 So our first speaker, Eric Pradikslis, sorry if I put you that, is a Rubenstein fellow at Duke University where he focuses on data science that spans medicine, policy, information technology, and security. Eric is also lecturer in biomedical informatics at Harvard Medical School and strategic innovations advisor to Mendicine Suns Frontieres. And now without further ado, please welcome Eric Pradikslis. Thanks, Eric. Thanks, everybody for coming. How's it going? There you go. Can you hear me OK? Usually I'm too short to stand in front of podiums. So Nina asked me not to wear my jacket. She's a little stiff for this crowd. It's like she's going to halfway through this talk. I beg you, she asked me to put the jacket on. So anyway, I'm going to start off. I'm going to talk about medicine. I'm going to talk about infectious diseases. I'm going to talk about medical misinformation. And I want to start off and say, if you can actually be dispassionate about medicine, which is something scientists and clinicians are taught to try to do, is be dispassionate, then medicine actually starts to look a lot like math in some ways while you're being dispassionate. So you all are coders and hackers. So you want a little math in the morning. We'll do a little math. What's 100,000? Odds of death from getting a measles vaccine. 1 in 3,000, odds of being hit by lightning. 1 in 250, odds of having identical twins. 1 in 50 children is me. I have a peanut allergy. 3 out of 300, odds of catching the measles of exposure after getting the vaccine. But the case will be much more mild. Not to catching measles if you're unvaccinated. Rare. That's side effects. No, just so you know, I'm actually not in love with these numbers. And I'm going to pick on this a little bit. And so I don't think that telling somebody something is rare is that helpful when you're a sleep-deprived new parent with a baby and you're really worried about something that you read. So I think everything is benefit-risk. So whether it was my time as a drug developer, my time at FDA, or whatever it was, there's always a calculus in this. And it's actually really hard to do. But you've probably done it. If you ever tweaked your knee and had your ACL done, the anesthesia is like, well, there's a 1 in 20,000 chance you're going to react to this thing. And you know what? It really sucks if you're the 1 in 20,000. And so I wrote a paper in JAMA about using cyber techniques against medical misinformation. And I got a lot of shade and a lot of hate for it. But I read through them. And of course, I found at least one of the email that I knew I would get and that I wanted to talk to. And that's the poor guy whose child had a really horrible reaction he lost his child. So to that one person, that one child, that 1 in 200,000, that really sucks. And I could see how that could change and color your opinion for the whole rest of your life. How many of you have ever seen what malnourished babies with measles look like? OK, so we'll get to that in a second. So these are all about numbers. You know what the odds of 10 out of 10 are going online saying something and meeting an asshole? Because guaranteed you're going to meet an asshole online if you go online. Not only that, it's actually an asshole who's probably willing to do you harm or happy to see you come to harm if you don't agree with their ideology. And I think this is kind of the basis of what we're talking about here. Keyboard courage can be a really dangerous thing, especially to unknowing and unwilling folks. Now that said, there are different types of misinformation. And I'm going to be really careful about this. So I may say this 10 times. I actually have no problem with individual citizens, even if they're whacked, sharing their opinion that's their right. I actually have no problem with Gwyneth's being an idiot. I have no issue with that. Right? What I'm talking about, organized, anti-science, anti-population misinformation and mal-information. This is how they're described. So I spent September on Rohingya Service with Dr. Laub-Borders. And you know, these people were mowed down because they were updating their status on Facebook. Someone was looking for one of them. I'm okay. Where's this going on? And next thing you know, the government was using Facebook to target this population. Somali population in Minnesota was targeted by anti-vaxxers and had an outbreak because they're a marginalized population and they're vulnerable. So I'm going to talk a little bit about that. But 2015, I'm working with Partners in Health, my last Ebola mission in West Africa, and we're looking for the last patients. And it was actually something kind of new because no spacesuits. One of the first times we went into the village is why? Because you walk into the village in spacesuits and you take away sick people and they die. People run from people in spacesuits, right? This isn't good. Remember, a lot of these cultures are actually used to what we're learning, that governments can be overtly corrupt and overtly self-interested. Because if you grew up in Africa, that was your reality, depending on where you lived. Is that anybody with power would use it to hurt you and for their own gain. We're just starting to see that here in this country, but it's real. So these people look really happy to talk to me, right? And you know, in this particular village, there was a man who had a stroke. He didn't have Ebola. We got him transported. But we found a lot of kids with measles. We found a lot of cholera, right? Because there was a full quarantine, schools were closed. Everything was closed. The hospitals