Ben Goldacre is the British physician and “nerd cheerleader” who has done important, debunking work about everything from homeopathy to vaccine scares in his book Bad Science and in his Guardian newspaper-column of the same name. Now, Goldacre has turned his gaze to the pharmaceutical industry. With Bad Pharma, his new book, he painstakingly documents how “good science has been perverted on an industrial scale.”
Pharmaceutical companies frequently hide unflattering studies and selectively publish positive ones, which means a lot of science never sees the light of day and the medical community has a skewed view of the evidence base for treatments. As well, trials are sometimes flawed by design and done on unrepresentative patients so that outcomes are more favourable. Drugs are then marketed to doctors in ways that reflect a distorted reality. Meanwhile, regulators are asleep at the switch and the medical establishment is failing to address these well-documented problems. Medicine is, indeed, broken.
While the book is called Bad Pharma, it could have easily been called Bad Doctors, Bad Academics, Bad Health Regulators, or just Bad Medicine. “The book describes a huge interlocking ecosystem of problems that reinforce each other and it’s not just about drug companies being bad,” Goldacre told Science-ish. “It’s also about regulators lacking ambition, doctors being thoughtless and lazy, lots of people being quite reasonably self-interested in systems that have perverse incentives, and patients not getting sufficiently involved in evidence.”
At a time when celebrity doctors peddle junk science, Goldacre talks with Science-ish about why real science matters, the next revolution in medicine, and the paramount importance of fair tests and trials.
Q: You’ve popularized nerdy things like clinical trial design at a time when it’s been more fashionable, among pop-science writers, to cover astronomy or particle physics. Why has study design been such a big focus?
A: There are two reasons I do what I do: one is, if people are wrong and misleading patients, the public, and governments, then that’s bad and people need to talk about it and you can try to fix these problems. I also just think it’s interesting pop science. People do a lot of pop science about the stars, about physics, about the wonders of life but people seem to have neglected epidemiology and clinical trial design. And actually, I think the way you design a fair test, and the way fair tests can be corrupted and flawed by design in such a way that they exaggerate the benefits of treatments, are themselves interesting and fun pop science topics. Because it’s not just about how you design a fair test, but how you catch people out when they have designed one that’s unfair. It’s also about how medicine intersects with culture and how the different players in the world of medicine interact with each other.
Q: What was it like to move from covering straight-up quackery—like ear candling and homeopathy—in Bad Science to the more covert and complex tactics the pharmaceutical industry sometimes uses to distort science?
A: Quacks and drug companies both use exactly the same tricks to get customers but drug companies use a more sophisticated version of them because they are bamboozling a more sophisticated audience, which is doctors, and to an extent (health) regulators. The thing they have in common is the interplay between why they distort evidence, but also their motivation for distorting it and how they commit themselves to doing things and saying things which seem to be very helpful and untrue.
Q: Alternative medicine is frequently criticized for not being evidence-based and using poor-quality research to back health claims. And yet, you show in the book how the evidence for ‘mainstream medicine’ is similarly flawed and perverted, don’t you?
A: Let’s be clear about scale here. I think it’s fairly unusual that drugs get on to the market which do nothing, or which are worse than useless. It happens but not commonly. What’s much more common is that we are misled about the relative benefits of treatments. That’s a serious problem with a serious impact on patient care. If we are misled into thinking the new expensive drug is better than the old cheap drug when in reality the older, cheaper drug is better, then patients are given inferior treatments. And whenever patients are deprived of the best treatments, because someone is trying to sell them one that’s better than nothing but not the best, then those patients are harmed. Alternative therapies—the vitamin pill peddlers and in particular the homeopaths—are not a huge threat to public health or dangerous. I think they are more interesting than they are murderous. It tells us a lot about the role of medicine in society that people are willing to buy sugar pills with no medicine in them from a homeopath. That’s interesting, but not evil.
Q: Speaking of evil, you’re frequently accused of being anti-industry, but as you outline in the book, this is much more than an industry problem—it’s a systemic problem with medicine today.
A: Indeed, I don’t think drug companies are inherently evil. There is no medicine without medicines. We need drug companies to develop new drugs. I want to live in a world where they behave in an ethical, responsible fashion because I would like to collaborate with them too. It’s ridiculous that by being so permissive about the bad side of industry that we perpetuate a world in which it is unethical to collaborate with industry. People should be able to feel optimistic and positive about industry collaboration but these days, when doctors say they do work with industry, they look embarrassed about it. That’s an extraordinary state of affairs to have arisen. I think the real tragedy is the major medical and academic professional bodies failing to address these problems openly and by their cowardice, they have made it more difficult for good people to work with industry.
Q: While reading Bad Pharma, I couldn’t help but think that we will look back at the way medicine is practiced today in the same way we look at something like bloodletting now.
