At the European Medicines’ Agency meeting held on November 22nd convened to look at the issue of Access to Clinical Trial Data, the pharmaceutical companies came armed with an approach signaled a few weeks earlier by GSK’s Andrew Witty (see Won’t get Fooled Again). The industry panelists came from Lilly and UCB along with a representative from EurorDis Francois Houyez.
Possibly for the record, industry quietly raised their commercial interest in data secrecy in a jaw dropping argument that slipped quickly out of focus. Companies want to keep the data secret so that if for instance their drug gets rejected in Europe, they will still be able to submit the dossier for approval in Canada or Australia or India or wherever. Individual companies also want secrecy so that if they have wasted a lot of money finding out something doesn’t work, other companies won’t be able to save money through their experience.
But this was a footnote to the main show. Faced with a cogent and consistent case for access presented by Peter Gøtzsche, Ginny Barbour and Ben Goldacre, the response was that we in industry want to protect patient privacy and ensure good research governance – we can’t have wild men descending from the hills doing just anything they want with the data. These points were all made most clearly by Francois Houyez on more than one occasion. This was entirely predictable. See Won’t get fooled again and Access to RxISK Data.
Protecting vulnerable companies
The consent forms for industry trials do offer to protect your privacy. The data will be held by the company and only made available to regulators that a company might decide has an appropriate interest. No one who signs these forms realizes that this means we will be hiding your data potentially from everyone for ever while in some cases we will let regulators, who have proven singularly incompetent when it comes to detecting safety issues and whom we for the most part fund, see a part of the data. Even if they ask for all of it – we have a multitude of legal ways to frustrate such requests.
In months to come as the issue hots up expect to hear more from Eurordis. Eurordis will be covered in greater detail in a forthcoming post Access Wars. You will hear Eurordis or other rare diseases patients representatives telling you that as a professional patient group, one of the most influential in the field, with close links to regulators and companies, they are campaigning hard to protect your privacy. Most casual hearers of this message will likely be grateful their interests are being looked after – and who better to do it than a group speaking on behalf of patients.
Hiding the data
The problem here is that most people view the consent form as an invitation to participate in a scientific exercise. This invitation comes with an implicit understanding that the data that arises from the risks we take with medicines will be made available to the scientific community. Few of us would be likely to agree to have our data hidden. But this is what we are being asked to do in disguised form.
The disguise needs to be discarded. Now that the European Medicines’ Agency have agreed to make data publicly available, patients like Francois Houyez need to be able to opt out of scientific exercises if they wish. Heaven forbid that they get confused by the mixed messages being put in the public domain at the moment.
The disguise needs to be discarded so that you and I can be certain we are not putting our friends and families and communities in legal jeopardy. If we consent to have our data hidden we are putting them in a state of legal jeopardy. We in fact do worse than this – we threaten some of the fundamental rights of people we know and love by our participation in clinical trials where the hiding of our data by making it seem like the drug has no side effects in scientific studies makes it impossible for anyone injured by a drug ever to seek redress. Hiding our data makes it impossible for people we know in many cases even to achieve recognition that their treatment may have produced an adverse effect.
Only proper scientists should get to see the data
The second point industry tried to hammer home at the EMA meeting was that “we” need to ensure the right analyses are performed on the data. We need to make sure that patients don’t view the data and come to the naïve conclusion that an increased rate of adverse events on a drug means that there is an increased rate of adverse events on the drug. Even scientists are prone to making what industry call a Type 1 error – concluding that an increased rate of adverse events points to an increased rate of adverse events.
We go to great trouble to make sure any adverse events that crop up in our trials at an increased rate cannot be statistically significant and we don’t want all this work undone by someone saying something about new clothes.
Francois Houyez was very alarmed at the idea that a company or regulator might be saying one thing in public and someone like Ben Goldacre or Peter Gøtzsche might be saying another. This would confuse patients. It would dangerously add to the burden they already bear.
The supreme court v big pharma
But in fact the Supreme Court in the United States has ruled on just this issue. In January 2011 in the Matrixx Initiatives Case, Matrixx was accused of failing to declare anosmia (loss of the ability to smell) as a side effect of their cold remedy Zicam. The company claimed that as the data on anosmia was never shown to occur at a statistically significant rate, it was under no obligation to tell people. The Supreme Court disagreed it said people needed to see that data and needed to be able to make up their own mind whether an increase in the rate of anosmia was a risk worth taking.
They needed to be able to see the data and decide if the increased rate of anosmia was a risk worth taking if they were going to invest in Matrixx.
The rule therefore for you if you are an investor is you get to see the data and make up your own mind what it means. But if you are a patient, Francois Houyez, GSK, and other pharmaceutical companies are proposing a different set of rules for you. If you are taking risks with your life and health you can only have what the drug company decides you can have. The historical record shows that company perceptions of what information you should have are shaped more by their commercial interests than any concern for your wellbeing.
A medicine is…
A medicine is a chemical that comes with information. What is consumed is a combination of chemical and information. The information is what distinguishes a medicine from a chemical. If we are taking a medicine based on false information we are being duped into taking something other than what we might consent to take. Worse again we suspend the natural caution we would have about taking chemicals.
NEJM and JAMA as a source of spam
Its not clear there is any such thing as a right to privacy – governments and corporations snoop on us the whole time. But in so far as any of us have a “right” to a particular religious belief we have an equivalent right to make up our own minds or have a third party of our choosing make an assessment for us as to what the data on a medicine shows. If someone else in whom I do not have confidence makes the determination as to what the data shows, my rights to bodily integrity are at a risk of being violated in a manner that I have not consented to. My privacy rights are being at least as assailed as they are by spam-mailers and cold-callers. In this regard leading journals such as NEJM and JAMA have slipped into the position of being the greatest purveyors of Spam today (Selective Publication and Academic Miscarriages).
