Latest / CareTalk: Healthcare. Unfiltered. / How Randomness Impacts Healthcare with Anupam B. Jena

Transcript

How Randomness Impacts Healthcare with Anupam B. Jena · CareTalk: Healthcare. Unfiltered.

• 0:00
  Dr.
• 0:00
  Bapu Jena is professor of healthcare policy and medicine at Harvard Medical School and a physician at Mass General Hospital.
• 0:07
  He's co author of Random Acts of Medicine, The Hidden Forces That Sway Doctors, Impact Patients, and Shape Our Health.
• 0:14
  He knew he wanted to be a doctor, but got into economics by chance.
• 0:18
  That experience influenced his thinking about the role of randomness in healthcare, leading to all sorts of odd but insightful observations, which we'll cover in this episode.
• 0:34
  Welcome to Care Talk, America's home for incisive debate about healthcare business and policy.
• 0:39
  I'm David Williams, president of Health Business Group.
• 0:42
  And I'm John Driscoll, the president of Walgreens Health.
• 0:45
  Well, I have a favor to ask.
• 0:47
  If you're enjoying the show, could you please leave a rating or review of the CareTalk podcast on your favorite streaming service?
• 0:53
  That helps us stand out from the crowd and lets others find us.
• 0:56
  We need those reviews, people.
• 0:58
  Appu, welcome to the show.
• 1:00
  Thank you for having me.
• 1:02
  Listen, you wrote a healthcare policy book that's been described as fantastically entertaining And deeply thought-provoking.
• 1:10
  How is that vital?
• 1:11
  I don't, I, I don't even believe that, David, there's just no way.
• 1:15
  We, we spent too much time in this, first of all, that was my own description, so I don't you have to take that with a , with a grain of salt.
• 1:23
  Um, it's a, it, it's a fun book.
• 1:25
  If you've, if you've heard about Freakonomics, if you read the Freakonomics books, I would describe it as Freakonomics meets Medicine.
• 1:31
  So it is healthcare policy in, in some respect, but really it's about how.
• 1:37
  These strange things affect our health, these patterns that I've been studying for about a decade now.
• 1:42
  And I'm attracted to them because they're just interesting and fun, and they're a little bit out of the box.
• 1:48
  But as the book shows us, I think there's things that we can learn about healthcare more generally, uh, from those sort of observations.
• 1:55
  You know, a lot of what you talk about is through these, uh, what you call these natural experiments.
• 1:59
  I mean, maybe, John, for John's benefit, because I'm sure everybody else knows, but, you know, what is a natural experiment and why bother?
• 2:06
  So the short answer is, uh, it's, I describe it almost like a randomized trial.
• 2:10
  So when you, uh, take a new medication, it's typically.
• 2:14
  uh, approved by the FDA for that indication, or often is, and that comes out of a randomized trial where some investigator took a bunch of people and put half of them in on that drug and the other half on something else, and that's how we know the drug works or doesn't work.
• 2:29
  Now, we can't do that for a lot of things we might care about in healthcare.
• 2:33
  So, uh, if you want to know whether a certain type of doctor is better than another doctor, You can't conduct a randomized trial, and that's where these natural experiments enter.
• 2:42
  It turns out that there are these situations where people are, by chance, by nature, exposed to different paths of care, different experiences that can tell us something about what works and doesn't work in healthcare.
• 2:55
  But it's sort of nature's randomization.
• 3:00
  You know, Bapu, the, the, the freakonomic stuff was often looking And finding patterns that other, other people didn't see.
• 3:08
  I mean, what do you, what do you think that the, the, the sort of this approach to randomness allows you to find in terms of patterns that perhaps others either haven't paid attention to or, or not even looking for?
• 3:22
  So let, maybe I'll give you an example.
• 3:24
  Um, so one of the big questions that we face in medicine is when to do something and how much of it to do.
• 3:31
  So, if, uh, if you're hospitalized with a medical condition, there are a ton of different paths of care that a medical team could take.
• 3:40
  And some of those paths are informed by really good evidence, but most of it is sort of a black box.
• 3:46
  It's part of the art of medicine.
• 3:48
  So how do you figure out whether or not doing more of some medical procedure in general is beneficial versus not?
• 3:55
  Well, ideally you have a randomized trial to get you that information.
• 3:58
  And that's not always possible though.
• 3:59
  So, I'll give you an example from the book and there's a chapter called What Happens When All the Cardiologists Leave Town.