A: That’s right. There’s this great line from Muir Gray (chief knowledge officer of the National Health Service in the UK) about John Snow, the great-grandfather of epidemiology who spotted the cholera epidemic in London. Gray says that in the 19th century, we made huge leaps in medicine with clean, clear water. In the 21st century, we’ll make the same leaps with clean clear information. That is where the action is. People will look back and say, ‘What on earth were you doing? Why were you relying on small crappy trials in unrepresentative patients? How could you expect to know what the true benefits of your treatments were? You must all be out of your mind.’
Q: So what will this revolution in medicine look like?
A: I think we have failed at getting a competent information architecture for evidence-based medicine. So many of these problems are only problems because medicine isn’t very good at finding out what works, bringing all the evidence together, and then making sure we get that information to the right person—the doctor, the patient—at the right time. If we were good at synthesizing and disseminating evidence to clinicians, marketing would be irrelevant. The only reason it matters that the majority of continuing medical education for doctors is funded by the pharmaceutical industry is because there’s no other better way of disseminating information to clinicians. Similarly, the problems of clinical trials being done in small numbers of unrepresentative people, comparing new treatments against nothing instead of the currently best-available treatment—all those problems should really be solved by embedding randomized controlled trials into everyday, routine clinical practice. You can run trials comparing one statin against another for almost no cost, using (National Health Service) electronic health records. If we did that, frankly, the fact that industry wants to fund trials against nothing—it would become irrelevant.
Q: A big chunk of Bad Pharma is the story of missing data—how the pharmaceutical and even academic community has hidden data on medicines and medical interventions that may be less favourable, so that doctors and patients have access to mostly positive studies, and a skewed picture of the evidence base. Why did you focus on the missing data story?
A: In the past, medical progress was characterized by miracle cures. We were discovering things like cortisol and insulin and streptomycin, taking these life-threatening disasters and turning them around. Your chances of dying as a middle-aged man roughly halved over the past 30 years, not because of any single epoch-making breakthrough but because of a gradual accumulation of shavings and modifications of risk. New treatments tend to move ahead not with leaps and bounds but very modest steps. All together, that has a huge impact on mortality and morbidity and suffering. But each individual step is actually, often, a difference that is sometimes quite difficult to detect. That is why we are ruthlessly hygienic about having fair tests in medicine, comparing one treatment against another. Because the benefits we are attempting to detect are much more modest but all together they add up to something very big. That’s also why we need to be very careful that we are seeing the full picture of the evidence. Because again, the difference is that the benefits and risks we are trying to detect are very modest and much more vulnerable to distortion.
Q: You describe a culture of ‘fake fixes’ in medicine—problems the medical community has long known about such as missing data but failed to properly address. Why are fake fixes so pervasive in medicine?
A: One thing I’ve learned since this book came out is that the process of fake fixes that allow you to kick the problem into the long grass and not address it is no accident. It’s now my view that delay is the industry’s tactic. They want to slow-up people fixing these problems. In the UK, we just launched an enormous campaign on releasing access to clinical trials data called Alltrials.net. Since we launched, we have been very clear that we’re asking for all trials to be registered, going back 20 years, all the results of all trials that have been done and full clinical studies. We are not asking for individual patient data. We wrote a letter to the Association of the British Pharmaceutical Industry asking them to respond, and they wrote back saying that individual patient data brings up a lot of privacy issues. But we have not asked for individual patient data; we asked for summary results! We wrote back to them and said, ‘We want just the summary results and clinical reports.’ And they have just written back today saying the UK medicines agency is looking at this, and they need to wait for them to complete their process. That process doesn’t finish until 2014, and it’s looking at sharing individual patient data. It’s not looking at sharing clinical study reports and summary results. So I cannot view that as a casual misunderstanding. The only way I can interpret that is that they are deliberately delaying fixing this problem.
Q: Health regulators, at least in Canada, seem to employ similar delaying tactics in responding to public inquiries and answering questions about their actions and decisions…
A: It’s extraordinary. I agree. I have serious concerns about regulators around the world. They are very variable. Firstly, they have a huge conflict-of-interest here. They don’t necessarily want everybody to have access to all of the information about drugs which they have because that would allow independent, external scrutiny of their decisions. In some respects, this probably suits civil servants quite well. They can make decisions behind closed doors without public scrutiny. Secondly, regulators also have quite a strong thread of paternalism running through them. It’s still very common to find regulators who will say, ‘Look, it’s not up to doctors or patients to think about clinical evidence. We look at the evidence, decide if a drug works, and put it on the market. Don’t worry your pretty heads.’ That’s an old-fashioned form of paternalism.
Q: The kind of paternalism you describe among regulators is no longer acceptable in other areas of medicine, such as with the relationships between doctors and patients.
A: That’s right. (Regulators) are on the wrong side of history. It’s an anachronism. And I don’t think it can last. But more than anything, what I see in regulators is a lack of ambition. It feels often to me that regulators have decided it’s their job to follow a series of algorithms involving, ‘paper comes in, decision comes out,’ without a clear sense of the bigger picture, which is that we all need to make sure patients get the most effective and cost-effective treatments, which means that they have access to the best information, which means that good trials are done and all of them are reported, and they are then delivered to decision-makers. I think (regulators) have forgotten they are an important part of that system.
