Note: This was to be the last in the Lasagna posts that began with Not So Bad Pharma, April Fool, Tragedy of Lou Lasagna, Empire of Humbug: Bad Pharma, Empire of Humbug 2, and Brand Fascism. But the series will continue into “Marilyn’s Curse” and “When Does Yes Mean No.”
Faced with questions about the $3 Billion fine imposed on GSK – is it just the cost of doing business? – Andrew Witty snapped back:
“Although corporate malfeasance cases end up looking very big, they often have their origin in just… one or two people who didn’t quite do the right thing. It’s not about the big piece. The 100,000 people who work for GSK are just like you, right? I’m sure everybody who reads the BMJ has friends who work for drug companies. They’re normal people… Many of them are doctors.”
The current action against the European Medicines’ Agency (EMA) in the European Court is the first time Pharma has taken the regulator to a Supreme Court since the Panalba case (see Tragedy).
The 1962 amendments to the Food and Drugs Act required a determination of efficacy for any drugs marketed. In the case of drugs that had been licensed prior to 1962, FDA put in place a Drug Evaluation Study Initiative (DESI) to determine if they were likely to be efficacious. DESI convened panels of academics to adjudicate on these drugs.
FDA and many academic physicians, including Louis Lasagna, did not approve of combination products, whether antibiotics, analgesics or psychotropics of which there are plenty on the market. DESI recommended removing a large number of these. Lasagna’s career changed forever when he came out against this recommendation (see Tragedy).
In 1969, DESI recommended removing Upjohn’s Panalba, a combination of tetracycline and novobiocin. By FDA’s calculations Panalba was likely responsible for 12-15 deaths per year that wouldn’t have otherwise happened if doctors had used tetracycline alone. At the time Panalba was making $18 million per year for Upjohn – 12% of their income; $1 million per death. There was evidence that Upjohn knew of the risks the drug posed and in house evidence that Panalba was no more effective than tetracycline alone.
The options for Upjohn were:
- To pull the drug immediately from the market,
- To stop production but let doctors use up the supplies available,
- To continue to produce but stop marketing,
- To continue to produce and to promote until it was actually banned,
- To attempt to overturn FDA’s decision using the Government and Judicial System
The company chose option 5. It involved the Secretary of State for Health in the Nixon Administration and pursued a legal action all the way to the Supreme Court.
When the details of the case were presented some years later to business management students in the University of Pennsylvania, Scott Armstrong found that none of his class supported Upjohn and all bar two of the 71 found Upjohn’s behaviour irresponsible.
But when he got students to role play the Upjohn Board and faced them with the decision to pull the drug or not, he found to his astonishment that 79% of respondents opted to take the same course of action as Upjohn and none withdrew the drug. This result has been widely replicated in different national groups with socialist Swedes behaving the same as capitalist Americans, and the idealistic young in just the same way as their cynical elders.
Debriefing the subjects and attempting to adjust the board and the circumstance to see if behaviour could be modified, Armstrong and others were left with the conclusion that it is generally accepted that the primary duty of Board members is to stockholders rather than to society at large. Board behaviour can be modified by appointing stakeholders but even then few Boards opt to remove Panalba from the market.
Armstrong modified the role play experiments by introducing an Upjohn safety officer who told the Board that the company did have hidden evidence supporting the complaints – to no effect. In real life not even the death of the CEO’s wife from suicide, from the Company’s drug Prozac, has led to a decision to pull the drug or add warnings.
Putting people in a group where the unstated assumption is they will behave like a proper Board produces a situation which has commentators reaching for comparisons with the Milgram experiment where decent ordinary people in experimental settings can be induced to give life-threatening shocks to innocent strangers simply because they are asked to do so by the conductor of the experiment.
The only thing that appears to modify the likely behaviour of a Board is a threat of boycott – that is if maintaining the drug on the market led to a disenchantment that led the company share price to fall.
While accepting Andrew Witty’s suggestion that he and other employees of GSK are just like us, few of us can imagine being party to behaviours that warrant a $3 Billion fine. But from this study it seems we are more like him than he is like what we might like to think of ourselves.
GSK’s proposals for transparency need to be read against this background – is the company really offering to do the equivalent of remove their drugs from competition? Or are GSK just a little bit smarter than the competition?