• 4:06
  And it's based on a study that we did, uh, five, ten years ago now, which looked at what happened.
• 4:11
  Less interventional cardiology.
• 4:13
  Exactly.
• 4:14
  Less interventional cardiology.
• 4:15
  So what happens when all the patients go to the hospital when cardiologists are at the American Heart Association or American College of Cardiology meetings?
• 4:24
  You would think that they do worse.
• 4:26
  Because staffing might be lower if you were cardiologists around.
• 4:29
  But guess what?
• 4:30
  They do better.
• 4:31
  They do a lot better, in fact, and the other data point is that rates of certain procedures that cardiologists frequently use, they go down.
• 4:38
  So this is sort of a real world data point that tells you that sometimes we might be doing too much medical care, or we say less is more.
• 4:47
  So let's actually talk about some of the other examples from the book.
• 4:50
  So even, even though you wrote that, you know, about being entertaining and thought provoking, you wrote the rest of the book, too, and I can assist.
• 4:55
  Tell you that it is actually entertaining and thought provoking.
• 4:58
  So one thing that I looked at was, you know, this relative age effect, you talk about that and you know, that's another sounds like a kind of a technical term, but I hear it has to do with kindergartners.
• 5:08
  And Tom Brady, what's that about?
• 5:09
  Yeah, yeah, so the short answer is that, uh, in, in every state, and actually it's across the world, but in the state that, that, uh, I live in Massachusetts, the entry, uh, for, for kindergarten works as follows.
• 5:22
  If you're age 5 by September 1st, you can enter kindergarten.
• 5:26
  If you turn 5 on, let's say, September 3rd, just a couple days later, you have to wait a year.
• 5:32
  to enter kindergarten.
• 5:33
  Now, what does it have to do with ADHD?
• 5:35
  Well, it turns out if you look at any class, uh, kindergarten class, the August born kids are 30 percent more likely to be diagnosed and medically treated with a medication for ADHD than the September born kids.
• 5:51
  And these are kids who are just separated by birth, by an order of days.
• 5:56
  And because of some arbitrary threshold, we see this difference in the likelihood that they're diagnosed and treated for ADHD.
• 6:02
  Now, why is it happening?
• 6:03
  Well, it happens, it happens because the August born kids, they are the youngest kids in their class.
• 6:09
  So when they are fidgety or inattentive relative to their peers, uh, They're more likely to be diagnosed with ADHD when really it's just a maturity effect.
• 6:19
  If you just watch those kids for six months, some, some proportion of them would actually not exhibit signs of ADHD and you wouldn't want to diagnose them and, uh, with that condition.
• 6:30
  So it's the relative age of a child, the August versus September born kid in the same class.
• 6:36
  That's really important.
• 6:37
  And, by the way, you don't see this for diabetes because that's a condition where you have a blood test to be able to diagnose.
• 6:43
  It's a subjective condition.
• 6:44
  So, I, I, you know, I, I'm just trying to figure out how you got the greatest of all time quarterback, worked that into the kindergarten reference.
• 6:51
  I'm not quite sure, David, that, what's going on there?
• 6:53
  So it, it turns out that Brady was redshirted.
• 6:56
  So you, um, he went to the University of Michigan, um, he, uh, was redshirted, meaning he was sort of on a training team for an extra year.
• 7:03
  And some people might think, well, whether that, that extra year helped him.
• 7:07
  Become a more successful quarterback.
• 7:09
  And so it's the same idea here.
• 7:10
  It's a lot.
• 7:11
  It's a question that a lot of parents think about is, is my child ready to enter kindergarten or not?
• 7:14
  Patriots and Brady reference into a book to keep it, keep it lively.
• 7:18
  Exactly.
• 7:19
  Buying.
• 7:19
  I'm not buying that one at all.
• 7:21
  By the way, LeBron James did not have an age relative.
• 7:27
  You get into professional sports.
• 7:29
  I think you're, you know, Kobe Bryant.
• 7:31
  I mean, I don't know that I'm not sure this one holds up as well for.
• 7:34
  for great athletes.
• 7:36
  But, uh, and do you find that age related effect across other categories of illness and diagnosis?
• 7:46
  So, yes and no, and by the way, it does hold up.
• 7:48
  So there is a, a, a finding popularized by Malcolm Gladwell in his book, The Outliers, where he looks at, uh, NHL, National Hockey League players, and the same sort of phenomenon occurs.