had been shut down because of all the workers who had been infected with Ebola. And so the idea here is that messaging really matters. And trust me, I'm a doctor. No, trust me, I'm a scientist. No. And the really thing about this is that a lot of scientists and doctors don't get that. They don't get the fact that a lot of people don't realize it, are we kind of part of this issue? So just a few pictures of babies with measles. I'm only putting this up for one reason. And that's that. It was an interesting study that I was watching. So New England Journal published this thing a few weeks ago. It's all about measles, it's fine. This person asked some really good questions. And you know, really detailed questions. Like where did you get your safety data? Did you get it out of VAERS? Which is the FDA Virus Adverse Event Reporting System. If you got it out of VAERS, everybody thinks VAERS is wrong. Did you apply the multiplier that Harvard Medical School says you should apply? Blah, blah, blah. Funny thing is I was previewing this. And I'm gonna tease them with former FDA Commissioner Rob Califf before I flew in. And he got all hot about this. He was like, well, you know, if you share data, they just attack the data. I'm like, then don't share bad data. But you have to at least try to answer the question. If they pile on and they don't, then you back off, you don't have to re-answer again. Now in fairness, I actually looked yesterday and the author did put up the answers to these questions. So good for them. Good for the New England Journal. These were fair questions. And they stuck their neck back out and answered them regardless. The other thing about this is the idea of public health. The idea of public health is that you're protecting the public more so than the individual, right? So herd immunity matters. Here's a really sad story of a young man with leukemia who died because he was exposed to measles. So we're in a room right now. We're all together. We're all breathing the same air. You know, you kind of, if you knew you were like dripping flu, sneezing and blowing, would you come in and sit down next to everybody? Probably not. So there is an idea here. Now, public health, you know this. Why were we, when we come back during the last Ebola outbreak, why were they allowed us to put us in jail for three weeks every time we came back? Because it was a mandatory evacuation order for the countries we went to. We broke the law by going and so it was their right to imprison us when we came back and we did. They were nice. They put me on house arrest. I had friends that went through a lot worse for three weeks every time. But you know, we knew it when we went. So I think the idea here, public health and that where people need to help people. And also we have to, you know, you don't want to hurt somebody on accident. And so I get the whole, well it's my body, I'm going to do whatever, but you know, you get schools being closed in New York right now with little kids and stuff like that. So it's a tough issue. Interestingly, medical charlatanism is as old as it gets. So everybody's heard of caveat emptor, buyer beware. The first known case, 1603 in England was somebody basically selling an intestinal stone that allegedly had magical powers called a bazaar stone. Chandler versus Lewis, this is fascinating. It actually was medical charlatanism. And basically what they decided was because there was no warranty and because the buyer couldn't prove that it wasn't a magical bazaar stone, they found for the plaintiff. So the other thing to think about here is that is that in medicine, the customer isn't right, you know, and that you have to watch out for you. It's caveat emptor. So when we're looking at what goes on in healthcare, self-educating, making your own decisions, these aren't easy things because not everybody's gonna understand the science or even the beginnings of the science. The other thing is that medicine is always playing catch up. You can't regulate something bad if it's not illegal yet. So I was at the FDA for this. How many people remember the New England compound and pharmacy thing? These people committed mass murder, right? They really did, right? They had basically dirty drugs that get injected right in your spinal column and people died. Here's the thing, being on the inside of the FDA, when this is going on, a reporter of the commissioner actually knew the inside story on this and what was killing them about the inside story is, of course, as soon as this happened, FDA, what are you doing about it? The FDA had actually tried to sue these people multiple times on anything they could think of because they were pretty sure they were doing bad stuff and they lost every lawsuit they filed. And then after this, and then as you can see from eventually after this bad thing happens, well maybe you should regulate compounding pharmacies. In 1910, people were selling belts with radiation on them as a cure for diabetes and stuff like that. So there's always been charlatanism, you know, the vaping stuff today. If it was really about anti, if it really about smoking secession, why was the first flavor bubble gum or one of the first flavors if you weren't trying to attract kids? So the people have to beware and they've got to watch what they're doing. But if something isn't illegal yet, is it illegal? Right, so if is targeting a community on Facebook, going into that community with pamphlets, talking them into, talking them out of doing medical care that leads to injury, illness or death, a crime. Today, not yet. Should it be? Good news, I'm not a lawyer, I don't want to