How AI and Data Analysis Can Improve Cancer Treatment with Dr. Andrew Norden, Chief Medical Officer at Cota Healthcare

Thanks for tuning in to the Outcomes Rocket podcast where we chat with today’s most successful and inspiring health leaders. I want to personally invite you to our first inaugural Healthcare Thinkathon. It’s a conference that the Outcomes Rocket and the IU Center for Health Innovation and Implementation Sciences has teamed up on. We’re going to put together silo crushing practices just like we do here on the podcast except it’s going to be live with inspiring keynotes and panelists. To set the tone, we’re conducting a meeting where you can be part of drafting the blueprint for the future of healthcare. That’s right. You could be a founding member of this group of talented industry and practitioner leaders. Join me and 200 other inspiring health leaders for the first Inaugural Healthcare Thinkathon. It’s an event that you’re not going to want to miss. And since there’s only 200 tickets available you’re going to want to act soon. So how do you learn more? Just go to outcomesrocket.health/conference. For more details on how to attend that’s outcomesrocket.health/conference and you’ll be able to get all the info that you need on this amazing healthcare thinkathon. That’s outcomesrocket.health/conference.

: Welcome back once again to the Outcomes Rocket podcast. We’re chatting with some of the most successful leaders in healthcare today and I have a wonderful guest for you. His name is Dr. Andrew Norden. Andrew is a Neuro Oncologist and Physician Executive who joined Cota in 2017 as the Chief Medical Officer. Prior to joining Cota, Dr. Norden served as Deputy Chief Health Officer and lead physician for Oncology and Genomics at IBM Watson Health. Previously he worked at Dana Farber Cancer Institute in Boston in several capacities including Associate Chief Medical Officer and Medical Director of Satellites and Network Affiliates. Andrew served as Physician Leader for the Dana Farber community network for more than five years. He attended medical school at Yale School of Medicine before moving to Boston for residency training at Mass Genn and Brigham Women’s hospitals. He’s got a phenomenal track record in health and now he’s serving at Cota as CMO. I’m so excited to welcome you, Andrew. Thanks for taking time to being on the podcast.

: We’ll thank you for having me, Saul.

: It’s a pleasure. Now is there anything in that intro that I left out that maybe you want the listeners to know about you.

: Well frankly you covered almost everything. So I would say no, I think that was a great start and I appreciate it.

: Awesome Absolutely. My pleasure. Now Andrew why did you decide to get into health care? Why the medical sector?

: Well to be honest I never wanted to do anything other than healthcare from the time that I was a child. Now that said my initial thinking about that meant something much more traditional than where I’ve ended up in the sense that I thought I would be a physician taking care of patients my whole career. And I infected trained to do that and ultimately became a brain tumor specialist. I work at Dana Farber Cancer Institute in Boston for about ten years in a variety of roles some of which you mentioned and what I learned over time is that there are great rewards associated with taking care of patients. But the overall magnitude of the impact for me felt somewhat limited because of the way patient care involves one patient at a time and particularly in the area of cancer research and care that I was involved in brain tumors. The pace of progress was really slow so it was kind of with that lens that I began to look for ways to get involved in something that might have impact on a much broader scale. And I had always been really interested user of technology fairly sophisticated user of technology and had some opportunities to implement new technology systems in an administrative capacity I had leading Dana Farber’s network. So I enjoyed that work a great deal. And when I had an opportunity to join industry and work for IBM Watson Health as a leader of their physician leader of their cancer care offerings I took that leap couple of years ago now and then ultimately made the move to Cota.

: What a great journey and it sounds like you were very focused. You’ve always wanted to be part of this industry helping people but now you sort of took that next step and in your role now you’re able to influence and even broader subset of people in a more agile way. Cotais doing some pretty impressive things maybe to level set the listeners on what Cota does. Andrew, I’d love if you could just give us the quick just understanding of what they do.

: Yeah absolutely. I’ll try to keep it as brief as I can. So Cota is a kids cancer focused data and analytics company and we were founded by a group of oncologists a number of years ago who serve that doctors were spending an awful lot of their time entering data into the EMR. I think most people in healthcare today can identify with that perspective that the EMR has become ubiquitous and yet despite the fact that we spend all of this time working on it we as physicians are often unable to answer the most basic questions that our patients ask us questions like how many patients like me have been treated before and what have been the outcomes associated with the decisions you’ve made in those patients. And people want to understand how do you Dr. Norden compare to other doctors and other centers. We can’t answer those questions and that is really what Cota’s primary goal is. We connect with provider organizations. We pull out data from their electronic medical records and from other clinical systems and then we present it back in a way that’s useful and allows them to understand the way that they’re treating patients and what are the impacts of the various decisions that their providers are making. Day to Day. We also have a really unique method of grouping patients almost like a digital barcode. We call it a CNA or a Cota Nodal address which is a novel way of grouping patients based on all the known clinically meaningful factors and patients who enter our system electronically or are assigned to this. This barcode this CNA which then enables one to compare groups of patients who are clinically similar between doctors insights and particularly when we marry that information up with outcomes data that we track clinical outcomes data and cost data. We can reveal a lot of useful insights that help our customers.

