Hi. So, this chap here,

未来のことがわかると考えていました

he thinks he can tell you the future.

ノストラダムスです

His name is Nostradamus, although here the Sun have

サン紙では ショーン・コネリーのように見えますね (笑)

made him look a little bit like Sean Connery. (Laughter)

大半の人と同じく 私も未来のことがわかるとは思いません

And like most of you, I suspect, I don't really believe

大半の人と同じく 私も未来のことがわかるとは思いません

that people can see into the future.

予知は信じません ときおり 誰かが

I don't believe in precognition, and every now and then,

未来を予知していたという話もありますが

you hear that somebody has been able to predict something that happened in the future,

それはまぐれ当たりでしょう

and that's probably because it was a fluke, and we only

まぐれとか極端な話だけが話題になり

hear about the flukes and about the freaks.

失敗した場合については話題にならないのです

We don't hear about all the times that people got stuff wrong.

予言のようなどうでもいい話は

Now we expect that to happen with silly stories

そんなものでしょう しかし困ったことに

about precognition, but the problem is,

全く同じ問題が学術や医学の世界にも はびこっており

we have exactly the same problem in academia

時として命にも関わる問題になっているのです

and in medicine, and in this environment, it costs lives.

最初に 予言の話だけを考えてみます

So firstly, thinking just about precognition, as it turns out,

つい昨年 ダリル・ベンという研究者が

just last year a researcher called Daryl Bem conducted

つい昨年 ダリル・ベンという研究者が

a piece of research where he found evidence

学生達に予知能力があると証明する研究をまとめ

of precognitive powers in undergraduate students,

査読付きの学術論文として掲載されたことがわかりました

and this was published in a peer-reviewed academic journal

読者の大半の反応です

and most of the people who read this just said, "Okay, well,

「なるほど でもまぐれか例外だろう

fair enough, but I think that's a fluke, that's a freak, because I know

学生達に予知能力があるという

that if I did a study where I found no evidence

証拠が得られなかったという研究だったら

that undergraduate students had precognitive powers,

論文誌に載らないことは明らかなんだから」

it probably wouldn't get published in a journal.

そして まさにそのとおりだと判っています

And in fact, we know that that's true, because

別の複数のグループが

several different groups of research scientists tried

この予知の実験結果を再現しようと試みて

to replicate the findings of this precognition study,

同じ学術誌に投稿したときに

and when they submitted it to the exact same journal,

その雑誌からは「反復実験に興味はありません

the journal said, "No, we're not interested in publishing

ネガティブデータに興味はございません」と言われたのです

replication. We're not interested in your negative data."

この例は 学術誌において

So this is already evidence of how, in the academic

我々が目にする姿と 実際に行われた科学的研究の全体像とが

literature, we will see a biased sample of the true picture

ずれていることの証拠です

of all of the scientific studies that have been conducted.

心理学という生死に関わらない学術分野だけの 問題ではありません

But it doesn't just happen in the dry academic field of psychology.

例えば癌の研究でも同じ問題が起きます

It also happens in, for example, cancer research.

ほんの一ヶ月前の2012年3月に

So in March, 2012, just one month ago, some researchers

ネイチャー誌に出た報告です

reported in the journal Nature how they had tried

癌の治療ターゲット候補に関する53件の基礎研究の

to replicate 53 different basic science studies looking at

再現を試みた研究ですが

potential treatment targets in cancer,

53件のうち 再現できたのは

and out of those 53 studies, they were only able

わずか6件でした

to successfully replicate six.

53件中47件は再現できなかったのです

Forty-seven out of those 53 were unreplicable.

おそらく 特異的な結果が論文にされるので

And they say in their discussion that this is very likely

こうなるのだろうと論じられています

because freaks get published.

よってたかってたくさんの研究をして

People will do lots and lots and lots of different studies,

うまくいった研究は公表され

and the occasions when it works they will publish,

失敗すると公表されません

and the ones where it doesn't work they won't.

この問題に対処するため こんな提案がされています

And their first recommendation of how to fix this problem,

これは問題であり 行き止まりの袋小路に至る問題だからです

because it is a problem, because it sends us all down blind alleys,

この問題を回避するために

their first recommendation of how to fix this problem

科学的に失敗した結果の公表を簡単にすることに加え

is to make it easier to publish negative results in science,

科学者にネガティブデータの公表を奨励するような

and to change the incentives so that scientists are

インセンティブを与えるべきだと 提案されています

encouraged to post more of their negative results in public.

さらに この問題は癌の基礎研究という

But it doesn't just happen in the very dry world

治験前の基礎研究のみならず

of preclinical basic science cancer research.