• 7:58
  I, I, uh, and, and I, and I think the thing with Malcolm Gladwell is you need to check his math because it doesn't always hold up.
• 8:03
  Like the 10, 000 hour rule around, around expertise does not actually hold up the natural capabilities.
• 8:11
  Um, uh, so I, I, I, I, but he does, you're going to bring up the sainted essayist Malcolm Gladwell need to admit that not, not all of his facts are as well researched as yours.
• 8:22
  That, yeah, that might, that might, that might be true.
• 8:25
  There's other examples in tennis and baseball, which we actually talk about in the book where it seems like this phenomenon is going on.
• 8:31
  But John, you asked about whether or not you see this for other diseases.
• 8:34
  And so we do see this a little bit with things like depression.
• 8:37
  So there's a study that we refer to in the book.
• 8:40
  Uh, which shows a very similar relative age effect.
• 8:43
  And also, you see this with, uh, teen fertility.
• 8:47
  Now why is that?
• 8:48
  It's because in any, let's say, high school class, a, a teenage girl who's young for her grade might be exposed to peer pressure from her older peers.
• 8:58
  That's different than if the girl was older for her grade.
• 9:00
  And so you actually see increased rates of teen pregnancy among these young for age, uh, or young for grade.
• 9:07
  Uh, uh, girls.
• 9:09
  What's the, what's the impact on depression?
• 9:11
  Where do you, where do you, how do you, because that, that's obviously a hot topic since post COVID, it's exploding everywhere.
• 9:17
  What's the age related effect there?
• 9:19
  It's a similar thing.
• 9:20
  It's so that the people who are young for their grade were more likely to be diagnosed with depression.
• 9:25
  And again, think about the mechanism here.
• 9:27
  If you are always struggling relative to your peers, and that would be particularly true at early ages when, you know, your percentage life Your percentage time on Earth is really large.
• 9:37
  Mm-hmm.
• 9:38
  , right?
• 9:38
  There's not much difference between a 65 and a 66 year old, but there's a big difference between a six and seven or maybe even eight and nine year old in terms of maturity.
• 9:46
  And so you start to see differences in depression.
• 9:49
  The young kids, kids who are young for their grade relative to their peers, are more likely to be diagnosed, uh, with depression.
• 9:57
  So, so far I like, you know, we like to take the big picture takeaways here, and so far I would say, I'm going to go back to kindergarten for another year.
• 10:03
  I'm going to spend another year in, in college and then I'm going to spend all my time at the uh, at the cardiology conference to keep people safe.
• 10:10
  Yeah, exactly.
• 10:10
  So if we move on, from that, I also saw you have a chapter where I think you described E ZPass as basically a public health program.
• 10:17
  How could that be?
• 10:18
  Yes.
• 10:19
  So that is some work by, um, uh, Janet Curry, uh, and, uh, Reed Walker.
• 10:24
  It's a, it's a really interesting study that the question that they were trying to answer, it's what is the impact of pollution or air quality on health?
• 10:32
  And they were focused actually on, on, um, sort of maternal health and, and, um, uh, infant health.
• 10:39
  The, the question is actually, even though we might think we know the answer, it's hard to study because if you just look across parts of the country that are heavily polluted, and not heavily polluted.
• 10:49
  And you compare outcomes in those two parts of the country.
• 10:52
  It's not a proper analysis because the places that have worse air quality might have a lot of other things that are different in them that might cause poor infant health.
• 11:03
  And so you, you don't have a really good experiment there because you're not holding everything else constant.
• 11:07
  And so what, uh, they do in that study is they say, alright, well, what happens when E ZPass enters into a market?
• 11:14
  Well, one thing that E ZPass does is it reduces congestion and basically car stalling in highways.
• 11:21
  And so because the cars are allowed to just kind of go through the booths in an automated way as opposed to sitting there, there's less local pollution.
• 11:29
  And they look at what happens to infant outcomes in an area when E ZPass enters.
• 11:34
  And they look before and after what happens when E ZPass enters to infant outcomes.
• 11:37
  And they do the same thing in areas where E ZPass does not enter.
• 11:41
  And what they show is that Infant outcomes improve when E ZPass enters, and the mechanism is thought to be related to a reduction in pollution because the cars aren't just hanging out at the toll booth.
• 11:55
  And that's a pretty significant difference.
