almost any regulator or drug developer or clinician can tell you what a case-reported outcome is and increasingly they are being integrated into the delivery of cancer care and they are a part of almost every drug registration clinical trial.

Leonard:  So what's taken so long for this to come to the forefront? We're just starting to see that outcomes are really being affected by self-reporting and you've been a leader, including a study in JAMA think about a month ago around lung cancer and web reporting.  So what have you seen.  How do you see the future then and how this is going to in terms of what works best?

Dr. Basch:  Well I think that one of the things that have changed that has made PROs more to the forefront is culture. I'd say 10 years ago or more there was a sense by clinicians that they understood how their patients were doing and that they could capture this information and that we weren't really missing anything by not asking patients directly about their experiences. But a substantial body of literature has emerged showing that in fact we miss about half of our patients symptoms and that by missing that information our clinical trial data are incomplete, and our ability to take care of patients is also incomplete because we often miss things that are important to our patients. So that's been one change. I think the other change is that advent of better technologies for capturing and communicating this information have improved and it's simply made it more feasible to collect and analyze this information.

Leonard:  What about the study in JAMA, what did you find there in terms of web-based monitoring and did you find that it was better than paper-based monitoring or it's a little bit more about some inclusions there?

Dr. Basch:  So first I'll just address your question about paper versus electronic. So electronic data collection is superior to paper-based simply because we can do real-time data collection. So if information is missing we can act on it right away.

 In terms of accuracy they're almost identical. There have been many studies now showing that whether a person completes a question electronically or on paper they essentially give the same answer. So it's really more about ease and processing data capture, data accuracy. So from those perspectives electronic data capture is vastly superior.

 I'll comment on the recent papers in JAMA. There have been two JAMA papers, one published within the last couple of months and the other a little more than a year ago. Both of these were randomized controlled trials in which patients with advanced cancers were assigned to either self-report their own symptoms using an electronic system with alerts sent to nurses when concerning symptoms were reported compared to standard care which does not include any patient reporting and it was found in both of these trials that by including patients self-reporting in clinical care that numerous clinical outcomes were improved; 1)Symptoms were detected sooner, 2) Symptom control was better, 3) Patient’s overall quality of life and physical functioning was better, 4) People were able to stay on chemotherapy on average about two months longer because their symptoms are being better controlled and so they didn't come into clinic with difficult to manage symptoms. They were caught early so people could stay on their treatments. People had fewer emergency room visits but most strikingly, overall survival was improved in both of these trials and one of the trials by more than five months and the other trial by more than seven months and the belief in these trials is that because symptoms are being caught earlier and being intervened on earlier people's physical condition remains superior. They remain more physically active. They don't get into trouble. They don't show up in the emergency room in extremis and this translates into better survival.

Leonard:  I think it was the control group was about a one-year survival and the web-based group was almost doubled up, is that right? So nearly doubling survival, is that right?

Dr. Basch:  The median overall survival benefit was about seven months in the recent trial and about five months in the older trial which are very substantial survival benefits in patient populations with advanced cancers. Just to give you a sense of how meaningful those benefits are those benefits are greater, the survival benefits are greater than the vast majority of drugs that have been approved by the FDA in the last five years.

Leonard:  So what is stopping that then from becoming the standard of care because as you say if this were a drug it would be immediately or pretty soon become a standard of care with those kinds of results? Is it the economics that’s stopping this from being implemented more widely?

Dr. Basch:  I think it's multifold. I think to your point if these were drugs with these survival benefits and there was a pharmaceutical company that was behind the drug and there was financial support from insurance companies then there would be a non-issue. They would be implemented now but these are health system changes that are not necessarily reimbursable. They require changes in workflow and potentially changes in the electronic health record and those are challenges. Now those challenges are definitely not insurmountable. They are surmounted all the time and there are lots of places that are doing this but this has not become the de facto standard and I think you're right. I think largely it's because there has not been a strong regulatory or financial driver for implementation.

Leonard:  There was a study you may have seen in therapeutic innovation regulatory science in 2017 that said a hundred thousand investment and engagement could turn return 50X or about 5 million dollars and accelerating clinical trials through reduced protocol amendments, faster recruitment, lower dropout rates, higher adherence protocols in general, is that consistent with what you've seen and what the studies you've done? Is this something that could really benefit clinical trials in a major way?

Dr. Basch:  So patient engagement writ large it involves a number of different techniques. Patient-reported outcome is certainly closely related to patient engagement. We engage patients to self-reporting when patients self-report they become more engaged. The data around patient engagement and clinical trials is pretty compelling. It is easier both to recruit and to retain patients when they are engaged and their compliance with clinical trial procedures is much higher when they're engaged and engagement of course has to do with communication and meeting people where they are and helping them to understand the reasons for the research and so on. So it's not just patient-reported outcomes but I'd say that these are all related to what the FDA calls patient-focused drug development. What the 21st Century Cures Act is referring to when it focuses on patient experience. So I think there's a broad recognition that there is value to research in care procedures and engaging patients as are essentially partners in getting things done in healthcare and in research but also there's a lot of value in collecting information directly from patients in order to understand their experience and I think these two areas are very closely related.

