Poking your nose in where it’s not wanted: the dark side of investigating healthcare
The 25th HealthWatch Award was presented to the former BMJ Investigations Editor Deborah Cohen, in recognition of her courageous reporting of medical issues in the face of attack from vested interests. She received her award from HealthWatch president Nick Ross at the 2017 HealthWatch AGM on Tuesday 17 October 2017, and the following text is adapted from her presentation.
It has meant a lot to me to know that others recognise that you’re reporting in the face of vitriol and attacks from vested interests.
The first big story I did was in 2009. HealthWatch members will already be familiar with the Tamiflu story and how the Cochrane Collaboration struggled to access clinical trial data for their review of this flu drug. The investigation into the missing data ended in my in-tray. I have degrees in medicine and medical journalism, so it was something to get my teeth into, and it turned out to be an interesting journey. As you’ll recall, the lead authors of the clinical trials had told us they didn’t have access to the primary data. The manufacturers, Roche, gave us observational data. But not the clinical trial data.
One of the privileges of working at the BMJ is you have access to some really good academics who will drop everything to help you out – they know who they are. So we set them to work making sense of this observational data that we had. Roche maintained that Tamiflu not only reduces the incidence of secondary complications by 67%, but it also reduced the rate of hospital admissions by 61%. But when we re-analyzed the observational data we found it didn’t say that. We published our report,(1) and the allegations we made were quite controversial. No satisfactory explanations had been given for the missing clinical trial data. Yet there were journalists who’d been covering the flu pandemic for a while, who’d taken it at face value that Tamiflu must be effective enough to merit the UK government stockpiling it – at a cost of £424m between 2006 and 2013.
Conflicts and controversies
In the course of our search we came across whistleblowers who had written up the clinical trial paper that was published in The Lancet. We were told they’d been instructed to include key words and phrases that were important for the marketing of the drug. It also became apparent that the people who were involved with the companies marketing the product were also key advisors to the World Health Organization (WHO), the European Medicines Agency (EMA), and others. Roche, together with some other manufacturers of influenza drugs, are the funders of a group called the European Scientific Working Group on Influenza (ESWI). This industry-funded group wrote WHO’s first influenza pandemic preparedness plan.
So, as a result of this investigation we did a follow up piece about the conflicts of interest of the advisors.(2) This proved even more controversial than the first, and we came under attack. The WHO would not release any information about whether their advisors declared conflicts of interest or not. The scientists said they had, but WHO would not confirm or deny.
Embarrassingly I ended up on an American “shock jock” show, where they took the line that this is a big pharma conspiracy, which was very far from what we had intended with this story. We’d been very careful to make the point that when you are making policy decisions, which could have huge impact in terms of finances and patient harm, should you really be using the people who have been marketing the drug? You would think not.
But the attacks continued. Nature did an exposé,(3) and people tweeted that our research had been discredited. In our article there had been a sentence that maybe could have been worded more carefully, so we ended up conceding one or two points in relation to the story, and the result was taken as an admission that our story was wrong, starting a back and forth between Nature and the BMJ. It became one of the most read articles on the BMJ website at the time.
The next thing we looked into was medical device regulation. We started with the European Union directives, which took some reading. Ours was a joint investigation between the BMJ and Channel 4’s Dispatches, so we had a whole team to help us go through the directives. Then in 2011 a very useful study was published by a group at the Cleveland Clinic, in Ohio. They looked at the US Food and Drug Administration (FDA)’s regulatory system for medical devices. The FDA have two processes, one is pre-market authorisation, and the other is called 510(k). To get a device onto the market using pre-market authorisation you have to have clinical evidence, such as from randomised control trials, or a case control study. But for 510(k) you only need to show something called equivalence, which means that your new device is “substantially equivalent” to an old one, a predicate device, that is already authorised.
It is possible to trace the history of some 510(k)-authorised devices, and to track the tweaks and changes that have been made to the design over time, so that a device approved in 2015 might not have had any clinical studies to support it, and gone through so many changes since, say, 1980, when the original version was approved, that you can end up with virtually a whole new device but without any clinical evidence for it. The Cleveland team’s study found that two thirds of device recalls were in respect of devices that had gone through the 510(k) route, that is, there was no direct clinical evidence for those versions of the devices.(4)
Working with the Centre of Evidence based Medicine at Oxford University, we thought it would be interesting to replicate this study in Europe. But we couldn’t. Because we kept coming up against the words: “commercial, in confidence”.