• 11:57
  I mean, that, that was, I could understand why, like, you might hypothesize such an effect, and yet it seemed to be fairly profound.
• 12:04
  Is that, we believe, I know John Gladwell, but do those numbers hold up in your view?
• 12:09
  I would say so, and the nice thing is that we don't have to hang our head on only that particular finding.
• 12:14
  In the same, same chapter, we talk about some other work, uh, by Nolan Miller and, and others, which uses wind direction.
• 12:21
  Just a really beautiful study design.
• 12:23
  So, again, places where there's lots of pollution.
• 12:26
  should not be compared to places where there's less pollution to establish the effect of pollution on health because the places are different.
• 12:32
  So what they say is, all right, well, let's look at a single place and look at instances where the wind is blowing pollution into that area on any given day versus situations where in that same area, the wind is blowing pollution away from that area.
• 12:48
  on a different day.
• 12:49
  So it's literally something as random as wind direction.
• 12:53
  And what they show is that when pollutants are higher in your area because the wind is blowing the pollution downstream into your area on a given day, there's an increase in hospital admissions for outcomes that we think might be related to pollution.
• 13:06
  But on another day in that exact same area where the wind takes the pollution a different direction, you don't see it.
• 13:12
  So it's a, it's a sort of a more elegant study design to get at what is the causal effect of pollution on health.
• 13:20
  You know, we talked about, uh, your own experience of, I think, thinking you were going to get a PhD in biology and then ending up with economics for sort of quirky, uh, reasons of who you ran into.
• 13:29
  And that's, you know, both of those are good paths, but there's some other things where you've got a random.
• 13:33
  One might have a random encounter.
• 13:34
  Are you suggesting a random path to education, David?
• 13:38
  No, John.
• 13:38
  I'm going to talk about opioid use, actually.
• 13:40
  So one of the things that's in the, uh, you know, in, in the book is talking about how if you go to the emergency room, we're almost by definition, you're not going to go see your doctor.
• 13:49
  in the emergency room.
• 13:50
  Unless John has like the super concierge thing where actually he does have his own emergency doctor too, but that's a different story.
• 13:55
  The, um, That's just because I've got so many doctors to call, David.
• 13:58
  It's a, it's a, it's a function of having, having worked in this industry for way too long as I call a, call a friend.
• 14:04
  And as I recall, uh, At least one of your brothers is a doctor.
• 14:09
  So I guess we'll give you the same thing.
• 14:11
  Oh boy, John.
• 14:12
  All right.
• 14:13
  So anyway, well, next thing we'll have to talk about real estate on Nantucket or something.
• 14:16
  But John, the, um, one thing that I noticed is when we're going to, uh, the emergency department, which is really, you're not getting your choice of doctor, it seems like.
• 14:25
  The practicing patterns of the specific physician that you see could actually have an impact, not just right there in the emergency room, which is pretty obvious for that episode, but actually long term opioid use.
• 14:36
  Can you say more about that?
• 14:37
  Yeah.
• 14:37
  So let me just take a step back.
• 14:39
  We have used this variation in who you happen to see, whether it's in the emergency department or in the hospital in a lot of different ways, because it's totally random, right?
• 14:48
  If you go on Monday.
• 14:49
  you get Keith.
• 14:50
  If you go on Tuesday, you get Lisa.
• 14:52
  If you go on Wednesday, you get Timothy.
• 14:55
  And each one of these doctors, either in the ER or in the hospital, might have different ways that they like to practice medicine.
• 15:02
  And it allows you to uncover what is the effect of that way of caring for people on their health.
• 15:09
  Um, and so David, we did do some work on opioids and this is, uh, you know, we mentioned this in the book.
• 15:14
  There's a paper in the New England Journal of Medicine now more than five years ago.
• 15:17
  And the basic, I The basic question that we wanted to answer was whether or not, um, a single opioid prescription could lead, um, to long term opioid use.
• 15:28
  And I kind of think about this as, um, like the soccer mom analogy, which is to say someone who previously didn't have any problems with addiction, they sprained their ankle or they went to a dentist and were prescribed an opioid and then they become long term users of opioids.
• 15:45
  Does that happen, right?
• 15:46
  We talk about it happening, but how do you show?
• 15:49
  that it happens, and how do you quantify the effect?
• 15:51
  And so our insight was, well, you know what, in the emergency room, uh, people see doctors of all different types of proclivities.
