March 10-14, 2017; Austin, TX; Full Report – Draft

Executive Highlights

Coming at you from Austin, Texas with our report from SXSW (South by Southwest) Interactive 2017! This report includes highlights from two The diaTribe Foundation-organized panels – one on preventing a diabetes avalanche, and another on next-generation doctors and the future of healthcare teams. We also bring you an update on Sano’s wellness CGM, a panel on reinventing devices with OneDrop CEO Mr. Jeff Dachis, and a fascinating session on self-driving cars with many parallels to diabetes tech. On a broader note, we also have coverage of Surgeon General Dr. Vivek Murthy on emotional wellbeing, AOL co-founder Mr. Steve Case on today’s innovation climate, Verily CMO Dr. Jessica Mega on data, and two insightful gems related to motivation. Read on for our full coverage of this always fascinating conference.

Detailed Discussion and Commentary

One Nation Under Stress: How Social Connection Can Heal Us

Vivek Murthy, MD (Surgeon General, Washington D.C., MD)

Highly-respected Surgeon General Dr. Vivek Murthy, whom we admire deeply, engaged in a deeply personal and moving conversation with IDEO’s Mr. Fred Dust, touching on the harmful effects of stress, his own emotional experiences, and addressing emotional wellbeing. The ties between sustained stress and chronic disease are well documented – a heightened state of arousal over the long term can ignite an inflammatory cascade that damages the body, accelerating aging, cardiovascular disease, diabetes, and beyond. In Dr. Murthy’s tours around the country, he found that emotional well-being was the single most universally resonant issue, regardless of age, location, and other determinants (this also came up in the diaTribe Foundation’s Outcomes Beyond A1c survey). Notably, poverty, economic hardship, the experience of discrimination, violence, illness, lack of self-efficacy, and isolation were among factors that he noted exacerbate stress. The importance of social and cultural determinates permeate his work and we hope continue to gain steam as very important components for the system to address. Notably, Dr. Murthy views emotional well-being as a scale; not as the absence of mental illness, but as a positive state of mind facilitating creativity, productivity, and healthy relationships. On his own experiences with the issue, Dr. Murthy commented: “I have had dark moments where I felt sapped of energy and motivation, like I wasn’t connecting, and my enthusiasm was low across the board.” We loved this personal side to the discussion – if people as successful as Dr. Murthy continue to discuss their own experiences as momentary challenges but not weakness, then a culture shift will follow. In these moments, he has turned to physical activity, getting more sleep, and meditation to as ways of lifting his mood and mindset (watch his TEDMED talk here). On a broader level of the country’s emotional health, Dr. Murthy was hopeful about community-level, grassroots movements: “There’s a saying; if you want to feel optimistic about America, just stand on your head, because it looks better from the ground up.” He sees the most potential for change in parents, teachers, local YMCAs, adding that the practice of paying doctors is antiquated because the majority of what causes bad health outcomes does not happen in the clinic. There have been countless programs around the country – bringing meditation into schools, for example, with outstanding results – which are not the result of major federal legislation, but the community recognizing a problem and coming together to solve it.

• Dr. Murthy doesn’t believe it’s “socially justifiable” to build platforms (i.e., social media) without regulating the way they are used. “We have to be conscious about the way they are used, because that effects how things are perceived in the real world.” While he believes that social media is ultimately a good thing when used as a way station rather than a destination – used to augment relationships, not replace them – he expressed concern that it’s a tool that can magnify and spread both positive and negative experiences. He likened the effect to broken heart syndrome, where acute stress can be fatal. In 10^th grade, his maternal grandfather died of a heart attack. His grandfather’s brother visited, and upon seeing the body, he broke down, suffered a heart attack, and died. While this may not happen on social media to the same degree, we imagine that effects of good and bad news can permeate a social network and induce stressful responses.
  
  • 40% of adults report feeling isolated today, even though there is more technology than there has ever been. There is a strong correlation between social media use and isolation, though it is unclear which direction that causal train faces. Dr. Murthy added that the impact of isolation on mortality is equivalent to the impact of smoking and obesity, and “think how much tie we spend trying to get people to quit smoking.”
• Dr. Murthy emphasized the importance of pause and requested that designers and technologists think about pushing people to take moments to relax. This is so important, but so oft-overlooked in today’s fast-paced society. In Dr. Murthy’s words, “if we fill all the spaces of our lives, then we leave no opportunity for creativity and chance encounters.” He then used a heart analogy to put this importance into perspective: “The heart pumps in two phases, systole (pumping oxygen-rich blood to organs) and diastole (where the heart relaxes and fills with blood). Most people think that systole is the more important phase, but it’s actually during diastole that the coronary blood vessels full up, so it turns out that pausing is what actually sustains the heart.” If those who construct our environments focus on helping individuals to incorporate pause and be mindful, then we imagine that people will be much happier and healthier.

Capability, Comfort, Calm—High Quality Health Care

Elizabeth Teisberg, PhD (Value Institute at Dell Medical School, Austin, TX), Scott Wallace (Value Institute at Dell Medical School, Austin, TX)

Dr. Elizabeth Teisberg and Mr. Scott Wallace, both of the Value Institute at Dell Medical School, reminded attendees that consumers want health – capability, comfort, and calm – not healthcare as a service. The duo explained that the right questions are not being asked in today’s healthcare climate: At the end of a hospital stay, the patient is asked “did the TV work? Were the sheets clean? How were we? Would you recommend us?”, when the real question should be “how are you?” To unearth the right questions and understand patients’ unmet needs, they initiated “experience groups,” cohorts of patients with the same disease who meet to discuss the pros and cons of their healthcare (featured in a February WSJ article). According to Mr. Wallace, they have learned about “all kinds of things” that matter to patients that run counter to today’s healthcare. In one case, providers were frustrated that insulin-dependent patients were not taking their insulin at work, but it turned out that there were no sharps containers in the bathroom. In another extreme example of the wrong questions being asked, women who had received breast reconstruction following mastectomy said that their implants didn’t feel right – all of the doctors were satisfied simply because it looked great. One woman thought her implant felt like an Idaho potato, while another said it felt like every time she moved, her breast was doing the hula. “Patients are taught to thank doctors, not imagine how their care could be better or give feedback. Shake yourself free from thinking about incremental change from where we are today. No one ever complained that their landline couldn’t take good pictures of their kids, but the possibilities today are incredibly different. Imagine if the same thing happened in healthcare.” This all hearkens back to the push for outcomes beyond A1c – what outcomes really matter to patients – Time in range? Glycemic variability? Emotional wellbeing? – and how can we start tracking those outcomes, standardizing their reporting, validating their impact, getting them into regulation and product labels, and driving reimbursement? (For more, see our report from August’s FDA Diabetes Outcome Measures Beyond A1c Workshop).

• Dr. Teisberg also emphasized that all outcomes and cost data, down to the individual level, needs to be collected. She gave two examples of where data was not being collected but proved beneficial: A heart hospital didn’t realize that children in its care were dying at an inordinate rate, and only when they started tracking mortality in a segmented fashion did they make the connection. Once they had this data, they began sending pediatric patients to the nearby pediatric hospital, which had much better outcomes. In the other scenario, disease-specific mortality for patients with prostate cancer was nearly identical (~95%) between the best hospital in Germany and the average hospital in Germany. However, upon looking at outcomes besides mortality, there were big differences in the rates of erectile dysfunction and incontinence between the two hospitals, in favor of the best one.
• Like electricity and flight, Mr. Walker posited that healthcare should be viewed as an intermediate good: “You don’t want to experience electricity, you just want enough to run your lights, make sure your air conditioning works, and so forth. Airline travel isn’t what you seek – you want to get there on time, to the right place, and have polite and respectful flight attendants. In healthcare, you want to seek treatment because you want to get to a place of better health. Nobody goes to healthcare for the experience – it’s not the purpose.”

Patient-Centric Healthcare: The Future of Health

John Mattison, MD (Kaiser Permanente, Oakland, CA)

Kaiser’s Dr. John Mattison noted that the next frontier will be in personalizing technologies to make them uniquely motivating. He used the terms “motivation formulary” and “motivicon” and explained further: “Personalized motivation is often absent ... Some people are really motivated by humor. Other people are not. Some people want directives. Other people want to drill down to the scientific evidence. Others want shared decision making. The point is, we need to personalize the motivational framework. I call this a “motivicon”: it could be a video, a patient centric health literate translation, an in app reward, a text or email, etc. We’ve seen stunning results in preliminary research where people were not doing what was in their best interest vis-a-vis diabetes. We tried to understand what motivated them and then personalized the motivation. I envision having this motivation formulary: if you want to tell someone to do x, y, z, we have five different ways of doing it, or motivational profiles. We prescribe the medium and message in a format that is personalized to their motivational profile.” We LOVE this idea, though in searching high and low online, we could not find anything more on it! We have emailed Dr. Mattison and hope to learn more – there is serious brilliance here.

Moral Issues in Designing for Behavior Change

Raphaela O’Day, PhD (J&J Health and Wellness Solutions, New Brunswick, NJ), Amy Bucher, PhD (MadPow, Boston, MA)

Dr. Raphaela O’Day (J&J Health and Wellness Solutions) and Dr. Amy Bucher (MadPow, a design firm) shared two useful motivation and behavior change models (COM-B and self-determination theory), both with important applications to diabetes and chronic disease. The COM-B model (Michie et al.) breaks behaviors down into three areas: capability, opportunity, and motivation. All three are required for a behavior to happen, and none of these are linear – they change from time to time, behavior to behavior, and day to day. Dr. O’Day noted that for a couch potato trying to exercise, simply saying, “Walking is effective and cheap” isn’t useful for changing behavior. While someone may have the physical ability and even the motivation to exercise, their opportunity (surrounding environment) may not align – e.g., no side walks, unsafe streets, etc. Her slide applying the COM-B model to medication adherence (see below) illustrates how much goes into something as simple as taking a pill. The other fascinating model (self-determination theory) put motivation on a continuum: amotivated (“I have no desire to do this”), external motivation (“Someone told me I have to do this”), introjected (“I’ve internalized the nagging: I better do this”), identified (“Doing this will help me achieve goals I really value”), integrated (“Doing this is part of whom I am”), and intrinsic (“I love doing this; it feels great!). The last three categories are where long-term change happens, while the first three categories are very susceptible to obstacles – when things get difficult, the behavior is not likely to be accomplished. This makes a lot of sense in diabetes, where patients are often told what to do but the reasons are not internalized. How can this change with insight from these models? We hope to see a lot more focus on these best behavior change practices in diabetes technology, especially in BGM and CGM.

I’ll Show You Mine If You Show Me Yours: Data For Good

Jessica Mega, MD (Verily, San Francisco, CA), Karen Horgan (Val Health, Stamford, CT), Sean Lane (CrossChx, Colombus, OH)

Verily CMO Dr. Jessica Mega, Val Health President Ms. Karen Horgan, and CrossChx founder Mr. Sean Lane deliberated on ways to make the hordes of available biometric and phenotypic data actionable for physicians and patients. Right off the bat, Dr. Mega made the point that data itself will never be enough to assist a provider or elicit behavior change in a patient. Ms. Korgan said that providers like to have data from the three- to six-month period in between visits, but even then, it’s still often masses of raw data with little added value. Furthermore, Mr. Lane added that providers don’t have the right tools for an outcomes-based world, and until they move on from the “monolithic EMR systems built for the fee-for-service world,” they will be “underwater and overwhelmed.” He compared the scenario to his tours of duty in Iraq and Afghanistan: At first, soldiers were given weapons that were useful against the Russians, but were not fit to take on guerilla insurgents. Ms. Horgan added that providers don’t have the training or tools to treat patients in a preventative, risk-based manner. Both issues will need to be addressed or the transition from fee-for-service may not be a smooth one. On the consumer side, all of the panelists were certain that people will never manually track anything, but you can pull powerful levers if you can capture data passively. For example, using machine learning, you can A/B test different word choices, lengths, and times of day for an email on different population segments to try to maximize enrollment in a diabetes management plan. Dr. Mega nodded as Mr. Lane explained the hope for digital phenotyping, creating phenotypically similar cohorts based on behavioral, genomic, and other data – this would allow consumers to compare their own metrics to others like them and predict the effect of a certain intervention on their disease state with more certainty.

