and image. called mrs. president by her detractors, she was outspoken about her views on slavery and women's rights. one of the most prolific writers of any of first ladies, she provides a unique window into colonial america and her life with john adams. join in the conversation at 9:00 p.m. eastern on c-span, c-span radio and c-span.org. ..

this -- there is no the readers health advice. there are no kind of vital conspiracy theories held drug companies are trying to kill us. this is a story about flaws and how we gather and synthesize evidence in medicine and i think the technical flaws in that important technical process are very well documented in the medical academic professional literature, and what i'm hoping to do is share that more broadly with the public in particular because there are very well documented programs which we have failed as a profession to fix. and so i think we need to help more than anything else.

they are people like you, they are birds and lawyers and doctors and policy wonks and all the rest. so we will talk about problems in the three domains. we will talk about how randomized controlled trials we like to think our fair tests are not created alike. there can be structural design flaws the way we from trials so they are no longer fair tests which treatment is best. we will talk out marketing although not a lot of people have talked about marketing a lot but a good doctor would ignore marketing and see the quality academic from journals and so on but first and most important we will talk about the problem of missing a trial data because it is the unspoken dirty secret if you like that it can be conducted and completed for the treatment we use today from

our best possible estimate have not been published are twice as likely to be published as trials with negative results. now this will be a huge impact on the information we used to make as doctors and patients about which treatment is best. but i would like to show you how it also has an impact even earlier on in the process of developing that. so a lot of your young hipsters so you don't know what these all are but a very long time ago in the dawn of grooviness, you would have the sort of punk scene or music or art or fashion, and this is the phase one clinical trial participants. when you develop a new molecule coming you first of all the

trial it in six healthy young men to make sure nothing horrible happens to them and they are often volunteers and wages. has anyone been a phase one participant? college students do it for money. [laughter] there is a kind of weird scene in the towns of anarchist people living these lifestyles becoming full-time denney pecos. doctors like to think of this as being very precise and clinical but there is a sort of real, practical culture to this stuff as well. and danny pay hero they talk about are bleeding pissing work and they give reports out of different places. they talk about unionizing guinea pigs to get better conditions. they talk about the history of the phase one trial participants and human guinea pigs and self experimentation and it's fantastic stuff.

[inaudible] it's like over-the-counter speed. i don't know why i did that in a welsh accent. [laughter] this is the world of the phase one clinical trial. so this chap here used to be a plumber and he was a participant in the phase one trial in the u.k. called teaching 1412 which you know because it went tragically wrong. so there were six young men come and they were all given a new biologic, and was a whole new kind of intervention really, to anything that had been tried before. they were given it and everything seemed fine but very rapidly things began to go downhill and they ended up on intensive care, they were incubated, they were ventilated, their whole blood was replaced, and then they began to shut down, their fingers and toes went pale and dark and then they went black and they began to fall of. these pictures are from the

newspaper coverage at the time. now, mistakes happen in medicine, and what's important is to distinguish between avoidable and unavoidable mistakes. in the u.k. requirements what happened is there are two observations made and to recommendations. the first recommendation has since been adopted. they said if you are getting a new intervention with a completely unpredictable mode of action and it's very unpredictable what is going to happen next, then don't give it to all six people at once. stagger it every three hours, six hours, maybe a great wealth and you wouldn't want to go first or why wouldn't by not participating in the phase one clinical trials. i don't understand how anyone can have that kind of relationship with risk. [laughter] and that has been acted upon. the second recommendation has not, because it turns out that in some respects this was

foreseeable, not entirely, but there was extra ground for concern because a similar intervention had been tried in just one person, and it had an adverse outcome. it was conquered and with what we saw. but the result working on 1412 the results of the early trial hadn't been shared with the research community. and as of the recommendation of the u.k. government is the result of the phase one trials should be disseminated more widely. but this hasn't happened. today only one out of tennessee is one clinical trials as published within a year and only one in five is published within five years. but the problem of the withheld trial data goes beyond just phase one and beyond posing a risk to phase one trial participants. this is a molecule called look tonight, it is a drug that

prevents abnormal heart rhythms. in the 1980's they would use these medications much more freely than we use them today and much broad categories. so everybody that had a type one arrhythmia or heart attack or a premature ventricular your heart attack would likely have been given a class one anti-affirmative mick drug. we knew these stopped at a normal heart rates so they would make it go away because people that have abnormal heart rhythms are at the risk of cardiac death that would be a good thing that we rely on the fact the jobs for getting rid of the abnormal heart rhythm and that was going to stop them from dying. it turns out this assumption was wrong. when we finally did something called the cast trial that reported the findings in 1992 we discovered infect patients that were given these drugs for those particular types of abnormal heart rhythms after a heart attack or an increased risk of

dying. this is an interesting sort of thing to grasp in that firstly nobody was being bad or malicious or evil treated to an extent what we were doing is just being sloppy and not thinking. so the intervention that had a positive an impact on the serious debate cause serious outcome, that successfully stopped people from having abnormal heart rhythms was necessarily also going to have a beneficial impact on the real world of come, which is death. but we were wrong. and because these prescribing practices were so widespread, it is so common, this drug was used frequently, it turned out some people have estimated that it caused unnecessary or potentially avoidable deaths around 140,000 u.s. citizens before we gathered a real world outcome data. it is a little bit mean and i don't want you to think this

about -- [inaudible] [laughter] there are a couple academic papers that say 130,000 people died which is more than the whole vietnam war. that is only true if you count deaths from the american side. anyway. now, again, to what extent was this foreseeable? it was certainly an important week of call for doctors that we need to make sure we don't rely on process of comes on surrogate outcomes. but interestingly, they're had been an earlier study conducted in 1980 before any of this happened. which hadn't been published. this is a remarkable paper written in 1984 after we find out about this. he says when we carried out of a study in 1980 we thought the increased death rate that occurred was an effect of chance. it wasn't one of the ones that was used in the prescreener practice in the 80's and he's

actually being a little optimistic when he felt that it was an effect of chance. there were 100 people in the trial, two groups of 50. nine deaths in the group that got lorcainide and nine deaths that had a sugar pill. anyway, the development of lorcainide was abandoned. you are a nerdy crowd. [laughter] this is very common and because of this for commercial reasons the study was never published and nobody knew this had happened. and to this day the industry in particular is a key to maintain the results of the trials on treatments which don't come to market with the fda certainly will share that information. it's a good example of publication by yes is a phenomenon whereby the - trials of unflattering results can go a missing and be erased from the academic record. the results described might have

provided an early warning of trouble ahead. just to remember again this understatement he means that potentially avoidable deaths are an estimated 130,000 u.s. citizens. as a, is this unusual clocks well, no. it has an impact on the everyday treatments that we make, the everyday decisions the we make. you kind of need to understand the very basic evidence and how we use evidence as doctors in order to make decisions and to do that you need to understand the systematic review is kind of embarrassing that was invented a so recently. up until the 80's if you wanted to write a book chapter for a medical textbook or review article about the treatment of,

you know, let's say treatment of depression in patients with physical illness, you might go okay well i can remember a few trials that i've read about in the past. i will keep my eyes open over the next nine months until the paper is due and when i come across a trial that looks rather that i will to give a copy and i will put that in the box for when i write my chapter to get me your friends will go you are writing that chapter on the depression, treatment and people with physical illness. have you seen these? so they are kind of ad hoc and in a systematic way much like an 19-year-old student might surround somebody around them and throw and essay. do it producer medical textbook article that would become comical and even the was vulnerable to a all kinds of unconscious by aziz but also perhaps even through the kind of the leverett cherry picking if you had a kind favorite idea about what you thought was the right answer. succumb in the 1980's people

