Dr Lloyd Minor: "Ten Things I Know to be True" | Talks at Google
Hello. Googlers, both, here in Sunnyvale. And around, the world good afternoon, my. Name is greg moore i lead healthcare here at google cloud and. We, have a very very, special guest with us this afternoon. It. Gives me great pleasure to introduce to you today dr. lloyd, b-minor, MD. Doctor, minor is the Carl and Elizabeth nauman Dean of the Stanford University School of Medicine, under. His leadership dr.. Miner is established, and is now executing. A. Strategic. Vision to lead a biomedical, revolution both. Nationally, and globally in, precision, health. Dr.. Minors a professor, of Otolaryngology, head. Neck surgery at Stanford University, with. Additional appointments, both in bio engineering and. Neurobiology. He's. A prolific author, over. 140, published papers and chapters, widely. Recognized. Authority. In the area of balance inner. Ear disorders, for. His contributions, in 2012. Dr. miner was elected to the National Academy of Medicine. So. We're really pleased to have dr. miner with us here today let's give him a warm welcome. It's. Like. Well. Thank you Greg it's really, a privilege and an honor to be here today and, I was as I was thinking about my remarks. You. Know Google is about to celebrate its, 20th, anniversary we we, Stanford, and Google obviously share, a lot of history right I mean Google began, in two dorm rooms at Stanford, and early. On the. Founders, articulated. What. Principles. That they described, as 10. Things we know to be true so. I thought I would parallel, that a bit and I think you're gonna recognize some, distinct. Similarities between. 10. Things that I know to, be true about health, and healthcare and, what. The founders, had in mind in establishing. Google and what is very much reflected, today and the, great work that, all of you are doing. So. Principle. Number one very. Similar to principle number one for, Google. For Google of course focus. On the user and all else will follow for. Us in health and health care focus, on the patient, and oils, will follow, and. This may seem to be self-evident. But. It's actually, only recently. That I think healthcare professionals. Have begun to embrace this, as the, true north. Only. Recently because, unlike. You. Know managing. Our finances. Or, deciding. What car, we're gonna buy or what bicycle. We're gonna buy or how, we're what, school we're gonna send our kids to you. Know health and health care in the past has, not always, garnered. The degree, of, engagement. From, consumers. That, almost every, other aspect, of life has. All-too-common. There's, been the attitude, well if I, get sick I'll go to the doctor and. They'll. Fix me or I'll. Take a medicine it'll cure me and, so the notion that, and. That I think. There. Are lots of reasons for that but I think one reason is that we, haven't we as healthcare professionals haven't. Focused on involving. Our patients. The people who entrust, us with. Their. Care and who look to us is to be advisors. As. Well as to bring our expertise, to their treatment, we, haven't always focused, on the way that we can engage patients, in the way we, need to. And. Yet really, if you go back to the founding. Of modern medicine, and some of the original greats, who helped, get us started on a more scientific, path in in. Medicine, you, know William Osler who was the first physician in chief at Johns Hopkins Hospital. Sort. Of a founder, of modern internal, medicine, he. Famously. Said you. Know listen, to your patients they're. Trying to tell you their diagnosis. And, so. Today this. Principle. Of, focus. On the. Patient, and all else will follow has. Never ever been more applicable than it is today and partly. And what I'll talk about a bit more and what many of you are engaged in everyday a is that our opportunity. To use technology. To. Focus on our patient, to, use technology, for, high-tech to enable high-touch Healthcare has, never been greater than it is today and is going to continue to increase exponentially. In the weeks, months and years ahead. So. Second principle the future of healthcare is proactive. You. Know related, to the first one, I, just spoke about in, the, past we've tended to think well if we, get sick we go. And we get fixed we either have an operation, we take a medication, we've. Had far less attention. On how. We can prevent disease in the first place how we can be much more proactive us.