know. I know what I personally think that I see a lot of sick people and I prefer that people don't get sick. But at the same time, it's a tough thing to do. But is it harm? Is it harm like yelling fire in a theater? Or is it harm like plowing your car into those same Somalis in Minnesota that were targeted? Something to think about when people go in and you drop vaccine rates like this in a marginalized community and then have people get very, very sick, what do you do? And again, I'm not talking about my aunt who may want to say, do not brush your teeth with that water if it has chlorine in it. She can tell anybody she wants about that. Or she can advertise anything she wants. I'm talking about organized campaigns of disinformation and malice information. Here's the thing. So as somebody who's faculty at Duke, Med School and Harvard, we're a big part of the problem. That whole rare reaction is rare. What does that mean? How rare? I mean, are we giving people? So I really like the line that what we have is a medical problem with a media dimension, not an information problem. That trust in medicine, trust in science has flagged over time. Then, and sometimes for good reason, who knows if it's really smart and healthy to eat eggs? Dietetic epidemiology, are you kidding me? I'm 52 years old, at least 10 times in my life I've been told the opposite of what I have been told. Now, the fact that science changes that we learn and grow doesn't make science bad. It does make it imperfect. It does open it to criticism. So living in an open society where it's okay to challenge whether or not eat eggs, I think that's fine. But I do think that we have to understand, the public has to understand that the World Health Organization leads patient harm as the 14th leading cause of illness and disease on the planet. There was a study out of Johns Hopkins a few years ago that said it might be his highest third leading cause of death in hospitals. So people do get hurt in healthcare. So we in medicine have to be better. The hospital, we have to figure this stuff. We've got to own part of it and not just discount people out of hand. Now, what I like about this is I love the Facebook post from BCH, not just some faculty there too, on the right here, because what the mom really says is, you know, they treat us intelligently. That's what we like about this place. They listen to us and our voices matter. They're not talking at us. So I do think that even scientists and doctors can learn how to do this stuff. So spent most of my time, I was a J&J for 20 years. I was an FDA and went to doctors above borders and places like that forever. And I've worked on a lot of compliance. The CIO would change, I was a CIO at FDA, so I've worked on a lot of compliance. You know what happens when things go wrong, at least this is my opinion, is that most of the time it's a mistake and then sometimes there's a few bad guys, right? Most of you work at device companies like that. Things go wrong. So I'm gonna forget something, there's a slip, there's a faulty part, whatever. There's a breach. Someone wasn't trained well. They made a mistake. I mean, when I was, I worked on the harvesting organs and tissues when I was in undergrad. And I actually shipped off expired tissue one day. I had to reread the SOP and do that, figure it out, right? You make mistakes. But what about the bad guys? So what I wanna talk about the rest of the talk is what should we be doing about the bad guys? I know they're in the minority, but they might be really bad. So can you do it by a counter messaging? This is an interesting thing, right? You know, in general, and I'm no expert on this stuff. I'll be the first one to say it. If for anyone else here, there's better experts in this. Most people actually say that counter messaging isn't that effective. A lot of people will say, and actually some of y'all that were in the device thing yesterday, we said it, right? We said, don't answer question number two because it validates the question if you try to answer it, right? So you back away from that. But, you know, the best study that I think on this has actually been anti-terrorism. How do people become radicalized? So counter-marriage messaging, we know because we've seen really good people try to do it. And I spent three years going back and forth in Arabia building a cancer center in the 2010 timeframe for the only large cancer center in the Arab world that imams getting up and saying violence is bad and hasn't stopped extremists from doing violence as much as they respect their imams, it hasn't happened. So counter-messaging, what appears to work is more like alternative messaging. You tell a different story instead, but don't directly go at what's going at. So credible voices, all of these types of things are somewhat difficult. My favorite headline in a while, read the two tweets, it's the best. A lady posting on an anti-vax site, what do I do to protect my child now that there's a measles outbreak? And then I don't think this, while this gentleman's response is funny, I don't know that it's all that useful for this poor lady who's now worried about her kid. And by the way, still deserves to get to protect her kid. Even if she is on a different website. And what really is going on here is everything else, it's also polarized that anybody that wants to be in the middle and be normal or be smart and be reasonable, they get shouted down by the two extremes that are trying to fight it. How do we get kind of towards the middle? So positive case studies work, learning by history and role model tend to work. Alternative messages and work. If