finally got a grip on the idea that this was not quite right. and we invented something called a systematic review. the systematic review you describe exactly how you search for your evidence so you say what database you go to, you say is ackley what search terms you type in and try to identify is randomized trials as well as randomized controlled trials you type in the placebo or whatever randomized controlled trials and you would get the most systematic and complete summary of all of the evidence that could possibly be drawn together and extract all of the numbers from that from a treatment actually did. you put that all in one table and one spread sheet and you do a bit and now this is where you at all the numbers to get the and you get the forest plot that looks like this and this is the logo of the collaboration that is a global nonprofit academic collaboration about 14,000 academics who are not producing

systematic reviews and a gold standard summary is used by doctors worldwide to make treatment decisions. now, would you like me to explain how it works? somebody said no. so each of these lines is one trial. if it is over to the left that shows the benefit and if it is over to the right that means the treatment was harmful. if it is in the middle, if it touches the line of no effect and it goes up and down and that means it showed no statistically significant benefit. now, you will see that some of these are quite narrow because they are big trials that are less vulnerable to the random area that gives an accurate estimate of the benefit of the treatment where they are quite broad and that is because they are much more vulnerable so you get a broad estimate. now, these are real world trials. this is the first systematic review and it's on a really important question.

it's on the question of if you give a steroid injection to a mother that is about to deliver a premature baby does that stop the baby from buying so this is important stuff. we call this the big day -- big dead baby graph to be at you have to kind of force people to fill in the connection between the abstractions of evidence and all of the kind of works that i am describing here. you have to feel it that there's a connection between that and the real world of suffering and pain and flesh and blood and death. this is the meat. so what we see here is some show that was beneficial. so that would be one there. that showed the steroid

injections stop babies from dying. there was no benefit. when you add them all together at the bottom of the diamond that shows the of all this is a beneficial treatment. the overall it saved lives. now, why this matters is all of these existed for many years and even though if you add them up together this stuff is effective there were many doctors that didn't used steroids when mothers were about to deliver premature babies because there were positive trials and looks like it is pretty useful but others would say why should i expose people to that by getting the steroids i'm not going to use this you must be crazy. because of that, because they didn't use steroids even though the evidence was there, babies died.

babies died not because we didn't have the evidence, but because we had all of these trials. they died because doctors at that stage for not savvy enough to realize the importance of synthesizing the evidence that we have together in one place to get the right answer and i think that is extraordinary. the flaw in what we ought to call the architecture of the evidence meant we had the information that we didn't bring it together in a way that meant we could make informed decisions you get the importance of this and this is now. so this whole project of bringing together the evidence in one place to get the best closest most accurate estimate we can of the benefits and risks of treatment, this breaks if we

don't have all of the evidence. if some of the trials are missing and we get an exaggerated impression of treatment so we might go -- we might say you were getting horrible side effects but it's really effective said you must stay on it. then they would attenuate the benefit and we would miss out and using the treatment, as we have to have all of the studies but we know that is not what happens and we can demonstrate that they've gone missing in action and they've been withheld into different areas you can do it with stories which was the traditional way of communicating to the audience or you can do it with statistics which is beyond aggressive invention because i'm very tired and i have to punish you. it's a very clever way of spotting the trials that have gone missing in action committee and relies on a very interesting and clever phenomenon, which is first it's easier for small

child to disappear in a massive controlled trial being conducted in the different hospitals in ten different countries and planning on hundred 80 different people with $20 million. so, small of trials easier for those to brush under the carpet and big trials. now each of these is one trial. okay? if the dog is over here it means the intervention as beneficial to lead its over here it means that the intervention did harm and if it is in the middle this kind of normal. and, as you go from the bottom to the top, you are getting bigger trials of the top and smaller at the bottom. okay? now, you can probably already seen that but at the top, big trials give you a lot more accurate estimate of what the science is come as a that is why we always like the big surveys of 23 people. we want the big surveys because they will give more, they are less vulnerable so we have all

these big trials we should cluster around the true effective treatment to get as you come down to the smallest, they are much more spread out. much more random error in the results. over here to the right, you have positive ones purely from the chart in the middle and then some negative ones purely. that means - studies will be missing from this picture but it won't be the big ones, it will be the smaller ones and so there will be botts missing from the draft and where will they be missing clocks what would be missing here will they be missing here or here? okay. this is a plot you can see they seem to be up here and you can see as you go down they are not

there and this is a regression and it's not a statistically significant deviation from the final shape that he would expect to be and this is only a subset of studies, but it is a plot looking for evidence and publication by is in the studies of publication by yes which i think is the funniest. [laughter] but you can also demonstrate and sort of show that this is an issue with stories. so this is a drug i myself have prescribed because it had an unusual mode of action and patient said it would better on one treatment that might be a good idea to try something with a different mode of action and perhaps more likely they will respond. i found a couple of trials that showed that reboxetinmesilat was better than the sugar hill and i found a couple trials that showed it was better than any of

the anti-depressant. so i signed my name on a prescription giving it to a patient. in 2010 a paper was published at the german cost effectiveness agency and they described an extraordinary battle they had where they said we will not look at this drug. we will not that people look at this for approval until you give us all of the trials that you've done on it and pfizer said no and they said we are not going to have anything to do with this medicine then. finally, they got ahold of them and what they found were there were three studies comparing at that show was beneficial and will publish, but there were about five more studies showing that it was no better than a placebo and they hadn't been published. two trials showing it was much better than any other antidepressant that the data was collected in three times as many patients showing it was no better, i'm sorry, shoving it was worse than other anti-depressant and those had not been published. i as a doctor as a consequence

of this i feel misled because i did everything you are supposed to do before i prescribed this medicine to my patient. i read the deed and i made an informed decision and i praised the trials and concluded they were well designed. and yet, i was still overall misled. when you add up all of the data depending on how you do it, it is either by some people's estimates completely useless and exposes people to side effects and it is marginally beneficial but that's not enough and this is an important thing we have to grasp because i'm not saying there are lots of drugs on the market and actively harmful. i think it's actually quite unusual. but the real problem that we have one half of the evidence is censored from the medical literature the problem we have is we don't know which treatment is best and when you get an inferior treatment even if it does something you are being deprived of a truly effective treatment, deprived of a better treatment and that is genuine

harm. this is a difficult thing for people the especially in the history to grasp. so, most drugs -- most treatments and medicine we have to be humble and recognize there are huge amounts of good. so a middle-aged man in the three or four decades not because of any breakthrough but because of the gradual accumulation of a whole bunch of different risks and benefits. now, let's say there is one treatment nellis av eight lives out of 101 that would save six lives out of 100. if you can get people to use the treatment which will only save six lives out of 100, then you'll sort of reward yourself a point because it's giving people what you believe is an effective treatment but if they are using the because you have made it look as if it is better than the one that would actually save eight lives out of 100 through a combination of selective publication, trial design and aggressive marketing, then the overall net effect is down to out of 100i think the difficult thing here is that we've got