As Individuals, thinking about our health and healthcare, professionals, thinking about how we develop, the science around. Prediction. And prevention, and. As. Greg mentioned before. We've, had a strategic, vision at Stanford. Medicine on focusing, on what we call precision health, now. All of you have heard of precision medicine and we for sure practice, precision, medicine at Stanford. Precision. Medicine is about, using genomics. And. Informatics. To improve, the, care we deliver to patients, with cancer, so. That the care is truly, personalized. Individualized. Based, upon the molecular genetic, characteristics, of the tumor and for. Sure when, we have cancer, or heart disease, we, need precision, medicine. But. Wouldn't it be great, if we pre prevented. Those conditions, to begin with or if. We diagnosed them much earlier than we are we did today. So. The way we think about the difference is that precision medicine, is about sick, care. Precision. Health is about, health, care the. Goal of precision, health is to, predict, to, prevent, and cure. Precisely. But. Really in that order prediction. Prevention. And then, cure, when when. That doesn't you know when we when, we still are not able to, adequately. Prevented. Disease from occurring in the first place, and. We're seeing all sorts, of examples, where, that vision, is coming. Into reality today. And. Of course the way we know we, will know 10 years from now that we've succeeded in this vision is that, the need for precision medicine the, need for the ultra. Complex, acute, care will, be reduced, because, we will have prevented. Or diagnosed, early and therefore more effectively, treated, diseases. Before they need the, complex, tertiary, and quarternary care. Principle. Number three. Data. Holds, the secret, to. Better health. There's. A tremendous, amount of health data that exists today embedded. With electronic, medical records, coming. In from wearables, so. When I think about digital. Health I think about two. Related. But, somewhat distinct, verticals, one. Vertical, focused. On consumer. Facing devices. The. Things that enable us to get a lot more information about, how our body is working and then.
The, Other. Closely. Related but distinct area is the, analytics, related to, complex. Datasets so how do we bring in data. From those consumer facing devices. Combined. With the, data that's sitting, by. And large you. Know unaddressed, in electronic. Medical records today in order, to, inform. Decision-making, and engage, people more. In their health in their healthcare. There's. Tremendous opportunity. In both areas both in terms of consumer, facing devices. And technologies, and also, in analytics, let. Me just give you one example. You. Know earlier in my career. As. Greg, mentioned I an. Ear expert, I focused on inner ear balance disorders. And in 1998, I described, a clinical syndrome an inner ear disorder it's. A disorder with one of the balance canals of the inner ear it's. A disorder caused by the bone that should cover that balance canal getting, eroded over time and then, the balance canal itself, begins to respond to sound and pressure, so. These. Patients that have this disorder experienced certain. Very distinct. Recognizable. Symptoms, and signs and, one. Of those is that they, have noise induced, dizziness, and, so if you open the, Google search engine right now and you type noise, dizziness. On the, first page you'll. Get several websites, that describe this, syndrome, superior. Canal dehiscence syndrome as well as a host of other conditions that, can cause related. Symptoms and signs if. You type those same two words in. The chief, complaint, line of one of the commonly, used electronic. Medical records systems. Today in doctors. Offices and hospitals around the country, you, get as a provider, no diagnostic, assistance. At all so. If you're someone in my specialty. You've certainly seen patients, with the disorder before you know to be suspicious but, if you're a primary care physician who's, maybe seeing a patient with this disorder for the first time you, may not know that that's the diagnosis. And so. It's still the most common. Way that patients, with this disorder being, diagnosed, today is by. Searching the, symptoms, and the signs themselves, usually. On Google and finding. A website, that describes, the condition, and then going to their physician, and saying I think I have this you.