• We’ve heard Tidepool’s Mr. Howard Look champion OAuth2 many times (most recently at November’s DTM), and Mr. Lane appeared to be an equally passionate advocate for it in healthcare. OAuth2 allows consumers to access without inputting passwords (the “Sign in with Facebook” or “Sign in with Google” buttons that occasionally appear when logging into a platform). Ten years from now, Mr. Lane envisions a world where fitness and genomics data will be included in the medical record, transferrable through OAuth2. “Data should work the same – log in with one password for any health app, and your data will be there.” This will enable the vibrant ecosystem of development and apps that Mr. Look so frequently refers to (though Mr. Look also lists JSON and RESTful URLs and APIs as prerequisites for such an ecosystem).
• In Q&A, an audience member asserted that the traditional ways of establishing the credibility of science is breaking down as technology and digital health solutions roll out so quickly – responses from the panel were mixed. Mr. Lane responded that if we slow down innovation because we’re worried about un-verified apps and slow regulatory, then “we’re missing the boat completely. It’s the FDA’s job to keep up with us.” Dr. Mega gave a more conservative answer, replying that she comes from a clinical trials background and all new apps should be held to the same standard. We don’t disagree, but we’d love to pick her brain for ideas on how to guarantee outcomes and value proposition in digital health – many feel that RCTs would simply take too long to conduct, such that technology would be outdated by the end, and that RCTs may be too old world, not accurately reflecting real-world performance of an intervention. This was a topic of debate at last week’s Digital Diabetes Congress.
• Quotable quotes from this panel:
  
  • “If data isn’t relative, then it’s not meaningful. When it’s not relative to other people, it means less. I want to know how I stack up to others. That’s super impactful.” – Mr. Lane
  • “Google has seven products that touch a billion users, and people use them because they provide value. I used the map feature to get here this morning. It’d be interesting if we could provide that same value in health.” – Dr. Mega
  •  “When I was in Iraq and Afghanistan, at the beginning, we were given tools to fight the insurgency. They were like tools that should be used to fight Russia. They were not the right tools. We are giving doctors a huge responsibility – they are becoming responsible for people’s health, but with the same tools as before. It is no longer fee-for-service, but becoming risk-based. But the tools are the same giant monolithic EMR systems built for the fee-for-service world. Until we give physicians tools to manage these incredibly sick patients with difficult social determinants, they’ll feel underwater and overwhelmed.” – Mr. Lane
  •  “The power of information is both awesome and scary. We work really hard to figure out what’s real and what’s fake. But we can’t slow down progress in information access, if we slow down because we’re worried about the credibility of science breaking down, then we’re missing the boat completely.” – Mr. Lane
  • “I have a background of clinical trials, and all new apps should be held to the same standard. If tools don’t improve outcomes and add value, then they’re really doing a disservice. Partnering traditional healthcare workers with developers is key to make sure digital solutions have clinical guidance.” – Dr. Mega
  • “I take the nerdy approach – I envision a developer ecosystem in 10 years, and it’ll be amazing. All of your data will be together to establish your identity, your quantified self, it’ll all be together. Fitness and genomics data won’t be separate from your medical record. That’ll all be transferrable through OAuth2. Health data should work the same way as other data. You should be able to log in with one password for any health app, and all data will be there. There will be so many developers in their basement that’ll blow your mind. And we can’t even imagine what that’ll be like. Because data equity will change everything.” – Mr. Lane
  • “I think the demand for the vibrant data ecosystem will come from the consumer, but they’re going to need to see some magic from solutions. There’s no magic right now when it comes to healthcare information. As soon as we do that, I think we’ll have an amazing amount of demand.”
  • “A lot of the time, we are healthy until we have a disease. How can we get ahead of the curve, and get as engaged as we are when we’re healthy as we are when we have a disease? Until then, it will be seen as separate box.” – Dr. Mega      
  • “I think the idea that information is somehow on its own going to help us do something proactive is not the right way to think about it. The device itself will never take us through the final mile. You can collect and organize data, but unless you activate it and make it useful to you, then you won’t see outcomes.” – Dr. Mega
  •  “Despair stinks when it comes to health, you stop caring until your jacket gets tight.” – Mr. Lane

How do we Prevent a Diabetes Avalanche?

Adam Brown (diaTribe Foundation, San Francisco, CA), Marie Schiller (Eli Lilly and Company, Indianapolis, IN), Sarah Mummah, PhD (IDEO, Stanford, CA), Phyllisa Deroze, PhD (Black Diabetic Info, Pompano Beach, FL)

Our own Adam Brown led an inspiring and actionable discussion on the diabetes avalanche facing our nation (and the world), touching on behavior change, urgency, nightmare stories of diagnosis, meeting patients where they are, changing our language, problem vs. solution-oriented research, and much more. Adam kicked off the discussion and grabbed the audience’s attention with some harrowing statistics: If everyone with type 2 diabetes was one country, it would be the world’s third largest, and while one person dies from a traffic accident every 25 seconds (1.25 million per year), someone dies from diabetes every six (!) seconds (5 million per year) – a good reminder for the self-driving-car-obsessed SXSW crowd. Furthermore, while a disease afflicting 14.1 million people in the country rightfully warrants a Vice Presidential moonshot (Dr. Biden also presented during this timeslot), there are twice as many people with diabetes in the US, and a shocking 86 million with prediabetes. Asked Adam, “Where’s the diabetes and prediabetes moonshot? And if we had such a moonshot, what would we do?” [You can view The diaTribe Foundation’s slides here].

• We need more solutions testing in prediabetes and type 2 diabetes; problem-focused research isn’t actionable enough. This theme came up again and again, since we all tend to narrow in on problems – what’s causing type 2 diabetes and prediabetes? Dr. Mummah, Ms. Schiller, and Adam were adamant that not enough things are being piloted and tried in prediabetes and type 2 diabetes. Dr. Mummah pointed out that solutions-oriented research is more powerful, since it gives data on what does/doesn’t work and allows for iteration – the difference between “why don’t people exercise in a community” vs. “does improving street lights on the sidewalks make people exercise more” is very powerful. Problem-focused research does not tell you what to do. Ms. Schiller also pointed out that prevention research studies have better recruitment and retention when they actually test an intervention. By contrast, natural history studies (like TEDDY) cost an enormous amount, may not give anything useful in the end, and retention/recruitment are harder. We may not have all the data on what causes type 2 diabetes and prediabetes, but that shouldn’t stop us from trying as many novel solutions as possible, learning from them, and iterating on solutions.
• Go offline. SXSW is a very tech-y meeting, but the need for in-person accountability came up repeatedly during the panel. Yes, technology is useful, but for something as personal as diabetes, it may not be enough. How can interventions blend the two, or better, personalize the approach depending on what someone needs?
• Panelists shared effective behavior change strategies for improving health: incremental steps and tiny wins. Dr. Deroze recounted her experience with a nutritionist many years ago – while she initially refused to overhaul her diet, she finally took the clever recommendations presented to her when the nutritionist met her in the middle, suggesting that she mix her bowl with a tiny bit of high fiber cereal and her usual fruit loops; over time, she made it to half and half, then mostly high fiber, than eliminated fruit loops completely. Meanwhile, Dr. Mummah emphasized the value of making behavior change countable (counting numbers of fruits vs. grams of sugar), positively framed (“Eat more vegetables” rather than “Eat less sugar”), and limited in scope (manageable). Adam emphasized that breaking down changes into tiny steps is key, whereas dumping a huge health change on someone may lead to him/her shutting down.
• Poor patient-provider communication without personalization was a common addressable thread throughout the discussion. Dr. Deroze recounted how, when she told her doctor that she was frequently urinating and thirsty all the time, he replied that she was exercising too much and needed to drink more Gatorade! Weeks later, she passed out in her bathtub and found out she had type 2 diabetes (and a blood glucose of 593!) at the ER. The poor communication continued as she was provided preliminary informational materials, which suggested to her that she was dying, rather than going on to live as a person with diabetes. Diagnosed as a young person, she could not relate to the pictures of elderly people in the brochures. A negative, un-personalized message is a truly bad combination for motivating change, and that is exactly what people with diabetes and prediabetes get all the time. 
• Ms. Schiller called for more attention paid to prediabetes at a population level. She emphasized that 90% of people with prediabetes don’t know they have it because there is no ingrained screening system. If she had $10 million to spend on prevention, she said she’d go to a company like Weight Watchers, a touch point for people who might be at risk for prediabetes, and ask them to partner and offer a screen for prediabetes. We think this is a brilliant idea! (Adam even pointed out that Weight Watchers is actually reimbursed in the NHS, no small feat! Could it scale?) One of the issues, which she illuminated in Q&A, is that no one organization has championed the prediabetes cause – ADA is focused on helping people with existing diabetes, and JDRF focuses on type 1. Dr. Deroze added that the average person doesn’t think diabetes is coming for them, much like she didn’t think it was coming for her. In one of her YouTube videos, she characterizes diabetes as the Apocalypse – it’s coming for you, and if you don’t become a zombie, then your neighbor does. This analogy may seem sensationalist and overboard, but it may take sensationalism to make prediabetes as urgent as it should be.
  