any of you followed on Twitter, there was a great piece that was put up by the Hunger Games Cast in 2015 that I put up. You can look at my Twitter and you can find it. What was awesome about it was they basically just said how they're trying to calm down all the hype, you're gonna close the airports, they're gonna close the port. There was a story of like in Ohio of someone who came back and went into a wedding dress shop for bridesmaid dress and that shop actually went out of business because nobody would shop there because somebody who had done a bowl of service had been there. And the whole thing, so the Hunger Games Cast put out this kind of cool message. What was really cool about it and what was first about it for me is one, Paul Farmer, my boss at PIH was in it. More importantly, my daughter was 12 and couldn't tell anybody in her school, middle school where I was. Could you imagine what Playground Talk was like when all that hype was going on on the table? She could say I weren't anybody. Her best friend's parents knew, we knew. And so that video was awesome. Gender, faith, ideology-based messaging and need for specificity. This last one's, the second and last one's really important. How do we match the volume and velocity? Because I've seen this. I have colleagues that will get an article in JAMA saying look, at the end of the day your statins are probably good for you and they get 5,000 people banging on them. What are you, one person, what do you do? I think we need a troll army of med students but I'll talk about that in a few minutes. There they are. How to accomplish better messaging. So at this point, the rest of this talk is my shameless ask for your help. You people know this stuff better than I do. I am looking for, I'm gonna be running some experiments. We need to try some things. There's a lot of people like me, academics, getting up and saying why this shit is terrible and find anybody actually doing something besides saying it's terrible. And I think the cyber community can be one of the communities that can actually do something that can make a difference. Oops, sorry. So how do you fix it? You know, how should patient groups convey their concern? I was looking at Andrea here. How do they communicate? How do they do things like that? How do you know when it's going on? So one of the experiments we're gonna run is a great guy, colleague of mine at Boston Children's Hospital named John Brownstein, built something called the vaccine centimeter. It measures a centimeter of vaccines on all social media platforms. Built it in 2014, shared it, nobody would fund it. So if you looked at my little chart about measles has gone up, I wish the CDC had funded it, or FDA or somebody had funded it, but we'll fund it. So go into these communities, measure what's going again. Not so much to say this community is bad or good because of sentiment. I wanna understand signals. I wanna understand that if I sort the social media by communities, am I seeing targeting? Am I seeing ethnic groups being picked out? Am I seeing, that's what I wanna see. I wanna know if that's going on because that's a signal that's maybe worth managing. What's the correct oversight? Who watches this stuff? One of the things, one of the comments Jim had took out of my paper, I say, so how does the AMA deal with the KGB if it happens to be state sponsored? That's a one-sided fight, isn't it? How does that really happen? So there's that. So we're gonna do consistent messaging pilots, counter messaging, trolling and adaptive pilots, and then measurement to see how things go on. And I'm looking for people to help with that. I'm looking for people to teach for that. I will put you in rooms full of med students that wanna do that. So do we need a CVE list for healthcare? Right? Chris is here. He's been helpful to explain this. Chris from MITRE explained the CVE list to me and how it works. It's not exactly, but do we need to know? Especially again, I'm talking about the bad guys. I'm talking about targeting of groups. So if you look at the breakout in the Orthodox Jewish community in New York, if you look at the Somalis in Minnesota, while they weren't targeted online, those were low tech. That was pamphlets, that was talks, that was phone calls. How do you think the people organized to do it? That was online. So you would have seen, should have seen leading indications into it. So this concept, how would you do it? So this is one of the pilots I'd like to talk to people. How do we set, what would you do? What would you watch? What would you change? We've got the epidemiology side. People have mapped the whole ethnography of some of these groups. How do you know when it's going on? What if it does look like it's state sponsored? Right, I mean, Commissioner Califf has said, look, if you want to kill Americans, you know, talking them out of their statins will be a certain way to kill a lot of older Americans. You know, you're gonna win an election that way? Maybe, I don't know. But if we see it, do we recognize it? And then once in a while, you gotta hack somebody. So if we find something's really, really bad, right? What do we do? If we see a community's being targeted by a group, what do we do? I actually have no problem dreaming of a world where their websites go off for 24 hours and I have no problem dreaming of that world. They have their right to do whatever, but if I see targeting or something like that, I don't have any issue with that. Let's talk about it. So there is the idea of saying, what do you actually do? We are doing honeypot experiments, working with some experts and honey