Know Look at this webpage we. Can do better than that we can do a lot better than that and I. Think the work that's going on in Google cloud as well as a lot, of different other areas is going to get us more, towards the type of environment, that can, really. Enable. Us to fulfill. This vision of precision health. Fourth. Principle, to. Change health we. Have to change behavior. You. Know when. We look at the determinants. Of health of, wellness and well-being yes. There's medical, care and there's genomics. But. If we look at health and well-being as being a pie. Then. Genomics. And medical. Care represent, maybe 25, to 30 percent of that pie. The. Majority, of the determinants, of health. Our behavioral. Social. And environmental factors, and, yet. In the past we. Focused, far less attention and not by we I hold, ourselves in academic. Medicine very much accountable, for this we focused far less attention. On behavior. And social, environmental, factors, than we have on the medical and genomic, factors, and. Today. We obviously, we have the opportunity, to change that now. Let me mention one project, that I'm really excited about going on today started. By a team, of Stanford, investigators. It. Goes by the acronym, sphere. So, at Stanford, just as at Google acronyms, play, a big role in how we communicate. Sphere. Is short for Stanford, precision, Health for, ethnic, and racial, equity, and. One. Of the projects. Sponsored. By this is funded by an NIH grant well the project sponsored, by. This grant looks at obesity, and Latino. American, children, focusing, initially in Santa Clara County, so. Over 200, children and their families, right now are in Santa Clara County are rolled in this study and. Yes. We are looking at their genomics, that their microbiome, at other metabolic, parameters, but. We're also looking at what types of interventions such. As, changing. The size of the plates in the home rather, than giving large plates furnishing. The family with smaller plates how. Can exercise, and physical activity be, encouraged, either in the school environment the home environment or getting to and from school. So. Both an. Approach, that looks at yes what are the traditional. Medical. Genomic, factors, that, are involved in obesity but also looking.
At The social environmental, and behavioral factors, and, it's. Through that combination, really looking at the whole pie, of health and well-being that we're going to be. Much better able to, have an impact with the. People who, entrust their care to us have, an impact on health and well-being. You. Don't need to, be at a hospital to get care I think, that's obvious. Today but it's gonna become more obvious in the future. We've. Rolled out a program to, people. Who are enrolled in our accountable, care organization, Stanford Health Care Alliance called. Click well. Click. Well let you do fairly. Common, simple things like schedule, an appointment online. But. It also allows you to schedule a telemedicine, visit, with a healthcare provider or, with. A lifestyle, coach or nutritionist. And what. Better way to talk, with the nutritionist, rather, than coming into an office, and sitting across a desk, from, a nutritionist, why, not have that, interaction. In your kitchen with, your spouse or significant, other and talk, in real-time about. Meal planning and cooking. And nutrition. So. We're, just scratching, the surface today, of the possibilities. Of telemedicine. And also, of how, people. Are able to engage. In. Engage. With the healthcare delivery system in ways. Other than physically. Being in an office or a hospital but. In the end part. Of this. Greater. Accountability. And, engagement. Of people in health and well-being. Involves. Making, it more, accessible. And if you will easier, for. Each of us to learn about our health and be more proactively. Engaged, in our health than. The old model where, we would go in and see our doctor once a year get some lab tests done and then, forget, about everything in between. Number. Six fundamental, discovery, pays off. You. Know it's fascinating to think that the transistor. Was only developed in 1947. Right and when. John Bardeen, developed, it it was considered a lab curiosity. I mean. How could he or his contemporaries. Have imagined. What was going to result just, in a couple of generations right. And. Also. We know that GPS, was developed, you know as a variant, on an atomic clock to. Test Einstein's. Theory of relativity and, look how GPS is being used to so. We never really, at, the outset, know where. A fundamental, discovery is going to go but. I've yet to see a fundamental discovery that hasn't had, some. Profound impact. An. Example, from our own world we. Wear, today very fortunate, the school of medicine have seven Nobel laureates on our faculty, one, of those Michael Levitt won the Nobel Prize in Chemistry a few years ago for work that he did more. Than 20 years ago Michael was the first we'll really, were the first computational. Structural, biologists, he. Used computer. Programs, and algorithms. To, predict. How the structure. Of proteins, change, in different conditions. Those. Formative, algorithms. The approach that he took in his very, original creative work are now in use today in drug, companies, to allow them to predict, how changing. The structure. Of a drug will, alter its interactions, with the receptor, so. An entire field of drug discovery was. Born. Based, upon some very fundamental, computational. Work designed, to understand, the structure of proteins. Number. Seven. One. That we're very pleased, to. Collaborate. And embrace. With you academia. And industry must collaborate. It's. Always been true if we, look at for example one. Of the great success, stories of the past 20. Years in, medicine, and that, is, changing. HIV. From. Being a 100%, fatal. Diagnosis. Now. To being, in the vast majority of cases a. Successfully. Chronic, disease. Successfully. Managed on. Multiple. Antiviral, therapy, that, success, story required. Discovery. Based research at universities with. Pride clinical trials at universities, but, would not have occurred without, the, type of support, and the. Type of focus and resources. That Pharma. Brings to problems as well and, the. Same is going to be true in the digital world for sure and I'm, really pleased that we collaborate. With Greg and team at Stanford in using. Google Cloud for our genomics, platform, and the. Analytics, that you have developed here and please, that in another branch of alphabet, verily, we're.
