by Ben Goldacre
That’s an interesting finding, because there is no other explanation for it. There was no difference in methodological rigour, or quality, between the government-funded research and the industry-funded research. There was no difference in the size of the samples used in the studies. And there is no difference in where people submit their articles: everybody wants to get into a big, famous journal, and everybody tries them first. If they get rejected they will try lesser and lesser journals, until someone takes the paper. It’s possible that the industry-funded researchers were simply more dogged, or more shameless; and that maybe, when they were rejected by one major journal, they hawked their paper to other equally large ones. It’s possible that they could do this more rapidly than those without industry funding, because they had administrative assistance from professional writers to deal with the tedious bureaucracy of each journal’s submission system, and tolerated the long delay in publication that this strategy would cause. Or perhaps lucrative industry-funded studies are simply favoured by editors.
Either way, getting published in a higher-impact journal is a huge advantage, for a number of reasons. Firstly, it is prestigious, and implies that your research is regarded as higher-quality. But secondly, papers in bigger journals are simply more likely to be read. As we’ve already seen, our systems for disseminating knowledge are ad hoc and antiquated, built on centuries-old platforms where science is presented in essay form, and printed on paper, with no clear mechanism for getting the right information to the right doctor at the right time. In a world where the information architecture of medicine is so massively flawed, simply getting under someone’s nose counts for a lot.
This brings us on to one final, dismal tale. In medicine, appearances are important: the appearance of an independent study, the appearance of lots of individual papers all saying the same thing, can help build a case in the minds of busy prescribing doctors. We have seen how individual academic papers can be ghostwritten. But in 2009 a court case in Australia involving Merck revealed a much stranger new game.
Elsevier, the respected international academic publisher, was producing, on behalf of Merck, a whole range of journals, entirely as advertising projects for that one company. These publications looked like academic journals, and they were presented as academic journals, published by the academic journal publisher Elsevier and containing academic journal articles. But they only contained reprinted articles, or summaries of other articles, almost all of which were about Merck’s drugs. In issue 2 of the Australasian Journal of Bone and Joint Medicine, for example, nine of the twenty-nine articles were about Merck’s Vioxx, and twelve of the remainder were about Fosamax, another Merck drug. All of these articles presented positive conclusions, and some were bizarre, including a review article containing just two references.
As well as specialist ‘journals’, Elsevier also produced a journal aimed at family doctors, which was sent to every single GP in Australia. Again, it looked like an academic journal, but was actually promotional material for one company’s products.
In a statement to the Scientist magazine after only one of these journals had been uncovered, Elsevier tried to defend itself by arguing that it ‘does not…consider a compilation of reprinted articles a “Journal”’. This defence was optimistic at best. We are talking about a collection of academic journal articles, published by the academic journal publisher Elsevier, in an academic-journal-shaped package, laid out like an academic journal, with an academic journal name: the Australasian Journal of Bone and Joint Medicine. It has since been discovered that Elsevier put out six journals like this, all sponsored by industry.99 Chief Executive Michael Hansen finally issued a statement admitting that these were made to look like journals, and lacked proper disclosure.100
As we’ve seen, it has been estimated that it would take six hundred hours a month to read the thousands of academic articles relevant to being a GP alone. So doctors skim, they take shortcuts, they rely on summaries, or worse. The simple and predictable consequence of these journals sent out by Merck – and all the other distortions we have seen, from adverts to drug reps, ghostwriting and so on – is that a misleading picture of the research on these drugs will lodge in doctors’ memories.
A quarter of the pharmaceutical industry’s revenue is spent on marketing, twice as much as it spends on research and development, and this all comes from your money, for your drugs. We pay 25 per cent more than we need to, an enormous extra mark-up in price, so that tens of billions of pounds can be spent every year producing material that actively confuses doctors, and undermines evidence-based medicine. This is a very odd state of affairs.
What can be done?
Journals should publish all advertising revenue from each individual drug company annually, and for each individual issue.
Journals should publish all reprint orders retrospectively for all papers at the end of each year, disclosing income for each; and for each new industry paper, they should declare how much they have previously made in reprint orders from that company.
Editors should anonymously disclose all cases where pressure is applied for commercial reasons.
Editors should declare their own conflicts of interest, funding sources if they are working academics, stocks, and so on.
More research should be done looking at whether projected advertising and reprint income has an impact on journals accepting papers.
Pharma’s medical school
At the beginning of this chapter, I presented you with what I hope is a chilling thought: the most senior doctors working today qualified in the 1960s. Today’s medical students will qualify at the age of twenty-four, and will then work for five decades. When you’re at medical school you’re told which treatments work best, in lectures and textbooks, and then you’re tested on it. A few years later you’re still doing specialist exams, and training in a safe and constrained world, with smart people actively teaching you. Then, suddenly, you’re out on your own, seeing patients and getting on with it. Medicine changes around you, unrecognisably over the course of decades: whole new classes of drugs are invented, whole new ways of diagnosing people, and even whole new diseases. But nobody sets you an exam, nobody gives you a reading list, Prof. MacAllister doesn’t tell you what works and how. You’re alone.