  • Asked Ms. Schiller, “what if we could give a screen to someone [with prediabetes], and know whether they could respond to coaching or a diet?” This sort of segmentation is one of the big promises of precision medicine and machine learning. Taking a patient’s genotype, phenotype, and behavior type, pairing it with a learning algorithm, and asking questions like – would this patient respond to coaching? How often should I check in with this patient to optimize outcomes? What first-line intervention should I try with this patient? What makes this patient tick psychologically and physiologically? This should ideally increase the effectiveness of and reduce inefficiencies in care.
• The full panel is included below, and here are some of our favorite quotes:
  
  • [Upon being diagnosed with T2D] “All the information I got was dismal. The images were a turnoff. All the information for African Americans suggested that they got more amputations and more blindness. All of the information made it seem like I had gotten a diagnosis and that I was dying rather than that I would be a person living with diabetes. This depiction prohibited me from learning how to manage diabetes.” – Dr. Deroze
  •  “One of the things that motivates a lot of us in this space is trying to help people live healthier lives. We are motivated by big, long term outcomes like bringing down rates of heart disease. But those are really big challenges. That’s what gets us out of bed. But if you are a person who needs to experience the solution or the intervention, suddenly those long-term outcomes aren’t enough to get you out of bed. Initially, it might be motivating. But two weeks or a month in, when you’re making a decision between a donut and a salad, you won’t think about how it will affect your long-term diabetes risk. We have to focus on the process and making the process rewarding in and of itself.” – Dr. Mummah
  •  “We often approach chronic disease with a view of avoiding long term negative outcomes. People hate hearing negatives and it’s hard to prioritize the long game. One of the things that we argue in diaTribe, and in my upcoming book (Bright Spots & Landmines) is why you should prioritize something right now. How does an in-range blood sugar make me feel today – more energy, better sleep, etc. If we gave people more action-oriented NOW messages, we could make a bigger difference.” – Adam Brown
  • “Many physicians are looking forward but end up using scare tactics. If you have impaired glucose or prediabetes, they’re saying that you should change because you’re going to progress to diabetes. But people just want to go back to where they were; where they didn’t have prediabetes. We need to bridge that, it’s a communication problem.” – Ms. Schiller
  • “The nutritionist wanted me to eat Raisin Bran, and I wanted Fruit Loops. I had chicken thighs, and she wanted me to eat chicken breasts. She made a plan with me to eat half chicken breasts and half chicken thighs, and when the thighs were out, to not buy them anymore. Now you’re talking! She told me to buy cereal with at least five grams of fiber and mix it with Fruit Loops or Apple Jacks, and when those were out, to not buy them anymore. Her advice changed my palate. She met me where I was.” – Dr. Deroze
  • “If you give people really big challenges, you undermine their confidence in being effective. If you can start with tiny behaviors that build, you set people up for success.”  – Dr. Mummah
  • “With $10 million, I’d definitely go offline. There’s a digital divide that still exists – many in the community don’t have access to the DOC (diabetes online community). Annually, I have workshops at a local library and I bring education that I’m getting from the DOC to my town community… I have the ability to see a nutritionist, but many in my community don’t. We spend a lot of time in church, so I say jump around. It’s a good way to burn calories and praise god. Walking at night is just not necessarily the easiest thing to do in my community. Where do you go? WalMart? Do two laps around before you do your shopping.” – Dr. Deroze
  • Read the full conversation below – highly recommended!

No More Apps: Why Reinventing Devices is Key

Jeff Dachis (One Drop, New York, NY), Stuart Blitz (Seventh Sense, Medford, MA)

An insightful panel discussion on reinventing devices touched on the payer landscape (barriers and strategies around it), why “hardware is hard,” why the future of devices is in the ecosystem (hardware + software), and why FDA is a “friend.” One Drop CEO Jeff Dachis and Seventh Sense’s Stuart Blitz (formerly of AgaMatrix) shared plenty of diabetes examples, arguing for a relentless focus on the user experience and the importance of building hardware that enhances the long-term value of apps. One Drop now has “more than 150,000” users globally, up from ~120,000 as of December when the company’s unlimited strips Premium model launched ($33-$39.95 per month, cash pay). We’re not sure how many are using One Drop Premium, but Mr. Dachis shared some valuable quotes on the payer environment (he’s right on, in our view) and why the company has approached reimbursement pathways in a measured way. Notably, One Drop has done several clinical studies and will be publishing data soon – no further details were shared, but assuming outcomes are positive, this should open up reimbursement pathways. Mr. Blitz also shared some stories from the early days of AgaMatrix, when the company was really ahead of its time and ran up against payer barriers. (He is now working at Seventh Sense, which has developed a user friendly, push-button blood draw device that is FDA cleared and self-administered.) See some of our favorite quotes from this session below.

• “Insurers’ financial interests are not always aligned with long-term health. 12-month premium renewals can drive decisions instead of a focus on longer-term health. I wish it wasn’t the case. I wish we could say we have the healthiest solution. That is not what insurers bet on right now. They bet on 12-month renewals. With that window, digital healthcare companies should focus on the cost-benefit from near-term savings. Going to market with near-term cost savings and longer term health benefits will allow companies to penetrate those plans more effectively and help insurers align with your interests. As we start to move into ACOs and outcomes-based payments, are we [One Drop] willing to go at risk? Absolutely. If we can find insurers that are willing to go to market with true outcomes-based payment solutions, we’re there all day long.” – Jeff Dachis in response to a payer representative from Anthem who asked the question. We loved this candor and the Anthem rep seemed to agree.
• “I think payers are a difficult group to navigate for digital healthcare companies. We are working with some of the largest payers in the world right now, but we chose a path where we could provide enough value to the end user that that they would be willing to pay for it. That meant getting to market at a price point that was competitive to what they are paying even with insurance. We looked at the payer, provider, and formulary channel as the final channel we wanted to go into vs. the first one... The third party payer system is super complex and very bureaucratic. If you choose that route, add 75% to the cost base to navigate that... We’ve completed several clinical studies, and as data is published, we hope to get on formularies over time. But you cannot get on formularies until you have evidence-based proof that your solution works, is safe, and effective to deploy at scale.” – Jeff Dachis
• “Payers are incredibly political and complicated. When we were in diabetes early on, our product was way better and half as expensive. Everyone said, ‘Shouldn’t you just clean up the market?’ The answer was NO at the time. Big payers have big contracts with J&J and Roche. J&J sells payers thousands and thousands of products. They’re not going to screw up the relationship for all these other categories. The market has changed since then. How can we be successful? You have to know how you are going to get paid and make sure you have a way forward.” – Stuart Blitz
• “At my previous company in diabetes [AgaMatrix] about 7-8 years ago, we envisioned what Jeff is building now. We were almost too early. We built an early prototype version. This was way before the trend...We were doing all this hard work building devices, in addition to the app that goes with it. I think there is something to that. It’s hard and it takes time; it’s not like consumer tech, were you code and release within two weeks.” – Stuart Blitz
• “The #NoMoreApps tagline is tongue in cheek; I think apps are really important, but apps are most valuable as part of an overall ecosystem: hardware plus an app. The companies doing that are going to be the most impactful on the overall ecosystem. I’ve advised tons of startups over last number of years. They say, “I’m going to create the uber app for healthcare.” That’s challenging to do on its own, because you have to engage with it all the time. You must give people hardware around it.” – Stuart Blitz
• “Hardware is hard, and we’ve made some mistakes. But above all, a digital healthcare company must be relentlessly focused on the user experience ... You need to be cognizant of all the pain points – price, packaging, how people feel about it, how many steps they have to go through – and you have to sand all the edges off and make it as frictionless as possible. In a notoriously fragmented industry, you can’t focus on the piece, you have to focus on the puzzle. If you’re not making the whole experience better, you’re just making another piece of hardware. Talk to users about every aspect of their journey. Relentlessly focus on the user experience before you build anything.” – Jeff Dachis
• “Regulation is your friend. Don’t look at the regulatory environment as a burden or hurdle. Being HIPAA compliant, FDA approved, CE marked is a meaningful and successful milestone that says “now you’re able to operate in the industry with a safe, and effective product.” I also recommend focusing on what parts of the product are regulated. We made a conscious decision about what’s in the app. We like to push a build on our app every three weeks which would be impossible to do if that component of our product platform was regulated. What part of this ecosystem do I want to have my friend with me (regulated)?” – Jeff Dachis
• “Sometimes the best way to get really clear feedback is actually to submit an application to FDA. They have to give you feedback. You meet with them in person. They’ll tell you, “We like this and this and we don’t like this.” The FDA wants to work with you to get medical hardware and new innovation cleared.” – Stuart Blitz

From Holodeck to Healthcare

Ashwin Pushpala (Sano, San Francisco, CA)

Sano CEO Ashwin Pushpala showed off new pictures of the company’s minimally invasive wellness CGM patch for people without diabetes. He did not comment on timing or accuracy (as usual), but we’ve confirmed with the company that a beta launch is now expected in 2017. The plan is factory calibration, very low cost in the “few hundred dollars per year” range, built-in Bluetooth to an app, and peer-reviewed clinical data before launch. Each sensor has just 24-hour wear, while the transmitter battery is reused and lasts seven days – an interesting approach. The beta launch timing is well back of the previous plans to launch in 2016, but there are now finally pictures on Sano’s updated website (see below). The low-profile, round, on-body patch houses the 24-hour wear silicon glucose sensor, which can reportedly be manufactured at high volume for lower costs. Each sensor contains hundreds of minimally invasive structures that read interstitial fluid (less than 0.5 mm each). The sensor is applied with an applicator, and the whole patch looks slightly larger than FreeStyle Libre in terms of on-body form factor. Mr. Pushpala was wearing the sensor on his inner forearm, a unique location we don’t usually see. The company remains on the “wellness” path, planning to bring CGM to people without diabetes via a direct-to-consumer model and under a lower regulatory burden (we assume Class I or Class II; discussions are ongoing with the FDA). Mr. Pushpala mentioned that “several studies” are ongoing in different populations, and we assume prediabetes and obesity are the logical targets. Some of the data displays are novel (e.g., “350 / 1000 minutes in target”), but the app also shows the more familiar current and historical glucose and trend information (pictures below). We’re not sure how a real-time CGM that reports a number could be considered a “wellness” device, given the obvious overlap with Class III CGM devices indicated for diabetes. (Perhaps the actual real-time number will be stripped out in the consumer version, leaving only trends? Plans for this device continue to be challenging to understand.) Sano does plan to pursue a diabetes indication, but not for this first consumer wellness product. It’s unclear if accuracy has improved; in our previous write-up, the sensor’s MARD was reportedly ~15-20% vs. fingersticks, though the 24-hour sensor wear seems like an accuracy limitation.

• Overall, we love the idea of bringing low-cost CGM to the masses, but are not sure of the accuracy, manufacturing, regulatory path, clinical efficacy, execution, and value-add/willingness to pay for the target market. We assume one of the larger players will do this better, though who does it faster will be interesting to watch. We’ll be interested to see what kind of outcomes are possible with Sano’s device in non-diabetes populations, assuming the peer-reviewed data tested the system in this way. Some precision and accuracy are still needed in a non-diabetes population to drive meaningful behavior change – is an MARD of ~15-20% good enough? How much will the average consumer be willing to pay out of pocket for Sano? Assuming Sano is in the “few hundred dollars per year” range, it would translate to only about $30 per month (~75% cheaper than FreeStyle Libre). Will people with diabetes try to get on Sano, especially if they cannot afford current CGM? How will the company label the product to avoid liability? Will the FDA consider this off-label use?

• Sano has raised ~$20 million to date (last round in summer 2015), and the team is now ~25 people (up from 20 as of 2015). It’s hard to imagine such little capital and such a small team could bring a novel CGM device to market, but perhaps Sano’s technology allows for more scalability and lower development costs. The company has an interesting investor base that includes True Ventures and Felicis Ventures, both majority owners of Fitbit, along with Intel Capital, First Round, Elevation Capital, Floodgate, and Rock Health. See our previous write-ups here (fundraising) and here (Rock Health’s 2015 Summit).

Collaborative Innovation in the Digital Health Age

Alex Gorsky (Johnson & Johnson, New Brunswick, NJ), Ginni Rometty (IBM, North Castle, NY), Clay Johnston (Dell Medical School, Austin, TX)

J&J and IBM CEOs Mr. Alex Gorsky and Ms. Ginni Rometty’s discussion with Dell Medical School Dean Dr. Clay Johnston spanned a host of hot topics in healthcare today: Personalized/value-based care, prevention (especially in the workplace), security and privacy (including Blockchain), partnership, and artificial intelligence.