Working, On what. We call the the. Google baseline store study left over from when Vera Lee was Google so, baseline, is a 10,000, person longitudinal, cohort study, cohort. In that it's, 10,000, people and everything you and I can imagine about their health will be measured, whole genome sequencing. Immuno. Microbiome. Barely, developed a new wearable, to, track, these, enrollees. Over. A long, period of time initially at least four years and, for. The first time we'll, have enough baseline, data about. The health of these individuals, to be able to print to. Look back and to say well when. Someone develops a disease later on well here were the early indicators, rarely. Do we have that sort of information today, so. Some of the patients enrolled in in, baseline. Are gonna be healthy today some. Will, have, a diagnosis, of cancer already, and some already a diagnosis, of heart disease and, that, information, gathered. In this, fashion very. Much powered. By the, type of analytics, that you and your colleagues do every day it's, going to give us insights we've never ever had before in health, and health care. Number. Eight diversity. Drives. Innovation. In. Academia, and in industry there's, a lot of thought about well do, you need specialists. Or generalists, and of course you need some specialists, in some generalists, but. Oftentimes the, real advances, come from what, are described as t-shaped. People, that. Is people who are specialists. In an, area but, also have a broad horizontal, component. Cutting, across multiple different areas I know, you have many such, people, here. At Google. So. There's that sort of diversity in terms of diversity of ability, and thought, but. Also and in particular in, health and health care there. We have to focus more on ethnic, racial socio-economic. Versity. You. Know over half the children under, the age of five in, the, United States today are, from. What. Previously were considered to be minority. Racial, and ethnic groups and the. Majority, of the u.s. population, will. Be that, minority, coming. Ahead in the next 40, to 50 years and, yet. Today, only. About 5.5, percent of the physicians in the United States identify. Themselves. As African, American and only. About six point three percent identify. Themselves, as Hispanic. We. Have to do better in terms of representation. In our. Profession. It's. Been a big passion, of mine since coming to Stanford five years ago I'm pleased, that in our medical school class we've moved from five. Years ago having 14%. From underrepresented groups. To now having 26%. Entering, our medical school and in, our Biosciences, ph.d, program we've, moved from 12%, from underrepresented groups, five years ago now, to 25%. But. We can't take our eye off that. Ball. We, can't for. A moment, back away from our commitment to, achieving greater diversity, in our profession, I think it applies certainly. Across the board in in in professions. In the country today. One. That's familiar took, me a while to get used to and that is you can be serious without a suit. There's. The obvious implication of this and I'll tell. You a funny story when, I was looking. At this job and was making trips out from the East Coast and of course searches. Of this nature are done pretty.