Leonard:  It sounds like the studies that you've done in JAMA were really about communication though as much as patient-reported outcomes, right? When you're managing those symptoms earlier it's because of communication right?

Dr. Basch:  Absolutely. It starts again from that early recognition that about half of people's symptoms are missed and so these are simply interventions to try to directly address that. If we don't miss those symptoms what happens? If we know about what's going on with our patients we can intervene and presumably improve care. So these are really just enhancements to what we already do. Symptom management and monitoring patients is already a standard part of how we deliver care and these are simply enhancements based as you say on improved communication to help us improve how we do things.

Leonard:  I want to talk a little bit about this information and the communication and how it's improving and how that might help to create more personalized care. Reminds me of a piece by Tim O'Reilly back in 2012 where he said that healthcare could solve the Wanamaker problem. Wanamaker problem is driven by a 20th century department store magnate and famously said, and a lot of people have heard, “I know my advertising budget doesn't work. The problem is right now that half of my advertising budget doesn't work. I just don't know which half.” Same could be said for a lot of patient populations too that we know that some treatments work particularly in clinical trial settings we just don't know for which patients it works. So in terms of the kinds of data that you can now collect at home or between office visits what kind of information do you see it's not genomic data, it's not proteomic data, what other kinds of data that's just kind of daily living data do you think will be interesting for defining which patients do better and which ones don't going forward?

Dr. Basch:  Information about symptoms and physical functioning can be very informative both for understanding if a product is effective; so did the pain of the cancer improve and if it did improve and stay away then that's it actually a fairly precise measure of the effectiveness of a product. This information can also give us information about side effects. If a drug is causing terrible nausea or terrible diarrhea getting that information directly from patients again is the most precise way to collect it. So it is a piece of the picture. It adds to the information that we gather from imaging, from biomarkers in order to understand what the impact of treatments are both on improving the symptoms of disease but also the negative side in terms of side effects. And so that's where these data can be very helpful. Now in addition to that, if we collect this information we can also act on it so for example, if a patient is receiving a drug that's effective but it has adverse side effects if we can better manage the side effects the patient will stay on the drug longer and will experience the benefits of that drug over a greater period of time. But if we don't know that the patient is experiencing the side effects then we can't act on them and that may actually lead to the patient either electing to discontinue or simply not being able to tolerate the drug for as long.

Leonard:  Yes side effects haven't really been monitored much in clinical trials I think you even mentioned that in an earlier interview you had done. Is there a historical reason for that? It seems like common sense to look at what side effects are measured. Is it because there was fear that a drug might be less likely to get approved if they start capturing side effects?

Dr. Basch:  Side effect information has been collected historically in clinical trials. It's actually a requirement, regulatory requirements. In oncology trials generally this is done through staff reporting. So either the physician investigator or the staff working with a physician investigator will systematically report the side effects that show up in those could be things like high blood pressure or an abnormal lab or it can be a symptom. I think what wasn't recognized until more recently is that when the providers, the investigators or the staff members report this information they miss a lot. Again we come back to this statistic that professionals miss about half of patient’s symptoms and side effects and so about half of the side effects that are symptomatic were being missed by this method. Nonetheless, this method had remained the standard approach but I think the reason that it was done for so long is that people simply didn't recognize the deficit in that approach and so now we're seeing a change but it takes time to make these changes.

Leonard:  We’re seeing that if you're reporting something every day through your app you're more likely to remember what's happened. Once you start monitoring yourself you have a lot more context to your patients or their symptoms on a day-to-day level.

Dr. Basch:  Yup.

Leonard:  In terms of integrating all this in the clinical workflow are you seeing any changes I think there's been new billing codes for care coordination and new attempts to try and drive more at home care, are you seeing any changes there?

Dr. Basch: I certainly think that PROs are a big part of care coordination and so if one is using billing codes for care coordination the use of PROs certainly can fall into that category. I think it would be even better if they were discrete billing codes for use of case-reported outcomes.

Leonard: Is that something you've talked about some of your groups?

Dr. Basch: Yeah I think this is a nascent area but there would absolutely be value in having those kinds of code.

Leonard: Great I agree. Well thank you for joining us. Is there anything else you'd like to add?

Dr. Basch: No, thanks so much for your interest in this emerging area which I and others hope can truly improve that patient centered-ness of care delivery and product development.

Leonard: Absolutely. Well thank you for all the work you're doing in highlighting just how big of an impact that can have on patient’s lives and on outcomes and so let’s keep it rolling into the future. Thanks a lot.

Dr. Basch: Absolutely, thank you.

Will AI and Machine Learning bring refocus on the individual, but defined by the patient journey?

Will AI and Machine Learning bring refocus on the individual, but defined by the patient journey?

Are We Grossly Underestimating the Returns on Engagement in Clinical Research?

Are We Grossly Underestimating the Returns on Engagement in Clinical Research?