Doctors need to learn about new drugs all the time, but we leave them to get on with it by themselves. Privately organised professional education is extremely expensive – hundreds or thousands of pounds for every course – so individuals tend not to pay for it themselves. The state doesn’t want to pay for it either. So the pharmaceutical industry pays instead.
The Department of Health spends a few million pounds a year providing independent medicines information to doctors. The industry spends tens of billions on providing biased information. This presents a bizarre situation: doctors’ continuing education is paid for, almost exclusively, by the industry whose products they buy with public money, and by the industry that has been shown routinely to mislead them.
In fact, in the UK, doctors are now actively forced to collect Continuing Medical Education (CME) points, which are counted up each year. This has been tightened up since the changes at the GMC on account of a GP called Harold Shipman, who turned out to be a serial killer, murdering older women with overdose injections of opiates. In a rather odd game of consequences, this means that a set of new regulations, brought in to prevent doctors from murdering people, in reality has simply shepherded them even more into the hands of expensive industry-sponsored promotional activity, where they are misled about the benefits of expensive medicines, and so harm patients.
The basic design of the pharmaceutical industry’s medical school for qualified doctors is simple: doctors who already like a company’s drug are identified by local drug reps, and then given a platform. In detail, this process can take on many different shapes. Sometimes a company will pay for its favoured doctor to give a talk to other local doctors. If they’re good, it will pay for them to give a tal
k to other doctors further away. If they’re reasonably senior, or influential, or have some kind of academic track record, it will pay for them to go to conferences, or give lectures around the world. Sometimes these lectures will be part of the conference sessions, but sometimes there is a whole separate industry strand, with an eerie edge to it.
In fact, it’s worth noting that the look of a ‘medical conference’ is what most people in other industries would recognise as a ‘trade fair’, and in some respects it’s odd that we don’t call them that now in medicine. The hall outside the lecture theatre is filled with promotional stands in which nice stuff is given away, brightly coloured floor-to-ceiling banners advertising various products, and attractive drug reps stepping imperceptibly into your path to engage you in conversation about their wares. This is what a trade fair looks like, although sometimes the cues are easy to miss.
I recently found myself eating some salmon at a boring doctors’ conference in Cardiff, out near the academic poster displays. It was pretty good salmon, but gradually I noticed that I was standing, eating, in a kind of temporary autonomous zone, denoted by a change in the colour of the carpet, and some brightly coloured promotional displays. I was approached by an attractive, smiling woman in a suit. She asked me where I worked, and whether I saw patients who might use her company’s drug; it was only then that I realised whose fork was in my mouth. The food was for people going to some special lectures, paid for by a drug company, in a parallel session, featuring its own chosen speakers. There was no drama, nothing rude, the rep was happy to chat, and the food was really good. She just wanted my contact details.
The paid speakers at these events are the ‘key opinion leaders’ (KOLs) we met earlier, and it’s an odd scene, not just for the audiences, but also for the KOLs themselves. Nobody is obliged to change their views in return for money, in an overt act of corruption, though that may well happen: for the most part, these people are simply saying what they already thought about the drug anyway. But views favouring industry are given a platform, a microphone, and a nice projector for their slides; while those less favourable to industry are left to fend for themselves. In this way, as with negative results being buried, a biased picture is created of the overall swathe of viewpoints and evidence; but no individual doctor or academic has done anything they would regard as unethical.
I have good friends, around the same age as me, just finishing research work and entering their first medical consultant jobs, who give paid talks as KOLs. For them, it isn’t about the money, which is often no better than working for an extra day on locum rates. It’s not even about the other benefits, like top-flight training in nice venues on how to write and deliver a smooth presentation. I’m going to quote an ex here, and I’m sorry if that’s weird, but I’m a doctor, so this kind of KOL activity is happening, all around me, all the time. Here’s what she told me: ‘None of those benefits matter. I do it because, in the speakers’ room at a conference, or at a country hotel to learn about presentations, I’m spending time with the giants of my field. I’m thirty-six, and I’m getting drunk with the people who write the clinical guidelines! There’s no way I could do that, unless I was a KOL.’
This is not unusual: often at conferences there will be a posh evening party hosted by one company. Only the people that company knows and likes are invited. If you go to its party, you get to meet top, influential people; if you don’t, you don’t, and this can have a corrosive effect on a conference generally, by sucking out the top ranks. I have a friend who complains that since vaccines have been invented for the disease he works on (which is mostly found in the developing world) the conferences are suddenly held in much more expensive hotels, and most of the senior figures disappear in the evenings to expensive restaurants, paid for by drug companies. Previously – am I being too utopian here? – they’d be falling over drunk with the junior researchers.