Quietly, Because, of. Factors. That the institution moving to and factors were you coming from so, I was coming out on a Saturday, and I got a call from the office organizing, the search and they, said you know you're you're meeting with the search committee on Saturday, and, the. Men will not be wearing a necktie and, and. I went home and, that, evening said, to my wife you know I got this call and I said did you know the guys coming. On Saturday won't be wearing a necktie and I, think that means I'm not supposed, to wear a neck tie and, she. Said yeah you know that's definitely what it means and that's gonna be hard for you because you wear a necktie when you take a shower in the morning so. I have gotten used to not wearing a necktie. I. Think, the more serious side of this statement is, that. We. Shouldn't be taking ourselves too, seriously you know, the beginning of, a decline, of an, organization. Usually. Centers around two things. Complacency. And hubris and in. Particular in, health care we. For sure can't be complacent and in, the past I think I would, say the physician profession, in particular has. Had more than its share of hubris, and that, has to change moving, forward. Before. Medicine. Became, more scientifically. Grounded, what. Physicians offer, their patients, was. An. Ear, a. Shoulder. To rest a head on an. Understanding. Empathetic. Voice. But. There wasn't all that much that, physicians, could bring to the table that's. Changed, today fortunately. And we wouldn't want to go back, but. That desire in that need for empathy has never been greater than it is today you, know, I mentioned before that we, started. This. That. In our accountable care organization. We allow people to have virtual visits, and that's great and many people take advantage of them we. Also give, people the option of receiving, all of their healthcare virtually. If they wish and, almost, no one elects, to have that they. Want to, have. A primary, care provider who, knows them who. They, rely upon as a primary, care provider but, they understand, that that person may not be available at 8 o'clock at night when they have a question. So. And that. Really then. Gets back to, the, crux of the talk and the discussion, and that is how can high tech. Be. Better able to enable the high touch that. I think everyone, expects, expects, from us. And. Then so, that's on the par, about making sure there's no hubris, but, then the part about, complacency. And how. It can spell the the downturn, of any organization. Really. Relates to principle number 10 exactly. The same as Google's principle, number 10 great, just isn't good enough and, certainly, when we're talking about health, and well-being when.
We Have this crying, need in the United States today you know 18%. Of our, GDP being. Spent on health care and yet, our outcomes, are very, poor, compared, to other OECD countries. You. Know we have to be doing better than we're doing today and, I think we always have to embrace the, notion, and, live. The notion, that great, isn't good enough, thank. You very much I'd certainly welcome your questions I'm honored to be here today. Happy. To take questions here, and because I have the microphone I get to ask the first one, over. So. Many. Of its of the engineers, here at Google are engaged in machine learning that's that's the engine the. Fuel. For that engine is data your. Point number three. That. Data does hold. Secrets. As, you said secrets, that can help us help patients. Globally. And. To unlock those secrets for discovery, in for, research how. Can I a. The. Academic. Medical centers one of which you run an, industry collaborate, to begin, to use that data more effectively because that's always been. A friction point and, there's, consequences to delay of sharing that data of, course that impact, the patient's, sure. I think the the. Data sharing problem has several. Root causes. There, are increasingly. Areas, of care that are covered by bundled, payments, namely that the, health system the physician, everyone, gets paid a fixed, amount and. Not anything more but. Still by and large it's a fee-for-service backbone. And there for it hasn't, necessarily been, perceived, to be to, the advantage, of health care delivery systems to make data easily, portable. That's, one reason, second. Is there is the legitimate, concern about the privacy, and security of data and. It's a real concern that you. Know I've read that it's, estimated, about 42, percent of Americans, have had in some way their, health records breached. But. You know that's not, all that difficult to. See how that happens, and, you were no better than I but when, you have data in multiple. Different platforms sitting in servers, all over the place, manage with various, different security. Regimens, well, that's kind of a recipe isn't it for for, having data breached. So, I think, one. Thing that you do obviously, you do really really well is to, manage the security of data I mean it is your, entire, business. And, I I know you have, lots. And lots of people working. On that every day so. I think that, what. What you bring to the table obviously analytics. And data security. Most. People, win their privacy, when. They trust that their privacy will be protected. Are happy. To share at least portions. Of their health record, for, the greater benefit, of. Discovery. And for the greater benefit of care even, though it may not immediately, benefit, them but. I think we, haven't had the incentives, in the system to, communicate, that nor, have we had the portals. To. Enable people to elect, to bring. Their data in or, for. Organizations. Like yours to, be able to aggregate the, data in a sufficiently meaningful, way that. That. Truly leverages, the power of Big Data you know when I talk to my computer, science colleagues and I say that we have over two million unique, patients. In the, Stanford medical record, and we do have a de-identified. Version. Of that record their. Comment is oh that's not big data you. Know call, us again when you have 20 million if you have 200 million yeah then we would be interested so we. Have to figure out ways to aggregate. Together for sure to. Increase. The security of data and. Then to convey the benefits, of being able to study the data in the way that that we're. Capable of doing, thank. You I think we had a question here and I'll ask you to come back and use a microphone if you don't mind. With. Those remote. My. Question is that related to the technological. Advances, like machine, learning there, might be a move towards. Diagnosing. Patients, through, technology, do you see that as something, that's actually going to develop in the near future maybe with diagnostic. Applications replacing. Doctors, or do you see technology. More as augmenting, the power of doctors I think. It initially, will be augmenting. The power of doctors, and already. In an. Interpretation, of radiology. You know imaging studies we're. Seeing the tremendous, power of.