• Mr. Gorsky characterized the healthcare climate today as “Dickens-esque: The best of times, the worst of times”. On the one hand, health systems around the world are being overwhelmed by surging demographics looking for care, but at the same time, incredible technologies and service models are being developed. Therapies predicated on data are making it possible to personalize treatment, help patients understand outcomes, and shift toward new reimbursement models. Ms. Rometty aptly chose IBM Watson’s collaboration with Medtronic to develop Sugar.IQ as an example of how data facilitates predictive analytics down to an n of 1: “If I have diabetes and I have this app, then it can predict hypoglycemia in advance. But we’re taking all of the data. It’s also based on my behavior – what I eat and do – they can get confident enough to tell me that they can keep me out of the hospital. They can provide care at a certain level. That’s toward value-based care, and I think that’s already happening today.” (Sugar.IQ was expected to enter a wider preview late last month, with a full launch slated for May-October.)
• Mr. Gorsky on prevention: “We think a lot about the interception and prevention of diseases, whether its cancer, cardiovascular disease, Alzheimer’s disease – we’re getting to these diseases in the ninth inning. We need to get there in the first or second inning.” Though he didn’t mention diabetes specifically, we know J&J is thinking about the earlier stages of disease; in 3Q16, it announced plans for a trial investigating the potential cardioprotective effects of SGLT-2 inhibitor Invokana in people with prediabetes.
  
  • As large companies invested in healthcare ­– IBM has 380,000 employees and has labeled healthcare as its moonshot, while J&J has 130,000 employees – Dr. Johnston suggested that they may be leaders in the innovation of workplace wellness initiatives. Mr. Gorsky responded that J&J has been focused on prevention since the 1900s, and tries to think about healthcare as an investment in its employees: “For every dollar we put in for prevention and wellness, we think we get $4 on the other side.” Further, ~90% of J&J employees know their heart rate, blood pressure, blood glucose, as part of an initiative to encourage them to know and think about outcomes. The company has also invested in health facilities and wellness (fitness center, nutritious options) on site, so that employees are more engaged with their health and feel as though the organization is taking care of them. Ms. Rometty echoed Mr. Gorsky’s perspective: “We want a healthy workforce so we can be at our best. We spend over $1 billion per year in healthcare, and we want our employees to take part in understanding the data for themselves, be it for chronic disease or acute issues.” IBM even offers Watson Health’s services to workers for free in some cases. Workplace wellness is a prime example of aligned initiatives and why many digital health companies seem to be going this route (Omada, Livongo, etc.).
• Ms. Rometty noted that IBM’s brand, when it comes down to it, stands for trust, and that Blockchain will do for trusted data transactions what the internet did for communication. Dr. Johnston pushed her on the privacy issue, claiming we need an ATM-like system in healthcare, where information on patient allergies, medications, and previous treatment is all stored in the cloud and shared. Ms. Rometty countered that security is the number one issue for healthcare companies – patient data must be owned by the patient and only shared when the patient desires, and the cloud must ensure that “your insights are your insights. Otherwise you won’t get very far.”
  
  • “Blockchain would make any supply chain more secure. The analogy is that Blockchain would do for trusted transaction what internet did for communication. It allows anything – device, money, data ­– to be transferred between individuals and trusted. It will have a rapid adoption. It is already at over 400 clients with work underway. In healthcare, it’ll be, one, around compliance. How to assure compliance is adhered to? Second is just the supply chain of anything you ship anywhere throughout medicine, which has so many moving parts. What you can get is supply chain visibility. Also, in an EMR – who sees what part of the medical record. Oh, and clinical trial matching is the other area.”
  • What on earth is blockchain? IBM Watson Health partnered with the FDA in January on blockchain. According to that press release: “By keeping an audit trail of all transactions on an unalterable distributed ledger, blockchain technology establishes accountability and transparency in the data exchange process ... IBM and the FDA will explore how a blockchain framework can potentially provide benefits to public health by supporting important use cases for information exchange across a wide variety of data types, including clinical trials and “real world” evidence data ... IBM Watson Health and the FDA plan to share initial research findings in 2017.”
• According to both CEOs, partnering is critical to success in healthcare, reminding us of AOL co-founder Steve Case’s talk on Day 3 of SXSW. Mr. Gorsky said it’s almost impossible to not be partnering, and Ms. Rometty added that healthcare problems can’t be solved alone. In fact, according to Mr. Gorsky, ~50% of J&J’s innovation is done externally. Collaboration is key in digital health, since disparate players – e.g., drug vs. device vs. software – often need to work together to drive the best solutions.
• The panelists were optimistic about AI, but Ms. Rometty wished that it were called “augmented intelligence.” “The fact is, people make the right decision one-third of the time, an ok decision one-third of the time, and a wrong decision one-third of the time. That is statistically true in every field. And with all of this extra data being generated, people will be subjected to cognitive overload. And yes, pieces of jobs will be done differently. Doctors want to spend more time doctoring than gathering data and putting it together … Mostly, you’ll be able to do better and more things, and it’ll assist man.” Mr. Gorsky added that AI is great, but healthcare also requires the human element. “AI can help make good surgeons great surgeons, and can make the system more effective so physicians have more one on one time with patients.” From the physician perspective, Dr. Johnston concluded that providers need help to be more effective and that they’re distracted with sub-par technologies today (i.e., EHR), and that AI can help, but the placebo effect driven by the provider-patient relationship is very real – “the emotional state of the patient impacts outcomes more than the skill of the surgeon.”

Embrace Digital Disruption to Reinvent Healthcare

Nick Reddy (Baylor Scott & White Health, Dallas, TX)

Mr. Nick Reddy of Baylor Scott & White Health led a panel asking why healthcare has lagged behind other industries in adopting new, disruptive technologies. Proteus’s Ms. Molly O’Neill hit the nail on the head when she said products have to follow the “life flow” and the workflow to scale. While this statement may feel obvious, it’s often overlooked amidst the hype. Is the innovation able to save time? Does it fit into people’s days? How easily can it be adopted across a single setting? Meanwhile, when asked about the future of artificial intelligence, Pager’s Dr. Oscar Salazar (the founding CTO of Uber) said that if he had to say which industry he’d “bet the house on” with AI, it’s “definitely healthcare.” He noted that he’s much more afraid of human stupidity than AI. Still, he believes that empathy is going to be the hardest part to simulate in AI. “It’s easy to simulate or emulate intelligence. Empathy is another story.” (As a counterpoint, there are anecdotes of people referring to AI chatbots as “he” or “she” and feeling like they “understand them.”) We are on board that AI could have disruptive power in diabetes and healthcare, especially in distilling data and reading images. Read on for some quotable quotes from the panel.

• “If you’ve seen one health system, you’ve only seen one health system. You need physician champions and someone invested in how care is delivered [to scale].” – Ms. O’Neill
• “To get consumers to use something, you have to have a ‘wow’ factor. This isn’t ordering a pizza. You have to have a high level of assurance about the trust and quality associated with the product.” – Mr. Jon O’Sullivan (PediaQ, Dallas, TX).
• “I think you will see robots performing common surgeries, like appendectomies. It’s a machine you can train. It’s a task that’s very repetitive.” – Dr. Salazar

A #)$%(*% Crisis: Training the Newest Gen of Docs

Susan Cox, MD (Dell Medical School, Austin, TX), Sarah Kim, MD (UCSF, San Francisco, CA), Bon Ku, MD (Thomas Jefferson University, Philadelphia, PA), Erin Kane, MD (Johns Hopkins, Baltimore, MD)

Dell Medical School’s Dr. Susan Cox led a diaTribe-organized panel featuring UCSF’s Dr. Sarah Kim, TJU’s Dr. Bon Ku, and Johns Hopkins’ Dr. Erin Kane on the deficiencies in and opportunities for medical education. Panelists emphasized that medical school should focus more on cultivating crucial interpersonal skills (collaboration, empathy) and knowledge beyond the physiological aspects of disease and medicine, such as even basic management training.

• Dr. Cox kicked off the panel by describing the “perfect storm for a crisis” in medical education: In 2012, ~50% of people in the US had at least one chronic condition, and ~25% had multiple. Yet medical education – usually consisting of two years of basic science followed by two years of in-patient training and guiding students toward “sick care” – has changed little. Dr. Cox said current education models need more emphasis on ambulatory training, advocacy, cultural diversity, information management, teamwork, and leadership. There’s no way that providers can keep up with medical knowledge, which doubles every 3.5 years (as of 2010), and by 2020, is expected to double every 0.2 years (source). Plus, with knowledge easily accessible online, the physician skillset must change drastically to include teamwork, collaboration, and leadership.
• An impassioned Dr. Ku called for a complete overhaul of medical design and education, all the way down to who gets accepted in to medical school. On the latter, he believes that looking at GPA and MCAT scores alone may be too limited. Instead, he said students should be admitted based on communication skills and other personality traits, “not whether or not they got a C in physics freshman year”. Dr. Kane later suggested that communication skills can and should be taught, but we imagine that a strong baseline level would make it far easier to learn. Once students are in medical school, Dr. Ku advocated for a shorter incubation period and less rote memorization; he noted that he spent 15 years studying to be a doctor, which he views as too long because of the speed at which healthcare changes. Instead, he would’ve liked to be taught more creative problem solving – “we’re taught very binary solutions and we don’t do well with ambiguity. But a lot of events in healthcare don’t have binary solutions.” In response to a Q&A comment about a medical curriculum consisting of virtual lectures and apprenticeships, Dr. Ku said models involving digital education can be disruptive. He had the most feedback on the role of the doctor: above all, he would like to see the healthcare system design “with patients,” not “for them” and discussed TJU’s human-centered design program. His dream is to have all providers care about and advocate for the culture of health in their communities – of course, the motivating incentive for this outside-of-the-clinic work would have to be a greater shift to outcomes-based models. Lastly, he asserted that doctors don’t need to quarterback all treatment, citing blood pressure management as an example; they can serve more as team players and care coordinators, and medical school needs to prepare them for that. Presumably nursing and community health worker training would encompass more management.
  
  • Two Close Concerns alums, Dr. Kane and audience member Ms. (soon to be Dr.) Lisa Rotenstein were worried about medical student burnout. The resident work week is capped at 80 hours, but Dr. Kane held that it’s more than just the cumulative hours: “When you let a patient down, you feel bad, and that feeling contributes to burnout. We need to create more supportive spaces, encourage them, and help them take care of patients more safely.” Ms. Rotenstein noted that empathy often declines from the start of medical school to the end, and that decline is correlated with mental wellness and depression. Dr. Kane responded that the value add and cheap labor aspect of residency needs to be reconsidered. Hear, hear!
  • Dr. Kane doesn’t believe that intensive technological education is needed in medical school, but that technologies should be rolled out in the context of physicians’ workflow.  This reminds us of comments from Onset Ventures’ Mr. John Ryan at Digital Diabetes Congress 2017 about how so many digital health technologies take the best IT innovations and force them into patients’ and providers’ lives, as opposed to first asking, “what does the patient o provider need?” That said, Drs. Kim and Ku highlighted the value of education around a tool like telemedicine, which could revolutionize the way care is administered, particularly to remote areas.
• Dr. Kim championed the idea of seeing eye to eye with patients, which would require teaching collaborative exercises and skills like motivational interviewing in medical training. Though a physician may have a goal of getting a patient to an A1c below 7%, he/she must also be able to understand that the patient simply wants to get through the work day without a severe hypoglycemic episode and adjust care accordingly. Dr. Ku pointed out that much of the problem stems from the fact that so much of patient-physician interactions are done within the context of the clinic – “we don’t see the human behind that person,” which makes it difficult to empathize. Dr. Kim delivers care to many publicly-insured patients, where affording copays or getting to appointments remain a challenge. “In med school, we’re not taught to think about these types of determinants of health. As we move to chronic disease, it will be important to think not just of what goes on in doctor’s office, but what happens outside as well.”
  