さらに生々しく肉と血を相手にする医療の学術分野でも

It also happens in the very real, flesh and blood

起きる問題です 1980年に

of academic medicine. So in 1980,

ロルカイニドと呼ばれる抗不整脈剤の

some researchers did a study on a drug called lorcainide,

研究が行われました

and this was an anti-arrhythmic drug,

心拍の異常を抑える薬です

a drug that suppresses abnormal heart rhythms,

心臓発作のあとに

and the idea was, after people have had a heart attack,

不整脈が現われることが多いので

they're quite likely to have abnormal heart rhythms,

抗不整脈剤を投与すれば

so if we give them a drug that suppresses abnormal heart

生存率が向上するだろうという考えです

rhythms, this will increase the chances of them surviving.

開発初期に極めて小規模の治験を行い

Early on its development, they did a very small trial,

100人の患者を対象にしました

just under a hundred patients.

50人にロルカイニドを投与し 10人が死亡しました

Fifty patients got lorcainide, and of those patients, 10 died.

残りの50人には薬効成分を含まない砂糖でできた

Another 50 patients got a dummy placebo sugar pill

偽薬を投与し 死亡したのは一人でした

with no active ingredient, and only one of them died.

研究者たちは この薬は駄目だと直ちに判断し

So they rightly regarded this drug as a failure,

新薬開発は中止されました

and its commercial development was stopped, and because

新薬開発は中止されたので 治験の結果は公表されませんでした

its commercial development was stopped, this trial was never published.

不幸にして その後五年 十年のうちに

Unfortunately, over the course of the next five, 10 years,

他の会社も同じように 心臓発作の後に投与する

other companies had the same idea about drugs that would

抗不整脈剤のことを考えついたのです

prevent arrhythmias in people who have had heart attacks.

これらの薬は上市され

These drugs were brought to market. They were prescribed

何しろ心臓発作は多いだけに たくさん処方されました

very widely because heart attacks are a very common thing,

これらの上市された薬もまた 死亡率を高めてしまうことが

and it took so long for us to find out that these drugs

判明するまでには長い時間がかかったので

also caused an increased rate of death

その赤信号に気付くまでに

that before we detected that safety signal,

アメリカでは死ななくてよかったはずの人が 10万人以上も亡くなりました

over 100,000 people died unnecessarily in America

抗不整脈剤を処方されたための死者です

from the prescription of anti-arrhythmic drugs.

さて1993年になると

Now actually, in 1993,

1980年に初期の研究をした研究チームは

the researchers who did that 1980 study, that early study,

科学界への告解文というべき論文でこう述べています

published a mea culpa, an apology to the scientific community,

「1980年に我々の研究を行ったときに

in which they said, "When we carried out our study in 1980,

ロルカイニドを投与した群における

we thought that the increased death rate that occurred

死亡率上昇は偶然のものと考えていた

in the lorcainide group was an effect of chance."

ロルカイニドの開発は事業判断により中止され 結果は公にならなかった」

The development of lorcainide was abandoned for commercial reasons,

ロルカイニドの開発は事業判断により中止され 結果は公にならなかった」

and this study was never published;

これは「公表バイアス」のわかりやすい例です

it's now a good example of publication bias.

この専門用語は

That's the technical term for the phenomenon where

嬉しくないデータが失われたり 公表も

unflattering data gets lost, gets unpublished, is left

対応もされずに放置される現象を示します

missing in action, and they say the results described here

論文には「後から生じた問題は 早期に警告できるはずだった」と書かれています

"might have provided an early warning of trouble ahead."

ここまでの話は20−30年前の

Now these are stories from basic science.

基礎科学におけるエピソードでした

These are stories from 20, 30 years ago.

今では学術出版の環境もすっかり変わりました

The academic publishing environment is very different now.

オープンアクセスの「トライアル」のような学術誌は

There are academic journals like "Trials," the open access journal,

結果のいかんを問わず人間を対象とした治験を

which will publish any trial conducted in humans

掲載しようという方針です

regardless of whether it has a positive or a negative result.

それでも否定的な結果が失われがちという問題は

But this problem of negative results that go missing in action

やはり広く見られるものであり

is still very prevalent. In fact it's so prevalent

「根拠に基づいた医療」の中核にかかわる問題です

that it cuts to the core of evidence-based medicine.

これはレボキセチンという薬で

So this is a drug called reboxetine, and this is a drug

私も処方したことがある抗うつ剤です

that I myself have prescribed. It's an antidepressant.