• 15:59
  Some are very likely to prescribe opioids, and some are not.
• 16:03
  And as a patient, when you go to the doctor because of a sprained ankle, or you fell, or whatever it may be, you don't know who's going to be in the ER that day.
• 16:10
  And what we showed is two things.
• 16:12
  One is that in any given emergency department, the likelihood that a, that a physician prescribes an opioid.
• 16:19
  An ER physician varies from something like 5% of the time to almost 20% of the time.
• 16:25
  And that's all the doctor.
• 16:26
  The doctor just has a different threshold for prescribing an opioid.
• 16:30
  Uh, so if you go on a day when there is a high prescribing doctor, you're more likely to walk out with an opioid prescription.
• 16:36
  Not rocket science.
• 16:37
  The interesting thing though, is if you look at those same people who by chance happened to see that high prescriber and you follow them out, let's say six months or a year later, they're about 20 to 30% more likely to be on long term opioids.
• 16:51
  So it helps kind of show.
• 16:53
  This idea that something totally random, like who you happen to see in the ER, might affect your likelihood of getting an opioid at that visit, but also becoming someone who's long term, uh, opioid using.
• 17:06
  Got it.
• 17:07
  You know, in a lot of the book, it seems like you're, you're obviously building on the work of others and in other fields, and, and in some places you're definitely channeling.
• 17:13
  Conneman and Tversky.
• 17:15
  I think about some of these things you use, you know, representativeness, heuristic and availability bias.
• 17:19
  How do those sort of And do we want to, do we want to just smarty pants Harvard person explain who Conneman and Danny Conneman and Amos Tversky are and what their background is?
• 17:29
  Or do you just want to just throw that academic reference out there via the Nobel prize winners who are the founders of, of understandable behavioral economics, uh, in order to showcase it?
• 17:42
  Your elite education, David.
• 17:43
  I like John.
• 17:44
  I like how you are both asking and answering the question at the same time.
• 17:48
  Spoken like a true Harvard man, I might add.
• 17:51
  Um, so no, I, I think I'll leave it at that.
• 17:54
  But, uh, uh, yes, it seems like there's some of these same terms that we may have heard in other contexts.
• 18:00
  So, so one of the, one of the more famous, uh, Danny Kahneman and Amos Tversky, uh, examples is, is whether it's, I think the Israeli Judge's example where the, if you, if you are, if you sort of before, depending on the time of day, a particular perp was in front of a judge and whether it was before or after lunch, uh, it's substantially affected, uh, very consistently more on time as opposed to judge or crime.
• 18:30
  much time the, uh, the, the guilty party, uh, was sentenced for.
• 18:35
  And so, again, looking at, uh, patterns in everyday life that would not necessarily be pertinent to the decision making criteria that you'd think would be Uh, judge and, and the law, it was consistent, even if it was administered inconsistently related to, you know, whether someone had a pastrami sandwich.
• 18:57
  Well, John, the thing is, you know, I know, you know, a lot about healthcare and doctors, but you know, there's a big judicial reform debate going on in Israel.
• 19:03
  I don't think all the people are out on the street because of the time of day element.
• 19:06
  I like the example better.
• 19:08
  No, actually.
• 19:09
  But let's come on, let's, why don't you share something from thinking fast and slow since you were, you were referring to it without giving me, giving the title.
• 19:16
  I'll share something from before that.
• 19:18
  So one of the things they did is they actually were training, they were coaching pilots on landing the air force.
• 19:24
  And what they would find is that.
• 19:25
  You know, they would, the flight instructors would say, you know, praise is bad because whenever I praise a good landing, they'd always do it worse next time.
• 19:31
  And it was actually just used to explain regression to the mean, which is essentially that, yeah, if you have like an incredible landing, there's a lot of randomness and it's probably going to be worse the next time.
• 19:39
  So they would necessarily see that and they would develop that, uh, that bias over time.
• 19:43
  And it turns out even in Israel and the Air Force, you can tell people nice job and they actually do better.
• 19:48
  So that's not going to comment on the, on, on, on, on, on, on, on motivational terms in, in, in Israel, but, but Bapu, I guess I I'd say what are the, you're looking at really important things like when you should see a cardiologist and, and impacts in public health, what are the most important questions in healthcare where people aren't doing the right research to identify patterns of effectively patients, correctable behavior or structural reform.
