What doctors don't know about the drugs they prescribe | Ben Goldacre

Translator: Joseph Geni Reviewer: Morton Bast

Translator: Anders Finn Jørgensen Reviewer: Niels Justus

Hi. So, this chap here,

Hej. Så fyren her

he thinks he can tell you the future.

tror at han kan fortælle dig om fremtiden.

His name is Nostradamus, although here the Sun have

Hans navn er Nostradamus, selvom The Sun får

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

ham til at ligne Sean Connery.

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

Og ligesom flertallet af jer, tror jeg, så tror jeg ikke

that people can see into the future.

på at folk kan se ind i fremtiden.

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

Jeg tror ikke på forudsigelser og hver gang

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

I hører at nogen har kunne forudsige noget som er sket i fremtiden,

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

så har det sandsynligvis været et lykketræf, og vi hører kun

hear about the flukes and about the freaks.

om lykketræf og særtilfælde.

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

Vi hører ikke om alle de gange hvor folk tager fejl.

Now we expect that to happen with silly stories

Nu forventer vi at den slags sker med fjollede historier

about precognition, but the problem is,

om forudsigelser, men problemet er at

we have exactly the same problem in academia

vi har præcist det samme problem i den akademiske verden og

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

i lægevidenskaben og i dette miljø koster det liv.

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

Tænk først på forudsigelser, og som det viser sig,

just last year a researcher called Daryl Bem conducted

så udførte en forsker kaldet Daryl Bem sidste år

a piece of research where he found evidence

et stykke forskning, hvor han fandt evidens for

of precognitive powers in undergraduate students,

synske evner hos studerende

and this was published in a peer-reviewed academic journal

og fik det publiceret i et peer-reviewed akademisk tidsskrift

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

og de fleste som læste det sagde bare: "Okay, godt nok

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

jeg tror det er et lykketræf, et særsyn, fordi jeg ved at

that if I did a study where I found no evidence

hvis jeg udførte det studie, ville jeg ikke finde nogen evidens

that undergraduate students had precognitive powers,

for at studerende har synske evner og det vil

it probably wouldn't get published in a journal.

sandsynligvis ikke blive publiceret i noget tidsskrift.

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

Og faktisk, vi ved at det er rigtigt, fordi

several different groups of research scientists tried

adskillelige grupper af forskere forsøgte

to replicate the findings of this precognition study,

at gentage resultaterne af dette forudsigelses-forsøg,

and when they submitted it to the exact same journal,

og da de sendte det ind til samme tidsskrift,

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

svarede tidsskriftet: "Nej tak, vi er ikke interesseret i at udgive

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

gentagelser. Vi er ikke interesseret i negative data."

So this is already evidence of how, in the academic

Så her har vi allerede bevis for hvordan den akademiske

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

litteratur giver en skævvredet billede af

of all of the scientific studies that have been conducted.

hvilke videnskabelige studier der er blevet udført.

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

Det sker ikke kun indenfor den støvede akademiske psykologi.

It also happens in, for example, cancer research.

Det sker også indenfor, for eksempel, kræftforskning.

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

Så i marts 20012, for bare en måned siden, rapporterede

reported in the journal Nature how they had tried

nogle forskere i tidsskriftet Nature at de havde forsøgt

to replicate 53 different basic science studies looking at

at gentage 53 forskellige basale videnskabelige forsøg indenfor

potential treatment targets in cancer,

mulige behandlingsmuligheder for kræft.

and out of those 53 studies, they were only able

Og ud af disse 53 forsøg, kunne de kun

to successfully replicate six.

gentage seks forsøg med succes.

Forty-seven out of those 53 were unreplicable.

I Syvogfyrre tilfælde af disse 53 forsøg kunne resultatet ikke genskabes.

And they say in their discussion that this is very likely

Og de sagde i deres diskussion at årsagen muligvis var at kun

because freaks get published.

særtilfælde bliver publiceret.

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

Folk laver masser af forskellige forsøg,

and the occasions when it works they will publish,

og i de tilfælde hvor de virker, vil det blive publiceret

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

og dem som ikke virker, vil ikke blive det.

