To understand what this result means, one needs to read the whole review carefully. It turns out that the average age of the participants was 57 years and that they weren’t that healthy to begin with. Some trials only recruited patients with diabetes, hypertension or increased lipids, and some included in addition some patients with previous cardiovascular disease. Further, the rate of smokers ranged from 10% to 44% in the trials that provided such data. One also needs to know after how long the benefit was obtained, and most trials ran for several years. Finally, what I always look for is whether the trials were funded by industry or by public funds, as many industry trials never get published if the results are disappointing. Only one of the trials that provided data on all-cause mortality was publicly funded. It seems to me, which the authors of the review confirmed in the Discussion section, that the 16% reduction in all-cause mortality is much exaggerated. For example, a large, publicly funded trial, the ALLHAT-LLT trial, which was not included in the review because more than 10% of the patients had pre-existing cardiovascular disease, didn’t find a reduction in mortality, risk ratio 0.99 (95% confidence interval 0.89 to 1.11, which means that we are 95% certain that the true effect lies somewhere between an 11% reduction and an 11% increase in total mortality).
The authors advised caution in using statins for primary prevention arguing that some trials were stopped early when the benefit was large, and that selective reporting of outcomes was common. They further noted that many trials didn’t report any adverse events, although it’s unlikely there weren’t any. Unfortunately, the abstract of the review, which is the only part most people read, gives a different impression. It notes that there was a reduction in all-cause mortality and that there was no clear evidence of any significant harm caused by statin prescription or of effects on the quality of life, and that there was no excess of muscle pain.
This information is not reliable. Statins cause muscle pain and weakness, and I shall again draw on my experience from the golf course. One of my partners, a physicist scientist, told me that he needed to take a statin for the rest of his life because he had had a heart attack. It bothered him a great deal and his muscle pain made it difficult for him to walk 18 holes. He also remarked that everybody else he knew who were on a statin suffered from muscle pain or weakness, or both. He had looked in the research literature and was perplexed to find that few people in the trials had reported muscle pain. At this point, I revealed that I am a medical researcher, and he asked why there was such a huge discrepancy between what the patients experienced and what the literature said. I explained how tremendously the drug industry manipulates their trials, particularly when it comes to the harms of their drugs. He wasn’t the least surprised.
Actually, my golf partner’s experience was more truthful than the randomised trials. In 2012, I found a paper about the impact of statins on energy and exertional fatigue.13 It said that, although many observational reports had cited fatigue and exertional fatigue with statin use, no randomised trials had addressed this issue. The paper reported the results of such a trial that found that 20% of the men and 40% of the women experienced a worsening in either energy or exertional fatigue. I have never heard any of my enthusiastic colleagues who advocate that most of us should take a statin for the rest of our lives, no matter what our cholesterol is, say anything about this. In fact, their arguments for irrigating the population with statins are that they work and have no side effects.
For many drugs, it’s relatively easy to overcome the fundamental problem with breaking of the blind through side effects by using so called ‘active placebos’. The term is somewhat misleading, as the idea is not that the placebo should contain a substance that is active against the disease, only a substance that gives a similar side effect as the active drug. For antidepressants, trials have been performed where the placebo contained atropine, which causes dryness in the mouth like the active drugs. As expected, such trials showed a considerably smaller difference between drug and placebo than trials that didn’t use an ‘active placebo’.14
The bias introduced by insufficient blinding can be aggravated by the fact that doctors and patients don’t always do what is expected of them. Psychiatrists are usually paid per patient enrolled and may not bother to go through all the items on Hamilton’s depression scale with the patient, as it takes time, but may use their overall impression to score some of the items without having asked, or to score later based on memory.9
Some patients participate in depression trials without being depressed just to cash the money, as a healthy person told a doctor on a train ride:15
‘I’m not depressed … the trials are advertised, the best pay about £100 a day to volunteers. For a 20 day trial that’s £2000 … it’s nice to see your regular friends.’
The atropine trials were performed a long time ago, and ‘active placebos’ are no longer used. The reason for this is clear. By far most placebo-controlled trials are performed by drug companies and they have no interest in showing that their drugs don’t work. I believe we should require active placebos and flatly refuse to approve drugs based on trials with conventional placebos, at least in areas where the expected effects are modest and the outcomes are subjective.
The companies go further than this. They often refuse to provide inactive placebos to independent researchers who wish to do their own studies.16 When Novo Nordisk did this, the researcher had no other option than do to the study without placebo, which was criticised as a great weakness when the study was published. In another case, Novo required that the authors dropped their idea of studying whether liraglutide (Victoza, a diabetes drug) reduced overweight, and the company also required changes to the part of their study that concerned a possible beneficial effect on psoriasis. It might have played a role that Novo was trying to get Victoza approved for treatment of overweight, and if independent researchers found other results, or more harms, than Novo reported, it wouldn’t be to Novo’s advantage.