this failure of ambition where we haven't noticed that that's not particularly awesome outcome and that is an important nuance. what i am saying is the flaw in the information architecture on the medicine are such we cannot be certain that what we think is best is necessarily the best service okay until we have it until 11 tonight. [laughter] is the story unusual? it turns out the answer is no to read the best evidence comes from the systematic review. this is a summary, systematic summary of all of the studies ever done looking at the evidence in publication published in 2010 its current, and shows that on average around half of the trials that are conducted and completed don't want to get published and they are twice as likely to get

published as the trials with negative results. to give you an illustration of the kind of research that comes from, if the trials have simply disappeared without looking at whether they had a positive or negative results you need a list of trials that have been conducted and completed as you can get that in all kind of ways in the local ethics committee and a list of all of the trials that have been approved to be conducted in the hospital or in an area in the clinic and then follow them up and see if they've been published. you can also use the regulatory document. so, this is a peter with a total of the trials of the fda held on the edge of anti-depressant and these are all of the antidepressant that you have heard of that have been approved over the course of just under a decade and they were in the drug approval pack. this is and all of the trials still preposterously a distant dream but it is an example of the trials because it was done

before the drug was approved, before it got on the market and were used to get approval from the fda. all kinds of fiddling and buffing around and for waste sometimes you forget you even apply and i only mention this because i had dinner with the man the dispute yesterday. and so, in the regulatory documents and the story of what actually happened there were 74 trials and then they went and looked at the academic journals to see what was there and what they found was markedly different. overall, the free - results and it's even worse than that because seven of the negative results published as if they had positive results with all kind of data driven techniques that we can discuss that's okay because we have until one in the morning. [laughter]

it seems to me that overall the picture where you go from 38 results and 36 - results to the three - results, this is very simply research misconduct. from here has done some kind of research project even school research like a school biology project? okay, now if when doing that school research project you deleted half of the data points on your own project to make the line will the right way the way that you wanted to go we would agree that you are guilty of research fraud, is this correct? again, this is a component. so, what i find fascinating is that we have this strange cultural blind spot in academia where for some reason is considered extensible that of the whole study with negative results go missing in action. even though we know that when that happens the overall impact on the overall apparent effect

of the treatment which is what we use in the real world of flesh and blood and suffering and pain and death, hold trials are deleted from that record, for some reason we don't regard that almost no medical bodies around the world have that in their code of conduct it's a strange oversight and again, we have until three -- [laughter] this is in the story about the trying to make money this is about flaws in the information architecture and its problems which we've known about for 30 years in the case of publication and failed to fix. worse than that we have a series of fixes which allow people into a false sense of security. you must be very excited that

you have come out when you could be asked to leave to about dancing for having sex, whatever it is you do these days in seattle -- [laughter] instead, you're listening to me talking about the international committee on the 2005 in which they said we will never again publish a clinical trial unless it is posted on the register and this was an interesting moment in the history because everybody does a trial and if they want to publish at in the most academic journal they can find because they want to get on their cd so therefore if we are the editors of the journal then we can tell people that you are not allowed to publish unless you've registered before hand and if we force everybody to pose the existence in the register that

doesn't necessarily get the results but at least it creates an opportunity for the trial where we can see what is being conducted and repeated where is the other half? so we have a chance of spotting the missing data. so everything is fantastic and fixed. a debate in the u.k., the u.k. head of merc and the health care authority he said this is rubbish. this is being fixed. the medical journal won't publish anything unless it's been properly registered and then you say but there is a paper in the -- and that's the end of the debate. but since we have the floor until four in the morning -- [laughter] [inaudible] [laughter] so 4:00 is a hard line? [laughter] that joke is over now. i shot it. five years after this was put in

place, regulation put in place in the code of conduct, five years later no -- was finally discovered that it has been ludicrously widely ignored. half of the trials published in the top ten journals in the fields of medicine were improperly written. and it's not surprising it's been completely ignored because academic journals themselves have a bunch of conflict. they are not necessarily an is and financial but they are cultural and professional. for example, an academic journal wants to to get funding from the industry trial because they will know that would get cited lot of times by others. so that would pump up the impact factor of the journal which is derived from a number of citations that it gets and also lets remember the academic journal is a full-time job being an academic articles and they make lots and lots of money from the reprint that most people

have never heard of but when it's published in the academic medical journal, then the drug company sales representative that by lots and lots of reprints of these on $5, $20 a pop and then they pushed those under the nose and they give those to the doctors in order to encourage them to use that treatment. so, that means academic journals have a huge financial conflict of interest to get to your has published on the academic journal? okay, great. when you do that there is an intersection at the bottom of the paper and you have what kind of things? i'm not asking you what your conflict are but what kind of stuff does one have to write? finance is above all to get some things like how much money you have had for giving talks and free conference travel, whether the funded research or this stuff. as a country is what i find interesting. last year i published a colleague paper or we went to the big five journals in the world and we said look can you tell us of these trials how many have you got and they said no.

that's confidential information. from the brief medical journal they knew they wouldn't give it to them. i will tell you what i find interesting about this. they are perfectionists about making sure that you give a clear declaration of interest when you publish in their journal. and yet, when we ask them a single direct question could you let us know how much money we got for the trials because we are interested in the conflict interest and the whole process since this is a summation of evidence which is really important. it's confidential. it's just interesting. anyway. [laughter] so, the fix is something called the fda amend an act of 2007. and that says you have to post the results of any clinical trial that has to reduce the part of the marketing approval. if they publish the results within one year of completion to the web site called clinicaltry

il's.gov it is a $10,000 fine, that sounds like a lot but it would be a bit of a struggle for most of you and i am not judging i'm just saying -- [laughter] but, $3.5 million a year is a bit of a parking ticket if you are a big organization that has a drug that's making a couple billion dollars. now, no routine audit. when it's finally collected and published in the medical journal in 2012 just last year, the rate of compliance is genuinely astonishing which everybody said this has all been fixed it's all been fixed because now everybody knows you have to post everything in one year. the rate of compliance was 22%. four of five trials ignored the act which is the single thing that is most commonly cited by people who want to poo poo this.

22% was the rate of compliance. yet no fine has ever been levied in the history of the amendment act that's bizarre. i'm not surprised it didn't work. the combination of it didn't work and everybody is pretending that it did. but even more preposterous than that, even if it had been uniformly perfectly adhered to, it still would have done absolutely nothing to improve the evidence based on which we practice medicine today. because about 80 to 85% of the treatment that we described in medicine today came on the market more than ten years ago. so, most of the research gets done at the time the drug is approved and, you know, fixing everything for the trials after 2008, that might have an impact on the madison to become a lesson in the year 2029. but it's not going to do anything for us now. what we need is access

retrospectively to the clinical reports and as a result of the trials that have been conned conducted on the medicine we are using today and that stuff is all out there. it's in the jul storage archives in the nuclear bunkers or whenever. so, this is a jury long and detailed story about the european medical agency which i am not going to tell you. [laughter] sometimes you can get so embedded in a set of problems that you lose your orientation and a sense of what is normal. when i was doing the book i kept having to go in my right to think this is really bad? and my wife or my friends would just say yes! and this is a good example of that.

