Print BCTV: Profiles: Smarter Trials -- Determining healthcare quality measures; organizing collaborative trials

Profiles: Smarter Trials

Transcript of BioCentury This Week TV Episode 197




Dr. Charlotte Yeh, Chief Medical Officer, AARP Services

Dan Mendelson, CEO, Avalere Health

Dr. Laura Esserman, Director of the UCSF Carol Franc Buck Breast Care Center

Dr. Maria Freire, President and Executive Director, Foundation for the NIH




Dr. Janet Woodcock, Director, Center for Drug Evaluation and Research, FDA

David Wholley, Director, The Biomarkers Consortium, Foundation for the NIH

Susan Edgman-Levitan, Executive Director, John D. Stoeckle Center for Primary Care Innovation, Massachusetts General Hospital


Veterans Administration

SWOG (formerly, Southwest Oncology Group)

Friends of Cancer Research

Journal of Health Affairs



Steve Usdin, Senior Editor




STEVE USDIN: Quality is the goal for many healthcare reforms. What is quality? And can the pursuit of quality actually harm patients? And in Profiles in Innovation, how new ways to search for scientific truth are bringing better medicines to patients faster. I'm Steve Usdin. Welcome to BioCentury This Week.


NARRATOR: Connecting patients, scientists, innovators, and policymakers to the future of medicine. BioCentury This Week.


NARRATOR: America's healthcare system is undergoing massive change. And many stakeholders, patients, hospitals, insurance, and drug companies are moving in different directions. There's one thing nearly everyone agrees on, a primary goal in healthcare should be quality. But what is quality? And what does it mean to patients? Defining quality is more than a semantic exercise. The right quality measures can be powerful incentives for improving health.


Bad or irrelevant quality measures can create perverse incentives and checklist medicine that hurts patients and adds unnecessary costs.


STEVE USDIN: To discuss quality in health care, I'm joined by Dan Mendelson, CEO of Avalere Consulting, a consulting company that's helped design quality metrics. And by Charlotte Yeh, Chief Medical Officer for AARP Services. So, Dan, what is quality?


DAN MENDELSON: Well what's, I think, important to know is that increasingly, the way that we are paying doctors and hospitals is around these new measures that try to measure quality. And so quality is really what the payer or what the government says it is. And providers are starting to align around those measures that we are putting in front of them and asking them to achieve.


STEVE USDIN: So, Charlotte, you have a story from the patient perspective. Dan mentioned quality is what payers want. But at some point, if it's going to be real medicine, it has to shift to what patients want. Can you tell a little bit of what's your story briefly?


DR. CHARLOTTE YEH: Sure. I'm an emergency physician, so I've been very active in the world of health care. I was crossing the street in Washington, DC with a green light in a crosswalk when I heard a thump. I went, oh, my God. That's me. I got hit by a car.


Rushed to the nearest emergency department, tested with everything, and how is your pain managed, et cetera. And at the end of all of that, they looked at all of my studies and said, nothing is broken. You can go home now. They didn't check to see could I walk, could I even go home. And it was only after realizing, when they tried to get me up that I fell to the floor, that they said, oh. I guess your tests were normal. Everything is normal. But I guess if you can't walk, we'll have to admit you.


Now where in the quality metrics do we measure, it's not the test, but it's actually, can you walk?


STEVE USDIN: So, Dan, it's an example of teaching to the test for medicine.


DAN MENDELSON: Absolutely.


STEVE USDIN: How do you bring real quality, the things that really matter to people, into the system?


DAN MENDELSON: This is definitely one of the biggest limitations in quality research right now and in the way that we are paying providers. And a lot of the quality measures that we have right now are oriented towards things that are important but really do not measure the patient experience. So, for example, do we immunize patients? Do we do colorectal screening for cancer? Do we do routine mammography when indicated for patients?


And all of these things are really important, but they don't get to some of the qualitative measures that we really need to start building in. I think that that's really why there needs to be more of a focus on these patient-reported outcomes and a lot of work done to try to figure out how to measure these important things that are more qualitative and more difficult to measure.


STEVE USDIN: So, very briefly -- sorry you wanted to say something, Charlotte -- but what is a patient-reported outcome?


DAN MENDELSON: A patient-reported outcome is really just getting the patient's feedback on whether their goals were met, very simply.