Is the content of this industry-funded teaching systematically biased? One study took a ‘mystery shopper’ approach, sending attenders along to some industry-sponsored CME teaching about calcium channel blockers, a class of blood-pressure drug.101 Usefully, lots of companies make their own version of this kind of drug, and there were two pieces of sponsored teaching on them within a year of each other: one on one company’s drug, the other on another’s. The researchers went to both, and recorded every mention of every drug, noting whether it was positive, negative or equivocal. On each course, the sponsor’s drug was mentioned more frequently, and much more positively: scoring three times as many positive mentions as negative ones. When the rival company’s drug was mentioned, it got a much tougher ride: on the first course, these mentions were more likely to be negative, and on the second they were more likely to be equivocal. In the few statements directly comparing the sponsor’s drug with the competitor’s, the teacher usually said that the sponsor’s drug was best.
The university supervising this teaching had a clear policy on how bias should be excluded. It obviously didn’t work very well. These policies don’t, which is why I don’t ever take them at face value, unless there is very good evidence that they’re followed. In the second study, researchers followed up doctors’ prescribing patterns after they attended some industry-sponsored CME teaching, again on blood-pressure medication.102 This research found that after doctors attended the sponsored course, their prescription of the sponsor’s drug increased.
Are these two studies perfect? No, but only because they have one simple flaw: they were both conducted twenty-five years ago, and nobody has done anything similar since.
This is extraordinary to me. It was established that the most senior doctors in the profession were receiving money to give talks that were, in effect, promotional, under the guise of educational activity; it was established that this distorted content changed prescribing behaviour; and then we just left it alone. The industry says, with no evidence, that everything has changed. I see absolutely no reason at all to believe that. Is a drug company really going to pay for a KOL to be shipped around the country, at great cost, to tell audiences of doctors that a cheap off-patent drug is the most effective first-line treatment for hypertension? The industry regards this activity as promotional; that’s why it pays for it. In almost any medical circle you’ll find stories of biased local consultants who give these talks, and who always seem to prefer one company’s drugs. Leaving them unmonitored, without even the most basic ‘mystery shopper’ research to monitor their content, is a collective scandal.
And these teaching sessions aren’t even just about the benefits of specific drugs: in recent years the manufacturers of the antipsychotic drug olanzapine, for example, have had lawyers running special teaching sessions for doctors.103 They aren’t about the medicine: they are to reassure doctors that they are unlikely to be successfully pursued in the courts over side effects from the drug.
So how big is this scene? Amazingly, although there is extensive work documenting KOL and CME activity in the US, there are very few openly available figures for Britain, because of our secretive regulations. As with drug reps, it’s often a matter of two cultures: some doctors and clinics engage in industry-sponsored education all the time, as a matter of routine; while some never do, and think the whole idea is laughable. I can tell you that it is absolutely routine, at a conference run by a Royal College, to find a sponsored section, with posh food and parallel sessions of lectures, given by doctors and academics paid by the sponsoring companies. I can tell you that friends’ and colleagues’ travel and hotels and registration fees are routinely paid by drug companies. I can tell you that low-key local events, sponsored by a drug company, where a ‘key opinion leader’ gives a talk about a subject area and a drug, are commonplace (and those speakers do always seem to love the sponsor’s drug). But in Europe we have very incomplete data.104
In the US, governments are more interested in transparency. As a result we can see much more, and there is little reason to believe that industry marketing activity there is a
ny different to how it is in the UK. So we know that $30–40 billion is spent by the industry on drug marketing in America, of which only 15 per cent goes on marketing to patients, even though TV drug adverts are permitted in the US. We know their spending priorities are likely to reflect their own research on what marketing activities bear the best fruit, so it’s clear that marketing to doctors is effective. In 2008 the US industry body, the Accreditation Council for Continuing Medical Education (ACCME), reported that CME companies – the private firms acting as intermediaries between industry and some teaching – offered 100,000 teaching activities, amounting to more than 760,000 hours in total.105 More than half of this was paid for directly by industry.
In case you think the US is a very different country to Britain, we can talk about Europe. In France, as of 2008, three quarters of all CME activity is paid for by the pharmaceutical industry, and of the 159 accredited providers, two thirds receive industry money.106 In Germany, a researcher conducted an anonymous survey of members of a major medical society attending an international conference, and got 78 per cent of them to respond.107 Two thirds said they’d got an allowance to attend from a drug company, most of them said they couldn’t have travelled to attend it without that money, and two thirds said they had no ethical concerns about taking the cash. Similarly, they were sure it would have no effect on their prescribing behaviour. They were, as we have seen, wrong: doctors attending a conference paid for by a pharmaceutical company are significantly more likely to prescribe and request that company’s drugs in future.