so this was sent by the european is an agency in response to the research that said we are worried about these weight-loss drugs. we know these harmful trial stock market published and we would like to get ahold of this. could you tell us what you hold on it and they sent less. a room full of people come just everyday people in all walks of life the moment they see that they spontaneously laugh. what was going through their heads? were they laughing when they put it in the envelope? it is a strange, strange business. this is interesting because now the european madison agency will give you full clinical study reports they hold on request. they invite you to kind of award them a point for this as if they gave this freely and fabulously out of the glory of their own

generosity. but the reality is they did it in response to one of the most damning findings of the european ombudsmen. the european that is an agency they said was to get a coherent or even consistent account of why they were for holding this information come extraordinary episode but apparently we have to stop at 4 o'clock according to this lady. [laughter] so finally just to put a nail in the coffin for those that pretend stuff is being fixed, by the way the european medical agency dingley hold clinical study reports they think they approve and the only approved things to the very few last year's and again, it's no good if we get the full of inspector for the treatment that has come on line in the last four years. we need to start in 2002, 1990, 1992 and so on. finally, tamiflu. tamiflu is a government to drug

the spend billions on stockpiling. it reduces the rate of complications of influenza infection. it's a medical use for death. when you use the word complication. we still don't have all the evidence on where exactly does that. the nonprofit academic body around the world produce the summary is used by doctors and patient to make informed decisions about the treatment works best. they were asked by the british industry and government in 2008 to update the review on tamiflu -- we don't have time for that destruction [inaudible] [laughter] no, seriously we are like one-fifth of the way through. so, half of the trial data hadn't been published. they got in touch with the company and said look we would like to see not just the result full clinical study reports and

the steady imports are interesting and medicine because we just started to realize and you will see in a moment why it's important. but in normal trials they are created alike and the brief summaries of clinical trials you get in the regulatory filings and also even an academic journal articles can often be incomplete, inaccurate and sometimes even misleading. and so we need a full report so we can see the meat is exactly what we've done, what was measured, what was analyzed and so on. in the summer of 2000 by the promised we would hand this over on tamiflu, and still three years and three months later they have failed to do so. this was an extraordinary thing. the drug that government have spent billions, the u.k. loan 500 million pounds, that is 5% of the u.k. annual drug budget. and just gone on this stuff where we don't even know if it is any better than aspirin or not. it's utterly mind bending. so that is the first one third of my talk.

[laughter] now i can't see what time it is -- 20 after eight, that is 820 in english. [laughter] okay, look we can do the budget marketing and about seven minutes if i took extremely fast. would you like that or should be shut up and have questions? >> [inaudible] >> are you sure? [laughter] okay. i will try to speak a bit fast for this. [laughter] if i do in american accent is that easier? [laughter] i just sound welch. okay. so we like to imagine all trials are fair tests and imagine the are created alike and perfect

but in reality there can be flaws that mean the are no longer getting accurate estimates of the risks and benefits of the treatment. so, i will tell you about some of the simple ones and there is a really important it's not a caveat but important background for you, which is the business of missing a clinical trial data. there is no cherry picking and there is no anecdotes. the single story i gave you were illustrations of the wide phenomenon, and that wider phenomenon and that is evidenced by systematic reviews of dozens of studies ...

it's better than nothing. not that people necessarily use it, but it's important to get over the fantasy that some people have and every charter proved his worth using. as a doctor in a patient coming in the not interested in the abstract question emphasis treatment other than that he peered your interested is this better than the treatment we have right now because that's a

decision nurturing to make. and yeah, 2011 after every drug approved from 2002 by 2010, dirty% have only been compared against a placebo when they were approved, even though there is an effective treatment for those conditions. so it's a widespread phenomenon of getting on the market. you have to guess where the ball is. you put down a dollar. if you guess where the ball is coming to get back. how many cups are your led to the thunder in the scandal? one. if you look under two, jury cheat and i'm not going to give you $3 anymore. and yet we tolerate exactly this. so for example, let's say i'm running to china of new

antidepressant. so not going to measure, and you get better in one way? i'm going to measure anxiety and depression scales than that its upscale and i'm also going to use an activity of daily living scale and maybe i'm going to sit back to work scale and a general questionnaire that got told different things than measuring. if i measuring 12 different things in a study and i allow myself to say if anyone shows a significant benefit at the end of the study with the kind of people, that means the treatment caused an industry that was success. if you measure 12 different things, you have a 50/50 chance, even if it's statistical noise in the data, you have a 50/50 chance of finding a benefit when there's nothing happening there at all and this is preposterously common.

the academic journals are the gatekeepers on a commonly passed through for the common outcome before the trial began is completely different in the primary outcome in the published paper. anyone who's ever published a journal just exhausted tedious stuff about them or we comment in the discussion question in your if you take a serious anyways. but people obsess over that staff and they failed to catch the most basic things. did you search the primary outcome? you can run a trial and freakish patients. it shows there are vulnerabilities in the whole fabric of this stuff. we've made trails and appropriately expensive. we've made them this material exception and cover them in a

minister different tape and because there's a unit cost per patient come the runs mall populations. people are desperate to do everything they can to ramp up chances of finding a benefit and one ways to trust an ideal population patient. he may try or truck in people younger, healthier, people on no other terms of no other medical problems and discredit much greater chance of showing a benefit. for once you've done this test in this perfect population, i'm left with a doctor going are the results of this trial really applicable to my population consisting of a lady in her 70s at the little bit of renal failure whose son for their medications with three other diagnoses. how does this strike work here? it feels a little passive aggressive and so preposterous

that the scale disparity between the trial and the reality. so there is 5.59 people in the u.k. with asthma. it's a common problem. they took 179 patients or family.as in the u.k. whose treatment was being managed in accordance with the current best practice guidelines. the decision treatment are the clinical trial. he said we've got 179 real-world patients being treated in accordance with the results. let's see what proportion would've been eligible to participate. 6%. it's not difficult to find people witit's not difficult tod people with asthma should you trust on an as we don't have time for this, but it's great on page 172 of the book.

so lastly, is it going to marketing, seeding trials are fantastic and tardive you have to salute to hang on for the. this is your word. so, let's say you've got your migraine drug approved anyone to do a trial in the real world population of 2000 patients. migraine come is that common or rare? it's common. so, in a neurology outpatient you probably get, how many migraine cases in the year do you make in? hundreds, maybe thousands, like this is bread and butter. so you recruit 2000 migraine patients. every time you have an extra site in your trial can you have to go through physics and getting everybody up to speed on

codes of conduct and lots of stuff. so the rational way to design your trial is maybe four states recruiting 500 patients. something around that. so what do we think what we see a child that recruits to patient in each of the thousand sites would say. this is a seeding trial. seeding trials are fantastic because you dare not say -- to be a very brave person because someone could say no, this is awful what you're saying. or just creating broadly from the population. we want to spread the risk. it's probably nonsense, but you can show us a seeding trial if you have access to that court documents. so that's what we have for most of the trials. and this is why companies are often so desperate to settle out of court because the court case

might be in a headline problem, but once you get access can you get people going this is interesting. remember that trial were they recruited to patients in excess and clinics? truth at the marketing department to find it. the idea of a seeding trial is not to find out what it does pay to get large numbers of doctors familiar. the thing i love so much is tedious to imagine the doctors, maybe posting, going you should try -- i'm a co-investor and we have a lot of good results and you think, you embarrassing human being. [laughter] so lastly, marketing. a classic example of a recurring

theme in this interlocking ecosystem of problems all reinforce each other and vulnerabilities in the system and perverse and then case. marketing shouldn't matter. it is only an issue because her so bad at disseminating to decision makers to ensure we have rapid expensive treatments and i'm totally up for that and also rational investment of things that don't work. and were incredibly bad at it. it's been completely neglect it. if you put me in charge of the one world government and the medical research budget for the universe, for one year, and i can reassure you this will never happen, but i can cancel all primary research for one year. we could live without another clinical trial to treat ends.

