by Ben Goldacre
But some industry interests are not so perfectly aligned with those of patients, as we already know. A company might want to increase sales of a product through the conventional covert marketing we’ve seen, but also by expanding the diagnostic limits of the disease, to enlarge its market. It has a special interest in selling new drugs, even though, as we’ve seen, these are the very products for which we know the least about the risks and benefits to patients, and for which the cost is – perversely, given the lack of information – highest.
If you read the pharmaceutical industry’s own commercial literature, it’s easy to see how it views relationships with patient organisations. Here is a PR company’s health-care strategist writing in the magazine Pharmaceutical Executive. This isn’t a smoking gun, but rather, a banal corporate explanation of why drug companies give patient groups money:
Years before a new drug is launched, pharma companies and advocacy teams should map out how strong ties can advance corporate goals and brand objectives. Product managers see advocacy groups as allies to help advance brand objectives, like increasing disease awareness, building demand for new treatments, and helping facilitate FDA clearance of their drug…But there are a few things to remember: Some advocacy groups, especially the more established ones, will not endorse one product over another. Companies need to determine boundaries early to avoid getting in trouble.35
And so on.
How prevalent is industry funding? The health campaigning organisation Health Action International (HAI) looked at the patient groups working with the EMA, the European drugs regulator.36 Two thirds of them received funding from the pharmaceutical industry, with the average donation rising from €185,500 in 2006 to €321,230 in 2008, which generally represented about half the running costs for each organisation. Most worryingly, it also found that many of them had failed to declare this income clearly. In 2005 the EMA introduced ‘transparency guidelines’, but even by March 2010 only three patient groups had reported income from as far back as 2006 online. Despite this failure, all the organisations were invited back by the EMA to participate in stakeholder meetings.*
Is there any evidence that this funding changes behaviour? I think so, and I think you will agree, though researching this has not been a priority for funders, despite the influence of these groups on practice. As one example, we can look at the ongoing cat-and-mouse game between industry and regulators over whether companies should be allowed to advertise, or give information directly to patients. This is recognised by most companies as an effective way to increase use of their pills, so they are keen to see the laws liberalised. Can we see any trace of this agenda in the patient groups funded by pharma? Another report by HAI, from 2011, looks at patient groups lobbying the European Commission and their lobbying patterns.37 All those that did not receive any funding from a pharmaceutical company wanted to keep the current regulations, preventing companies from promoting their drugs to patients. Groups receiving money from drug companies were significantly more likely to think that the industry should have a larger role in providing information about drugs to patients.
This is a worrying finding in itself, but it also undermines the very purpose of having independent patient charities involved on ‘multi-stakeholder’ policy-making forums: these already contain industry voices, formally representing themselves; the patient groups are supposed to represent patients.
But this correlation between voting patterns and industry income may not be simple evidence of foul play. While there will obviously be episodes of bad behaviour – people altering their views simply to attract industry money – I think there’s something much more interesting happening here. Patient groups’ interests and industry interests can legitimately overlap, as we’ve seen. So there is no need for people to explicitly change their views for the overall voice of patient groups to be distorted: the industry can simply give funding, and therefore a more prominent platform, to people who spontaneously express the views that it prefers. In this way, everyone can feel OK about themselves, while still participating in a broader system that produces a biased and distorted picture of patient opinions. This helps to explain why patient charities receiving industry funding are so angry and baffled if you suggest that their output is biased; even though, overall, it’s clear that their sector’s output is biased.
But that moral sop doesn’t change the reality we see on a regular basis, some of which is frankly ugly. As an example, the Independent newspaper recently examined some major media outcries around patient groups attacking NICE, and married up the attacks with the funding.38 When NICE advised against expensive arthritis drugs, the Arthritis and Musculoskeletal Alliance (ArMA) organised a letter criticising it in The Times, signed by ten rheumatology professors. Half of the charity’s income is from drug companies, and it didn’t raise a single word to criticise its industry funders over the cost of these drugs, despite this being a major policy issue, and plainly relevant to NICE’s decision. The National Rheumatoid Arthritis Society (NRAS) launched an appeal against the same decision, standing alongside ArMA and three drug companies, calling it ‘another nail in the coffin’ for its patients. NRAS receives over £100,000 a year from the industry, and it didn’t criticise the industry over pricing policy either. The National Kidney Federation attacked NICE over its rejection of new, hugely expensive, marginally beneficial treatments.39 Its press release was vicious, describing the decision as ‘barbaric’, ‘damaging’ and ‘unacceptable’. Half of the National Kidney Federation’s annual £300,000 budget comes from the pharmaceutical industry. Nowhere does the press release criticise the company charging tens of thousands of pounds for each person receiving its drug.