Who owns your baby?
In the face of issues like this, many have been asking for years “who owns the data that arise in clinical trials”? It seems very clear that Industry don’t or don’t in all circumstances as in certain legal cases they have taken care to ship data off the US mainland. Ownership of the data is not as much an issue in Europe where there is almost no possibility of taking a successful legal action against a pharmaceutical company.
But the ice is very thin here. All right minded people – which for most of us means people who think like me – might instinctively feel that the people who take risks with medicines in clinical trials at the very least own their bit of the data. But if we do, we need to make sure we don’t mean something like ownership in the sense of property rights. This is a Trojan Horse moment – beware Pharma bearing gifts. It would suit Industry nicely to magnaminously offer patients like Francois Houyez ownership of their data if that meant property rights. Another wonderful press release for GSK. Except in this case we would all end up in a market where data gets sold to the highest bidder.
Ask not what GSK can do for you, but what you can do for GSK?
The paradox here is that almost all the advances in medicine and in science come from sharing rather than ownership. As Annemarie Mol makes clear in one of the greatest books written about medicine – The Logic of Care – it is a calamitous mistake to view patients as consumers. It is nearly as bad a mistake to view them as citizens – although it sounds far less problematic to ask what you can do for your country.
The idea of progress through sharing might sound like something close to socialism. It’s not intended to be. This is certainly not the usual social response to market aberrations which is to call for more regulation. Introducing concepts like ownership rights might sound like a good way forward but it risks being a recipe for handing over the latest rule book to the most powerful. The answer to unaccountable power is usually co-operation.
Aside from a belief that clinical trials are simply not the right way to investigate adverse events, RxISK is founded on the principle of co-operation. The model, if there is one, is more like the one that underpins the voluntary sector or better again the relationship between a mother and child or daughter and parents. This produces extraordinary economic benefits without a need for contracts or rights.
“The paradox here is that almost all the advances in medicine and in science come from sharing rather than ownership.”
“The idea of progress through sharing might sound like something close to socialism. It’s not intended to be.”
You’re so right!
“If you are taking risks with your life and health you can only have what the drug company decides you can have. The historical record shows that company perceptions of what information you should have are shaped more by their commercial interests than any concern for your wellbeing.”
Isn’t this the problem for consumers/patients? It’s all about the filthy lucre. How can we in Europe change this? We vote for governments who promise reforms here there and everywhere and who ultimately settle for pharma big bucks and fail their citizens at their most vulnerable. Shame and more shame on all of them.
If GSK and big Pharma feel sufficiently threatened by the European Medicines’ Agency or the FDA obliging them to surrender data, they will simply move their game up a notch. This shift upwards may not happen overnight and be a bit more expensive for them, but whatever their costs they are always recoverable.
Pharma might decide to simplify their lives and improve their image, after all shipping data offshore might stir up suspicions of wrongdoing, never mind the perception of guilt; –while the strategy of ghostwriting articles has been exposed for what it is, a scam that has hopefully been put to rest (but don’t blink …). Question: might those arguably misguided adventures be seen by industry as just part of their own growing pains?
Pharma might decide to move back ‘up the food chain’ to the genuine old RCTs, absorbing and then recovering the costs that such a step backward might involve: a strategy involving a backwards move while planning the next move forwards, so to speak. This revolves around the idea that data can be altered at source, at the point it is collected, in order to generate a desired result.
But what about data verification? Data verification in RCTs only confirms the validity of essential data as transcribed from source, there being no verification of data that has been fabricated from thin air, and most people don’t know the difference. The internal validity of a trial can therefore be tossed together like a garden salad rather than be generated from a scientific discovery process. And external validity is something different again.
The opportunities for falsification of data at source are many and these manipulations can be done off site. But there is a difference: one thing is transcribing from extant patient records (patients’ files) thereby leaving an inconvenient paper trail, and it is another to collect data electronically (EDC) on electronic CRFs (eCRFs), where the ‘meaningful’ part of yet-to-be-derived data can be falsified as it is captured; –data is not fabricated because the participants actually do exist and there is compliance with the protocol.
Should additional tweaks be needed a computer literate, perhaps even well intentioned summer student can be paid a few dollars an hour to access the database, code to find ‘every elevated LDL level collected on 45-49 year old men taken on alternate Thursdays’, adjust the readings upwards by 8.5 percent, close and leave without a trace, all of which could be undetectable. So this now becomes genuinized data that can be released to the world and no one would know.
The reason Pharma is currently fighting the release of their data is probably that they know very well their data manipulation has been somewhat clumsy having dealt with it at the back-end, rather than genuinizing it at the front-end, and that those mistakes are going to cost them.
The only way to stop corruption and fraud in the long run is by regulating the activity of clinical research. This does not mean regulating the trialing process itself. So that individuals in any way connected to a clinical research activity would have to be trained, qualified, certified, registered and licensed in clinical research by a regulatory college, as professionals are in every other health related activity. They might then be held accountable as individuals. Clinician-investigators would presumably start howling against any such proposition; meanwhile Pharma continues to laugh all the way to the bank.
I had an interesting conversation with my local pharmacist today. I live in a small community in which everyone seems to know everyone else. His comments re SSRIs: “I haven’t seen anyone get better on them.” “They’re no better than placebo.” “I have only two patients on MAOIs – there should be more.” “Why don’t they just prescribe a placebo – it would be just as good.”
My experience in major teaching hospitals has been that physicians listen to the pharmacist more than to anyone else. In the community – at least this community – he’s whistling in the wind.