just for one year. i would cancel all travis and spend every single panel we have on creating a better infrastructure for the synthesis of the evidence we have, to get it to decision makers and put it into this because current me it's a low status occupation. we spent hundreds of millions of dollars on one trial and yet i'm equally important business of getting it out, but completely dropped the ball. you get some prescribing advice about what happens if you do refuse and geriatric patient and does it reduce pharmacy and it's nonsense. fiddling and hopeless. because we don't do that staff, we are vulnerable to marketing and i can't play in the industry for wanting to disseminate information about what works.

i just think we should be above it. so this is a person called jordan and i haven't replaced it with the american equivalent, but she has a lovely face and who would you here? uncrowded thirsting. you're telling me there's no equivalence. you're about opera singers and novelists. anyway, it comes as no surprise that when this person publishes, it probably wasn't written by them. that's no great surprise. it's the consent issue. you probably prefer is involved in the creation of the work. the funny thing is the same thing happens in meds. so they are hard to spot and there are shades of gray, though with reasonably frequency, drug

companies will write an article for submission to a journal and take it to a.your academic essay which elect a pitcher name on this? was funded to the journal and you get your name on an academic journal. now, sometimes the commercial system has declared that it's not declared very fun buoyantly. perhaps just a mention of a company somewhere. and sometimes it completely hidden. i think they heinousness is twofold. firstly obviously destroys the content of the relationship because there's a bunch of people with money to make every services to deploy who spent time and effort in getting their message. they are part of the overall

discourse and that can't be right and good. there's a second concern but i think it's almost a bigger one than the medical workforce. so who works in academia here? idea get ahead get ahead in academia? you publish, okay. so you could academic publications and that's how you become part of the department and respect it in so on. now what's interesting is where people are being lots and lots of help, there's people are selectively propelled up to hierarchy and academic bennison comest that we are selectively promoting the people most willing to participate in what a lot of us might regard as a morally gray area.

we are selectively promoting these people specifically into positions in which they accumulate the appearance of independent absolutely bizarre and peculiar, especially because he's getting an article accepted, said it easy difficult thing to do? difficult. it's a pain, difficult business. if somebody could bring that to you. that's better than giving me $3000. i'll probably give you $3000 for that. maybe i should employ in medical writing company. interestingly you get a whole journal. so they just appeared in australia a few years ago. another one when not to general patients with tens of thousands of that is and it turns out they

were full of reprints of articles about how fantastic merck drugs were. the only reason i put this picture a is because when it came out, he went these were presented academic commercials. this is what an academic commercial looks like. ridiculous. next, job reps. they go down and talk to doctors to convince them and they give you free samples, cherry picked information. of course it's not entirely brilliant unbiased dissemination of evidence. i wouldn't see a drug rep. i feel like i'm not very clever. i can run perfect, but i'm not sure for years after he landed i could necessarily recall is his providence. so not sure i trust myself to

have pictures of the risk and benefits of treatments whispered into my inner by an attractive drug company sales rep at a revealing top. cannot wait to get involved in that whole circuit. also this comes from a systematic review published in 2011 in summarizing 50s studies. doctors who see drug reps over all are either the same servers prescribers. unsurprisingly how might that why would people spend billions of dollars on the staff? it goes without saying of course it works. if doctors want to do it, that's fine. i'm not going to ban it, but we should talk about it and we should be saying to doctors who see drug reps, be sure that's really on. and finally another fantastic illustration of how this isn't

about bad people. it's about small tweaks in the corners and end up in the overall process that crucifies dissemination. so you criticize a doctor in your 20 as a specialist in your 30s and then you practice independently for the next area for decades. during that period, a doctor who qualified in the 70s is today treating medicines that were almost universally in rented after they were trained. they were self-taught for all that stuff. doctors are keen and paying undocumented medication. the state isn't going to pay for this outcome is the industry steps in advance for a doctor's professional education than anybody else does.

so doctors are top which treatments work best by industry, the people who make them. i think it should be a relevant and should be okay to talk about it. of course it's well-regulated and people can point a big books. but we've seen whether they actually work. of course the evidence shows when you send mystery shoppers, of course it's biased a little bit. and it's not corrupt people. first an interesting thing to grasp. just sort of understand the moral framework in which all of this is happening. let's say there's three doctors. they look like hospitals. [inaudible] >> a pill how. okay, pirate guys listed as one

rheumatology clinic with readout errors and let's see two of them racket for what particular generic drug, just as good as the current branding one, but one doctor has a genuine relief and a defensible and reasonable literature. the one will say i think this particular single branded drug is better because the mechanism of action a genuine belief that one particular drug is better. not corrupt, vicious but he genuinely thinks. so the local drucker companies like search a and they literally take that doctorate platform and a microphone. in the message will be an essay to the medical community. so that a few hundred dollars to teach local doctors and maybe thousands for teaching

specialists, a few thousand for conferences and a resort for teaching people internationally in these programs. not that dr. hasn't changed their view for money. as a person who genuinely loves their drug and now we've done is we think she's a charismatic person. bill shepherd and around and help him share his message. a lot of the time -- were trying to be nice here. so it's not necessary that people are bad and evil. the theory commentary here exceptions, i draw the line when people deny the existence of problems. i draw the line when people elicit annunciation's appears a systematic review.

it's a systematic review. i draw the line when people say these real problems, just shut up. you're not allowed to talk. but i don't think these people are bad. the incentives are misaligned in the information architecture has several interesting technical flaws and we should fix them. we should fix the vulnerabilities in the system that levodopa to manipulation or bishop force people to publish clinical trials currently in use. it is arranged that we don't. i genuinely believe that the people of the future, the doctors of the future will look on this era of 97 and the way we tolerate clinical trials going missing in action. they will look back and go, you

spend hundreds of millions of dollars on producing these carefully designed clinical trials to detect modern modest differences between one treatment on the other. and several times over he completely throw the baby out by letting half of them: published. you know this is a problem for decades and you do nothing about it and are unable to believe how we are. thank you very much. [applause] >> i just want to remind everyone of you have a question to conquer this microphone. you can form a line that they silently have 20 minutes for questions and because we have a

full house, keep it concise questions semiconductor as many as possible. >> to share questions. while the observation is that it takes -- and of biochemistry and pharmacology five, six, seven years to produce one main script where i have found many who produce half a dozen per year, so half a dozen reviews. could you speak to that? also there's this enormous proliferation of journals every week i received two or three invitations to be an editor at the african journal of the stern from genomic science, so there's this huge profit driven explosion of thousands of journals trying to get you to become part of the editorial board to lend credence, but it's virtual journals receiving profit from that model. could you speak to this?