Drug companies may try to avoid to be seen as uncooperative by demanding ludicrous sums for the placebos, although the cost for producing them is close to zero, knowing that academic researchers would not be supported by a public funder for such excesses. On one occasion, the largest drug company in the world said that the placebos would cost about €40 000, which was enough to block an otherwise well-motivated trial.
Please consider this: doctors and patients help the companies with their trials but companies won’t help doctors and patients with their trials. This asymmetry is immoral, just as it was immoral when the imperial powers exploited the colonies. We should make it obligatory for companies to deliver placebos for independent research at low cost, i.e. the manufacturing cost, as a condition for having a product on the market.
Drug companies may abort important studies that threaten their income in other ways. Ciprofloxacin is an antibiotic that is prone to develop resistance. In 2000, when a bacteriologist asked Bayer for a supply of pure ciprofloxacin for his research into antibiotic resistance, he was asked to sign a document stating that he would not publish without written permission from Bayer. He wrote to the European Commission but was told that the only thing the Commission could do was to remind companies of ‘the potential public interest of this type of research’.17 Again, we should not accept this state of affairs but make it obligatory for companies to deliver the pure drug to independent research at the manufacturing cost. I have heard many stories of blank refusals to give away or sell a pure drug sample.
References
1 Arroll B, Elley CR, Fishman T, et al. Antidepressants versus placebo for depression in primary care. Cochrane Database Syst Rev. 2009; 3: CD007954.
2 Hróbjartsson A, Gøtzsche PC. Is the placebo powerless? An analysis of clinical trials comparing placebo with no treatment. N Engl J Med. 2001; 344: 1594–602.
3 Hróbjartsson A, Gøtzsche PC. Placebo interventions for all clinical conditions. Cochrane Database Syst Rev. 2010; 1: CD003974.
4 Blumenthal DS, Burke
R, Shapiro AK. The validity of ‘identical matching placebos’. Arch Gen Psychiatry. 1974; 31: 214–15.
5 Gøtzsche PC. Believability of relative risks and odds ratios in abstracts: cross-sectional study. BMJ. 2006; 333: 231–4.
6 Hróbjartsson A, Thomsen AS, Emanuelsson F, et al. Observer bias in randomised clinical trials with binary outcomes: systematic review of trials with both blinded and non-blinded outcome assessors. BMJ. 2012; 344: e1119.
7 Angell M. The Truth about the Drug Companies: how they deceive us and what to do about it. New York: Random House; 2004.
8 Moynihan R, Cassels A. Selling Sickness: how the world’s biggest pharmaceutical companies are turning us all into patients. New York: Nation Books; 2005.
9 Healy D. Let Them Eat Prozac. New York: New York University Press; 2004.
10 Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet. 1994; 344: 1383–9.
11 Stead LF, Bergson G, Lancaster T. Physician advice for smoking cessation. Cochrane Database Syst Rev. 2008; 2: CD000165.
12 Taylor F, Ward K, Moore THM, et al. Statins for the primary prevention of cardiovascular disease. Cochrane Database Syst Rev. 2011; 1: CD004816.
13 Golomb BA, Evans MA, Dimsdale JE, et al. Effects of statins on energy and fatigue with exertion: results from a randomized controlled trial. Arch Intern Med. 2012; 172: 1180–2.
14 Moncrieff J, Wessely S, Hardy R. Active placebos versus antidepressants for depression. Cochrane Database Syst Rev. 2004; 1: CD003012.
15 Boyd R. A view from the man in the seat opposite. BMJ. 1998; 317: 410.
16 Villesen K, Rottbøll E. [Drug industry blocks free research]. Information. 2012 Feb 3.
17 The tightening grip of big pharma. Lancet. 2001; 357: 1141.
5
Clinical trials, a broken social contract with patients
If clinical trials become a commercial venture in which self-interest overrules public interest and desire overrules science, then the social contract which allows research on human subjects in return for medical advances is broke.
Jonathan Quick, WHO, director of essential drugs and medicines policy1
The social contract between researchers and patients was broken long before the WHO director warned against this in 2002. Epidemiologist Jan Vandenbroucke has described why industry-sponsored drug trials are not research but marketing:2
In usual clinical or epidemiologic research, studies are repeated by others, in different settings and by different means, looking for biases, flaws, and ways of remedying them, endlessly arguing whether the biases are remedied or not. That is the essence of open scientific debate and criticism, which is the only guarantee for progress. That is no longer possible with pharmaceutical products because the monopoly of the pharmaceutical industry of studies of its own products leads to persistently one-sided studies that can no longer be questioned by studies from other sides. Moreover, the one-sidedness cannot be seen from the public record, that is the published papers. Without the possibility of open debate, science simply ceases to exist … all data submitted to drug regulatory authorities should become public because these data are different from the published papers. Even better would be independent funds for clinical research.
Science philosopher Karl Popper would have come to the same conclusion.3 In The Open Society and Its Enemies, he depicts the totalitarian, closed society as a rigidly ordered state in which freedom of expression and discussion of crucial issues are ruthlessly suppressed.