The head of NICE, Professor Sir Michael Rawlins, points out that the cost of manufacturing these drugs is often a tenth of the price for which they are sold, that we pay high prices in part for marketing (some of which goes directly to the patient groups), and that when we spend money on one thing, we can’t spend it on something else.40 This last gritty reality, which would present itself in any medical system without infinite money, is not often welcomed by patients and the public. Repeatedly, we come back to the same circle: we pay high prices for drugs; a quarter of what we pay goes on marketing; our money is then spent on things like patient groups; who in turn insist that we should pay very high prices for these drugs, undermining the very groups, like NICE, that try to determine the best choices for patients overall.
What can you do?
Drug advertisements do not serve to inform the public, and should be banned. Their expansion into Europe should be resisted.
If drug companies really want to help inform patients about health, they could pay into a central, independent repository, which can give grants to people with a good track record of giving evidence-based information to the public.
Patients, journalists and the public should be wary of people selling new diseases if they are also selling the cure.
Any firm running a disease-awareness campaign should declare in its advertisements that it is doing so because it is developing or marketing a product to treat it.
All educational materials should bear the same declaration.
Adverts to doctors
Addressing doctors directly is the most tangible way that drug companies try to influence prescribing practice, and this usually happens through print adverts, in academic journals. As with most marketing activity in medicine, we can be fairly sure that if the companies spend money on it, they know it has some value. The published academic evidence supports that view, as far as it goes, but again, this has not been a funding priority.41 So, drugs are used more after their advertising programmes start, and less when they stop. Doctors who recognise the advert for a drug are more likely to prescribe it. Econometric models – as far as any mortal can follow them through – suggest that marketing has more influence on drug-usage patterns than the publication of new evidence, and so on.42
As you might imagine, drug adverts are supposed to b
e regulated for things like truthfulness and accuracy, but there are good reasons to worry that this is not done well. In the UK, the Prescription Medicines Code of Practice Authority administers the Association of the British Pharmaceutical Industry’s code of self-regulation. To find out more about the general tone of adverts, a Health Select Committee looking at the influence of the pharmaceutical industry in 2005 got the Institute of Social Marketing to examine a sample, and it found that the goalposts were mobile. While ads for drugs are supposed to contain ‘objective and unambiguous’ information, in reality they associated them with attributes you’d associate with any other product: ‘energetic’, ‘passionate’, ‘desirable’, ‘sexy’, ‘romantic’, ‘intimate’ and ‘relaxed’. The PMCPA clarified that ‘emotional messages’ were OK, if the material was ‘factual [and] balanced’.
But this is a side issue: what we care about most is whether adverts make assertions that are factually correct, and supported by good-quality evidence. That’s a simple thing to assess: you just gather together the claims from a representative sample of adverts, then check them against the available evidence, and one good example of this kind of study was published in 2010.43 Some researchers from Holland went through all the biggest medical journals in the world – the JAMA, the Lancet, the New England Journal of Medicine, and so on – between 2003 and 2005. Every advert that appeared in that time was included, once each, if it made a claim about the effect of a drug. Then they checked the references for all the claims in the ads, found the trials they referred to, and passed them to an easily exploited workforce of assessors (250 medical students who’d just finished their evidence-based-medicine training).
Each student independently checked the methods of two trials, and the associated advert, using objective criteria, like a well-established scoring system for assessing the quality of trials. Medical students are cheap, but they might not be reliable raters, so each trial was scored by between two and six students, and if there was a discrepancy in scores, the trial was reviewed by a panel of four academics. The results were abysmal. Only half of the claims in the adverts were supported by the trials the adverts themselves referenced as evidence; only half of the trials got a score of ‘high-quality’; and less than half of the adverts – in the leading medical journals in the world – referenced a high-quality trial which supported their claim.
This is just one study, but it’s fully representative of what has been found before. Another study from the Lancet in 2003 looked at all the claims from cardiac medication adverts in six Spanish medical journals: of the 102 references the researchers could trace, 44 per cent simply didn’t support the promotional statement.44 Another from 2008 on psychiatric drug ads found similar results.45 The story is the same for rheumatology drugs.46 Am I cherry-picking? The best current systematic review is free to read, well worth the time, and found twenty-four similar studies.47 Overall, it found that only 67 per cent of the claims in adverts are supported by a systematic review, a meta-analysis or a randomised-control trial.
Despite this overwhelming evidence, the British Department of Health has rejected calls for drug companies to be forced to publish corrective statements when they’ve been found to have made incorrect claims in their adverts.48 So doctors will never know when they have been misled.
As far back as 1995, around half of all medical journal editors responding to a survey agreed that they should check the content of the ads they accepted for accuracy, and even submit them for peer review.49 In reality, this almost never happens.50 If the factual claims in these adverts aren’t reliably backed up by evidence, then you already know all you need to know about whether the regulations governing these claims work: around the world, they don’t.