DR. CHARLOTTE YEH: So there are two parts to this. First is I really love the immunization measures and all the very important clinical, how is your diabetes managed, that we are measuring physicians in hospitals. But we also realize they must be customized. So if you're 60 years old, and you've got diabetes, heart disease, cancer, severe lung problems, you're so busy trying to manage your life, you might not get that mammogram that you're supposed to get every year.


And yet the physician will be dinged for not getting a mammogram when, in fact, it was the right thing for that patient. So that balance between that technical measure, the process measure for payment purposes, but making sure it aligns with living well.


DAN MENDELSON: Yeah. Absolutely. I will tell you, from the experience of being on the board of a publicly-traded managed care company for many years, that when the Medicare program started measuring things like immunization, our resources started focusing on those things very intently. And when we put it into the bonus plans of all of our Chief Medical Officers, they really started focusing on it. And it really is true, and Steve, you were saying before about teaching to the test, that the care managers will do what we ask them to. And they will do what we incent them to do through payment systems.


We just have to make sure that the payment systems are balanced enough to also begin integrating things like compassion in care.


STEVE USDIN: Well we're going to talk more about that, and really how can you get compassion into the system, in just a moment. Charlotte Yeh described her experience as a patient and her view on how quality metrics can go wrong in the Journal of Health Affairs. Here's a look.




STEVE USDIN: We're discussing quality healthcare with Dan Mendelson and Charlotte Yeh. So Dan, you mentioned compassion there. And I'm wondering -- there's this mantra, if you can't measure it, then it doesn't count -- but can you measure compassion?


DAN MENDELSON: Yeah, I think you can. I think you can measure compassion by asking patients how their experience was with clinicians. You might also be able to do more innovative measurement and look at social media, for example, of whether patients say positive or negative things about their experience in a given institution. These are all things we have to figure out how to measure, because, again, what's going to happen in this system is that providers will ultimately do what we ask them to do. And if we don't ask them to provide compassionate care, it's not going to happen.


STEVE USDIN: So how do you get the patient's voice into that and into these measures?


CHARLOTTE YEH: So I think one of the important things that you mentioned earlier is imagine if you went into the doctor's office. You asked about goals. And I know Suzanne Edgman-Levitan, who does a lot of primary care work at the Mass General. She said, what if you walked into the physician's office -- and we always say, what's the matter? We want to know, how's your diabetes, how's your high blood pressure?


But what if the doctor said, what matters to you? And then customizes that care around whether you want to live longer, whether you want to be functional and more active, or whether you want to be pain free? So if you centered around that goal and measured how well -- as Dartmouth does -- how well did the physician elicit your goal? How well did the physician help you understand how to achieve your goal? And how well did the physician incorporate a care plan with your goal in mind?


DAN MENDELSON: Mm-hm. Very important.


CHARLOTTE YEH: That is how we could really capture that patient voice.


DAN MENDELSON: I agree. And I think also that what Charlotte is saying points out a couple of things about patient-reported outcomes. Not all patients want the same outcomes. You go in for knee surgery, the outcome might be very different if you plan to run a marathon versus if you just want to be able to spend time with your grandkids. And so your version of success might actually look very different from the patient next to you. And I think also that it is very difficult to measure a lot of the things that we associate with very high quality healthcare.


CHARLOTTE YEH: But you mentioned the knee surgery. I think it's really interesting, when we actually look what we measure, it's mortality, morbidity, infection rates. But you know what a patient wants to know? How long will I be out of work? Will I need help when I go home?


Will I be able to do things better than before? Where are those kinds of measures? We've actually been very poor at capturing what really matters to a patient -- how well do I live at home?


DAN MENDELSON: Absolutely. And the things that are important to patients are often very different from the things that are important to health plans or health systems. And both the patient and the health systems have legitimate interests. So it's important to keep the quality of care high, the cost of care low, and also be able to provide the patient with the things that they need.


STEVE USDIN: So there's also some examples, I think, from the Affordable Care Act, the way that it's driving the use of quality measures. Can you describe some of the --


DAN MENDELSON: Absolutely. The Affordable Care Act, for the first time, put in place very strong incentives for quality, both for hospitals -- so for example, there are penalties if a lot of patients are readmitted into your hospital. And if you have a lot of readmissions, you're going to get paid less. And likewise, for health plans that are operating for the Medicare program, there's a system called the star rating system, where health plans are paid on the basis of whether they achieve a lot of the outcomes that we were talking about before -- immunization, mammography, colorectal screening, and the like.