Devices come in different risk categories. In Class I, you have low-risk items like plasters and syringes; in Class II there are insulin pumps. Class III is for hip implants, pacemakers, things that need surgery to get them in place. I phoned the Medicines and Healthcare Regulatory Authority (MHRA) and asked if they had a list of Class III devices that are used in the NHS.
And they said “No”. No such list exists.
We knew that some implants must have been causing problems, so we put in a freedom of information request to look at reported harms. Again, “commercial in confidence”. We learned that under the Freedom of Information Act you can get clinical data related to pharmaceutical trials, but you can’t get it for medical devices.
So, we were thwarted from the outset by a lack of transparency. We wrote a story(5) about a hip implant called the ASR, which is made by DePuy, a subsidiary of Johnson & Johnson, and which was recalled in 2010. The ASR was used for two types of orthopaedic surgery – total hip replacement (ASR XL) and hip resurfacing (ASR resurfacing). The ASR in the United States had gone through 510(k) for total hip replacements. For the hip-resurfacing application it should have gone through the full pre-market authorisation process, but the FDA had called a halt to the approval because of problems with the trial. That didn’t happen in Europe. It has been used in tens of thousands of patients.
We later looked at all large diameter metal on metal total hip replacements. We reviewed the medical literature, and whole 510(k) daisy chains, and found that among an entire class of implants there was no publicly available clinical data to support their use. Now, bear in mind that these are made of cobalt and chromium. In the 1970’s and 1980’s surgeons were debating the toxic effects of cobalt and chromium, and in 2007 a comment piece in the Lancet(6) pointed out that little is known about the transport, distribution, excretion of metal ions in the body; toxic effect thresholds have not been characterised. So wouldn’t you think there would be a need to proceed with caution? We know a lot about the local effects of metal ions in the body, but about the systemic effects we know very little because if we’re talking about pharmacovigilance of medical devices like you have with drugs, well, it just doesn’t happen.
My question was, in this article, surely we should proceed cautiously with these devices, given that there is no known toxic threshold for these compounds? In the media it became, the BMJ hip replacement cancer scare, which had not been our intention at all. Suffice to say, that the bullying I encountered as a result gave me a glimpse into the business side of orthopaedic surgery and how things run when there is a great deal of money at stake, and some of it is very unsavoury.
After that, we went under cover. I should point out that it is actually really hard to get clearance to go under cover. There has to be a strong public interest reason, and you have to demonstrate that you can’t get the information any other way. We had a good reason – this was health. We created a fake metal hip. We called it the TMH (total metal hip) and tried to get our metal hip onto the market. We had a fake dossier, and gave it some disastrous data, failures, ions all over the place.
The system here in Europe is that you have to go through what are called notified bodies. There were about 70 of these, they are private organisations, and these are where manufacturers go to get a CE mark. Once they have this mark they can sell the device anywhere across Europe. The various regulators in the different countries then regulate those bodies. It’s an incredibly complex system.
You have to pay these bodies to review your dossier, before they decide if it should receive a CE certificate. That already seems like a conflict of interest, as these companies are paid for approving the device and then are paid again to do follow up audits every few years. An officer at one of the notified bodies in the Czech Republic admitted to us that it is “on the side of manufacturer and their products, not on the side of patients.”
We set up a fictitious company called Changi, and we hawked our new device around notified bodies, pretending to be their PR people. We found that not only are regulators out-sourcing the regulation of medical devices to private companies, but some of these private companies are further out-sourcing their device regulation activities to other private companies. We found ourselves at a branch of a Czech notified body which was in South Korea, where a small group of companies came together to operate as a one-stop-shop. Device manufacturers would go to one for advice about how to create their dossier, then another to receive the CE certificate, and then the next company offered marketing. We were told that it is easy to get products approved in Europe, and they are not worried about inspections by European officials, though they were really scared of the FDA.
We got our design approved for a CE certificate for our non-existent hip implant. That investigation had an impact.(7) We presented our evidence at the Science and Technology Committee and at the European Parliament when they were updating their medical device regulations.