オタク気味の医者なので この薬に関して

And I'm a very nerdy doctor, so I read all of the studies

読みうる全ての論文を読みました レボキセチンは

that I could on this drug. I read the one study that was published

偽薬よりも良いと示す論文を1本 そして

that showed that reboxetine was better than placebo,

別の3つの論文ではレボキセチンは

and I read the other three studies that were published

他の抗うつ剤と同等の効果を認められていました

that showed that reboxetine was just as good as any other antidepressant,

私の患者には他の薬は効かなかったので

and because this patient hadn't done well on those other antidepressants,

レボキセチンが同等というなら試すべきだと考えました

I thought, well, reboxetine is just as good. It's one to try.

それは誤った判断とわかりました 実際は

But it turned out that I was misled. In fact,

砂糖で作った偽薬とレボキセチンとの

seven trials were conducted comparing reboxetine

治験は7つ行われ そのうち1つの治験では

against a dummy placebo sugar pill. One of them

効果が認められ論文となりました

was positive and that was published, but six of them

他の6つの治験では否定的な結果となり 公開されないまま終わったのです

were negative and they were left unpublished.

レボキセチンと他の抗うつ剤を比較した—

Three trials were published comparing reboxetine

3つの治験が公開され 同等の効果が

against other antidepressants in which reboxetine

認められていますが

was just as good, and they were published,

その3倍もの患者数を調べ レボキセチンは 他の抗うつ剤よりも劣るとわかったのに

but three times as many patients' worth of data was collected

その3倍もの患者数を調べ レボキセチンは 他の抗うつ剤よりも劣るとわかったのに

which showed that reboxetine was worse than

これらの処置や治験については論文になりませんでした

those other treatments, and those trials were not published.

私は誤った判断に導かれたようです

I felt misled.

極端に例外的なケースと

Now you might say, well, that's an extremely unusual example,

思われるかもしれません

and I wouldn't want to be guilty of the same kind of

今 問題にしている人たちのように都合のいいデータを

cherry-picking and selective referencing

選ぶようなことを 私はしたくありません

that I'm accusing other people of.

結局 公表バイアスという現象は

But it turns out that this phenomenon of publication bias

これまで大いに研究されてきたものです

has actually been very, very well studied.

たとえばこんなふうに調べます

So here is one example of how you approach it.

古典的モデルですが 実施されて

The classic model is, you get a bunch of studies where

完了した治験を集めて

you know that they've been conducted and completed,

これがどこかの学術雑誌に

and then you go and see if they've been published anywhere

出版されているかどうかを調べます さてここに示すのは

in the academic literature. So this took all of the trials

この15年間に FDA が承認した

that had ever been conducted on antidepressants

抗うつ薬の治験の全てです

that were approved over a 15-year period by the FDA.

承認手続きの一環としてFDAに 提出された治験を集めたのです

They took all of the trials which were submitted to the FDA as part of the approval package.

実際にどれだけの治験が行われたかを 知ることはできないので

So that's not all of the trials that were ever conducted on these drugs,

全てが網羅されているとは限りませんが

because we can never know if we have those,

製品を承認させるために行われた治験なのです

but it is the ones that were conducted in order to get the marketing authorization.

さてこれらの治験について 査読付学術誌での出版状況を

And then they went to see if these trials had been published

調べたところ こんなことがわかりました

in the peer-reviewed academic literature. And this is what they found.

治験の成績は五分五分で 治験の半分は成功

It was pretty much a 50-50 split. Half of these trials

半分は否定的な結果というのが実状でしたが

were positive, half of them were negative, in reality.

これらの治験についての査読付き論文を探してみると

But when they went to look for these trials in the peer-reviewed academic literature,

違う描像が浮かび上がってきます

what they found was a very different picture.

否定的な治験で論文公開されているのはわずか3報です

Only three of the negative trials were published,

肯定的な治験は1件を除いて全て論文公開されています

but all but one of the positive trials were published.

この２つの結果をよく見比べると

Now if we just flick back and forth between those two,

驚愕すべき違いがわかります

you can see what a staggering difference there was

医師や患者や

between reality and what doctors, patients,

健康保険の理事や学会の人々が

commissioners of health services, and academics

査読付き論文誌を通して知る姿と現実との間のギャップです

were able to see in the peer-reviewed academic literature.

我々は誤った判断に導かれます

We were misled, and this is a systematic flaw

これは医学の根本にあるシステム的な欠陥です

in the core of medicine.

実際 公開バイアスに関してはこれまでに100件以上の

In fact, there have been so many studies conducted on

研究が行われてきましたが

publication bias now, over a hundred, that they've been

さらにその体系的なレビューが2010年に出版されました

collected in a systematic review, published in 2010,

公開バイアスに関してのあらゆる研究を

that took every single study on publication bias

取扱ったものです

that they could find.

公開バイアスは医学のあらゆる領域に影響しています

Publication bias affects every field of medicine.