STEVE USDIN: And also, patients can see those star ratings, right? So they can decide which ones that they want to use.


CHARLOTTE YEH: Right. But again, if you think of the star ratings, they don't necessarily resonate with the heart and mind of the consumer, because the consumer's always looking for what's next. There is more -- it's not about disability, but what can I do. And the measures don't measure that. And if you're getting immunization, but you're busy caring for your father and your grandparents and your children, you may not have time for that mammography. And how do we incorporate that lifestyle, also, as part of the quality metrics?


STEVE USDIN: Let's follow up on that in just a moment. Quality metrics are a major focus for the Agency for Healthcare Research. Here's how you can learn more.




NARRATOR: Now back to BioCentury This Week.


STEVE USDIN: We're wrapping up with Dan Mendelson and Charlotte Yeh. So, Dan, the idea of quality measures, it isn't going away. And we've talked a little bit about some of the potential problems and pitfalls. And one of the solutions we've talked about is bringing patients into it more. How do you make that real?


DAN MENDELSON: Yeah. I think that's the important point, Steve, is that these measures are not going away. We are going to be paying providers on the basis of quality measures. And we're not going to abandon the important measures that we've discussed so far like immunization. And so then the question is, how do you balance these measures with patient-reported measures? And the measures that are really important to patient experience? And how do we come together and bring together all the different diverse stakeholders who need to have a voice in creating those measures? And we have a lot of work to do, I think, over the coming years to make these outcomes measures better.


STEVE USDIN: And Charlotte, from your perspective, you're at AARP now. And you have a career with a lot of different experiences. How do you actually make it real? How do you bring the patient voice into creating these measures?


DR. CHARLOTTE YEH: So I think the real gap we have now with measures, besides patient-report outcomes, is it's all about consumer engagement. How do you capture the hearts and minds of the person? Because if you want people to have healthier lifestyles, live better, and do better, you want to be able to engage.


And what matters to them? It's not the hemoglobin A1c. I'm sorry. It's about going to church. So how do we capture what's next? We're living longer than ever. And we learned at AARP that when we talk about programs -- I have this great program that's going to help you manage your health better. Engagement rate, not much.


But when we said, we have a program that's going to help you live independently at home, the way you want to, enjoying your family and your friends, the engagement rate was faster than we could keep up with. That's where the measures need to go, on independent living and living well every day.


DAN MENDELSON: And they need to be simple, too. So we need to be able to measure things simply and directly without burdening patients with a lot of forms. And some of that, I think, really is involved in working with patients, and clinicians, and plans to come up with measures that are easily identifiable and also can be folded into the kinds of electronic medical records that are currently being collected and used by providers.


STEVE USDIN: And is it also maybe part of it, integrating other kinds of technology? The fitness technology and things that people wear on their wrists, those kinds of things? So that people see the reality of it in their own lives?


DAN MENDELSON: Bringing that data into the electronic medical record is absolutely key for really being able to capture a lot of patient experience.


DR. CHARLOTTE YEH: And how you're doing at home, your connectivity with others. We found if you're lonely, you're more likely depressed, your medical costs go up. So if we can begin to capture how well are you living? And what is your goal in life? And how do we get you there? We will achieve truly the best outcome of all.


STEVE USDIN: And with your patient population, I imagine a lot of it also has to do with how you get these measures into Medicare. Right? Because most of your population is Medicare.


DR. CHARLOTTE YEH: Right. And most of the measures now were started in younger, commercial population. And then translated to older. But when you're 80 years old -- we were studying depression. And when you're 80 years old, and you're talking about end of life, we thought it was, oh, my god. We've got to send you to the suicide hotline.


And it turned out the suicide hotline said, they're not talking about suicide. They're talking about legacy. What are they leaving their children? How they're financially planning. That's where the measures need to go.


STEVE USDIN: So, Charlotte, one of the things I get also that has to happen to make all this work is a difference in the way that we train physicians, isn't it?


DR. CHARLOTTE YEH: Right. So what I worry about is this balance as we have all this great, sophisticated technology and science, is that we are training a generation to the test. Did they follow the quality metric and not listen to the patient? So that will be one of our challenges going forward is, in my case, your test is normal. Therefore you must be normal. There's nothing wrong.


Oh, you can't walk? Sorry. Your test is normal. You can still go home.