Augmenting. The diagnosis. And of using. Machine. Learning initially. On an imaging study, to. Give a reading and then having, that confirmed. Or not, by, by. A radiologist, you. Know I don't think that the. Need for I. Think the role of physicians is going to change for sure and, there, may be some specialties, more affected, by those, changes, than others but, the need to have a human interface in, in. Healthcare, I don't think is going away any. Time time, in my lifetime I certainly. Think the majority, of physicians. Welcome. The opportunity to, more constructively. Engage. Technology. You. Know one of the problems today is that. Electronic. Medical records, are oftentimes. Cited, as a. Or. The leading, contributor. To physician burnout in America, so, over, 50% of, physicians, in America today by a variety of different metrics are. Determined. To be burned out now, that is really really concerning, and is an organization. That trains, physicians its, enormous, Lee concerning, to me, and you. Know average, primary care physician. Is spending two additional hours a day, doing. Documentation, in, the electronic, medical record at home or after. Hours in addition, to the care that she or he is providing, to patients, and furthermore. During. Their, encounters, with patients, they may be at a terminal, typing. Not, looking, someone in the eye and, really, trying to have an. Empathic. Conversation. With them so. In that sense technology, gets a bad rap that's not really you know the bad rap is on the structure, of the systems and there are a whole host of reasons for that it's not technology, per se so. I embrace. What, is occurring. Today in, in, health, related technology, and what I know is going to occur, even more in the future. Thank. You I think we're gonna take a question remotely, from Dorie and the, highest voted one is at the top their medical. Technology, improvements, are great but often come with a very high price tag we're. Increasingly, faced, with healthcare occupying a greater proportion. Of GDP, how. Can a health care system like Stanford, deliver. Precision in a cost-effective, manner that is accessible, to the populace, right. It's, certainly, true that most. Technologies. Particularly when they're first introduced. Increase. The cost of healthcare. It's. Not true across the board though let me give you a counter, example that, I think.
Increasingly. Will be paradigmatic. For the type of, cost. Efficiencies, that can be occur, that can occur with, advances, in technology. You. Know it used to be that the, way to die, the only way to diagnose, a chromosomal, abnormality. In a fetus was, with a procedure called amniocentesis. That's. Inserting, a needle, through. The skin of the. Abdomen and, in. The mom and. Withdrawing. Amniotic. Fluid, and then looking and doing a chromosomal analysis, from the amniotic, fluid, it's. An invasive procedure it. Carries a risk to the fetus, and. It's an expensive, procedure and. Fortunately. It's, now virtually. Been eliminated. Because, it. Was found that you can as. Accurately. Diagnose. A fetal, chromosomal, abnormality, from, the circulating, DNA in a blood test from the mom as you. Can from withdrawing, amniotic fluid, so. Now a simple, blood test can. Be used to diagnose, chromosomal, abnormalities. That's. Tremendous. Savings in terms of of the. Dollar sign but even more importantly, in terms of, reducing the risk and. There, gonna be a host of other advances. Under that umbrella but. A lot of the problem, a lot, of the reason. I think for. You. Know medical technology, driving up costs is. That we by and large introduced. Medical, technology, in. The, most severe, complex. Diseases, and as. We start seeing technology. Transform. In, the. Areas of prediction, and prevention. Technology. Should actually be driving down cost and there, are examples of that today examples. Of companies. For example that, that. That. In diabetes. Management by. Measuring. Glucose, either continuous, monitoring, or intermittent. Monitoring and then, having coaches that receive the information as well as the individual, the. Management, of diabetes and a lot of people is being dramatically, improved, their. Savings. In terms of costs of health care in those, people, with diabetes. Far. Outstrips, the cost of the implementation, of the of the technology, so, it's. Both a problem, that the. Technology, has been focused, on the. Most. Severe acute diseases. Acute, and chronic diseases, and. The. Technology, has not made, its way into. The. Prediction and prevention realm and the way I think it will very soon. Thank. You question. Back here locally. Just. Question on changing. Healthcare, from reactive, to proactive using. A precision, health so like precision medicine things like genome-wide, Association, studies are great for diagnosing. Metabolic, pathways but. How do we change those, into, actionable. Things. Physicians, can do so for an example if I find LDL, cholesterol is, associated, with aortic stenosis do, I start prescribing statins, right away. To. A person who has LDL. High, cholesterol, rate so, it, seems it seems almost like invasive, to start becoming proactive. Yeah. Let me well let me mention an example. Similar. To the one you.