Q: You have talked about the concept of ‘do-it-yourself regulation.’ Can you explain what that means?
A: Regulators shouldn’t be regarded as the ultimate single authority on whether a drug works or not. They make a decision at a very low bar about whether a drug should come on to the market. We often approve crap drugs and we want to do that because if patients can’t tolerate the most effective drug in a class because of idiosyncratic side effects then we want to move on to a drug that’s second or third best, but for that patient is the best treatment. Regulators have a low bar for saying something works but they aren’t the only or best people for doing that. They don’t give transparent, reliable information in the way, say, the (non-profit academic collaborative) Cochrane Collaboration does. So we need to get away from this idea that medicine is this cannon of drugs that work or don’t work because regulators either say yes or no. (We need to) recognize the reality of drugs is that we make difficult decisions about the relative benefits and risks of different treatments based on judgement calls and statistical analyses where there’s often a healthy debate about which is the best way of thinking about the question. That all has to happen in the public domain. There are many people in medicine who have this old-fashioned paternalistic idea that regulators make decisions and doctors prescribe what they are told. That doesn’t make sense any more. Regulators have proved they are not perfect. All the big scandals that have been spotted in the last decade—Vioxx, Tamiflu—they were not spotted by regulators but by academics.
Q: Besides the failures of regulators, you also document the failure of doctors to demand better evidence, and you describe their future role as ‘expert personal shoppers.’ I’m sure some doctors won’t be too pleased by that…
A: The role of doctors has changed a lot of over the past 40 to 50 years. There’s been a huge explosion in medical treatments but also medical knowledge. There’s a huge amount of information people need to have and need to know, and increasingly, it’s much more difficult for everyone to have all that information in their head. We’ve seen a real change in the way doctors and patients relate to each other. In the past, we told patients what to do, what to take. We also often lied to patients. We wouldn’t tell them they had cancer, for example. Now, the line from the handout in our medical student teaching course is that you should work collaboratively with the patient toward an optimum health outcome. This means, a lot the time, doctors are doing the job of the pop-science writer. They are talking to people, establishing what their intents are, then having a conversation with them, explaining to them risks and benefits of different treatments, in a way that is appropriate to their level of knowledge and understanding, and the things they are most interested in. Increasingly, doctors will find themselves in a position of being good-quality communicators on risks and benefits and that’s no bad thing. It sits alongside all the other things doctors do.
Q: How has your own practice changed how your view the role of doctors in medicine?
A: The funny thing is, when you first qualify as a doctor, the first time you treat somebody in the emergency room who has come in with a life-threatening asthma attack, and you put the self-breathing nebulizer on their face and they start breathing normally, you feel like a superhero. You saved someone’s life. By time you’ve done it 100 times, it gets boring. But somewhere between those two points, you realize, I didn’t invent salbutamol, I didn’t invent the physiology of the lung, I didn’t invent nebulizers: I am merely the interface between the presentation of this patient at this moment in time and this giant pile of medical knowledge and this giant pile of medical technology. That’s what doctors are. We are the interface. We turn a patient’s very human description of what they are experiencing into a diagnosis that fits within the framework of pathology and treatments. And then we come up with something that will do some good. And we travel back down from that rarefied abstract academic level, to the real world of flesh and blood and pain and pills and syringes, and we do something that will make a difference. The most significant part of being that interface is listening to patients, helping them make decisions that most reflect what they want. That means knowing how to find good evidence and understanding it.
Q: Since the book came out, you’ve been leading a global charge to address the problems with the medical establishment you’ve set out, in particular, the missing data scandal. What’s progress like so far?
A: I am amazed and delighted by the support Alltrials.net has had in the last three weeks since we launched it. Huge organizations have signed up (most recently, the pharmaceutical company GlaxoSmithKline). In the next couple of weeks, a campaign will be launched by junior doctors, taking on professional bodies over dodgy (medical) guidelines. There are various ways to get people to do what you want, or to do the right thing: shame, logic, violence, love, and money are five. I suppose shame and logic are the two tools most available to me, and so far I’ve tried shame and logic in the book. The thing that makes me most happy is seeing somebody in a big organization in a senior role doing the right thing. The thing that makes me really, really sad is seeing very senior people in medicine saying things that aren’t true. It would be ridiculously optimistic and ambitious and unrealistic to imagine I would write a book like this and people would suddenly turn around, immediately read 400 pages, and then instantly go and change their behaviour. But hopefully by setting out the case reasonably, it will help shift things in the right direction.
Science-ish is a joint project of Maclean’s, the Medical Post and the McMaster Health Forum. Julia Belluz is the senior editor at the Medical Post. Got a tip? Seen something that’s Science-ish? Message her at firstname.lastname@example.org or on Twitter @juliaoftoronto