STEVE USDIN: So Dan, how do you change the way that you measure things so you incorporate medicine, basically?


DAN MENDELSON: Providers are going to have to do both. And in medical training, we're going to have to make it very clear that they need to do both. I think one of the other things is we have to also acknowledge that these quality measures are always going to have limitations. So, for example, there might be a quality measure about reducing cholesterol. But there might be a drug that is able to do it more rapidly. And it might be that that drug is better for a certain subset of patients than others.


And so we have to make sure that the quality metrics are there, that they're balanced, but they're also not the be all and end all in terms of how we're ultimately evaluating the way the providers are performing their services.


STEVE USDIN: Isn't there also an element to it that the metrics have to make sense? For example, in the VA health system scandal, one of the reasons for the scandal is that there was a metric that everybody should see a physician within two weeks. And that wasn't always necessary. But in order to meet that metric, there was gaming. And there was fraud.


DAN MENDELSON: Right. We have to fine tune the measures. But ultimately, I do think -- and I think this is implied in Charlotte's comments as well -- that the humanity of the physician can't be measured. And we'll have to really just work to make sure that it's very high.


STEVE USDIN: And patient satisfaction, is that a surrogate for humanity of the physician?


DR. CHARLOTTE YEH: Yeah. Patient satisfaction, how well did they communicate, how well were you thought of as a person, how well did he understand your life as he incorporated the science into that art? So maybe it's the art that is the hard part to measure. But I think we can get there if we bring in the patient's voice, the patient satisfaction, and the patient experience.


STEVE USDIN: And is there an element also of feedback? That individual patients, providers, hospitals, payers can take that feedback from the patients and improve care?


DAN MENDELSON: Absolutely. We need to be much better about bringing that feedback in and incorporating it back, and making sure the providers are hearing the feedback in a way that doesn't burden them.


STEVE USDIN: Thanks a lot, Charlotte and Dan, for a fascinating conversation. Coming up in Profiles and Innovations, how collaboration can improve the search for the truth.


NARRATOR: Now in its 22nd year, visit for the most in-depth biotech news and analysis. And visit for exclusive, free content.




STEVE USDIN: Medical innovations are big advances that help prevent, cure, or treat illness. Often this means translating scientific discoveries into drugs, devices, or diagnostics. Sometimes an innovation is a novel way of organizing a search for scientific truth.


This week, we'll learn how pioneers from universities, companies, and government are working with patient advocates to create new ways to determine the truth about experimental drugs. The collaborative clinical trial movement could slash years off the time it takes to get breakthroughs to patients, turn personalized medicine from the exception into the rule, and make millions of patients partners in medical research.


NARRATOR: It takes about a decade and costs billions of dollars to go from the discovery of a possible drug to an approved product. Tests conducted in humans -- clinical trials -- are the longest and by far the costliest part of that journey. Clinical trials are essential to learning how hope overlaps with reality. But the time and expense of trials has increased to the point that they become a major stumbling block to medical progress.


Doctor Janet Woodcock, the top drug regulator at FDA, says it doesn't need to be so cumbersome.


DR. JANET WOODCOCK: You need to turn the paradigm on its head. Right now, a company will say they want to do an efficacy trial. They'll get a protocol together. They'll get thought leaders. They'll get PIs, investigators. Then they'll have to negotiate agreements with each university.


STEVE USDIN: And it all takes a couple years.


DR. JANET WOODCOCK: Years. Yes. To give an analogy, if you wanted to build an airplane, you said, well I'm going to get 200 independent contractors to build each part. Or 2000. And then we'll try to put it together and make sure it flies. And then the next airplane you build, you're going to get 200 different contractors. And get them to make the part. And, by the way, they all have other day jobs.


STEVE USDIN: Collaborative trials move the search for scientific truth from handcrafted trials studying one drug at a time to ongoing trials or master protocols simultaneously looking at many drugs for specific conditions. They expand opportunities for patients to participate in and benefit from clinical research.


NARRATOR: Collaborative trials are run by independent researchers, not companies that have a financial interest in the results. Multiple drug candidates are tested. Once enough data's been collected, a decision is made that an experimental drug is likely to be safe and effective for a specific set of patients, and it can graduate from the trial. Or that it doesn't measure up and should be dropped.


