Dr. Vivian Lee, President of Health Platforms at Verily Life Sciences and author of The Long Fix, discusses the importance of personalization to overcome disparities in health care, and reveals the ongoing challenges of being an introvert at heart

 

Vivian S. Lee, M.D., Ph.D., M.B.A is President of Health Platforms at Verily Life Sciences. A physician and health care executive, Lee also serves as a senior lecturer at Harvard Medical School.

Prior to joining Verily, Lee served as the Dean of the Medical School and CEO of the University of Utah Health Care, an integrated health system with a budget of $3.6 billion. Dr. Lee previously was the inaugural Chief Scientific Officer of New York University’s Langone Medical Center.

Elected to the National Academy of Medicine with over 200 peer-reviewed publications, Lee serves on the Board of Directors of the Commonwealth Fund, the Board of Trustees of Boston Children’s Hospital, and is also a director on the board of Zions Bancorporation, a publicly traded company.

Dr. Lee is a magna cum laude graduate of Harvard, received a D.Phil in medical engineering from Oxford University as a Rhodes Scholar, earned her M.D. with honors from Harvard Medical School, and her MBA from NYU. She was named by Modern Healthcare as one of the 50 Most Influential Clinical Executives in 2020.

She is also the author of The Long Fix: Solving America’s Health Care Crisis with Strategies that Work for Everyone

 

 

What initially sparked your interest in medicine?

When I was in junior high school in Norman, Oklahoma, I had a very enterprising teacher who matched each of her students with a community leader in the hope that they would have some career mentoring. I happened to be matched with a prominent internal medicine physician and for many years, I would accompany him on hospital rounds on Saturday mornings. He really inspired my interest in medicine because we had no physicians in my family. My father is an electrical engineer, and my mother is a statistician and epidemiologist.

At the same time, my mother was part of a large study looking at diabetes and cardiovascular diseases in Native Americans in Oklahoma and other parts of the US. There was a high prevalence of diabetes leading to diabetic retinopathy, placing individuals in the community at high risk of blindness. It’s a very significant and serious complication and can be prevented if we take pictures of the retina regularly and spot the early abnormalities. My mother was leading an epidemiological study with a large number of these retinal images, but there was a shortage of ophthalmologists to interpret these pictures.

So my mother recruited my father to build one of the earliest neural network-based algorithms in the diagnosis of diabetic retinal abnormalities using these images. The project was eventually published in JAMA Ophthalmology. It’s the only paper I’m aware of that they ever wrote together, and it also happens to be my first exposure to AI.

Could your career have taken a different direction?

When I was young, I was very passionate about mathematics and engineering. I took most of my advanced classes in engineering during my pre-med years and I had the opportunity to go to the UK under the Rhodes Scholarship to focus on medical engineering at Oxford. Later, I also developed a strong interest in business. So my career can be defined at that intersection of engineering, medicine, and business. I could easily imagine having pivoted to engineering, as opposed to medicine. Fortunately, today, my work includes all of the above.

What was the most memorable thing about living and studying in Oxford?

I had a chance to dive deeply into medical engineering and work with the late Professor Brian Bellhouse. He was a wonderful person with an entrepreneurial spirit. He built many very successful companies from his innovations, which was unusual in those days. He was ahead of his time.

I was also one of the earliest graduate students of Professor Lionel Tarassenko, who was the first Director of the Oxford Institute of Biomedical Engineering. He was a terrific and highly energetic mentor. I will never forget going to the lab and working with other students and lab technicians to build devices for our projects. It was so much fun! On top of that, it was an incredible experience to live in another country. The UK attracted many people from across the world, giving me the opportunity to meet people from South Africa, Australia, and New Zealand.

In your new book, The Long Fix: Solving America’s Health Care Crisis with Strategies that Work for Everyone, you propose the prioritization of preventive care. Do you think technology, especially AI, is a good tool to facilitate preventive care?