When faced with the growing Fascism in healthcare, we can retreat to the wilder shores of conspiracy theory and claim the problems are down to the Germans, the Jews, or a Socialist cabal. Or we can attribute the problems to some evil people somewhere in pharmaceutical companies who break laws.
But if we accept that the Board of GSK and other companies are populated with people just like you and me, who are perhaps even less likely to break the law than you or I, the $3 Billion fine for GSK notwithstanding, then the problem must stem from and the remedy lie in the system.
(Accounts of what led to the 2017 Amendments to the Food and Drugs Act vary wildly. Some say the President’s first grandson was born with Tetralogy of Fallot linked to his daughter’s antidepressant intake. Conspiracy theories invoked Marilyn’s Curse. They noted things like the coincidence of initials between the unknown American Woman whose baby had a Tetralogy of Fallot and Andrew Witty (See AW and AW2). Others thought it was triggered by the fact that the previous President had to step down early because of cognitive failure linked to treatment with a Statin. Whatever the reason, Andrew Witty was recruited from GSK to become special advisor to the White House on how to change the system. Here is a précis of his option appraisal).
1) Regulation & Science
The licensing of drugs is a bureaucratic procedure that has nothing to do with science. As things stand, this procedure has enough of the appearances of science so that most doctors and patients taking part in it are fooled into thinking they are participating in scientific exercises when they are not.
Regulation is incompatible with science. Science seeks uncertainty, whereas regulation seeks to abolish uncertainty.
That Pharma does regulation not science should have been crystal clear in the legal action taken by Pharma against the EMA (2013), when Pharma took the option to stress that their “data” was simply the result of a regulatory exercise and regulators ordinarily maintain the confidentiality of submissions on products they regulate. This was the winning option.
There were doubts within Pharma about their ability to win using this argument while maintaining the illusion that companies do science. The argument was a bitter blow to the self esteem (amour propre) of FDA and EMA who liked to claim they based their actions on science. But medicine predictably noticed nothing.
2) The 2017 Amendments to the Food & Drugs Act
Regulation needs to be disentangled from science, but this cannot easily be done while the 1962 Amendments to the Food and Drugs Act contain an effectiveness criterion. This was a well-intentioned but unfortunate accident – a simple solution to a complex problem that has made things worse.
In the new regulations to mark the 55th anniversary of the 1962 regulations the word effectiveness needs to be replaced by phrasing to the effect a drug should have a clear effect on a structure or function of the body. This falls back on language already in the regulations so should not be unduly problematic.
This would have the merit of making it a lot cheaper to develop new drugs. As a result, the cost of drugs would fall, more new drugs would be brought on the market and after marketing a lot of discoveries would be made. We could look forward to a new era of Wonder Drugs to rival the 1950s.
There would be comparatively little clinical trial data and campaigns like BMJ and the Cochrane collaboration to access the data would likely wither on the vine. It would be safe to appoint Peter Doshi and Tom Jefferson as data access Csars.
3) Evidence based Medicine
The determination of treatment effectiveness is critical to the rational practice of therapeutics. It cannot be left to industry; it should not involve industry. If undertaken properly by medicine, as for instance in trials like the Women’s Health Initiative study of HRT or NIH studies of drugs from tolbutamide onwards, it would likely mean that a full appreciation of the risks and scale of the benefits to be obtained from a treatment would only be realized five years after the initial launch of a new compound.
4) Prescription-only Arrangements
Quite simply these need to be dismantled. Anyone who doesn’t understand this doesn’t understand marketing.
If prescription-only privileges are to be retained, only those who do understand the need to abolish them should be involved in any modification. There are a range of possible modifications such as requiring doctors to have a training post in a company marketing department before qualification. There is talk of rewarding them for reporting adverse events – better to penalize them if they do not report them at a certain rate per year.
Another is to maintain certain drugs on prescription-only status such as chemotherapies for cancer where patients are particularly vulnerable and the treatments particularly toxic.
5) Abolish Medicine
Although companies are engaged in exercises that are simply bureaucratic, they have captured the appearances of science, more successfully than alchemists, homeopaths or others, and using these appearances they have been able to infect the body of scientific therapeutics. The insertion of what are essentially sophisticated adverts into the medical literature has triggered a Clinical Auto-Immune Deficiency reaction leading many doctors and scientists to attack those who question the results of ghostwritten articles that purport to represent “studies” whose data is withheld.