When a drug graduates or is dropped, new drugs can take its place and the trial continues. One of the first collaborative trials is called I-SPY 2, launched in 2010 and led by University of California San Francisco researcher Laura Esserman. I-SPY 2 incorporates a number of novel features that are intended to learn much more, much faster about experimental breast cancer drugs.


DR. LAURA ESSERMAN: The problem is that it takes, in oncology, it's taking 10 to 15, probably 15 years, to go from an idea or the development of a targeted therapy to get it through the phases of testing and approved by the FDA and to patients. And the price tag is upwards of $2 billion. That is not sustainable. And the big losers here are patients.


I'm a surgeon. I'm a treating clinician. And when someone stands in front of me, they don't have 15 years for us to figure out what the next innovation is.


NARRATOR: I-SPY 2 is an adaptive trial. Patients are assigned to different experimental drugs based on the genetic profile of their tumor, a biomarker. As new patients enter the trial, researchers give them experimental treatments based on the outcomes of previous patients with the same biomarkers. I-SPY 2 is managed by the Foundation for NIH, a non-profit.


DAVID WHOLLEY: The I-SPY 2 trial is a large-scale partnership that is designed to use a targeted medicine approach. So patients are grouped according to their biomarker profiles. And then there's an adaptive engine that actually allows patients who come in later in the trial a better chance of getting a drug that works for them. So drugs that graduate from I-SPY 2 have a chance of going to a Phase III trial with as few as 300 patients and an 85% chance of success. And the entire process can take place in half the time.


DR. JANET WOODCOCK: So what I-SPY is doing is flipping the paradigm. And saying the trial, and the sites, and the patients are central. Right? Rather than a drug or question.


NARRATOR: To participate in I-SPY 2, drug companies have to give up control, a big change from standard drug development. They have to accept the fact that their drugs will be tested by independent, academic researchers alongside drugs from their competitors.


DR. LAURA ESSERMAN: The trial is a master protocol. So that the trial is up once. There's no inefficiency when another drug comes in. It seamlessly drops into an amendment. And we get it into the trial. We are increasingly understanding cancer and every disease to be a subset of diseases or a number of different diseases. So in a very short order, we're going to have a bunch of orphan diseases. And no one is going to invest $2 billion to get a therapeutic strategy for such a small group of patients.


So if we want to really realize the promise of personalized medicine, we're going to have to come up with a different strategy. The status quo simply will not do.


STEVE USDIN: Another collaborative trial has just started. The Lung Cancer Master Protocol, or LUNG-MAP, will test multiple experimental lung cancer drugs. Like I-SPY 2, it's being managed by the Foundation for NIH. LUNG-MAP was conceived by a patient advocacy group, Friends of Cancer Research, and is being conducted by SWOG, a cancer research cooperative group.


NARRATOR: The trial is optimized for personalized medicine. Tumors from all patients are sequenced. And most patients are assigned to a treatment that was designed for their tumor's genetic fingerprint. LUNG-MAP makes it possible to quickly test experimental drugs that are designed for a small percentage of lung cancer patients, according to Maria Freire, President of the Foundation for NIH.


DR. MARIA FREIRE: What we've done in this collaborative trial is we've established an umbrella platform, an umbrella trial. So that we can bring drugs in, we can take drugs out. If a drug works, it continues the process. If a drug doesn't work, it comes out. We're able to put another drug in.


You have the ability to recruit patients, which is very hard to do. If you look at the statistics, only about 4% of cancer patients are part of clinical trials. That's a very small percentage. If you add onto that percentage the fact that maybe 5% to 20% have a specific type of cancer, then you start realizing that in order to have enough patients that you can determine with certain statistical certainty that the drug is working, it could take two, three, five years to recruit.


So the first advantage of having this type of trial is that you enhance accrual. You can get more people into the trial faster. For the patient, it reduces the time of getting one drug from the moment we thought that this might in fact work to where they can actually use it. For people that have one year to live, this is incredibly important.


DR. LAURA ESSERMAN: The clinicians want better answers. Patients want better answers. We don't have to be stuck in this 10 to 15 year time frame. There's no reason for it.


STEVE USDIN: LUNG-MAP started enrolling patients in June. And I-SPY 3, a follow-on trial that could lead to FDA approvals will start soon. Collaborative trials are being designed for colon and other cancers. They're all based on the same premise: collaboration will bring better therapies to patients faster. That's our show for this week. I'm Steve Usdin. Thanks for watching.