tokens. If anybody wants to see that, we're gonna put up data, we're gonna see where the data goes. Anybody that has expertise and wants to see that, we gotta do it. Right now, everything I'm talking about is diagnostic. Not talking about any retaliation, I'm not talking about hacking, I'm not talking about any of that stuff. That's for law enforcement to decide. I'm talking about as a scientist, I wanna diagnose and I wanna figure stuff out. So because we don't learn our lesson, we've got a bowl again, right? Really bad. So guess where I'll be heading in a couple of weeks? So I'm gone again. I gotta say bye to my wife and kids because we don't learn lessons we should have learned two years ago. It's not easy to think about infectious diseases. My mentor was my great uncle. Graduated Harvard Medical School in 1943. We shipped right to the Battle of Britain. Was there at 48 concentration camps? Came back, practiced 50 years. Grew up right after Spanish flu, thought the world was gonna be taken out by an epidemic. This little guy, we're trained not to touch people. That's not me, that's my friend, Tim, holding that guy. Because the ambulance drivers were the first ones to go in Ebola, put the people in the back of the car, close the door, nobody opens it until it gets to the clinic and we can take them out. That little guy was with his mother, his father, a cousin and a grandmother and when we opened that ambulance he was the only one who was alive. And all that happened during the six hour ride. So we need to work on these problems and the world is a lot smaller than people say it is, right? And by the way, stuff works when it works. So here we are going over to help Africa with Ebola and you've got places like Rwanda where cervical cancer is the second leading cause of cancer death and they're killing it with campaigns. We know the HPV vaccine is a huge success. So anyway, there's that. I love this tweet. I haven't met Jenny, I'm stealing this tweet. I do think y'all are the immune system of healthcare. Thank you. That's kind of a great question. Complaints, criticisms. Okay, yeah, it's a great question. I don't have a definitive answer. All I can say- Keep the question, please. Yes, sir. I'll do it with a microphone. The question is, in the conversations with people, how do you meet people in common ground when you've got somebody who has strong opinions and they're trying to manage a lot of stuff, right? So it's a great question. My answer is what I've used as patients and I think that's why we need an army, a troll, you literally need an army of people and trust me, just so you know, I'm absolutely not the phenotype for some of these conversations but I've got friends and colleagues that are. Now, you can't get sucked in and you can't take forever with it but you've really got to look at it and be very fact-basic. This is what I think and this is kind of where it is. You also have to try to conduct something outside. If somebody is like, if somebody, I had no idea if I was going to have someone call me out in the first five minutes of this talk but if they did, I would answer two questions and ask them to shut up so I could finish my talk and then I would have bought them lunch but, you know, just, but I can do that. You can do that one-on-one. When you're managing a group of 20,000 you can't necessarily do that but I do think people have to get there because, again, that one voice can cause a lot of trouble and it really amplifies. So it's a great question. Sir. It's more a question about the campaign. We see issues in modern US medicine at a peer-to-peer level especially when we're looking at areas that are underserved. If we can get other physicians to agree and to come along with modern messaging and evidence-based medicine and sticking both practices they know work, how do we expect this is going to work in a moderately educated society? Yeah, that's a great question. So the question was is that it's hard to change everybody's mind, even experienced doctors and how do you do it in a thing? I actually think we have to land in a different place. That's why I like the idea of medicine is, medicine is a medical problem with immediate dimension. I don't think 10 years post the internet where anybody has the authority they used to have as a professional anymore. I don't care what your profession is. You can be an architect, you can be whatever. And so I think a lot of people aren't used to dealing with that but I do think it's more of a conversation now and that it's going to be a mixture of what people are doing and the next generation of clinicians as we train them have to figure that out because they're not going to have that white coat authority and I don't know that that's a bad thing. They're being trained by the ones who do, med students, residents, fellows through the program having the same traditional values in a world that's changing outside of them. There's a lot of... There's a lot but at least what I see in the classes, I work on, for example, at HMS we did a computationally medicine medical class where we actually had med students rotate into a pharma company and figure out value and pricing. I do think it's going to take stuff like that. I think we have to change the way we're medicine. I actually think that we should be more selective and opportunistic about what med students do in their third year. I think physician scientists get this all the way through but I think that third year is just a prime spot to go work with me at MSF, do whatever is your passion, figure it out and find it. It's a great question. Any other questions? Thanks guys, see you.