Sugar and water
We were at a meeting just before the 2012 Olympics, when sports medics told us that sports drinks are one of the most controversial thing ever. What could be controversial about sweetened water? Again working with the Centre of Evidence based Medicine, we decided to take the top 10 best selling UK and US sports and fitness magazines, and pulled out the adverts to see what health claims they were making for sports drinks. Then we looked at the references to see what evidence we could find to support them. Then we tried the manufacturers’ websites to find references, before approaching the companies for evidence to support their claims.
We were told things like, water doesn’t quench thirst.
Anecdotally, we heard that one company was telling kids that if they drank water rather than sports drinks they would get cerebral oedema. An idea had been created, that of ‘exercise induced dehydration’. Now, any doctor who has treated someone with genuine dehydration, well, it’s not quite about jumping on the treadmill and getting a bit thirsty. But this idea had been created that if you exercise you get exercise-induced dehydration, and the solution is to drink more sports drinks. This then led to a genuine health concern which is, hyponatremia – a reduction in the body’s electrolyte levels due to drinking too much, which was leading to cerebral oedema, and there was a study in the New England Journal of Medicine that looked at the Boston Marathon and people who had had exercise-induced cerebral oedema, because they’d actually been taking in too much fluid.(8)
The backlash against our “The truth about sports drinks”(9) was incredible. The attacks kept on coming. One company held up what they said were over 100 clinical trials that underpinned their science. We critically appraised all of them, and found that most were poor quality and didn’t support the main claims made.
We did a programme with BBC’s Panorama about in vitro fertilisation (IVF).(10) You get to a certain age and lots of your friends are going through IVF, and if you listen to their stories you find they’re all having different kinds of treatments and also being sold extras, on top of IVF, so-called “add-ons”. We again teamed up with Oxford University, we scraped lots of websites from fertility clinics looking at the claims made for add-on treatments, and then looked at the evidence. We found very few were able to support their claims that these add-ons could actually improve your chances of having a baby. We used undercover again – and came under criticism for this – to look at the lack of fully informed consent for these treatments, and the lack of evidence, and how poorly regulated they are by the HFEA. Needless to say we experienced a backlash from private clinics in the UK.
There are times when you get home, shut the door, and you’re looking at the door, wondering what’s going to happen next. You can take the professional attacks, but worse than that is the sexual harassment. As a female reporter on a heavyweight subject I have my authority questioned, because I’m young and female, and I am constantly challenged about my ability to carry out these investigations. It’s a constant battle, and that is probably the biggest challenge in this area.
Based on Deborah Cohen’s presentation at the HealthWatch 2017 AGM, adapted by Mandy Payne
1. Cohen D. Complications: tracking down the data on oseltamivir. BMJ 2009;339:b5387. http://www.bmj.com/content/339/bmj.b5387
2. Cohen D, Carter P. WHO and the pandemic flu “conspiracies”. BMJ 2010;340:c2912. http://www.bmj.com/content/340/bmj.c2912
3. Butler D. Flu experts rebut conflict claims. Nature 2010;465:672-673. http://www.nature.com/news/2010/100608/full/465672a.html
4. Zuckerman DM1, Brown P, Nissen SE. Medical device recalls and the FDA approval process. Arch Intern Med. 2011 Jun 13;171(11):1006-11. https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/227466
5. Cohen D. Out of joint: The story of the ASR. BMJ 2011;342:d2905. http://www.bmj.com/content/342/bmj.d2905.full.print
6. Learmouth ID, Case CP. Metallic debris from orthopaedic implants. The Lancet 2007;369(9561):542–544. http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(07)60251-7/fulltext
7. Cohen D. How a fake hip showed up failings in European device regulation. BMJ 2012;345:e7090. http://www.bmj.com/content/345/bmj.e7090
8. Almond CSD et al. Hyponatremia among Runners in the Boston Marathon. N Engl J Med 2005; 352:1550-1556.
9. Cohen D. The truth about sports drinks. BMJ 2012;345:e4737. http://www.bmj.com/content/345/bmj.e4737
10. BBC One Panorama. Inside Britain's Fertility Business. First broadcast Monday 28 November, 20:30. http://www.bbc.co.uk/programmes/b084ngkd (not currently available).