: What a great explanation there Andrew and sort of offering the providers an opportunity to gain insights on a siloed data. Your guys are helping liquify that data bring it to them in a presentable form but also offering a precision medicine type of approach by matching these patients that have similar conditions and backgrounds with outcomes.

: Exactly.

: It’s beautiful and you know what I’ve been having a lot of conversations about this and it’s striking that the important things like measuring outcomes need to happen more so I’m really thrilled to hear that this is a focus for you guys.

: Yeah. I mean you often hear the adage these days that you can’t manage what you don’t measure. And I think it’s really quite true in healthcare measuring outcomes just hasn’t been part of our routine work until recently. Our systems aren’t optimized for doing that. That is definitely a major focus of Kodos efforts today outstanding.

: Now we’ve already dove into several hot topics without even asking this but what would you say if you had to boil it down to one hot topic that needs to be on medical leaders agendas. What is that hot topic and how should they be approaching it?

: Well at the risk of using a cliche I mean I think the answer is data data is an absolutely hot topic. We’re drowning in it. There are systems that surround us. And you know if you ask a typical CIO of a healthcare organization how many systems he or she works with on a daily basis. How’s meaningful data pertaining to patients. The number is huge and sometimes I’ve heard it said as high as a hundred or more systems. I think we all need to be thinking about an approach to make meaning out of this data. There are forward thinking health care organizations provider organizations and other types of health care organizations that are doing this in-house and others who are working with vendors like us to do it. But my general sense is that if you’re not thinking about how you’re going to use your data to your advantage at this point I think it’s an important topic to jump onto as quickly as possible.

: That’s interesting and definitely spot on. Now give us an example how are you guys Andrew at Cota are creating results and improving outcomes by filtering and processing this data.

: Yeah I’ll do that for sure as I think about potential examples some of the ones that come to mind are things that are strikingly easy. If you have the right data at hand we work with a large hybrid or healthcare organization. It’s sort of a community academic hybrid and also with a major local player in that centers region and one of the things that we have looked at as we’ve assigned the CNA that I described to patients this organization is patterns of variability and what we see in oncology are that there are two fundamental patterns of variability. One is that there is variability in the treatment decisions that physicians make for similar patients. So you take two patients who have the virtually same set of attributes and often you see a wide variety of treatment choices being made. And unfortunately in oncology oftentimes those variable treatment choices are in fact all acceptable according to local consensus guidelines which is to say that our consensus guidelines are quite broad and they leave the provider perhaps in many circumstances too much leeway to make a variety of different choices. The second and perhaps more surprising variability pattern that emerges is that when you take similar patients in whom their physicians have made the same treatment choice you still see a great deal of variability in cost and that’s somewhat counter-intuitive because people often imagine that the treatment choice in oncology is the most critical driver of cost. But it turns out that it’s only one piece. The other piece relates to all of the elements of care that surround the treatment choice. How often does the doctor see his or her patient. What types of scans the doctor order and how often are those scans done. What types of labs and how often are those lab done when the patient has radiation therapy which specific radiation technology is used. All of those different things have really meaningful impacts on cost and in some circumstances they may also impact important patient reported outcomes like quality of life. So that said we’ve been interested in delving into this issue of how patients who are similar clinically and are treated similarly still have differential costs and so one of the insights that we uncovered at the provider that I mentioned is that in a common scenario in breast cancer patients need to receive a drug that is called Adramycin which is potentially toxic to the heart. As a result when patients are going to go on this therapy they need to have baseline cardiac testing to ensure that the heart muscle is functioning properly. The reason being that if there is some baseline dysfunction then this would be a potentially dangerous drug for the patients to receive. There are two common tests that physicians order that are equally effective in assessing cardiac function. One is the echocardiogram and one is a nuclear medicine test called the Mugga scan. In this particular scenario that I’m describing it turns out that the mugger’s scan costs hundreds of dollars more than the echocardiogram. So simply by looking at these patterns of care and cost impacts we uncovered that a set of physicians in this network were routinely ordering mogas scans getting no more meaningful information than the echocardiograms and contributing to hundreds of thousands of dollars of additional spend per year. So just by highlighting this for the system we were able to change practice and save money with frankly no resistance from the providers because they don’t care at all which scans they use. It’s simply a matter of what they’re used to doing. All we had to do is let them know that is one test was more expensive than another and practices changed right away. And we have uncovered just countless examples like this where you can simply identify almost meaningless differences that are driven by habit and physician behavior more than by data or any evidence of superiority. And you can save a lot of money for the system at large.