平均すると 全ての治験のうち半分がやりっぱなしで失われ

About half of all trials, on average, go missing in action,

肯定的な結果が論文になる割合は

and we know that positive findings are around twice as likely

否定的な結果の2倍となるとわかりました

to be published as negative findings.

これは根拠に基づく医学の根本におけるガンです

This is a cancer at the core of evidence-based medicine.

私がコインを100回投げてそれから

If I flipped a coin 100 times but then

その結果の半分を隠すことをしたら

withheld the results from you from half of those tosses,

いつも表の出るコインを持っていると 思わせることができますが

I could make it look as if I had a coin that always came up heads.

実際は違っています

But that wouldn't mean that I had a two-headed coin.

私はペテン師で

That would mean that I was a chancer

いんちきを許す皆さんがアホなのです(笑)

and you were an idiot for letting me get away with it. (Laughter)

でもこれが根拠に基づく医療において

But this is exactly what we blindly tolerate

見過ごされていることです

in the whole of evidence-based medicine.

私見ですが 不適切な研究とも言えます

And to me, this is research misconduct.

一つの研究として行った中で データの半分を

If I conducted one study and I withheld

隠してしまったら

half of the data points from that one study,

間違いなく不正な研究として非難されます

you would rightly accuse me, essentially, of research fraud.

しかしながら 理由はともあれ 10件の研究を行い

And yet, for some reason, if somebody conducts

自分の望む結果の得られた5件だけを論文にする人がいても

10 studies but only publishes the five that give the result that they want,

これを不正な研究とは呼ばないのです

we don't consider that to be research misconduct.

さらに その責任は

And when that responsibility is diffused between

研究者全体のネットワークや学会や

a whole network of researchers, academics,

支援する企業や 学術誌の編集者までにわたって 希薄に広がり 許容されがちです

industry sponsors, journal editors, for some reason

支援する企業や 学術誌の編集者までにわたって 希薄に広がり 許容されがちです

we find it more acceptable,

しかし 患者に及ぼされた影響こそが 逃れようもない証拠です

but the effect on patients is damning.

影響は今まさに生じていることなのです

And this is happening right now, today.

この薬はタミフルです

This is a drug called Tamiflu. Tamiflu is a drug

世界中の政府が何十億ドルもかけて

which governments around the world have spent billions

タミフルを備蓄してきました

and billions of dollars on stockpiling,

パニックに陥ったように競って備蓄してきました

and we've stockpiled Tamiflu in panic,

この薬がインフルエンザ合併症の割合を 低減するだろうと信じたのです

in the belief that it will reduce the rate of complications of influenza.

合併症というのは医学的な婉曲表現で肺炎と

Complications is a medical euphemism for pneumonia

死亡のことです（笑）

and death. (Laughter)

さてコクランの体系的レビュアーが

Now when the Cochrane systematic reviewers

タミフルが合併症を抑えるかどうかという

were trying to collect together all of the data from all

治験の全てのデータを収集しようとしたとき

of the trials that had ever been conducted on whether Tamiflu actually did this or not,

いくつかの治験の結果が 公表されていないことがわかりました

they found that several of those trials were unpublished.

レビュアーは結果を入手できないのです

The results were unavailable to them.

別の様々なルートを通じて詳細記録を集め始めましたが

And when they started obtaining the writeups of those trials through various different means,

情報公開法に基づいたり

through Freedom of Information Act requests, through

いくつもの組織に嫌がられながら集めたデータは 整合しませんでした

harassing various different organizations, what they found was inconsistent.

そして臨床研究の報告書を入手しようとしたとき

And when they tried to get a hold of the clinical study reports,

その文書は1万ページにもわたって

the 10,000-page long documents that have

この研究についての最良の記述をしているのに

the best possible rendition of the information,

これは渡せないと言われました

they were told they weren't allowed to have them.

このやりとりの全てと

And if you want to read the full correspondence

製薬会社による弁解と説明にご興味があれば

and the excuses and the explanations given by the drug company,

全ては今週の PLOS メディスンに

you can see that written up in this week's edition

掲載されています

of PLOS Medicine.

この中で私を何よりも愕然とさせたのは

And the most staggering thing of all of this, to me,

このことが問題だったことに留まらず

is that not only is this a problem, not only do we recognize

これが問題だとわかってなお 偽りの対策でごまかされていることです

that this is a problem, but we've had to suffer fake fixes.

この問題は解決済みだというふりをする人がいるのです

We've had people pretend that this is a problem that's been fixed.

最初に 治験を申請するときには口を揃えて

First of all, we had trials registers, and everybody said,

治験参加者全員に登録させ 実施要綱も公開させると言います

oh, it's okay. We'll get everyone to register their trials, they'll post the protocol,

申請時に言っていたとおりであれば

they'll say what they're going to do before they do it,