Yes, 100%. This is what makes me so excited about AI. It’s a tool that can extract meaningful insights from all the data we collect, something we couldn’t do in the pre-digital era. We can use the information coming from genomics and other ‘omics as well as an individual’s environment and lifestyle to personalize both the way we engage people and the recommendations we make about their care. Customization is crucial. We can identify who is at higher risk for certain diseases like cancer, suggesting changes to one’s diet and medications and engaging people in behavioral changes that need to take place, for example.

It’s like smartphones. The devices are similar, but our experiences are completely personalized. Smartphones engage us because they provide information that we want to have at the right time. How great would it be if health care could be like this too? We would have a personalized experience that is designed around what is important to us – whether it’s losing a few pounds, quitting smoking, or making sure we get our vaccinations. Ideally, we’d think, “keeping ourselves healthy has never been so easy!”

Do you think AI tools will end up like electronic medical records where there was so much hope at the beginning, but it created a whole new set of problems?

AI is like any tool. It really matters how it’s used, developed, and designed. That’s why we need an effort like AIMed to help attract some of the best and brightest people into the field. We also need to ensure people from other backgrounds have a deep appreciation of the most important healthcare challenges.

This is what I had in mind when I wrote The Long Fix. I want people who don’t know anything about healthcare to appreciate where the challenges and opportunities are, where some of the minefields lie, so that people who are armed with the expertise will build tools that are going to improve people’s lives. This is what I hope people will focus on. I wanted to show the opportunities for where new entrepreneurs can both do good and do well.

In what areas do you see the next big advances in medical AI?

I can give you a very specific example. Verily has a smart solution called Onduo. It was originally designed to help individuals with chronic diseases to manage their conditions at home such as people with Type 2 diabetes, who need to monitor their blood glucose regularly.

Part of the solution includes using a small continuous glucose monitor – applied to one’s arm or abdomen to measure blood sugar level 24/7. This eliminates the need to prick fingers multiple times a day. These devices come with a Bluetooth chip that can transmit data directly into one’s smartphone. Because it’s linked to their smartphones, users can also take pictures of what they have eaten and match it with their blood sugar level.

AI can help with insights about how each person’s diet affects their blood sugar level. With Onduo, the algorithms can recommend the most suitable food or meals. In addition, if the AI learns that the users are most responsive to certain challenges – diet adjustments, new exercises, and so on – the tool will begin customizing the experience to help people achieve their goals. The same AI tools can help coaches and physicians interact more effectively with the patients.

Through this example we can see the enormous potential we have for AI in preventive care. While these individuals may not necessarily have prevented diabetes, they are preventing complications of diabetes. This kind of experience can be useful for everyone. For instance, a colleague of mine likes to take a handful of M&M chocolates every time he passes the candy jar. But since he started tracking his blood sugar level, he immediately stopped doing that. We don’t often notice the consequences of our actions on our health. These sensors and AI algorithms are helping us with that.

With all the biases and disparities in health care, do you think technology, especially AI, is going to make them worse or better?

It’s not preordained that AI will make health care better, but I truly believe it can. Part of it falls back on how we use AI to create a more personalized experience, like Onduo does. If we can do that well, it’s you and your behavior that are generating the data, and our models are built for you.

Digital technologies powered by AI can potentially leapfrog many of the deeply embedded challenges that we are facing in healthcare, helping to bring us to a much more positive future. We have the opportunity to disrupt how people get care and also how they pay for care. At Verily, all our products are developed in partnership with health care systems, clinics, and community based clinical programs. We work hard to make sure all our partners are very diverse and are representative of the communities that we are trying to serve.

What’s the best piece of advice you ever received?

There are two pieces of advice that have stuck with me through the years. One comes from Bob Grossman, the Dean of NYU Grossman School of Medicine (now named after him) and CEO of NYU Langone Health. He said to me early in my career that I should get out there and expand my horizon and circle of friends.