Most of the time, when I have tried to publish unwelcome truths about the drug industry, I have been exposed to the journal’s lawyers, and even after I have documented that everything I say is correct and have been said before by others, I have often experienced that important bits have been removed or that my paper was rejected for no other reason than fear of litigation. This is one of the reasons I decided to write this book, as I have discovered that I have much more freedom when I write books.
Popper would have viewed the pharmaceutical industry as an enemy of the open society.3 Rigorous science should put itself at risk of being falsified and this practice should be protected against those who try to impede scientific understanding, as when the industry intimidates those who discover harms of its drugs (see Chapter 19). Protecting the hypotheses by ad hoc modifications, such as undeclared changes to the measured outcomes or the analysis plan once the sponsor has seen the results, or by designing trials that make them immune to refutation, puts the hypotheses in the same category as pseudoscience.3
In healthcare, the open democratic society has become an oligarchy of corporations whose interests serve the profit motive of the industry and shape public policy, including that of weakened regulatory agencies. Our governments have failed to regulate an industry, which has become more and more powerful and almighty, and failed to protect scientific objectivity and academic curiosity from commercial forces.
In the first half of the 20th century, drugs were very poorly researched before they were allowed onto the market. There weren’t any demands that they should have been demonstrated to have a therapeutic or prophylactic effect. What was most important was that they were not unduly harmful and not even that was adequately investigated. As a result, a number of drug catastrophes emerged and many dangerous drugs were withdrawn from the market after having harmed or killed many people.
The thalidomide disaster marked a watershed in drug regulation. The drug, which was produced by the German drug maker Grünenthal, was marketed for a broad range of indications including pregnancy-induced nausea, although it hadn’t been adequately tested in pregnant animals.4 Before long, the first reports of children being born with an extremely rare condition, phocomelia, which means lack of arms or legs, were submitted to Grünenthal. The company ignored the reports and didn’t take any action, although the reports continued to flow in. It was a classical case of profits before patients. It didn’t matter how seriously malformed the children were and how many they were, as long as the company managed to keep the reports secret.
In the United States, an astute FDA scientist had concerns about the drug and refused to recommend approval. Because of her well-placed stubbornness, the drug never made it to the US market, but its citizens weren’t totally spared, as the company had distributed samples of the drug all over the country even though it had not been approved. Thalidomide was withdrawn worldwide in 1962 and the disaster led to demands of extensive animal experiments and also that new drugs needed to have demonstrated their efficacy in randomised trials. These requirements had a major impact on the efficacy and safety of new drugs. Patients could now be more confident that the drugs their doctor prescribed were good for them. However, there was a huge backlog of drugs that had not been adequately tested and were still widely used. It took decades before most of these drugs disappeared and some of them are still with us today, although we don’t know whether they are effective and what their harms are.
Armed with its new powers, the FDA actually did nothing that could raise eyebrows in the drug industry. It came up with a new categorisation of drugs and required manufacturers to list in small print in their promotional materials: ‘The Food and Drug Administration has determined that this product is “possibly effective”.’ It would surely have been more honest to say that the old products were likely to be ineffective than to pull the wool over the public’s eyes. Drug epidemiologist Jerry Avorn has explained what this really meant:5
There is not a shred of solid evidence on the entire planet Earth that this drug is of any use whatsoever for any purpose known to man or beast, but the manufacturer has successfully demanded additional years to study it, and we don’t have the political clout to take it off the market until that unbearably lengthy foot-dragging process has run its course.
The main purpose of requiring randomised trials was to ensure that useless drugs didn’t make it onto the market. However, there was a problem with the regula
tory requirements, which is still with us today, 50 years later. All that is required to demonstrate that a drug is effective is that a statistically significant effect has been found in two placebo-controlled trials. As I showed in the previous chapter, this can often be accomplished for drugs that have no beneficial effects.
The drug companies give the impression that they play by the rules, seemingly following Good Clinical Practice guidelines and other requirements for randomised trials, e.g. using adequate randomisation procedures and blinding, and monitoring the trial sites to ensure that what is reported to the company is correct.
However, there are numerous ways in which a drug company can manipulate its clinical trials to ensure that the results become useful for its salespeople, no matter what an honest approach to science would have shown. The manipulations are so common and serious that one of my colleagues said that we should see published reports of industry trials as nothing else than advertisements for its drugs. To which I dryly remarked that industry trials do not even live up to EU requirements for advertising:6
‘No person shall issue an advertisement relating to a relevant medicinal product unless that advertisement encourages the rational use of that product by presenting it objectively and without exaggerating its properties.’
It is not surprising that the drug industry manipulates its results. The difference between an honest and a not-so-honest data analysis can be worth billions of Euros on the world market (see the CLASS study in Chapter 14). It is therefore naïve to expect the industry will perform disinterested research on its own products, with the aim of finding out whether its new drug is any better than placebo or much cheaper alternatives. If the industry truly had this aim, it would put its drugs at risk by comparing them with ‘active placebos’ and it would let independent researchers perform its trials.
Deadly Medicines and Organised Crime Page 10