Just. Very. Nicely described, we're, already, we're, seeing an, analytical. Approach making a difference. You. Know congestive. Heart failure heart, failure is one of the most common, causes of people, being admitted to the hospital and, now. In Medicare, for, patients with heart failure who, are covered by Medicare if. You're admitted to the hospital, and then you go home and you get readmitted, within 30 days the. Hospital, gets paid very little for that readmission. So. What. We need to do is figure out what. Are the parameters, of, the. Management, of that patient, to know. That they're optimized, before they go home and that they won't be coming back into the hospital and, using. An approach, that, I'm. Certain. Is similar. To what you're doing every day in your work Nigam. Shaw professor, at Stanford. Was able to look at multiple. Different factors, almost like a jiwa. Study, for. You. Know for physiological parameters and. Identify. In a weighted fashion, you, know what the predictors, of readmission, would be and that, algorithm. Can, now be implemented, used in real time you. Know as physicians. Are making their rounds in the morning to say well you, know mr. Jones, looks, like the, probability, of readmission is, is less than 10% he's, feeling well you, know he's ready to go home or no, he's still got these indicators, that would, are, these parameters, that would indicate there's, a very high probability of readmission so. You. Know for that to be successful required. Aggregation. Of a lot of data and then machine learning, and then, testing, the algorithms, to see, if they if, if, they're. Accurately. Predicting. Probability, of readmission. That. Model, will certainly generalize, to other diseases and conditions including. The one that you just mentioned, on cholesterol management. Thank. You. Hi. Thanks, for thanks, for your lecture my, questions about precision in health and preventive health and nutrition. Nutritional. Studies are notoriously, difficult to conduct and curious, about your thoughts on how, nutritional. Studies can be better designed. And, and funded. High-quality, studies funded to to. Until. We have the benefits, of the, knowledge thanks, I think. To take the second part of the question first as we move more, away. From a fee-for-service model. And more. Into a model, that places. Risk and accountability. For. Health on the. Shoulders, of health care providers and systems as well as on, the, shoulders of consumers. Receiving, healthcare then. It, will, be a much more appealing. Business, model, to really focus in on nutrition. And other factors, that. Can. Play into keeping, a person healthy so. The reason that that there hasn't been one of there are many reasons and you're right it is difficult. To, study. You know what we are eating and then sort. That out among the many variables, that impact health. But. There haven't been the incentives. To focus. On those. Types of studies as there, have been for, example, to, focus. On new. Classes, of drugs for you name the condition, now, it shouldn't be an either/or world, I mean we shouldn't abandon the, amazing, work that's going on today for example in cancer immunotherapy which, is truly transformative. So, it is, precision. Devoted, at the Cure a part, of the equation of precision. Health but. There will be more, and more economic. Emphasis, on understanding the roles of nutrition, as it, becomes an economic imperative, for. Health care delivery systems, to be engaged more, in thinking about maintaining. Health of the, who are enrolled. In their benefits, and health plans. So. I think that's one element, the. You. Know the other. Most. People, employed. Today are being. Employed by organizations, in in one form or another are self-insured. So. There's a lot that companies. And organizations. Can do in, the. Type of food they offer in cafeterias, in the, type of nutritional, education that's. Offered, as a part, of being an employee there's, a lot that can be done right, at the workplace, even, before, the. Healthcare delivery systems, start embracing, it into the models of care. Yeah. Thanks for your overview. Likings. Resonate, I'll, ask this slightly provocative question, just, because I can, and. You touch the palate but, I'll ask a little bit more so, I think people. Have noted, that health care is a field with a particular, high tolerance, to pain or disorganization, and you, know just look at which other field - people still work 24 hour shifts, or which.