Both Grossman and I are radiologists. When we talk about images like CT or MRI scans in radiology, we use the term, field of view (FOV), which refers to how big the picture of the body is that we are trying to take. I remember Bob giving me this advice, “Vivian, you might think about increasing your FOV.” I asked him what he meant, and he replied, “Friends of Vivian”. It was great advice that I’m still following today. At the time, I decided to give myself a few goals: speak up as much as possible at both local and national meetings, participate actively in committees – contributing meaningfully – and to challenge myself to make new friends and colleagues at events.

The other piece of advice came from Mario Capecchi who received the Nobel Prize for his groundbreaking work in developing knockout mice. When I became the Dean and CEO at the University of Utah Health, we decided to make some videos to showcase our faculty, including Capecchi. The video is still available online here. Capecchi shared a proverb that he was raised on, “The difficult we do right away, the impossible takes a little longer” and then he went on to say, “It takes the same amount of effort to work on big questions as little questions. So why not work on big questions”.

What advice would you give someone starting their career in medicine or medical AI?

I remember giving a talk to a group of first-year medical students who were extremely passionate about what they were doing but I thought, for them to have an impact, they need to understand the business of healthcare. I am not just speaking about the US model of healthcare; challenges exist in every healthcare system globally. Regardless of government subsidies or private healthcare, the relationships between patients, clinicians and payers are complicated.

I had very little understanding of this when I began my medical journey, and this is what prompted me to write The Long Fix. Medical, public health, business, technology and public policy professionals who are seeking to change healthcare need to have a good understanding of the business of healthcare, where the key opportunities are and the levers that can drive change most effectively.

One of those is technology and data science – not surprisingly, that’s what led me to take on a leadership role in health technology.

Have you ever felt discouraged that medicine is a lot more complex than you expected when you first entered the domain?

I don’t necessarily feel discouraged. What would be discouraging would be to know that people who have the talent and ability to make a difference don’t understand the problem well enough to be effective. I wrote The Long Fix to provide those insights and extra support to people who are curious about why healthcare isn’t serving society better and who want to do something about it. If anything, the urgency is only greater today. Look at what the COVID-19 pandemic has done to the world. It’s further revealing all of the system’s inherent weaknesses, including the lack of investment in public health and the huge inequities in our system. All these were already problems before the pandemic, but the public health crisis just makes them even more apparent. The crisis is attracting some of the best and brightest to science and health, and that cheers me greatly. We need people who have the talent and passions across various disciplines to work together.

If you could go back to the past, what would you change?

Personally, I wish I had taken economics when I was in college. That’s not something profound but it’s tied to the notion that understanding the underlying economics of healthcare is something I came to later in my career than it should have. Behavioral economics is another field I’ve been introduced to since then. Understanding human behavior is vital to improving health. Just how do we understand what motivates people and engage them in activities that improve rather than diminish health?

At Verily, we are building tools and tech for patients, clinicians, and payers. This brings together software engineers, product managers, user experience researchers and designers, quality control experts, and more. I always want to be learning more so I have just taken a course in data science, learning how to code with Python. But subjects like economics, computer science, statistics, data science, psychology, finance are all dimensions and sources of creative thinking that we desperately need. Although, at Verily, I get to work with wonderful teams of collaborators and business partners who have that expertise. We’re learning from one another, and that’s where the real sparks of innovation are flying!

 

Dr. Vivian Lee is a keynote at our Clinician Series running from September – December.

Lessons from the Long Fix: applications of AI in population health

The potential for digital technologies to solve some of the most deeply embedded challenges in healthcare is enormous. It is not, however, guaranteed; how tools are designed, developed, and used greatly impacts their utility, efficacy, and equity. Dr. Lee will discuss the potential of AI to advance health at the population level, and will highlight some of the key considerations necessary to ensure equitable outcomes across all sectors of the populace.

29 September

11am – 11:45am EDT

 Click here to find out more.