And their first recommendation of how to fix this problem,

Og deres første anbefaling for at fikse dette problem,

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

fordi det er et problem fordi det sender os alle ned ad blinde veje,

their first recommendation of how to fix this problem

deres første anbefaling til at løse dette problem,

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

er at gøre det nemmere at publicere negative resultater indenfor videnskab,

and to change the incentives so that scientists are

og ændre incitamenterne, så videnskabsfolk er

encouraged to post more of their negative results in public.

opfordret til at indsende flere af deres negative resultater til offentligheden.

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

Men det sker ikke kun i den meget tørre verden

of preclinical basic science cancer research.

af førklinisk basal videnskabelig kræftforskning.

It also happens in the very real, flesh and blood

Det sker også i virkeligheden, i kød og blod

of academic medicine. So in 1980,

indenfor lægevidenskaben. Så i 1980

some researchers did a study on a drug called lorcainide,

udførte nogle forskere et studie om en medicin kaldet lorcainide,

and this was an anti-arrhythmic drug,

som var et middel mod hjerteflimmer,

a drug that suppresses abnormal heart rhythms,

et lægemiddel som undertrykker unormale hjerterytmer,

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

og ideen var at folk som havde haft et hjerteanfald,

they're quite likely to have abnormal heart rhythms,

de har ofte også unormale hjerterytme,

so if we give them a drug that suppresses abnormal heart

så hvis man gav dem et middel som dæmpede unormal

rhythms, this will increase the chances of them surviving.

hjerterytmer, så ville det forøge deres chancer for at overleve.

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

Tidligt i deres udvikling, lavede de små forsøg,

just under a hundred patients.

med under hundrede patienter.

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

Halvtres patienter fik lorcainide, og af disse døde 10.

Another 50 patients got a dummy placebo sugar pill

Andre 50 patienter fik en placebo sukkerpille

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

uden nogen aktive stoffer og kun en af dem døde.

So they rightly regarded this drug as a failure,

Så de betragtede korrekt dette lægemiddel som en fiasko,

and its commercial development was stopped, and because

og det kommercielle udvikling blev stoppet og fordi

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

den kommercielle udvikling stoppede, blev deres forsøg aldrig publiceret.

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

Uheldigvis, i løbet af de næste fem, 10 år

other companies had the same idea about drugs that would

fik andre selskaber den samme ide om et lægemiddel

prevent arrhythmias in people who have had heart attacks.

mod hjerteflimmer til patienter der havde haft et hjerteanfald.

These drugs were brought to market. They were prescribed

Disse lægemidler kom på markedet. De blev udskrevet

very widely because heart attacks are a very common thing,

bredt fordi hjerteanfald er en meget almindelig ting,

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

og det tog lang tid at finde ud af at disse midler

also caused an increased rate of death

også forøgede dødsraten

that before we detected that safety signal,

og før vi opdagede det,

over 100,000 people died unnecessarily in America

døde over 100,000 folk unødvendigt i Amerika

from the prescription of anti-arrhythmic drugs.

på grund af udskrivning af medicin mod hjerteflimmer.

Now actually, in 1993,

Nu faktisk i 1993,

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

forskerne som lavede det oprindelige forsøg i 1980, det tidlige studie,

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

offentliggjorde et mea culpa, en undskyldning til det videnskabelige samfund

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

hvor de skrev: "Da vi udførte vores forsøg i 1980,

we thought that the increased death rate that occurred

troede vi at den forøgede dødsrate som skete

in the lorcainide group was an effect of chance."

i lorcainide-gruppen var på grund af et tilfælde"

The development of lorcainide was abandoned for commercial reasons,

Udviklingen af lorcainide var opgivet af kommercielle årsager,

and this study was never published;

og forsøget blev aldrig publiceret;

it's now a good example of publication bias.

Det er et godt eksempel på publiceringsbias.

That's the technical term for the phenomenon where

Det er den tekniske term for fænomenet hvor

unflattering data gets lost, gets unpublished, is left

uflatterende data forsvinder, forbliver upublicerede,

missing in action, and they say the results described here

og de siger de resultater, der er beskrevet her

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

"måske havde givet en tidlig varsling af problemer forude."

Now these are stories from basic science.

Nu er disse historier fra grundvidenskab.

These are stories from 20, 30 years ago.

Disse er historier fra 20, 30 år siden.

The academic publishing environment is very different now.

Det akademiske publiceringsmiljø er meget forskelligt nu.

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

Der er akademiske tidsskrifter som "Trials" open access tidskrift,

which will publish any trial conducted in humans

som vil offentliggøre ethvert forsøg på mennesker

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

uanset om den har et positivt eller et negativt resultat.