>> the issue of journals is an interesting different one. the thing i find interest in is the extent to which the entire architecture of medicine is this ad hoc ecosystem that's evolved on the basis of legacy systems that are demonstrably outdated. the volume of papers published today in tens of thousands of journalists and millions of manuscripts every year clearly requires a systematic approach. and yet this has put an incredibly low priority. the very fact to report the findings of clinical trials, which are identical experiments. all are pretty much the same except intervention and the and the outcome is different, but every trial is the same experiment. instead of forcing people to report them in a structured data

format, without people write them up in six page essay cycads 1876. it's utterly bizarre. the problems you describe are a part of the microcosm of how we haven't thought carefully enough about the infrastructure of scientific data. we need to be building frameworks instead of piling this stuff up in ruins. it's bizarre. >> thank you for coming to talk first of all. >> no one out there can see her eyebrows. sorry, carry on. >> usually you have to really the question. >> this is a little of a jump

off point from what we just talked about. in the mid-to thousands or so they were talking about bringing the nih is an actual administrator for all clinical trials and have been all fun and, all all result come out as opposed to this clinical trial and so forth. you see that an implementation of results of an ipo to achieve in the next five years to have more transparency. >> so firstly the problem of incomplete dissemination for my publication bias happens in academia, too. i wouldn't hold them up to the white knights who often hope. a lot of people have fantastic notions of how we can fix this to involve things like nationalizing the pharmaceutical

industry for making all research administered by putting money into a big pot in subminiature body does it. i'm not sure that's necessary because i'm not sure we tried the basic stuff yet. the missing clinical trial results haven't really tried asking these very carefully. nobody has yet been called off for not handing over. we weren't even asking for very much. when you look at trial registers, it's this extort marry non-overlapping or partially overlapping patchwork of incomplete list of trials. nowhere has anyone seen me the list of all trials conducted on treatments are using right now. so clinical trials.gov for use as part of the marketing approval for the last two years. it is taught entirely in secret until march 2011.

a unit paradox whose entire contents of the register is conducted and completed in europe since 2004. that's nobody should be. they should be running a complete list of all the trials that's ever been done or prescribed in europe and that's what the public would obviously expect us to be doing. it's obvious that blacked out document is laughable, right clicks so before we get to centralizing the entire research process which is billions and billions of dollars, a massive of people to to sense that, we could drive a forcing people to reports that is appropriately. the european agency should be sainted companies, well done. you've got a marketing approval, so that's a list of all the trials ever been done best on

the market in here at the forms. check out the trust them or pop them off of essay. there's a spectacularly bizarre ambition were we think it's okay to say can you tell us about the trials are starting from now, but don't worry about persaud, brescia, india and china even know that's growing by 30 here. it's hopeless. so i try asking on fixing the model before it completely turn everything upside down. >> this is probably a smaller part of the problem, but i wonder if anyone reviews the reviewers in particular i don't know if you're familiar with tom jefferson who has these

outspoken vaccine flu s. hazard views on our radio show, which is equivalent to church nor it coast-to-coast if anyone is familiar with that one. i seems that he's written where he compares the analysis of flu vaccine effectiveness, but then to the side effects not in the analysis, but numbers he pulls out. i've seen things or he said in the data it shows a mild affected in some very shows no effect. that was the wording and i'm just wondering if you're familiar with this guy so we can keep cochrane reviewers and chat. >> i don't know anything about the flu vaccine story, but i think is a pretty good guy. he did the team of flu review. any article like any piece of science to can comment on more

easily and the problem that was spotted in the systematic view they are was spotted after the team of flu review and then a japanese doctor came along and post a comment going hey, look, they're worried here because most of the trials in this review, which are just updating i understand, comes from a systematic review analysis that describes a dozen trials were all but two have never been published and i think that's weird because they normally would expect people to carefully and critically opposed the literature instead of relying on a systematic review done by someone else and then put those numbers into your own review. immediately there's an example

being done the right way. they want your absolutely right. that was sloppy. we're going to go and have a look and that's how the saga began because of somebody reviewing an article in going i think that's rubbish. you haven't done it properly. they went okay, so we're going to redo it properly now. it's you've got concern about the systematic reviews, he would have done a few years earlier. >> okay, so you know, post them online. i wouldn't regard anything is comical. posts and they're criticize it. is the question is who monitors cochran and is it any good, the service you, right clicks and

that's exactly how science should be. but it sounds like you have more to say, so say hello afterwards. or now. >> afterwards other to talk to you about specific things. >> my card is good till 4:00 a.m. >> hey, my question is when the perspective of health insurers. in our country is pretty fractured, but we do have access to quite a bit data over quite a bit of time and different information coming together. if you are doing research on behalf of the health insurer, what questions would you be asking of your work in order to steer the ship towards a better outcome and/or what kind of changes to the way that pay about policy or fantasy benefits

structures would try to affect. i don't have to be able to do it without shaft. i would answer really important relevant questions about which treatment is as that nobody is a financial interest in addressing. so you have the potential to hold huge administrative data about how patients are doing. in the u.k., i work on a project involving electronic health records. we got about 3 million health records in the gp research database and this is part of the socialist utopia. [laughter] it's used already for observational research and that kind of thing, but that's no

good if you want to find out which of two treatments is best. so we run a trial was two against each other to see which is the best number running for half a million were to do it the traditional way is tens of millions. that's a really long answer because apparently you can get an extension on the car till 5:00. so, our project is all in the market hasn't compelling it to see though to work in a sense to reduce your chances of dying. they've all been compared against each other, but not to say which is best. they're compared against each other in sure to be equivalent for how much they lower your cholesterol. to do a big trial from one against another with the really big expensive project because

you have to get follow-up data because one of the annoying things about medical research if people take ages to die. [laughter] so you need a way of following people for a long time. so our project is basically you go to the doctor. your doctor decides -- [inaudible] as they prescribed a big risk and compensate we don't know which is the best. push this big red button commendably assigned and you never think about it again. these are the trials which are doctors don't want to get involved because they no thought to come back for blood tests and also they're anxious they may be testing some new experimental thing. but they been shown to be safe and effective in millions of people around the world. then you get all of the follow-up data so you can see if they have a heart attack from

the data. you can see if they die, but the cause of death. you can pick up things about side effects and so on. the potential here by harnessing the data we have is that we could turn our health systems into machine that are constantly testing and learning and adapt team routine comparisons. so not proposing this testing some new chart where we don't know where what works best. nobody could care because nobody has any idea which is better. but if one turns out to only be 1% or 2% better than the other, would be massively worth finding out. his birth are interest is a government in all we have to do is persuade the population that

it's an okay thing to do and find a way of having doctors who have the humility to say i don't know which of these treatments is the best and that requires we hold the line because it is one doctor who says those guys are fools, i know which one is best for you, look at my cheesy smile, that dr. bush the whole game. when i say the evidence-based medicine is broken and it should have been the title of book except it's really not very catchy, you know, in another example of how we fail is that we're not constantly identifying uncertainty in resolving it. so you've got all this data. we could say any new treatment has just been approved for use that's really expensive it is not currently any data to show

whether it's better or worse than currently available treatment can only get it by agreeing to be randomized to a trial. you can do that. you say do you want this new treatment? we actually don't know if the works. if the amount of time were we continue to not know where people will be given a treatment which was subsequently turned out to be less effect it. i find that out as soon as possible harming anybody. could you do that? >> making health care less expensive for everybody doesn't make us more profitable. only if we can make more money relatively. >> you could compete on other aspects, but she could

collaborate on gathering data in resolving inserts see. >> i guess what i was asking was either report her for pharmaceutical manufacturers as part of how dodgy they are? do they do payments goes based on how dodgy they are quick >> yes. >> something interesting happened and i should've told you this. i'm a bad person but we were running out of time. do i look like a man -- [inaudible] >> actually we do have to move on to the signing at this moment unfortunately. one thing. i want to remind everyone that ben will be available to sign books after the talk. >> the book came out in the u.k. a little while ago and lots of nice things have been. this is what happens if i press

this and the health minister agrees it's a problem with that editorials in academic journals in the science and knowledge she committed into this problem and european mps decided me the proper campaign so he put it at charros.net. it's now 30,000 signatures conquering medical research, but also over 80 patient groups. if you're a member of any professional body, last week access inet, one of the biggest drug companies in the world come extraordinary really. [applause] and in answer, it's interesting because for a situation where you may possibly say there's two treatments. we don't know which is best. the apparent benefits of the two are equivalent, but one is a

come and he said were you everything. one is by a companies not willing to say that. once you've got a situation where there's an imbalance of how much transparency, doctors and payers might find themselves starting to say, if they're both apparently equivalent, the one has shared the evidence they've got, it's money transparency and nasa fixes it. thank you very much.