: Andrew so insightful what a great example. Very niche down to oncology very differentiated in the treatments that you’re able to tease out for your provider listening to this. I’d definitely bet the light bulbs going off in opportunity right to reduce costs to offer more resources to these patients that need different treatments. So don’t worry at the end of the podcast we’ll get you a way to connect with Andrew and Coda but definitely continue the conversation and you know one of the things that I really love about the intel that you’re sharing here Andrew is that it was derived by you and your partners which you guys are oncologists and in the grand scheme of sort of tech companies that come up in healthcare and you’ve got the tech focused people that think that hey since I built Uber I could solve healthcare problems. And then you have the providers that have the intel and are approaching this which is the category that you and your team fall into. And I think the insights that you’re able to make come from those with knowledge and experience and insights that are going to be able to be useful in a faster way. So kudos to you and your team for bringing up these examples. It sort of reminds me my wife and I we were in Hawaii several years ago and it was February. We wanted to see the whales and we would just drive we would see no whales. Then we got into the boat and the whale watchers sort of taught us what to look for the water that comes out of the spout. The fin that hits the water all these things that we should be looking out for. And Andrew before you know I would see whales everywhere like I went from not seeing anything to there’s whales everywhere. And when my wife and I got off of that boat and we drove up the mountain and we were driving back to the house that we rented. I swear to you from miles and miles away we saw whales everywhere.

: Wow.

: And it’s because we were educated and here you are with your colleagues. You guys are on that boat and you guys are the experts and I’m sure that when you talk to provider executives you’re like their whale watcher counterpart that helps them see these things.

: Absolutely. And you know what’s exciting for us is that the kind of information I just talked to you about comes from a data set that we build at a population level that maybe the chief of oncology or the administrative leader of oncology or the hospital leadership is looking at what we’re really excited about at Cota is that we’re now in the process of of offering Point of Care tools as well so that we can actually influence the way providers behave and the decisions they make at the point of care based on on these sorts of population level insights so we can bring it right past them at the time that they’re making decisions.

: Wow that is interesting and it sounds like you guys are making this to a more and more useful so so kudos for you guys. Now tell me this Andrew. You know it hasn’t always been smooth. Can you tell us about a setback that you had and what you guys learned from that setback?

: That’s a very good question and I think I’m going to take it from the standpoint of this industry in general.

: Absolutely.

: And that is to say there has been a great deal of interest in encouraging providers to do the all right thing more often. And there have been a variety of different non technological and technological tools that people have built to help nudge decisions in the right direction. My sense is that in healthcare a lot of what we have done in the realm of clinical decision support particularly around treatment decisions relates to so-called clinical pathways or the idea that you can bring together a group of experts to make recommendations for how physicians should. Act in particular scenarios and then push those expert recommendations into the hands of frontline decision making staff to change their behavior. There’s nothing wrong with that approach per se but you can imagine that there is a fair amount of pushback when oncology professional or any health care professional who’s trained his or her whole life to be able to make a particular set of decisions is told. You’re now going to follow this particular expert guidance. So our approach is distinct from that in that we bring actual historical data to the fore in a way that makes it really objective and non-threatening. When you’re a physician and I show you data that says the last 10 times you saw a patient like this and you gave treatment X and your colleague did saw very similar set of patients and chose treatment. Why. But in the case of that colleague the patients had invariably fewer side effects and the cost to the system was less. It makes providers think twice in a way that I think they’re much less likely to if you say here’s how you have to do it going forward. So we are taking that approach of presenting objective data that’s straightforward and easy to understand. And what we find is that physicians actually appreciate the opportunity to see the impact of their decisions in ways that they haven’t been able to before rather than taking the approach of sort of telling them what to do.

: Now that’s so interesting Andrew and you know the other thing too to call out here is is that when it comes from a peer it’s that much more well received. And they think that one of the things that differentiates your company from others out there is that it’s peers right. You guys are oncologists and you’re offering credibility. And with that they open up their ears and they listen.