But this problem of negative results that go missing in action

Men problemet med negative resultater, som ikke bliver udgivet

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

er stadig meget udbredt. Det er faktisk så fremherskende

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

at det går til kernen af evidensbaseret medicin.

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

Så dette er en medicin, der kaldes reboxetine, og dette er et middel

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

at jeg selv har udskrevet. Det er et antidepressiv.

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

Og jeg er en meget nørdet læge, så jeg læste alle undersøgelserne

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

at jeg kunne om dette middel. Jeg har læst den ene undersøgelse, der blev offentliggjort

that showed that reboxetine was better than placebo,

der viste at reboxetine var bedre end placebo,

and I read the other three studies that were published

og jeg har læst de andre tre undersøgelser, der blev offentliggjort

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

der viste at reboxetine var lige så godt som alle andre antidepressiver,

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

og fordi denne patient ikke haft det godt med de andre antidepressiver,

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

Jeg troede at hvis reboxetine er lige så godt, er det et forsøg værd.

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

Men det viste sig, at jeg blev vildledt. Faktisk

seven trials were conducted comparing reboxetine

syv forsøg hvor reboxetine blev sammenlignet

against a dummy placebo sugar pill. One of them

med en placebo sukker pille. Et af dem

was positive and that was published, but six of them

var positivt, og blev offentliggjort, men seks af dem

were negative and they were left unpublished.

var negative og de blev aldrig udgivet.

Three trials were published comparing reboxetine

Tre forsøg blev offentliggjort, som sammenlignede reboxetine

against other antidepressants in which reboxetine

med andre antidepressiver, hvor reboxetine

was just as good, and they were published,

var lige så godt, og de blev offentliggjort.

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

Men tre data med gange så mange patienter blev indsamlet

which showed that reboxetine was worse than

som viste at reboxetine var værre end

those other treatments, and those trials were not published.

disse andre behandlinger, og disse forsøg blev ikke offentliggjort.

I felt misled.

Jeg følte mig vildledt.

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

Nu du kan sige, godt, det er et yderst usædvanligt eksempel

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

og jeg ønsker ikke at være skyldig i det samme slags

cherry-picking and selective referencing

udvælgelse og selektive referencer

that I'm accusing other people of.

som jeg anklager andre folk for.

But it turns out that this phenomenon of publication bias

Men det viser sig, at fænomenet publikationsbias

has actually been very, very well studied.

er faktisk blevet meget, meget godt undersøgt.

So here is one example of how you approach it.

Så er her et eksempel på hvordan du undersøger det.

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

Den klassiske model er, at du får du en masse undersøgelser

you know that they've been conducted and completed,

hvor du ved, at de har gennemført og afsluttet det,

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

og derefter ser om de har været udgivet det

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

i den akademiske litteratur. Så tog her tog man alle forsøgene

that had ever been conducted on antidepressants

der nogensinde havde været gennemført på antidepressiver,

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

som er godkendt i en 15-årig periode af FDA.

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

De tog alle de forsøg, som blev indsendt til FDA som en del af godkendelsespakken.

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

Så det er ikke alle de forsøg, der nogensinde blev gennemført på disse medikamenter

because we can never know if we have those,

fordi vi kan aldrig vide, hvis vi har dem alle,

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

men det er dem, der blev gennemført for at få en markedsføringstilladelse.

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

Og derefter undersøgte man om disse forsøg var blevet offentliggjort

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

i peer-reviewed akademisk litteratur. Og resultatet var at

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

det var omtrent 50-50. Halvdelen af disse forsøg

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

var positive, og halvdelen af dem var negativ, i virkeligheden.

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

Men når søgte efter efter disse forsøg i den peer-reviewed akademisk litteratur,

what they found was a very different picture.

fandt de et helt andet billede.

Only three of the negative trials were published,

Kun tre af de negative forsøg blev offentliggjort,

but all but one of the positive trials were published.

Men alle undtagen én af de positive forsøg blev offentliggjort.

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

Nu, hvis vi bare bladre frem og tilbage mellem disse to,

you can see what a staggering difference there was

Du kan se hvilken rystende forskel der var

between reality and what doctors, patients,

mellem virkeligheden og hvad læger, patienter,

commissioners of health services, and academics

sundhedsvæsnet og akademikere

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

kunne se i peer-reviewed akademisk litteratur.