>> about 44,000 navajo indian contract these were used during world war ii. next, chester nez sits down with judith avila, his co-author, to talk about their book "code talkers: the first and only memoir by one of the original code talkers of wwii." mr. ness is the only surviving navajo code talker at 92 years old. >> this is chester's and my book, "code talkers: the first and only memoir by one of the original code talkers of wwii". and chester is the code talker. he's actually the last of the 29 original code talkers.

he turned 92 last week. a code talker for people who don't know what it is for someone who helped the marines in the pacific were in world war ii to be able to transmit messages without the japanese being able to intercept and decipher those messages. before we had navajo code talkers, the japanese are intercept and in deciphering messages. we're actually losing in the pacific. a man named philip johnston suggested to them that they use navajo. he had grown up on the reservation. his parents were missionaries and he knew kind of what they called traitor navajo and he knew enough to know how complicated it was anyway to the marines and said look, here's what i think you need to do. they said how can you prove this to us?

can you give us some sort of demonstration? he found for navajo man who worked in california and brought them to san diego and have them say a few things using a quick little code they devised. the marines were so impressed with how indecipherable it was and how foreign it all sounded to their ears if they said yes, were going to give this a chance. the marines recruited 30 young navajo man to become code talkers and defined codes that other navajos couldn't break. >> the reason was the navajo was the hardest to learn. tonight there's a couple of the reasons they chose to use navajo. was not at the tender but language,, so no one could buy a book and learn it.

and they were very unfamiliar to anyone who would have grown up with navajo. so it was almost impossible for anyone to learn. and greek, navajo people have always been very willing to participate in the military and defend their country because they raised his warriors. they were thought lot of young navajo man. [inaudible] -- when we hit guadalcanal, the code said japanese machine-gun on your right flank. and the kurds said [inaudible] >> the men who were assigned to develop a code at first were

shocked because in boarding school that had been severely punished if they spoke navajo. character had or had their teeth broached without around so. but they soon realized he was a pretty clever idea because they realize almost no one outside of the navajo reservation spoke navajo. there were maybe 30 people who could kind of speak navajo who hadn't grown up on the reservation. so what did it first is to develop an outfit that. and it was clever because it's a doubly encrypted alphabet. they chose an english word for every letter of the alphabet. they chose words that would be easy to remember, things turn their everyday life. it was aunt and he was spare and so on. but then they translated to english word to a navajo word.

so the navajo word really had nothing else to do with the letter in the english alphabet. so it can't become white sheet. then they came up with a bunch of different birds for pieces of equipment, military ranks, months of the year and things they would be saying that they didn't want to have to spell out, so they chose again items from their everyday lives that made them think i know a small and maneuverable propitious if you've ever seen hummingbirds, they'll attack each other. and so on, the words all made

sense. but other navajos made all sent. it was worth every in them. not even our government understood this code. there was no one other than the navajo code talkers who could understand the code. so the marines put great t. so the marines put great tressed and these young navajo man and it was also recommended by the same men who recognize the developer code using navajo went to the army had suggested they should do the same thing. but the army decided the navajo young man would be smart to handle that. they actually developed a code. a didn't just speak navajo. they developed a strategy in code that other navajos could never break and i think it's important people know that because that's the biggest misconception. people are said to me they spoke

navajo, big deal. but it was a big deal. they didn't just speak navajo. the code is not use today, but he was from the outside clinics. to chester and the other men, they thought it was easy. but as 30 or 35 pages of code words. i don't know how they did it, but they did. developing the code to the young men about 13 weeks and they were locked in a classroom. they let out for dinner and lunch was brought to god in the classroom and every night when i locked the classroom, all their working papers were locked in a safe. if they were given the congressional gold medal, in 2001 there were only five of them still alive at the time. the other code talkers who followed were given the congressional silver medal so

they were recognized as someone really important to the war effort. it was a long time after the war and fortunately, but it's nice the government did recognize them. >> war was very important to our tribe and the rest of the people. [inaudible] to be able to come home alive.

>> good morning, good morning, thank you. i moved up by five and i'm going to -- easier port literally right after he spent about 10 minutes here reading. i'm going to read something quite sure. on the theory that less is more, which is what i try to tell my writing students and speaking of them, one of the reasons i'm hurtling back is because they have told officers at the lovely little iv brats. so bus ticket home and sleep well or try to sleep although my wife and i say we haven't supposed to the jimmy carter industry shame. thank you so much.

you holding us up is the key and i bet i won't even have time to save thank you angered by two miles, so i'll just say two miles how eloquent to say great introductions have been and tell him good by an arlo and all the rest of you for coming. i'm supposed to read something. i'm fretting about what to be. i'm going to read from the end of the prologue. one of the things he was trying to stress in the toxic it yesterday and the panel he appeared on day before is that for all of the undeniable,

appalling dark side of the hemingway, there was also delayed. there was this bout of generosity. and sometimes they came out fast when a child was involved. not disenchant necessary and especially until child. who wouldn't respond to that. he seemed to respond in a special way. so i was thinking of reading something of a key west passage and i said no, that would be like a piece of cola offering the newcastle. so i'm not going to read that. i'm just going to read this little moment from the end of the prologue and indeed it's the end of ernest hemingway's life when everything was lost. but there is still something

there. book backwards 17 days from his death to june 15, 1961 in rochester, minnesota. a man in the psychiatric ward at saint mary's hospital at mayo clinic is writing a letter to a 9-year-old boy. the man writes on two small sheets metal paper and his big brown budgie bohan but this trademark slant, and irreversibly damaged his entire landscape now it paranoids nightmare has found within himself at the end of his life the kindness encouraged and momentary lucidity, not to say literary craze, to write 210 beautiful words to a kitty likes very much. whenever i begin to feel repulsion at hemingway speaker

and brash behavior towards other human beings, i like to take out a copy of this letter. 210 words with so much emotion tucked below the surface of the pros. the sentences pile driven by contained acute observation of the natural world would've been a a half decent output for a workday, even in a masters prime. the boy, his name is frederick g saviors, although everyone including hemingway posts and threads has a congenital heart condition. he is the son of hemingway's small-town doctor who is also one of hemingway's favorite duckhunting can pinon. in these last weeks, hemingway has oncpinon. in these last weeks, hemingway has once more from idaho for

treatment to nato. not long after this note to france, hemingway will fool his foolish doctors at the world-famous clinic into believing he is well enough to go to idaho and i must immediately the bosch shotgun would go off in the quiet of the house that fits a couple hundred yards for the steep slope from the west bank of the big wood river. the patient on the log toward its st. mary's on june 15, 1961, has just learned dr. savior son is in a denver hospital. in idaho, hemingway and fritz and fritz's father like talking about rainbow trout. none of that will ever be the same again.

st. mary's hospital, rochester, minnesota june 15, 1961. as terribly sorry to hear in a note from your father figure of a death in denver for a few days more to tell you how much i hope will be able to feel better. it's been very hot and muggy here in rochester, but the last two days it is turned and lovely the country is beautiful around here and i've had a chance to see wonderful country along the mississippi where they used to drive the logs annealed lumbering days in the trailer were the pioneers came north, south of the best jump in the river. it is really a very beautiful

country and there's financed 20,000 contacts in the fall. but not as many as in idaho. we will both be back there shortly and can joke about experiences together. best always to you old-timer from your good friend who misses you very much, mr. poppa. ps, best of all the family and feeling fine and cheerful about things in general and hope to see you all soon. >> no one knows for sure, but he seemed to be the last row sentences ernest hemingway sat down on paper. amid so much room in conestoga beauty. thank you very much.