  • Dr. Kane spoke about two “gaps” relevant to communication and interpersonal skills: (i) In the first- second-year of medical school, there is training in communication (i.e., motivational interviewing), but when doctors-in-training hit their third year and beyond, there’s much less tactical training. This is detrimental because people solidify their practice habits during this period and could use some reinforcement. (ii)  “Every day in the ER, I see a patient come in with high blood pressure. They couldn’t take medicines, couldn’t afford them, or didn’t understand why they are important. There’s a gap between writing prescriptions, which used to be enough, and getting from prescription to health. The task of this generation of physicians is going to be bridge gap between prescription and health.”
• Panelists felt that medical students need a broader understanding of the numerous factors influencing the healthcare system. Dr. Kane advocated for education about reimbursement, financials (both from the patient and hospital perspectives), business, and policy. A deeper understanding of these issues will help a doctor see eye to eye with a patient and advocate for social and political change that would help more people get access to quality care. 
• Here are a few of our favorite quotable quotes from the panel:
  
  • “I probably spent 20 hours memorizing the Krebs cycle in college, whereas the patient can pull it up in five seconds on a smartphone. No longer is knowledge my super power.” – Dr. Ku
  • “In med school, we’re not taught to think about these types of determinants of health. As we move to chronic disease, it will be important to think not just of what goes on in doctor’s office, but what happens outside as well.” – Dr. Kim
  • “My dream is to have physicians that care about the culture of health in their communities. We are not doing a good job of that. Those that care about their communities go into family medicine, internal medicine, pediatrics. Every doctor should have some skin in the game.” – Dr. Ku
  • “So much of our interactions are done in the context of the clinic, we don’t see the human behind that person...we don’t involve patients enough in the process. We design stuff for them, not with them.” – Dr. Ku
  • “You go to work every day, and when you did well, you feel good, and when you let a patient down, you feel bad, and that feeling contributes to burnout. We need to create more supportive spaces, encourage them, and help them take care of patients more safely.” – Dr. Kane
  • “We need to redesign how we choose future doctors. If you’re 24 or 25 and don’t know how to communicate yet, then maybe you shouldn’t be a doctor. We look at GPA and MCAT, not personality. I have some students who should not honestly be communicating with real people. If you’re an excellent communicator but got a C in physics freshman year, you will still likely be a good doctor.” – Dr. Ku
  • “Being a physician is not just prescribing therapy or ordering a blood test, but being there for a patient in a more holistic way.” – Dr. Kim
  • “There’s a gap between writing prescriptions, which used to be enough, and getting from prescription to health. The task of this generation of physicians is going to be bridge gap between prescription and health.” – Dr. Kane
  • “I wish I had learned more creative problem solving – we’re taught very binary solutions and we don’t do well with ambiguity. But a lot of events in healthcare don’t have binary solutions.” – Dr. Ku
  • “Med school teaches you to think there are doctors, patients, and hospitals, but there are other really huge forces in healthcare, and to be equipped to make change, med students need to understand the broader health system.” – Dr. Kane

Use Behavior Change Science to Improve Your Health

Daniel Goldstein, PhD (Microsoft, New York, NY), Michael Dermer (WellTok, New York, NY)

In a packed session entitled “Use Behavior Change Science to Improve Your Health,” Microsoft researcher (and a member of Britain’s “Nudge Unit”) Dr. Daniel Goldstein explained why it’s so hard to avoid negative behaviors in the present and introduced a virtual tool to help. “It’s easier to make yourself do things than to not do things that you want to do. The present self wants, but the future self doesn’t want, and it’s an unequal battle because the present self is in control right now. The future self doesn’t have an advocate when you reach for a cigarette.” Current research even suggests that we view our future self as a sort of a stranger – we can’t imagine our future selves or we have a false belief about what the future will hold. He first highlighted the Band-Aid solution: commitment devices. These allow the present self to restrict the future self’s choices. “Ever thrown away perfectly good cake?...With a writing workshop, you know you won’t learn anything, but you have to meet deadlines. It’s called paying for the deadlines. We take classes and go to school – unless we have a deadline that we pay for, we wouldn’t learn the material. Commit to doing something publicly. Limit the amount of money you’re taking to a casino. These all prove that our future self needs help.” Commitment devices work, but when the device goes away, Dr. Goldstein said, the behavior goes away. In his research, he tries to connect people to their future selves using a simulator. If we can see how actions today – such as continuing on a trend of weight gain or spending money – will impact our physiques and wallets months or years down the line, then Dr. Goldstein’s data shows our behavior is more likely to change. In one experiment, he showed participants a sliding scale of spending money now vs. saving money for retirement, along with two virtual depictions of the individual: one old and one young. As the bar slides from spending more to saving more, the younger self looks sadder, and the older self looks happier. Participants allotted 50% more money to their 401(k) when they could see the older self vs. when they couldn’t.

• We see strong applications for this research in prevention efforts – as Dr. Phylissa Deroze pointed out in a previous SXSW panel on preventing a diabetes avalanche, few believe type 2 diabetes is coming for them. Could this visualization help connect today actions to future health consequences?

• Meanwhile, WellTok’s Mr. Michael Dermer discussed financial incentives for good health, which Dr. Goldstein noted is really a commitment device. Mr. Dermer showed survey data that 91% of people would change their behavior if rewarded. Surprisingly, this wasn’t limited to young and financially troubled people – 78% of respondents with income over $200,000, and 85% over age 55 said they would respond to incentives. He believes that we will soon see major payers like Aetna and Cigna soon offering rewards, and if they don’t, “it’d be difficult to compete. Much like if we started a hotel or airline today without a loyalty program.” UnitedHealthcare’s Motion program (read our CES coverage) is one example of how payers are already incentivizing healthy behaviors with money: Motion is offered through employers and gives up to $4 per day and $1,500 per year for employees that that meet one or more of the “FIT” activity goals measured via a wrist-worn tracker: (i) Frequency (take 300 steps in five minutes, 6x per day, spaced an hour apart); (ii) Intensity (take 3,000 steps in 30 minutes once per day); and (iii) Tenacity (>10,000 steps per day). The program recently partnered with Fitbit so that Fitbit would provide the wrist-worn activity tracker.

Interactive Keynote: Adam Grant

Adam Grant, PhD (Wharton, Philadelphia, PA)

Brilliant author and Wharton professor Dr. Adam Grant gave an inspiring keynote exploring what it takes to be an original thinker, how to better communicate original ideas, and how to hire for an organization. Here are a few of our favorite quotes from the dynamic talk, which you can see for yourself here.

• “Culture fit becomes over time a proxy for group think. As the organization gets bigger, you’re no longer now built on a disruptive idea, you’re now dealing with disruption from all angles. And when you say we want people who are going to live our values, be passionate about our mission, that weeds out diversity of thought and you end up bringing in people who look and think the same way that everybody else already does. Some companies use culture fit as code for ‘do I want to have a beer with this person’ or ‘do I want to hang out with this person,’ which is not good for stimulating original thought…We all ought to step back and ask ourselves ‘what is absent from our cultures, and how do we bring in people with the cultural backgrounds, the functional experience, the demographic characteristics that are different from what’s dominant here? It can really add value and stretch our cultures in directions it needs to go.”
• “We don’t have a lack of creativity in world, but a lack of people who know how to champion their own ideas. Once you have an idea, what do you do?”
• “Take leader and managers out of the gatekeeper role. Instead, involve peers who are more likely to see potential in the wild, the maybe weird, but the slightly worth-a-look.”
• “It’s hard to get good ideas across to other people. It takes 10-20 exposures to a new idea before someone starts to get used to it.”

Navigating a Rapidly Changing and Connected World

Steve Case (Revolution LLC, McLean, VA)

According to AOL co-founder Mr. Steve Case, the “ignorance is bliss” mindset pervasive throughout the “second wave” of internet will not work in the “third wave.” For context, the ‘first wave” consisted of getting everyone on the internet so that they couldn’t live without it; the second wave consisted of building apps on top of the internet; and the third wave (the namesake for his book, The Third Wave: An Entrepreneur’s Vision of the Future) consists of integrating the internet in “seamless, pervasive, invisible ways throughout our lives to disrupt sectors of the economy that haven’t changed much, such as education, healthcare, and agriculture.” Whereas in wave two, entrepreneurs benefitted from thinking outside the box and coming up with great standalone concepts (e.g., Snapchat, Facebook and PayPal), in this third wave, Mr. Case believes that innovation will come from within the sector, from a deep understanding of the complex ways in which doctors and hospitals function – “without this, you probably won’t be able to revolutionize healthcare.” He also believes that figuring out partnerships will be a hallmark of the age of big data and connectivity. We are already seeing major partnerships in diabetes – Verily with Dexcom, Sanofi (Onduo), and Novartis; IBM Watson with Medtronic, Novo Nordisk, and ADA; and countless others – and hope that this can drive software, hardware, therapeutic, and business model innovation more quickly than solo efforts.

Accelerating the Race to Self-Driving Cars

Danny Shapiro (NVIDIA, San Francisco, CA)

In a packed session on self-driving cars, NVIDIA’s Danny Shapiro (a chip and software company) reminded us of just how many parallels there are to diabetes technology and closing the loop. We include some of our favorite quotes and points below, followed by implications for our field. Separately, it struck us as fascinating that a primary argument for self-driving cars is to save the 1.25 million people killed in traffic accidents each year (WHO). By comparison, 5.0 million people die from diabetes each year (IDF) – 4x the problem! – and it doesn’t receive nearly as much media attention as self-driving cars.

• “It’s mindboggling how far you can apply deep learning and AI [artificial intelligence]. You can apply it to pretty much any industry. In the automotive industry alone, eight car companies now have AI Cars: Audi, Tesla, Mercedes, Baidu, and Volvo.
  
  • How will diabetes technology companies leverage AI? AI is a hot topic at SXSW, and most believe it is going to revolutionize pretty much every industry, including healthcare. Where will it be implemented best in diabetes? Medtronic and IBM Watson are potential frontrunners here with Sugar.IQ as are any of Verily’s diabetes partners (Dexcom, Sanofi (Onduo), and Novartis). Still, it will be interesting to see how AI is used, who has the best data to train the machine, and how it is implemented in products.
• “Moving forward, the brand of a car company is going to be based on how safe their self-driving car is. Some car companies have huge software teams, some don’t. Our company [NVIDIA] has transformed too. It’s all about software now.”
  
  • Will the success of pump companies closing the loop be based on how safe their systems are?
  • Will diabetes device companies’ success be increasingly based on how good their software is?  How much upside is there from better hardware vs. better software, especially in CGM and insulin pumps? Which company has invested the most in great software?
• “Volvo has come out and said, ‘we are responsible for anything that happens when our car is in self-driving mode.”
  
  • Will pump companies be held responsible for things that happen while patients are under hybrid closed loop? Since hybrid closed loop systems will come to market first, who will take liability when patients interfere too much (e.g., a big frustrated correction bolus on top of increased basal insulin)?
• “With self-driving cars, ERs are not going to need as many surgeons. Mechanics are not going to have as much business. There won’t be as many speeding tickets.”
  
  • What are the unintended implications of AI and closed-loop insulin delivery for endocrinologists, diabetes care providers, and companies? Whose role will diminish and how quickly? Many speakers at SXSW have emphasized that technology is about extending human potential rather than replacing it. This is certainly true for the foreseeable future in diabetes, but where will it be in 10 or 15 years?
• “Government will have to create a bar – this is how safe you must be to be on the road. Is it going to be perfect? I don’t think anything is perfect. But it’s going to be safer than a human... Seat belts, air bags will be around for the foreseeable future. You still see Google’s self-driving cars getting hit all the time.”
  