Physicians have had decades to find a way to get this cuckoo’s egg out of the nest of science. They could have based clinical practice only on investigations whose data is publicly available. They could have used the Human Rights of their patients as a lever – it is not for instance possible for any doctor to prescribe any branded medicine with informed consent as things stand at present.
But they haven’t done anything like this. The branches of medicine linked to the prescribing of branded products are finished as a profession. They have no brand value. The market has now developed so that nurses, pharmacists, clinical psychologists and others could take over the role of prescribing and will be far less expensive.
If we believe in the market place, abolishing medicine will lead to a solution in that people and governments will realise that they need a set of experts who have brand value when it comes to good quality information on drugs and a new profession will be called into being.
There already was an effort in this respect by a Hungarian emigre to Canada, some ten years ago, to form a First Multinational Corporation of Physicians. This prototype could be dusted down, or another assembled.
Other physicians practicing in areas of medicine where branded products have had little penetration could be allowed to continue. One caveat here is that I have not considered whether areas of medicine that are heavy users of medical devices or medical procedures are as badly affected as mainstream medicine.
A further option might be a promotion of patient co-operatives. Groups like ACOR.org are making a significant difference to healthcare. They might find it far easier to work with nurse and pharmacist prescribers than with doctors.
One of the few things that might have influenced my behavior as CEO of GSK would have been a boycott. Record fines and even jail time are not deterrents. Almost everyone views these as a cost of doing business already. A jailed CEO might even boost share value.
We in GSK know that boycotts can hurt. The crisis over making ARVs available to South Africa for AIDs and the threat of a boycott was the only time our Board substantially altered its position.
Tied as they usually are to disease indications that don’t exist, and surrogate markers, and produced as they are for bureaucratic purposes only, company efficacy data are frankly close to worthless and are not what companies have been trying to hide anyway. The key data have always been the adverse event data.The boycott would be aimed at getting companies to hand over Adverse Event data.
7) Boycott – 2
In 1962, all of the wrong options from Kefauver’s proposals were picked – the effectiveness criterion, prescription-only arrangements, and controlled trials. The change he most favoured, a revision to the patent arrangements, was rejected.
Modifying patent arrangements is an indirect form of boycott that Congress should consider. Companies need to be rewarded but the current product patents reward then excessively for treatments that often shorten life and increase disability. There are alternate arrangements, such as process patients, that would make the various Panalbas less valuable and put a premium on diversifying the portfolio of compounds companies hold rather than have them dependent on a small number of blockbusters.
8) Access to Data
There are at present thousands or hundreds of thousands of drug trials that remain completely unregistered, where there are no issues of clinical confidentiality. These are studies undertaken in healthy volunteers – normals. There is a compelling cases for ensuring these data are made fully available. Companies will need an amnesty for some of the abuses of people that will come to light as these studies come into view.
9) Independent Adverse Event Reporting
This is the key step. The steps companies have taken to prevent decent reporting, or to denigrate reporting when it happens, and to manage the perception of risk rather than risks themselves, make everything else they have done pale in comparison.
For instance companies supported the development of an FDA MedWatcher App in 2013. Widely applauded, this was in fact a way for companies to reduce the expense of maintaining a pharmacovigilance department, reduce their legal liabilities, and transform adverse events into anecdotes. Companies you see companies have a duty to follow up and decide if their drug caused the problem – FDA don’t have this duty.
But what was astonishing about all this was that pretty well everyone completely bought the idea that when it comes to working out whether a drug has caused a problem, a bureaucrat in FDA who is there because they don’t like meeting patients, has never treated the condition you have and never used the drug you are on, would be better placed than a good team of doctor, pharmacist and informed patient perhaps in touch with other good teams, to work out what’s going on.
If people buy this, who knows; at a push they might even buy the idea the earth is flat.
Unless a patient organization gets involved in assembling real-time data and both doctors and patients combine to put risk mitigation programs in place, none of the other changes will work.
The bottom line is the average drug has at least 100 effects. Using clinical trials we have been hugely successful in hypnotizing doctors and patients to focus on one effect and to miss the other 99. This pharmagnosia is a major driver of Pharmageddon (See Marilyn’s Curse). If the climate change encroaching on healthcare is to be rolled back, we need to manage pharmagnosia.
We can still have a market solution but it needs to be a Comparative Safety market rather than a Comparative Efficacy market.