[applause] >> so knowing you have a covenant housing is a no-brainer. i felt thought privilege to write the forward because it gave me a chance to recognize the fact that my dad would've been homeless and so. he was born to a single mother, very poor. my father now exaggerates even more genetically. he couldn't afford the other two letters, but it was an extraordinary community is very intact and very watchful of the children and my father was taken in by another family who threw extraordinary love cat my father

forward. people in the trinity hospital for the first semester tuition so college became a reality. all these things i call the conspiracy of love have been it made me who i am today, the stars of the young people. what bothers me about our society as we talk so dramatic we in such a negative fashion about the adults who fill our prisons. we don't realize every one of those adults with a child who we could have done more for to prevent a lot of the challenges as an adult. it was easier to raise strong children and heal broken man. and so i just feel real urgency in america that we do not prioritize our children as much as we should.

>> team has been saying to a fair part of the book is your forward, which is lovely because we work for two and a half years on the book. but it's so moving because it's from the heart and this week you're doing something else, you're a geisha miss that challenge. maybe you can talk about why you're doing this. >> so my staff teases me. i was up late with micro friend twitter. when sma are going to get a life? it was a sunday night before those of you who use social media to throw out things that are done frankly. just as i was getting into an intellectual question about the role of government and the person said that government should not provide for the nutrition of children and it really struck a chord to me because i don't do people think about that would team.

we don't realize we live in a society do we make small amounts of investments early. we won't have to make big investments late family all in fact are deeply invested because of the more successful of the economy grows, artists, teachers, professors attach printers. the greatest natural resource we have in america is our chi3 the greatest natural resource we have in america is our children. so long story short, but it made for some insane space and she goes back and i say finally, i know we see what it's like to live on food stamps for this program. so i went to bed taking no big deal. i woke up and it was a big story. and so i called my staff and i said guess what i'm doing? but it is a powerful thing

because they were one of 14 cities as a food policy director. where dirty done a lot of work while trying to expand affordable healthy options. the more i talk with my food policy director may say this is a great thing. but could not only raise levels of compassion and understanding and dispel stereotypes and focus them on the reality of that and the policy changes we can make at the local level to address food insecurity, to address food deserts and more healthy options. that's really what we're doing this week. as many moments where we have to think of society as a whole. i pointed to a moment where he had security guards in my office about what talking with them because these are guys, some of them making $7 change in our.

many try to make more money still qualify for programs like snape. so here we are allowing employees, especially as i was saying behind the curtain. you notice that is the one curtained off here. the line across magazines. so you guys should put your book on the i/o. >> should've called it 50 shades of homelessness. >> you guys have such dirty minds. get back to the subject. he and. but these guys, the poignant testimonies are telling us because were in a society where frontline first responders intervening in petty crime had one of the buildings targeted by some people who i terroristic

intent. we can only pay $7 change in our have no benefit from in the retirement security. one guy worked for 10 years. he has to come work for the sickness. that's not the america i think it is. so i'm hoping this week, just to finish the overly long answer, just to bring more attention to these problems. right now the session congress says we need to debate cuts in a snap program. in this time of austerity we can't be done without things that provide long-term benefits that are really really investments in us and our society that we should prioritize these things federally as well as sections locally. >> mayor company you are speaking in your forward about small actions people to to help your father. we talk about small actions

people take that can help homeless young people. can you talk about how that works in the city? >> first of all, i have lots of conversations of people with famous people like tyler perry was homeless, living in a car to people i know throughout my community who have broken drug addictions, though it was her go, brutal he should because they came out of the closet at a young age. it's amazing to me that we have these stories about how one person's small act of kindness is a difference maker for them. it gives me chills to think we all have that power in the biggest thing we do probably could be a smaller kindness to someone else. so the fragility of life, you really get to see up close and personal in cities like ours in

new york and in newark, new jersey and how it doesn't take that much effort to be there for kids. >> reargument judge frederick block, author of "disrobed: an inside look at the life and work of a federal trial judge." you are brought onto the court in 1994. >> correct. i was nominated by president clinton, recommended by moynihan. i've been there the last 18 years. >> your subtitle, inside look at the life and work. give me a regular day in the court room. >> is no such thing as a regular day in the court room. we sent his people to jail for your four times a week. we have to share the high-profile trials because we set in new york city.

the life in new york city is a very dynamic type of judgeship because were really pretty much where the action side. i think terrorist cases. you name it, we've had it. >> heavy separate professionally from personal life. i guess this is a consuming position. >> is certified to believe i have been integrated by it. i'm a musician. i wrote this at the tender age of 78. the reason why did i decide to believe they need more transparency about what we're all about here at the public doesn't have a lot of awareness or understanding. the whole trick is to bridge that gap and a federal court judge right about that will be an entertaining book that at the same time people learn an awful lot about were all about, how you become a federal court judge. they're such a need for that. >> he sat on the bunch for a

long time now. tell us what are your biggest cases. >> i've had a lot of them. they're all in the book. did peter gotti trial which was a major case. i was one of the profound race rights they had in the history of new york city. i recently had the child with it to have guys were acquitted and i have sentenced a terrorist attack. had the trail involving new york city the madrid club scene and on and on. i can't believe i've had all these trials, but i feel blessed to have the opportunity to preside as a federal judge in new york. >> talking to frederic block, author of "disrobed: an inside look at the life and work of a federal trial judge." thank you very much. >> here's a look at books being published this week:

>> sometimes referred to as the left-handed grants. he was 10 years younger than grant and sherman. he was a dynamo, inspired his men with intensity and by his personal leadership. he died from the front, but he was also a careful planner. yet he promptly acted on a plan once it was made and was willing to change it if they can nations change on the battlefield.

but during the war, shared and became a household name because the great victories in the shenandoah valley, especially at cedar creek park are waging what was called a total warfare. he was one of the most dependable generals, so much so during the closing days became the de facto commander. few dispute sheraton is the most aggressive general the union had. they shared on the worse western theater of stones river, tennessee, alertness and tenacity saves you are william nostrand's army on the last day of 1862. his division strong missionary which in november 1863. in march 1864, grant are assured

that east. sheridan spent the last year of the war in virginia. after the war, sheridan carried out governments reconstruction policies in louisiana and texas. he wish he cold war in the mexican border during the plains indian war was a top indian fighter, eventually became commander-in-chief of the army and surprisingly saved yellowstone national park from exploitation. sheridan grew up in ohio and graduated from west point in 1853. when the civil war began in 1861, sheridan was scared 30-year-old infantry cat jane serving in the oregon