We were misled, and this is a systematic flaw

Vi blev vildledt, og dette er en systematisk fejl

in the core of medicine.

i kernen af lægevidenskaben.

In fact, there have been so many studies conducted on

Faktisk har der været så mange undersøgelser på

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

publikationsbias nu, over hundrede, at de har været

collected in a systematic review, published in 2010,

samlet i en systematisk gennemgang, offentliggjort i 2010,

that took every single study on publication bias

der tog hver enkelt undersøgelse af publikationsbias

that they could find.

at de kunne finde.

Publication bias affects every field of medicine.

Publikationsbias påvirker hvert felt af lægevidenskaben.

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

Ca. halvdelen af alle forsøg, i gennemsnit, forbliver upublliceret,

and we know that positive findings are around twice as likely

og vi ved, at positive resultater har omkring dobbelt så stor chance

to be published as negative findings.

for at offentliggøres end negative resultater.

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

Dette er en kræftknude i kernen af evidensbaseret medicin.

If I flipped a coin 100 times but then

Hvis jeg slog plat eller krone 100 gange men derefter

withheld the results from you from half of those tosses,

tilbageholdt resultaterne fra halvdelen af disse kast,

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

kunne jeg få det til at se ud som om jeg havde en mønt, der altid viste plat.

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

Men det ville betyde, at jeg havde en tokronet mønt.

That would mean that I was a chancer

Det ville betyde, at jeg var en svindler

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

og du var en idiot for lade mig slippe af sted med det. (Latter)

But this is exactly what we blindly tolerate

Men det er lige præcis hvad vi blindt tolererer

in the whole of evidence-based medicine.

i hele den evidensbaserede lægevidenskab.

And to me, this is research misconduct.

Og for mig, er det et forskningsbedrag.

If I conducted one study and I withheld

Hvis jeg gennemførte en undersøgelse og jeg tilbageholdt

half of the data points from that one study,

halvdelen af datapunkter fra én undersøgelse,

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

ville I med rette beskylde mig for uredelig forskning.

And yet, for some reason, if somebody conducts

Og alligevel af en eller anden grund, hvis nogen udfører

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

10 undersøgelser, men kun udgiver de fem, som giver det resultat, at de ønsker,

we don't consider that to be research misconduct.

mener vi ikke, at det er uredelighed.

And when that responsibility is diffused between

Og når dette ansvar er spredt mellem

a whole network of researchers, academics,

hele netværket af forskere, akademikere,

industry sponsors, journal editors, for some reason

industrisponsorer, tidsskriftsredaktører, finder vi det

we find it more acceptable,

af en aller anden grund mere acceptabelt,

but the effect on patients is damning.

men effekten på patienterne er ødelæggende.

And this is happening right now, today.

Og det sker lige nu, i dag.

This is a drug called Tamiflu. Tamiflu is a drug

Dette er et medikament, der kaldes Tamiflu. Tamiflu er en medicin

which governments around the world have spent billions

som regeringerne rundt om i verden har brugt milliarder

and billions of dollars on stockpiling,

og milliarder af dollars på lagre,

and we've stockpiled Tamiflu in panic,

og vi har oplagret Tamiflu i panik,

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

i troen på, at det vil reducere graden af komplikationer af influenza.

Complications is a medical euphemism for pneumonia

Komplikationer er en medicinsk eufemisme for lungebetændelse

and death. (Laughter)

og død. (Latter)

Now when the Cochrane systematic reviewers

Nu når Cochranes systematiske reviewere

were trying to collect together all of the data from all

forsøgte at indsamle alle data fra alle

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

forsøgene, som der nogensinde er blevet gennemført, om hvorvidt Tamiflu faktisk gjorde dette eller ikke,

they found that several of those trials were unpublished.

fandt de , at flere af disse forsøg var upublicerede.

The results were unavailable to them.

Resultaterne var ikke tilgængelig for dem.

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

Og når de begyndte at få referater af disse forsøg gennem forskellige forskellige kilder,

through Freedom of Information Act requests, through

gennem Freedom of Information Act anmodninger, gennem

harassing various different organizations, what they found was inconsistent.

pression på forskellige forskellige organisationer, var det de fandt inkonsistent.