Other Field do people actually go to school for 15 years and. So. My question is around tolerance. To broken, stuff and so, for example. You. Know it's, faster, and easier to actually search trillions, of documents on the web than, a few thousand of notes. About this individual patient or, we. Are probably more active in predicting, the, click on this ad then, nigga model, is I'm predicting really arean and, so. What. Would it take for physicians. Or, administrators. Or patients. To say this. Is not really good enough, why. Is this such a high tolerance to. Things. Not being good enough when. And this is going back to your point great isn't good enough yeah. Why, do people tolerate in this. Industry so much things, that are not good enough, well. It's. A great point I think it I don't, claim to have all the answers but, it. Relates, to something I said during the talk and that is you. Know why is it for, example good good. Example of people not tolerating. You, know chaos would be what's, happened in the financial services, industry you. Know each of us in this room. Has. An ATM card probably. And we, can use it at pretty much any ATM, on the planet, to, withdraw cash perform, financial transactions. Or, we've. Gotten rid of the ATM card and we use venmo or something like that you know but there are a whole host, of, technological, solutions that. In order for those to work, the. Financial services industries, had to, come, together and, say we will adopt uniform, platforms, and standards, right why, because. I wouldn't, you, know would I choose to bank with, a bank that only allowed me to use my, ATM card at their machines, and then, when I'm traveling, you know I can't know so, there, was this imperative. Generated. I think by. Consumers, and embrace, then by business. Leaders that nobody have to make this interoperable, interchangeable. So. Why not in healthcare I, think. Partly, is the. Attitudes, that, many. Individuals have, had about. Healthcare, as being. Somewhat, their responsibility. But but also not. So much not in the way that each, of us I think accepts, that our finances, are our responsibility. Again the thought that well if I get sick I'll go to the doctor I'll take a medicine I'll be better, and. And. Then, that has been added to by. The traditional, attitudes in the profession, which have not, engaged. Healthcare. Consumers. In. The ways that for, example something, like the financial services industry is, always, known. It needs to engage its consumer, base more of I'm, not saying we should become like financial, services but, it's a combination, of the, inherent attitudes, of, or. The previously. Dominant. Attitudes, of both, consumers. And of, the, businesses associated, with delivering healthcare and not having. Those converge, on a model. Which really puts. The patient first but also puts. More responsibility. And accountable. Both. The. Healthcare consumer, and the provider to. Maintain, health, and not just treat, disease. Thank. You and we'll just do one last question you sure we can finish up with that and. Just the top part how, do you see machine learning applying to precision, health use cases that. It that is how do we take vast. Quantities. Of information from many. Patients, and actually bring it down to a single patient using machine learning sure. Well. There. Are so. Many sources, of information about, health right and. Those. Are not being, brought together. In. The way that I think they could be brought together we talked, before about the. At, least in the past have been very few incentives for health, care systems, to make their data readily, interoperable. I mean, an example, I told you about an ATM card still if I want to get my, medical, records from my primary care doctor in, Palo, Alto to someone else I'm still taking them either as printed, copies or, through, a factum fax, machine, where, do you see fax machines, today other, than in healthcare do you have fax machines in your office is here of course not.
Well, Maybe you do okay I bet, you don't use them very much so. It. You. Know we have to get be able to bring the data together aggregate. The data and increasingly, though, as, wearables. Become more, common, as we. Can look at other sources, of health-related, data, like. The environment. And. Then bring all of those factors, together those. Algorithms. Like. The algorithm, for predicting, it readmission. Will. Become much. More accurate, and much more comprehensive, than, than, they are today because, they'll, be based on a broader, diversity. Of data, thank. You please, join me in thanking dr. Marr thank you very much. You.