  • How safe does an automated insulin delivery system have to be? Will systems be approved as fast as Medtronic’s MiniMed 670G? Has the FDA set a minimum bar? How long will it take before closed-loop is a widely accepted technology – similar to how long it took seat belts and airbags to be widespread?
• “The processing has to happen onboard the car. If you connect to the cloud to see if that is a pedestrian, it may be too late.”
  
  • How should devices leverage the cloud vs. storing algorithms and smarts on the device? What are the pros and cons from a safety and speed perspective?

Panel Discussions

How do we Prevent a Diabetes Avalanche?

Adam Brown (diaTribe Foundation, San Francisco, CA), Marie Schiller (Eli Lilly and Company, Indianapolis, IN), Sarah Mummah, PhD (IDEO, Stanford, CA), Phyllisa Deroze, PhD (Black Diabetic Info, Pompano Beach, FL)

Adam Brown: One of the things that came up in our discussion was this idea of problem-oriented vs. solution-oriented research. Do we need more data about what causes prediabetes and type 2 diabetes, or do we need to try more solutions? If you had a huge budget and could do whatever you wanted in diabetes, what would you focus on?

Dr. Mummah: From my perspective, we don’t necessarily need more data. We need to be testing solutions. There is a subtle distinction between problem- and solution-oriented research. In a problem-oriented research paradigm you ask questions like “Are children who live in communities with fewer parks and green spaces less physically active?” In solution-oriented research, instead, you ask “Does adding green parks increase physical activity or does limiting advertising to children on TV reduce obesity?” For example, if you look at neighborhood safety, it makes sense that you would be less physically active if you feel unsafe. You’re less likely to go outside your home if you’re likely to be mugged or putting yourself in danger. A solution-oriented approach to this problem would say, “Let’s increase street lighting, let’s hold a music festival one evening every month to bring people outdoors in the evening. Let’s create a neighborhood watch program.” Then if the experiment works, you have a solution that’s ready to implement and scale. It’s a subtle shift, but it’s powerful.

Ms. Schiller: I spent a year looking at type 1 prevention. The question was posed around whether we need more data on what causes diabetes. Do we know enough about type 1 diabetes to start an intervention study? No, but does that mean we shouldn’t start one? As we started the research, we figured out we don’t know enough. But if you an intervention, you’re actually learning about the disease along the way. It’s fascinating when you get in it. When we started asking about intervention studies, people thought recruitment may be a problem – people are at risk but there’s not enough research there to convince them to join. But it’s the opposite; intervention studies actually have an easier time recruiting and maintaining people in the study. When you start a natural history study there’s not an incentive to stay in the study – year one comes, year two comes, year three, and the dropout rate is a huge problem. But in this case it’s “Huh, look at this intervention, let’s see what the data keep suggesting.” At the end of the day, with the money, the intervention studies are an impactful way to learn. You don’t have to wait the 10 or 15 years to start the intervention. You can achieve both of those goals in one setting. Take the TEDDY study, The Environmental Determinants of Diabetes in the Young that was started in 2004. It’s funded by NIH and costs hundreds of millions of dollars – it’s been years and it’s questionable about what the impact of new learning has been from this study. It’s an important study that’s being done. But what if we didn’t design that study right? At the end, we will still not know what to do.

Adam: And you told me before that the people who do problem-oriented research are not the right people to do solution-oriented research.

Ms. Schiller: Even as you get into the intervention, there is a lot of analytic power surrounding the dataset. If you look at the individuals who are doing the problem-research, they don’t have the skillsets to mine the studies and manipulate the randomized controlled arm. Versus when you do interventions. The work we did at the Helmsley Charitable Trust will lead to the largest study in type 1 diabetes. And we have brought in individuals from the outside who like to do intervention work.

Adam: Dr. Deroze, we had conversations about when you were diagnosed with diabetes and what that was like. You mentioned remembering that the healthcare system dumped a lot of problems on your plate. Can you talk about what that was like and what you wish had been done differently – what solutions had been offered? 

Dr. Deroze: I was diagnosed with type 2 diabetes in 2011. I told the doctor I was frequently urinating. He said it was because I was working out frequently and I need to drink more Gatorade. I then passed out in the bathtub and went to the ER. The triage nurse asked me how long I had been diabetic. It was the first time I had heard about that, but my blood sugar was 593 mg/dl. All the information I got was dismal. The images were a turnoff. All the information for African Americans suggested that they got more amputations and more blindness. All of the information made it seem like I had gotten a diagnosis and that I was dying rather than that I would be a person living with diabetes. This depiction prohibited me from learning how to manage diabetes.

Adam: We often approach chronic disease with a view of avoiding long-term negative outcomes. We all hate hearing negatives and it’s hard to prioritize the long game. One of the things that we argue in diaTribe and my upcoming book (Bright Spots & Landmines) is why you should prioritize something right now. How does an in-range blood sugar make me feel today – more energy, better sleep, etc. If we gave people more action-oriented NOW messages, we could make a bigger difference.”

On a big picture note, if someone gave you $10 million right now, what would you invest in or test?

Dr. Mummah: I was struggling with this question, but ultimately thought about how we might help people eat healthier. That has a disproportionate effect on diabetes and obesity compared with other health behaviors. But healthier eating is really complex. It’s a range of lots of different behaviors. Who you choose to go to dinner with. What your company serves for lunch. What snacks your company has. What grocery store you go to. If we’re going to be successful in changing behaviors, we need to be more actionable and specific for folks. I have definitely had moments where I felt like I didn’t know where to start. One strategy is making an action really countable – like counting grams of sugar or fat that you are consuming. But that can be confusing when you’re cooking for a family. What does that entail? Counting pieces of fruit or vegetables is easier. Framing matters too. If you frame something as positive, people don’t feel like they are punishing themselves – eat more vegetables. Finally, restricting the scope helps. At Stanford, we tried to hone in on vegetable consumption and created an app around vegetable consumption.

Ms. Schiller: I’ve been thinking about this too. That’s a lot of money, and you need so much to make an impact. First, I’d try to get matching dollars to my $10 million! I’d go to government groups and try to get a 1/10^th of that, and then go to another group and try to get more. In DPP (diabetes prevention programs), we know they touch people, but not as many as we need. We need collaboration, those extra media and distribution channels. I would, for example, go to Weight Watchers, a touch point for people worried about their weight, give them a lot of money, and ask them to partner and screen for prediabetes. In weight loss surgery, you prevent and even cure diabetes in some instances, but none of these companies are looking at preventing diabetes. Someone like Weight Watchers, with massive installed groups motivated to make change that are just not educated or motivated, would be a good starting place.

Adam: Weight Watchers is actually fully reimbursed by NHS, which is pretty impressive. And I think that connects nicely with the weekly groups that changed your diabetes, Dr. Deroze?

Dr. Deroze: After I was diagnosed, I went to a weekly support group. In just nine months, I started managing diabetes with only diet and exercise for 2.5 years, until I moved to the Middle East. With $10 million, I’d definitely go offline. There’s a digital divide that still exists – many in the community don’t have access to the DOC (diabetes online community). Annually, I have workshops at a local library and I bring education that I’m getting from the DOC to my town community. For example, we have a “learn the plate” method day. We have a buffet, and they fix their plate with ox tails, chicken wings, collard greens, corn bread – whatever they normally do. Then I teach them how they should do it. I have the ability to see a nutritionist, but many in my community don’t. We spend a lot of time in church, so I say jump around. It’s a good way to burn calories and praise god. Walking at night is just not necessarily the easiest thing to do in my community. Where do you go? Walmart? Do two laps around before you do your shopping.

Adam: Yes, I think the way you phrased it to me was, “people need nuggets”: personalized advice that applies to their circumstances. I had this moment over the holidays where I realized the limitations of technology, poor design, and lack of personalization. I gave my uncle a Fitbit; he has prediabetes and I thought this would be a useful, motivating tool. But he had never downloaded an app in his life. I had to make him an iCloud account, make him a Fitbit account, and set the whole thing up. He had no idea how to do it. Then when he put on the device, his heart rate was so high that Fitbit was counting steps even when he was not moving. I have so much faith in technology to solve huge problems, but in that moment, I realized that the future will need a mix of approaches.

Ms. Schiller: I like the offline experience too. I won’t go online to talk to someone about my disease. I’ll call Adam, for example. Making the internet personal is hard to do. On the technology point, I think we’ll eventually get smarter. It would be really interesting if people could put on CGM for a week and could see what their glucose is doing. Just having that information from a few days would be powerful.

Adam: Dr. Deroze, speaking of advice that is personalized, what about the story you told me about the young nutritionist?

Dr. Deroze: I went to a nutritionist many years ago, and it helped. I’ve lost 85 pounds since. But at the time, she was giving me all these ideas and I kept saying no. Her mentor interrupted me frustrated and told me I was resistant to everything! The nutritionist wanted me to eat Raisin Bran. But I ate Fruit Loops. I had chicken thighs in the fridge, and she wanted me to eat chicken breasts. Eventually, the young nutritionist told her mentor to hold on and approached the situation differently. She made a plan with me to eat half chicken breasts and half chicken thighs. When the thighs were out, she told me not to buy them anymore. She told me to buy cereal at the store – any cereal I wanted – but it had to have at least five grams of fiber. I had to mix a little bit of it with Fruit Loops or Apple Jacks. Eventually I got to half and half, and then when those Fruit Loops were out, I didn’t buy them anymore. Her advice changed my palate. She met me where I was. 

Adam: Amazing. I think there are at least two lessons there: One, inciting behavior change with small, incremental steps is much more effective than dumping healthy behaviors on people. And the other is avoiding what I call “the shutdown response.” People get advice and don’t want to hear it because it’s a huge leap, so they just stop listening.

Ms. Schiller: Many physicians are looking forward but end up using scare tactics. If you have impaired glucose or prediabetes, they’re saying that you should change because you’re going to progress to diabetes. But people just want to go back to where they were; where they didn’t have prediabetes. We need to bridge that communication problem.

Dr. Mummah: I love that story. One of the other reasons that this strategy was successful was because it taught you that you can have tiny wins and that you’re actually effective. If a nutritionist tells you to replace every dinner with a salad, that’s a really drastic change. It may be hard to do that. So you admit defeat. If you give people really big challenges, you undermine their confidence in being effective. If you can start with tiny behaviors that build, you set people up for success. 

Adam: Look into BJ Fogg’s program at Stanford called Tiny Habits. He talks about flossing one tooth today. Then two teeth tomorrow. Then three the next day. It’s about making that first step so small it’s impossible not to do. And you should also pair a new behavior with an old behavior – that’s called an anchor. For instance, when I wanted to add stretching  into my routine, I started doing it after I shower every night. It really works. 

Dr. Mummah: That strategy helped me start a running habit. When I used to go out the door, I told myself I didn’t feel good unless I did a 30-45 minute run. But that’s daunting, so I put it off for weeks. I took the pressure off myself by telling myself I just needed to get my shoes on and out the door – if it’s a five minute run and I walk for a while, so be it.

Adam: Yes! I always say, “Five minutes of exercise beats zero minutes. That reframes what exercise requires. It’s a totally different mindset. Sarah, I think this relates to something we talked about too – the idea of long term motivators? This is such a huge issue in diabetes and prediabetes; what are the best practices?