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

Og da de forsøgte at få fat i rapporter vedrørende kliniske undersøgelser

the 10,000-page long documents that have

10,000-siders lange dokumenter, der har

the best possible rendition of the information,

den bedste muligt gengivelse af resultaterne,

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

blev de fortalt, de ikke kunne få lov at se dem.

And if you want to read the full correspondence

Og hvis du vil læse den fulde korrespondance

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

og undskyldninger og forklaringer fra medicinal virksomheden

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

Du kan læse det i denne uges udgave

of PLOS Medicine.

af PLOS Medicine.

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

Og den mest forbløffende ting af alt dette, for mig,

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

er, at ikke alene er dette et problem, som vi ved

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

er et problem, men vi har må lide under snydeløsninger.

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

Folk har foregivet, at dette er et problem, der er blevet løst.

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

For det første havde vi forsøgsregistre, og alle sagde,

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

Åh, er det okay. Vi får alle til at registrere deres forsøg, de vil sende resultaterne,

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

de vil fortælle hvad de vil undersøge før de gør det,

and then afterwards we'll be able to check and see if all the trials which

og så bagefter kan vi checke og se, hvis alle forsøgene,

have been conducted and completed have been published.

som er gennemført og afsluttet er blevet offentliggjort.

But people didn't bother to use those registers.

Men folk gider ikke at bruge disse registre.

And so then the International Committee of Medical Journal Editors came along,

Og så derefter kom den internationale sammenslutning af medicinske tidsskriftsredaktører,

and they said, oh, well, we will hold the line.

og de sagde, Åh, godt, vi vil holde linjen.

We won't publish any journals, we won't publish any trials,

Vi vil ikke offentliggøre nogen artikler, vi vil ikke offentliggøre eventuelle forsøg

unless they've been registered before they began.

medmindre de har været registreret før de begyndte.

But they didn't hold the line. In 2008, a study was conducted

Men de holde ikke linjen. I 2008 blev der gennemført en undersøgelse

which showed that half of all of trials published by journals

som viste, at halvdelen af alle forsøg offentliggjort af tidsskrifter

edited by members of the ICMJE

redigeret af medlemmer af ICMJE

weren't properly registered, and a quarter of them weren't registered at all.

ikke blev registreret korrekt, og en fjerdedel af dem ikke var registreret overhovedet.

And then finally, the FDA Amendment Act was passed

Og derefter endelig kom FDA med et krav,

a couple of years ago saying that everybody who conducts

for et par år siden, som kræver at alle der udfører

a trial must post the results of that trial within one year.

et forsøg, skal udgive resultaterne af dette forsøg inden for et år.

And in the BMJ, in the first edition of January, 2012,

Og i BMJ, i den første udgave af januar, 2012

you can see a study which looks to see if people kept

kan du se en undersøgelse, som ser på, hvorvidt folk overholdt

to that ruling, and it turns out that only one in five

denne regel, og det viser sig, at kun én ud af fem

have done so.

har gjort det.

This is a disaster.

Dette er en katastrofe.

We cannot know the true effects of the medicines

Vi kan ikke vide sande virkningerne af lægemidler,

that we prescribe if we do not have access

som vi udskriver, hvis vi ikke har adgang

to all of the information.

til alle oplysninger.

And this is not a difficult problem to fix.

Og det er ikke et svært problem at løse.

We need to force people to publish all trials

Vi har brug at tvinge folk til at udgive alle forsøg

conducted in humans, including the older trials,

udført på mennesker, herunder de ældre forsøg

because the FDA Amendment Act only asks that you publish the trials conducted after 2008,

fordi FDA kræver kun, at du udgiver de forsøg, der er gennemført efter 2008,

and I don't know what world it is in which we're only

og jeg ved ikke, hvad det er, som vi gør, hvis vi kun

practicing medicine on the basis of trials that completed in the past two years.

praktiserer lægevidenskab på grundlag af forsøg, som fuldført i de seneste to år.

We need to publish all trials in humans,

Vi har brug at udgive alle forsøg på mennesker,

including the older trials, for all drugs in current use,

herunder de ældre forsøg for al medicin som bruges,

and you need to tell everyone you know

og du skal fortælle alle du kender

that this is a problem and that it has not been fixed.

at dette er et problem, og at det er ikke blevet rettet.

Thank you very much. (Applause)

Mange tak. (Bifald)

(Applause)

(Bifald)