: Yeah I think that that’s really true in a lot of ways I think about it also that data is the language of science and medicine and expert consensus is valuable but it’s not as powerful and it’s certainly more emotion laden and sometimes controversial. So that’s our philosophical frame.

: Yeah I think it’s great. I mean it beats somebody from Google Sean up that knows nothing about oncology telling them what to do.

: Well no question no question about that although we all know that our colleagues at Google have been doing some smart interesting things in healthcare in months and years as well.

: Without a doubt. But you know you know I’m just saying that to make a point.

: Well taken.

: That for sure Andrew. So tell me what’s one of your proudest medical leadership experiences debate.

: Well I think unique in that I’ve had a variety of different careers in healthcare. I had a career as a clinician and clinical trial list and then a career as a hospital administrator and now a career in health I.T. and I could probably answer this question in a different way for each of those. But maybe what I would say is that in my current world where I work with certainly a number of folks with oncology experience but also a huge number of folks who don’t have oncology experience people who come from the technology world or the business world or fresh out of school or data science biostatistics. All of these are key disciplines that bring a lot to what we do. But for me what has been most gratifying is really helping our team at Cota to understand just how valuable data and simple visualization and analysis can be in healthcare. I think everyone would agree that healthcare is behind many other industries in terms of the way we use information technology day to day. We still are faxing records between provider organizations on a daily basis. We have limited ability to see population level analytics. All of these sorts of things. If you want to have if you show up in an emergency room the odds that the provider there has access to your recent test results remain really really low. So I’ve spent a lot of time at škoda educating our team about how even things that seem technologically unsophisticated can in fact make major major impacts. And so for us simply organizing the data and bringing it to the fore at the right time is so powerful. And I have found it really gratifying just to be able to show our team how that matters and how by doing that we can actually help to ensure that patients are getting the right care or not having duplicated tests or the unnecessary treatments provided and the like. So that is probably where my pride is the chief medical officer Cota comes from today.

: Andrew you brought up such a great point on this point that he didn’t have to be technologically amazing. It’s the simple fact that if you find a way to implement it you just find a way to implement whatever is out there and in a useful way that’s what makes a difference. I teamed up with the Center for Health and innovation and implementation sciences at Indiana University’s School of Medicine. We’re putting together a conference. It’s called the healthcare thinkathon. And our theme is innovation is implementation and health care.

: I love it.

: You don’t have to keep coming up with new things. And your point here is it folks if you’re curious about that Scott outcomesrocket.health/conference you’ll find more details about that there. But yeah Andrew I love this point you’re talking about something real here is that it doesn’t have to be complex, it has to be useful. Exactly. Most powerful man. So that’s exciting now. Now take us down the path of something exciting that you’re working on a project or a focus that you guys are working on at Cota today. What’s lighting your fire today?

: Oh sure I will. We have spent most of this conversation today talking about what we do that helps providers and patients and of course providers and patients are absolutely at the forefront of what we do Cota and in the healthcare industry in general. One of the really exciting things that we’re working on right now relates to the pharmaceutical industry and the drug development process. I think all of your listeners will agree that drug development takes too long and is too costly today and there is universal agreement that anything we can do to get effective drugs into the hands of providers to prescribe them and patients to receive them is warranted. That’s one of the things that we’re working on at Cota is called the concept of synthetic control arms. You may know that as a drug is developed when it’s promising it often needs to be tested in a randomized Phase 3 trial before the FDA or other regulatory agencies will approve the drug for use in people on a routine basis. And we also know that an unexpectedly large proportion of drugs fail in phase 3. So that is to say that perhaps they’ve been tested in comparison to historical controls in a Phase 2 study and the results look promising. So they go into Phase 3 where the drug is then compared against the existing standard of care or against the placebo and oftentimes after years and billions of dollars spent in Phase 3 we learned that in fact the drug is not more effective than placebo or not more effective than the standard therapy and drug development is abandoned. We think that there’s a great opportunity to change that reality by providing a contemporary high quality synthetic control arm in Phase 2 and I think it’s important to understand what often happens in Phase 2 studies is that the comparison group is a historical control group that may have been reported 5 10 even 20 years ago and the details of which may not be well published are well known at present. So we are working with a variety of healthcare stakeholders and agencies to develop a synthetic control arm concept in which we could statistically compare a drug in a small phase 2 study to a synthetic control arm created of for example historical patients in Cota’s real world data set so that if the drug still appears promising after Phase 2 it’s much much more likely to succeed in Phase 3 and ultimately be approved. Alternatively if it looks unlikely to be more effective than a synthetic controlled arm then drug development can be stopped at that point averting the need for additional years of Phase 3 patient enrollment and again billions of dollars spent. So this is a work in progress that a variety of different entities including us continue to work on but I think that the potential to really help speed up and make more cost efficient the drug development process. I think that’s really exciting and is going to be real.