Dr. Mummah: One of the things that motivates a lot of us in this space is trying to help people live healthier lives. We are motivated by big, long term outcomes like bringing down rates of heart disease. But those are really big challenges. That’s what gets us out of bed. But if you are a person who needs to experience the solution or the intervention, suddenly those long term outcomes aren’t enough to get you out of bed. Initially, it might be motivating. But two weeks or a month in, when you’re making a decision between a donut and a salad, you won’t think about how it will affect your long term diabetes risk. We have to focus on the process and making the process rewarding in and of itself. 

Ms. Schiller: I keep going back to the fact that 90% of people who have prediabetes don’t know. Interventions in prediabetes have good success – it works ~58% of the time. How will we make that impact though? We need to create awareness and collaborations. Now that we have more genetics information, it might be a good time to reset the thinking on type 2 diabetes prevention. It would be interesting to stratify people with prediabetes. What if we could give a screen to someone, and we knew that they could respond to coaching on a particular diet?

Q: We talk about awareness as a huge issue, and I’ve seen it with my father and his brothers – the surprise of being diagnosed with type 2 diabetes. I feel like there was an opportunity to tell him his whole life that he might be diagnosed – who’s failure is that? Breast cancer and ice bucket challenge have huge campaigns – where’s that awareness in diabetes? What don’t we have it?

Ms. Schiller: It’s not any one disease organization, but people providing funding – ADA wants to make a difference for those with diabetes. There is no foundation for those with prediabetes – no one has championed that cause. It’s the same for type 1 diabetes – JDRF is focused on type 1 diabetes. I don’t know about it in prediabetes – there are wellness campaigns and all of that, but that hasn’t moved far enough. And perhaps we are doing ourselves a disservice calling it “prediabetes”? They’ve tried with impaired glucose tolerance, but that was defined by a group and then sat in a box. On Google, I think you get 10 million hits from a prediabetes search, and 250 million for diabetes. I’m not sure we have that one champion for prediabetes.

Adam: There’s also a lot of fragmentation in diabetes – type 1 and type 2 diabetes. Payers are challenging too – for a payer to invest in prevention, it’s a tough sell. That person is going to be Medicare’s problem, so why invest today? Generic advice given to people at diagnosis is a problem too – “eat healthy and exercise and you’ll be fine.” PCPs are not equipped to handle this either – maybe they get half a lecture in medical school on nutrition, and on top of that, our understanding of nutrition continues to change.

Ms. Schiller: The example of what happened in hypertensive agents – the AHA said everyone with some level of LDL should be on one, and in a short period, there were millions of people on these agents. During a visit to my provider, the EMR screen literally wouldn’t advance unless I was prescribed a statin. My LDL is low, but it was coded in. There needs to be plugs, as descriptive as possible, about whatever we should do to prevent diabetes.

Dr. Deroze: Probably one of the biggest things is that the average person doesn’t think diabetes is coming for them. I have a video on YouTube where I characterize diabetes as the apocalypse. It’s coming for you, like in those zombie movies. If you’re not a zombie, your neighbor is. The average person thinks the apocalypse is coming, “but not for me.” We need to make people aware that it is coming for you. People are totally surprised that they were diagnosed, I was too. Rally people around prediabetes and diabetes online and offline.

Q: I’m a nutritionist. I like the practical things that you’re talking about, but I would find it difficult to mix Apple Jacks with Raisin Bran when neither of them are very healthy. I also worry about the fact that at the beginning, we give people medications in a drastic attempt to bring their blood sugar down, but what food? What about after that?

Dr. Deroze: It wasn’t the healthiest thing to do, but it’s what made a difference. I lost almost 100 pounds and I’m not eating those foods anymore. It was important to me that she met me where I was. On the medication point, I recently told my doctor I didn’t want to go back to a higher dose of medication because I felt like I was relying on medication. Most type 2s are not aware of behavioral changes being able to reduce dependency on medication.

Adam: Having people run their own food experiments is really powerful. Check your blood glucose, eat something, then check it again in 90 minutes. It’s a great way to see the different impact and learn from your own data.

Dr. Mummah: Running experiments is an interesting exercise. I wonder how we could make people make the right choices if there weren’t any associated health benefits. How could you try to make the process enjoyable?

Ms. Schiller: You can also help people think about it in terms of tradeoffs. If you like cereal, eat it and go for a run.

Dr. Mummah: The growth mindset (Dr. Carol Dweck), used in education psychology, applies to food as well. If kids believe their intelligence isn’t fixed, they will try harder in school and won’t be held back by setbacks. It’s similar to stretching palates. You can learn to love foods that are good for your bodies. If you are stuck with eating vegetables for a few weeks, you can eventually learn to love the taste. 

Q: My company wants to record every successful instance of diabetes prevention. We don’t know how many instances – how many people are diabetic, and how many have reversed it and how? We are Healthcoin, a blockchain-based health analyst company. We are about to run a solutions-oriented experiment. I was prediabetic a few years ago. I was eating low fat, running 30 miles a week, 4 glass of orange juice a day, and 3 bowls of cereal a day. The minute I cut orange juice and cereal out, my blood glucose improved. Society has decided that sugar is dangerous – manifesting in sugar taxes, cutting it down in education. Don’t we sometimes overcomplicate this question of healthy eating and exercise? Isn’t that a simple thing? Do you believe that, and what would it take to convince you?

Sarah: It’s simple in theory but making those changes is really hard over the long term.

Adam: If you read any of Gary Taubes’ books, it shows how one person (Ancel Keys) can sidetrack an entire field. Or how cholesterol, one measurement, can misguide the research in a field for decades. Virta Health and other companies are working on this, and they have the advantage of market incentives. Turning an oil tanker is slow, especially when certain professional organizations have given ineffective advice for decades. [Audience member: Hear, hear!]

Q: I worked on the Diabetes Prevention Program and we published the results in 2002. Guidelines have been set up by the ADA and AMA. I wonder what it would take to make providers follow the recommendations.

Ms. Schiller: I think building prediabetes into HEDIS scores will help with that. If it’s tracked, it tends to create results.

Adam: People respond to incentives. And on that note, we are unfortunately out of time. Thank you so much to everyone for coming!

A #)$%(*% Crisis: Training the Newest Gen of Docs

Susan Cox, MD (Dell Medical School, Austin, TX), Sarah Kim, MD (UCSF, San Francisco, CA), Bon Ku, MD (Thomas Jefferson University, Philadelphia, PA), Erin Kane, MD (Johns Hopkins, Baltimore, MD)

Dr. Cox: What are the most important trends in healthcare that will affect the work of current students? If you had $500M to invest in education, what would you do?

Dr. Kane: Value-based care. It’s a term that gets thrown around a lot. The definition for me has two elements – healthcare is at the intersection of cost and value. Did you get better? Do you have more health? In the current system, it’s like if you took your car to shop and they said they’d keep car for three days, it will cost a lot, they can’t tell me exactly how much, and it may or may not run better. That’s basically what we do in health. As an emergency room physician, I get asked about cost all the time, and I have no cost transparency. I can’t give information at all. Lack of information about cost and value is a problem. We’re not going to solve value at this panel discussion, but it’s important in medical education – do we give building blocks to students? We need to educate about cost – health insurance, discounts, copays, co-insurance, and really even just cost to hospital. There is active research where hospitals don’t understand their own cost, and the cost to the patient is different from the real cost to hospital. On quality, doctors are great at reading medical literature and interpreting data, but we don’t bring that into conversation with patients. Patients don’t want healthcare services, they want health. That is the most important next phase in healthcare.

Dr. Ku: The last major innovation in medical education occurred in 1910. It was a major change – industrial change. Then there was the digital revolution. That information, that technology, is going to be really shifting the role of the physician. I probably spent 20 hours memorizing the Krebs cycle in college, whereas the patient can pull it up in five seconds on a smartphone. No longer is knowledge my super power. We’ve been great memorizers. We had this knowledge advantage over patients. No longer is that going to be our super power. Now we’re doing telehealth, urgent care, 24-hour observations. My colleagues grumble and complain that things are shifting. They continue to move faster. I just see that as the most important disrupter.

Dr. Cox: Bon, have you ever seen a patient with a messed up Krebs Cycle?

Dr. Ku: No.

Dr. Cox: It’s one of those things that we learned but never use.

Dr. Kim: One of the important trends in healthcare is shifting from acute to chronic illness, and a lot of the top illnesses and top causes of mortality are lifestyle diseases – cardiovascular disease, type 2 diabetes. They are so profoundly influenced by social determinants of health – who you are, your day-to-day life. And in medical school, the way you’re taught about diabetes is about insulin resistance, failure to make insulin, you’re told about the 10 drugs you can use, and here are the lab tests you can use. All these are medical-type things, pharmaceutical interventions, blood tests, but when it comes down to it, I’m a diabetes healthcare provider for many publicly-insured people. I can tell you the reason diabetes is difficult to control in many people I see is not because we give them the wrong medicines or tests, but because they struggle to fill the prescription with a five-dollar copay they can’t afford, or they can’t make it to the office because they don’t have any reliable transportation. In med school, we’re not taught to think about these types of determinants of health. As we move to chronic disease, it will be important to think not just of what goes on in doctor’s office, but what happens outside as well.

Dr. Cox: We surveyed Dell students and residents from UCSF and Hopkins on how physicians should be trained. There was no difference in response between the groups: 36% said they needed to be trained in the business and economics of medicine. Inter-professional education was 28% and teamwork was 20%. Increased ambulatory training was rated at less than 10%. Yet that’s what we all feel we need to be going for.

Chronic disease care requires greater interaction between patients and healthcare providers. How can medical education close the chasm and ensure that doctors and patients are seeing eye to eye?

Dr. Kim: Patients and doctors often don’t see eye to eye. Students are being taught more shared-decision making as decisions are made in the office. What the physician decides should be a collaborative exercise that takes into account patient priorities. Motivational interviewing is the most important next step, from my perspective – seeing eye to eye, as a physician, is doing your best to see world through patients’ eyes. My goal for patients is to get an A1c, a marker for average glucose, under a certain threshold. 7% is standard. In some cases, however, my patients might just want to get through the work day without severe hypoglycemia.

Dr. Ku: At my medical school [TJU], we’ve created a human-centered design program to really understand the point of view of the end user of healthcare –patients. A lot of our training is done in a very hospital context. When I was a resident, there was a nursing home next to the hospital, so I thought all 80-year-olds were sick people. So much of our interactions are done in the context of clinic, and we don’t see the human behind that person. Often we try to involve the patient in the process of design, which is called participatory design. It’s seen not only in med ed, but also in medtech, biotech, and pharmaceuticals. We don’t involve patients enough in the process. We design stuff for them, not with them. We need to change the mindset, and there are ways we can do that. When we were redesigning part of the emergency room, we had physicians. I suggested bringing patients in, and they were taken aback a little –“whoa, that’s a novel idea. Never thought of that.”

Dr. Kane: In terms of medical education, specific communication skills can and should be taught. Motivational interviewing. I use the teach-back method. You can actually teach medical students and residents to do this. Especially if the diagnosis/instructions are important or complicated. “Can you say that back to me? did I explain it right?” It’s not just emphasizing the importance of communication.

Adam Brown (Close Concerns, San Francisco, CA): Communication skills are coming up a lot as a big need. What percentage of current med school curricula teach communication skills, and what percentage should it be?