• 20:16
  Are there, are there areas where you think this application of sort of, uh, it's not so much randomness, it's patterns that drive behavior or outcomes that people aren't paying attention to in a system, a healthcare system that costs too much and does too little, or maybe costs too much and does too much too.
• 20:34
  In the case of cardiologists, we should sort of like, it's, I feel like we're beating up on the cardiologists, David.
• 20:39
  Actually, cardiologists are pretty essential parts of the healthcare system.
• 20:42
  I don't want anyone to think that we are not, we are, we are saying bad things about cardiologists, but I'm, but I'm intrigued by the approach.
• 20:49
  And I'd be intrigued as to where, what, what are the questions you think we should be pursuing where we haven't necessarily looked for the right answers or patterns?
• 20:56
  Yeah.
• 20:57
  So I'd say a few things.
• 20:58
  One is, you know, this book is about.
• 21:00
  Ideas, right?
• 21:01
  It's not, it's not, it doesn't purport or, or, or cannot say, you know, how do we make health care altogether better?
• 21:08
  If I, if I knew that answer, I probably wouldn't have written a book.
• 21:11
  I'd have gone and done it.
• 21:12
  There's so many questions that, and so many things that we do in health care that are based on really poor quality data.
• 21:20
  So part of what we do in the book is to say, look, some of these questions are important to answer and we can't phone it in.
• 21:27
  We've got to be creative in the way that we approach those.
• 21:30
  those questions.
• 21:31
  And so one thing that we talk a lot about in medicine is, is, you know, when do we need to do something or do we need to do more of something?
• 21:38
  And there's so many low quality studies that would lead you down one path or the other that might be completely wrong because we just approached it from the wrong way.
• 21:44
  So part of the book is about trying to tell people, look, we've got to be a little bit more creative.
• 21:49
  Um, there is a part of, uh, of the discussion that we don't really touch on in the book at all.
• 21:55
  Um, because it's a hard area to study and that relates to it.
• 21:59
  to the economics of medical innovation.
• 22:01
  It's something I've thought a lot about.
• 22:03
  Uh, we don't talk about it in the book.
• 22:05
  To me, that's sort of one of the first order questions, right?
• 22:08
  We, we think about problems with cost in the, in this country, think about quality, meaning even if medications are free, 50% of people are not taking the medications as prescribed.
• 22:19
  So that's not a cost problem.
• 22:20
  That's a completely different problem that we've got to get a handle on.
• 22:23
  And the last thing is, What is it about healthcare that is different now than it was 50 years ago?
• 22:29
  Uh, we don't have the types of medications, we did not have the types of medications 50 years ago that we have now.
• 22:35
  And 50 years from now, what's going to separate our lives from a health perspective is new medical technology.
• 22:41
  And one of the things that I think we need to really struggle with is how do we make sure we get more of those innovations in the future, but also make sure that we don't bankrupt the system in the process.
• 22:51
  And people in policy spend a lot of time thinking about that second question of reining in health care costs.
• 22:57
  But I don't think that they give enough weight to how do we make sure that the future has the kinds of medical technologies that are crucial.
• 23:06
  Because we can get part of the way by making sure that we give everybody the technologies that exist today.
• 23:11
  And that'll improve health.
• 23:12
  No doubt about it.
• 23:13
  But we don't have a cure for Alzheimer's.
• 23:14
  So this is not a situation where people are just not getting appropriately treated.
• 23:18
  We don't have a cure for Alzheimer's.
• 23:20
  We don't have a cure for many cancers.
• 23:22
  We don't have cures for heart disease.
• 23:24
  So there's a lot of life expectancy that's out there on the table that we haven't touched yet.
• 23:28
  And so I think a lot about that problem, which I think a lot of policymakers do not spend enough time thinking about.
• 23:35
  So it sounds like another book may be coming, at which point we'll make sure to have you back on.
• 23:40
  Yeah, exactly.
• 23:41
  So that's a good place to end it.
• 23:42
  Dr.
• 23:43
  Bapu Jenna, we've been speaking to you today about your book.
• 23:45
  Random acts of medicine, hidden forces that sway doctors, impact patients, and shape our health.
• 23:51
  I'm David Williams, President of Health Business Group.
• 23:54
  And I'm John Driscoll, President of Walgreens Health.
• 23:57
  If you like what you heard or you didn't, please subscribe on your favorite service.
• 24:00
  And thank you, Bapu!