: There’s no doubt you guys are focused on improving outcomes and cutting costs and I think it’s all being done in a very thoughtful way. Andrew so thanks for sharing that listeners if any of this is resonating with you. I encourage you to check out Cota on a deeper level. Go to www.cotahealthcare.com and you’ll be able to find out their solutions are technology more about the company and their hiring too. So if this seems like, right you’re telling me at the beginning of the before we started recording folks. They’re hiring engineers they’re hiring medical directors, they’re hiring delivery and human resources so check out Cota as a workplace. But also if you’re a provider check them out to see what they’re offering because I think you’ll be very intrigued and pleased to know that they may have something for you and help you and what you’re working on today. So this has been fun. We’re getting close to the end here. So I’ve got this medical leadership course you and I are going to build. It’s what it takes to be successful in health care the 101 of Dr. Andrew Norden. So we’ve got four questions. Lightning round style followed by a book and a podcast that you recommend to the listeners. You ready.

: OK.

: Here we go. Lightning round. What is the best way to improve health care outcomes?

: I would say measure what matters and use data intelligently.

: What is the biggest mistake or pitfall to avoid?

: Remember that hype and results are not the same thing. And it’s important to ensure that you’re able to do what you say you can do before you go out into the world and start announcing it.

: Love that. How do you stay relevant as an organization despite constant change?

: I think the answer there is you need to focus. You need to pick one problem to solve and make sure you can solve it better than anybody else out there.

: I love that Andrew one of my favorite acronyms as focus. Follow one course until success.

: I like that too.

: What’s one area of focus that drives everything in a health care organization?

: Well I’ll tell you in our company it’s about having a high quality data set and being sure that it’s collected in a way that that means it’s real.

: Beautiful. What book and what podcasts would you recommend to the listeners on the syllabus.

: So my favorite book not about health care at all but has been meaningful to me throughout my professional life is the Dale Carnegie book. I suspect many of your listeners are familiar with it How to Win Friends and Influence People.

: Yes a great book. What a great one. And folks listen. One of Andrew’s favorite books being Dale Carnegie’s. Then surprise me why he’s done the amazing things that he’s done and has had the career shifts that he has had. So if you haven’t checked that one now be sure to check it out. How about a podcast?

: I’ll tell you I also have really been enjoying the A16Z podcast put out by the Silicon Valley VC firm Andreessen Horowitz they talk about technology and healthcare entrepreneurship and I find all sorts of tidbits of wisdom in there when I listen.

: What a great recommendation listeners. All the things that we discussed today the podcast that he recommended, the book, and all of the show notes and transcripts as well as a link to the company. All of that could be found at outcomesrocket.health/cota so visit that and check out all the links there and resources. But before we conclude I’d love if you could just share a closing thought Andrew. And then the best place for the listeners could get in touch or follow you.

: Yes you bet. So as we close you know I would say that a key to the next step in health care data and analytics is going to be around how we capture high quality data in the course of our daily work. I think Cota and others like us are beginning to do a really nice job of trying to make meaning of the data that exists in the EMR. But wouldn’t it be nice if the EMR were constructed in a way that the data that flowed freely from it was high quality and helpful from the get go. I think that’s going to be the next frontier for this work. I would be delighted to hear from your listeners. I’m at andrew@cotahealthcare.com. I think he spelled it out before but cota.healthcare.com and it’s just andrew@cotahealthcare.com.

: Outstanding Andrew. What a great close there that you left us with that aspirational EMR I think is a destination that we’re all going to enjoy getting to. And with the efforts that you and your colleagues over at Cota I know it’ll probably make it faster so just want to appreciate all that you’ve done for us today and sharing your thoughts. And we’re looking forward to staying in touch.

: Thanks a million. I enjoyed it

Thanks for tuning in to the outcomes rocket podcast if you want the show notes, inspiration, transcripts and everything that we talk about on this episode just go to outcomesrocket.health. And again don’t forget to check out the amazing healthcare thinkathon where we could get together to form the blueprint for the future of healthcare. You can find more information on that and how to get involved in our theme which is implementation is innovation. Just go to outcomesrocket.health/conference that’s outcomesrocket.health/conference be one of the 200 that will participate. Looking forward to seeing you there.

Automatically convert audio to text with Sonix