Dr. Cox: Probably a very large percentage attempt to, at least. It’s one of the core competencies for residency programs, and LCME (Liaison Committee on Medical Education) is looking for. Probably mainly around doctoring courses, people don’t learn how to interact with patients until we show them how to do it.

Dr. Kane: There’s a gap. In the first- and second-year courses, there is training in this stuff, but in fourth year and in residency, there’s much less of it, and that’s where people solidify practice habits and that’s where we need to enforce it.

Dr. Ku: Too late. We need to redesign how we choose future doctors. If you’re 24 or 35 and don’t know how to communicate yet, then maybe you shouldn’t be a doctor. We look at GPA and MCAT, not personality. I have some students who probably should not be communicating with real people. If you’re an excellent communicator but got a C in physics freshman year, you will still likely be a good doctor.

Dr. Kane: A lot of people who are excellent communicators don’t think they have to use a certain tool, while there are some who do you use it. Some use line-by-line techniques. When someone’s family member dies, look them in the eyes, look for emotion, and respond appropriately. They walk themselves through these steps. It’s easy to pull off.

Q: We want people who are great communicators, but we want someone who is great at precision medicine on that team as well. We need a case manager who knows a little about the population and who could speak about money, and we need a technology expert, and you need the human element too. What are we really training? Better hospitalists and PCPs still want in depth people who can’t really communicate.

Dr. Kim: At UCSF, as in many medicals schools, we have an MD/PhD program. I interview people all the time. We are looking for a fairly good communicator. But there are people who are brilliant at science, and they are likely going to be research people. They may not be best communicators, but these individuals may only spend half a day in clinic per week, just to keep clinical relevance up. We need all sorts of facets of doctors trained.

Dr. Ku: The incentives often aren’t there in academic centers. Researchers bring in NIH money. If you’re a good clinician, you may be less likely to get promoted. You’re often not incentivized to be a great clinician. We need to start rewarding giving great clinical care. 

Q: A hospital in India does more bypass surgeries than the Mayo Clinic and with better outcomes, and the head of the hospital has proposed for a global medical school that makes GPs in five years using something like Khan Academy for technical skills and apprenticeships in any hospital with 300 beds or more. With modern healthcare education, 80% of people won’t be able to afford it. What do you think of this combination of internet training and apprenticeship?

Dr. Ku: I totally agree. If you include college, I spent 15 years studying to become a physician. That incubation phase is way too long; I didn’t need 15 years. I think these models that involve digital education can be disruptive. My med students, a lot of them don’t go to lecture but stay at home and watch the lecture video on 1.5x. So they cut a 60-minute lecture down to 40 minutes.   So it’s already happening. And to keep physicians in low-resource countries – 90% leave from Sierra Leone and don’t return. This could be way to cheapen education and keep them in their homes.

Dr. Cox: I wonder with that strategy, if we’re not allowing team to develop. They haven’t learned that interaction with and how to be a team leader.

Comment: They form amazing teams in these apprenticeships. The apprenticeship seems to be working.

Ms. Lisa Rotenstein (Harvard Medical School/Harvard Business School, Cambridge/Boston, MA): I’m a medical student, and I’m curious about the concept of student wellness in relation to empathy. Data shows that empathy often declines from the start of medical school to the end. It’s correlated to mental wellness and crazy statistics on depression. What should we do to curricula to keep students well?

Dr. Kane: I will speak to the resident part – in my view, I agree, there’s a lot of burnout, pushing people to the limits of sleep and their own wellness so that they can perform residency functions. There’s a lot of waste in residency. We have work hour restrictions, letting people sleep, using physician extenders, nurse practitioners to do discharge summaries that you’re really just doing because you’re cheap labor for a hospital. We need to separate the value add and cheap labor pieces.

Dr. Cox: Yet the incidence of burnout and depression have increased, even since the 80-hour workweek. In my day, it was 120-hour work week, yet depression rates are higher now.

Dr. Kane: It’s more than just hours. You go to work every day, and when you did well, you feel good, and when you let a patient down, you feel bad, and that feeling contributes to burnout. We need to create more supportive spaces, encourage them, help them take care of patients more safely.

Comment:  It’s my first year at Dell – and what we’ve seen above all is the culture. It’s a major component, making sure you feel like wellness is a priority. Where  you can maintain wellbeing and can bring any concerns over issues to faculty. There is also a lot of integrating human aspects into our curriculum. For example, The Blanton (art) museum – in doctoring courses, we take trips to museums, see art, and connect back to the medical practice. We see what we’re doing outside of core science, take it to a larger level, and make sure empathy is maintained. We’re also stepping outside the science. We have a  variety of initiatives through the wellness committee. We have a cooking setup. Three students cook in the cafeteria every week – the most recent challenges were diabetes-friendly and gluten free meals. I am a nutritionist and thought this would be a cool way to empathize with patients. We cooked meals, served to faculty, and everyone who wanted to could come down.

Q: What is the current state of med ed in terms of graduating doctors that are ready to tackle chronic disease? Where have we improved and where do we need work?

Dr. Ku: My dream is to have physicians that care about the culture of health in their communities. We are not doing a good job of that. Those that care about their communities go into family medicine, internal medicine, pediatrics. Every doctor should have some skin in the game. Every physician needs to care about communities. Academic health centers need to care about the cities and communities they are a part of. The incentives are not there. Healthcare delivery is providing sick care. It is not thinking about making our populations healthier. I’m an ER physician. I wasn’t trained about making my community healthier. Your problem is non-medical factors, which contribute to ~60% of health. I am just taking that little portion of ~20% (medical factors) and trying to make a difference. I think there’s a huge opportunity to get skin in the game.

Dr. Kim: We can identify the things that need to be improved in a community. We don’t have the knowledge or skills to change any of that. We could train medical students to work with local government, how to change government policies, have them turn the lights on at parks so people can feel safe to exercise at night. This is a big example of why diabetes is out of control. Public health measures can improve. The other thing is that we can learn how to better collaborate with members of the community. You often may not be of the same culture. We make assumptions about what the community needs. What are your health priorities? Learning tools of community-based research and collaboration are really important.

Dr. Kane: For chronic disease, we have a technical solution: The device or medication, and then the implementation. Every day in the ER, I see a patient come in with high blood pressure. They couldn’t take the medicines, couldn’t afford them, or didn’t understand why they are important. There’s a gap between writing prescriptions, which used to be enough, and getting from prescription to health. The task of this generation of physicians is going to be bridge gap between prescription and health.

Q: We’re seeing a lot of changes in the workflow of physicians. Not only in content, but what it means to play the role. How do you see the roles changing, and how do you see technology as a part of that?

Dr. Kane: I think people talk a lot about how we can get physicians comfortable with tech, but it’s about setting up the workflow and the system to help physicians. Doctors know how to use email, have time to respond to emails over a secure server, and maybe even get reimbursed – so they will use email. It’s not just teach them; it’s build the solution for the day to day.

Dr. Kim: I think the physician will be more of a team player, and a leader in some case. They will be less of a lone character who saves the day.

Q: In Colorado, we have many rural and mountain areas, so no local provider. How can we do a better job of arranging the system to provide access to care for those who don’t have access? Both architecting the system and training providers to offer remote care; not just telemedicine, but training remote providers to assess health.

Dr. Kim: It’d be interesting to start to incorporate telemedicine. We’re doing more of that in my hospital. It’s a different skillset to give advice over electronic written communication or by video conferencing. We’re just learning the nuts and bolts, but in med ed, it would be pretty critical.

Dr. Cox: In Minnesota, medical students are assigned to rural areas for their longitudinal clerkships. One year spent there, so they are more likely, when they finish training, to do primary care and be PCPs in that community where they have relationships.

Dr. Ku: How often do we have to be part of the solution? Because we’re very expensive. Do we need to quarterback everything? Blood pressure management doesn’t need to be done by a physician – it’s probably better by a nurse practitioner. It’s about less of central role and more of a team player, and having future physicians be comfortable with that.

Q: I’m a proud medical school dropout turned public health professional. I work for Blue Cross Blue Shield. I just have to make the observation that the things that medical students need is what you get in public health school. Physicians who are MD/MPHs don’t integrate well as doctors. We have a lot to teach you. We just don’t get paid. In terms of envisioning a future with population health, how do we rebalance between these public health professionals, who are not physicians, or people who left medical school, who have that knowledge, and physicians who are paid very well, but not receiving this type of education? How do we advocate to finance that?

Dr. Kane: There are so many elements from different disciplines – business, economics, public policy. There’s actually a lot of things we’re trying to take from. New England Journal of Medicine and Harvard Business Review Catalyst is a good example. Take a cross-disciplinary approach, which applies more broadly than public health.

Dr. Ku: I think we’ll see a major shift once payment shifts from sick care to value-based care. Help us be on the hook for population care – we’re going to need your voice and those collaborations will be a lot more critical. We’ve been trained to give sick care.

Dr. Cox: In asking residents and students what needed to be done about preventative care and partnerships, 27% said Medicare/Medicaid, 28% said academic medical centers, 10% said advocacy groups, and 21% said other (community, payers, FDA).

What is the most important thing you learned in med school, and what do you wish you’d learned?

Dr. Kim: I went to Tufts and we had a relationship with Eastern Maine Medical Center and I spent three months living there doing a pediatric rotation. I was rotating in hematology, an oncology clinic, and there was a child who was visiting her oncologist and going through chemo for leukemia. After the nuts and bolts of the visit, the mom and the child’s major question was “and so what do you recommend for my really dry feet?” and the doctor said ‘Burt’s Bees coconut foot cream – it’s the best, go get it.’ As a med student, I was surprised that this was part of a medical visit. Being a physician is not just prescribing therapy or ordering a blood test, but being there for a patient in a more holistic way. For that child, it was about fixing her really dry feet and the doctor compassionately addressed that concern. I still have a bottle of that cream in my medicine cabinet to remind me.

Dr. Ku: I think what I saw was that the future of healthcare won’t be changed unless medical school is redesigned. I was in a group of disgruntled med students. I was getting burnt out. I studied classics in college and then entered in this machine. I thought one day I will get out and try to redesign it. It inspired me to go back into the process because that incubation phase is so long, so we have to also predict what healthcare will look like in 15 years. I wish I had learned more creative problem solving – we’re taught very binary solutions and we don’t do well with ambiguity. But a lot of events in healthcare don’t have binary solutions. Reducing mortality is a tough thing – so we’re taught to do very precise outcomes that are easy, but what does that mean for a patient to be healthier?

Dr. Kane: I will just share about my personal story. When I finished med school, I went to the consulting side because I didn’t understand business, pharmaceuticals, insurance, etc. Med school teaches you to think there are doctors, patients, and hospitals, but there are other really huge forces in healthcare, and to be equipped to make change, med students need to understand the broader health system.

Dr. Cox: And when we asked what the most important skill for a doctor was, 36% said collaboration, 29% said empathy, 8.5% said autonomy, and 3.4% said technologically savvy.

Comment: There’s been so much around empathy and mindfulness. About three years ago, I created a company with training in all these areas, and we’ve been working with hospitals on compassion fatigue. I’d like to offer the product for free to everyone here.

Q: I’m a trauma surgeon just finishing my fellowship. I think one thing, this is a poll for med students and residents, and it’s hard to teach someone empathy. But one important point is that collaboration is a huge part of what we do. When we round in ICU, we round with nutritionists and pharmacists and so many others. I didn’t realize that in training – just how important that was. So maybe you can’t teach empathy, but it kind of just develops and you can see that by collaboration you see.

-- by Brian Levine, Lisa Rotenstein, Adam Brown, and Kelly Close