‘Drug reps’
Drug reps are the people who visit doctors in their offices, and try to convince them in person that their company’s drugs are the best. These people are often young and attractive; they also bear gifts, and the promise of a long, mutually beneficial relationship with a drug company. It’s hard to know how well these interactions are policed: as with all relationships, they are built incrementally, on mutual trust, so the most egregious behaviour will happen between friends. Here, because this world is harder to penetrate, I’ve stepped away from the drier world of evidence, and spoken to some drug reps in confidence: if you’re feeling melodramatic, we could call them whistleblowers, though I don’t think they’ve said anything to me that you wouldn’t hear from them in a pub.51
First, though, before we look at how they operate, there is already a wealth of published evidence on their activities. This is a huge business: the overwhelming majority of the industry’s promotional budget goes on influencing doctors, rather than patients, and about half of that gets spent on drug reps. They are not cheap, and though their numbers fluctuate, they have doubled in the past two decades,52 with one rep for every three to six doctors, depending on how you measure it.53 A systematic review found that the majority of medical students have contact with drug reps before they even qualify.54 Because the industry spends so much money on drug reps, you can be sure they influence prescribing.
Doctors repeatedly assert, in both qualitative and quantitative research – not to mention when you chat with them socially – that drug reps have no impact on their prescribing (many claim that they improve it).55 Cheeringly, they also report that their own behaviour won’t be changed by interactions with drug reps, but that other doctors’ behaviour probably will be.56 And the more drug reps you meet, the more likely you are to think they’re not having an effect on you at all.57
This is naïve arrogance. From the most current systematic review, there have been twenty-nine studies looking at the impact of drug rep visits.58 Seventeen of those twenty-nine studies found that doctors who see drug reps are more likely to prescribe the promoted drug (six had mixed results, the rest show no difference, and none show a drop in prescribing). Doctors who see drug reps also tend to have higher prescribing costs, and are less likely to follow best-practice prescribing guidelines.
To give a flavour of this research, one classic study took forty doctors who had requested that a drug should be added to their hospital formulary – the list of locally approved drugs – in the preceding two years.59 Eighty doctors from the same places who hadn’t applied to put a drug onto the formulary were then randomly selected, and the contact these two groups had had with the industry was compared. The doctors asking for new drugs to be made locally available were thirteen times more likely to have met drug reps, and nineteen times more likely to have directly accepted money from drug companies.
These visits – repeatedly shown to distort prescribing practice – take place on time that patients have paid for, and generally without the blessing of the people commissioning local services, who know that such activities increase costs, through foolish prescribing. They’re also spreading: since the new ‘nurse practitioners’ are now able to prescribe drugs in many places (a development I welcome, although it annoys many doctors), they too have become a target for promotional activity. The most recent US study in this new area found that 96 per cent of nurse prescribers reported regular contact with drug reps, and the overwhelming majority thought such contact was ‘helpful’.60
Individual visits aren’t the only way drug reps can get time to persuade doctors. One of the most prevalent exposures – and one of the hardest to avoid – is at meetings. ‘Grand Round’, for example, is a tradition in most hospitals, where one medical team presents a complex or interesting patient for discussion in front of the rest of the hospital. This is a big deal – especially for the quaking junior doctor who presents the basic history of the patient being discussed – and it’s attended by everyone, from medical students to professors, as an educational event. Grand Round generally happens at lunchtime, with sandwiches by the door, and is sponsored by a drug company: it either presents for a minute or two at the beginning from the stage, or runs a stall, with reps on hand to engage doctors in discussio
n.
I wouldn’t say that working hospital doctors are either particularly rich or particularly poor, compared to other graduates with similar abilities and qualifications. The UK scales are all publicly accessible: junior doctors are paid between £25,000 and £40,000 per annum for the first five or ten years, and then consultants go on to around £70,000. It’s a gritty world, without the kind of glossy corporate perks you might see in the City; but then, it’s a different kind of world. However you cut it, doctors can afford either to buy or to make their own sandwiches, and don’t need these paid commercial advert breaks embedded in normal hospital work. In the many surveys that have been published, junior doctors attend between 1.5 and eight industry-sponsored lunches or rounds each month.61
The problem isn’t just that this kind of sponsorship looks bad. Junior doctors are more likely to choose a sponsor’s drug, even if it is inappropriate, after seeing a drug rep present on it at a Grand Round meeting.62 These interactions generally start at medical school, and doctors can be very naïve about the interest being shown in their career and their well-being.63 To really understand the human impact drug reps have in the workplace, we have to veer – against my better judgement – into personal stories.
When I was doing a junior job in the middle of nowhere, I went for a team meal that was paid for by a drug rep. The evidence, going back many years, shows that people who go on drug-company dinners are more likely to prescribe that company’s drugs.64 But every other junior doctor was going, and since we all lived in hospital accommodation, if I hadn’t gone I’d have been sat in an undecorated institutional bedroom on my own with nobody to talk to. This is not a sob story, but rather a description of how objections are eroded. At the end of the meal, the friendly rep asked where everyone was going next, because we were all soon moving on to our new training jobs. Applying for these was all we’d thought about for weeks, and everyone was bubbling with information.