Deadly Medicines and Organised Crime
Page 16
A rare admission that doctors’ opinions are for sale to the highest bidder was provided by Canadian rheumatologist Peter Tugwell, who wrote a letter to several major companies soliciting funds for CME conferences on behalf of an organisation called OMERACT:43
We think that support for such a meeting would be very profitable for a company with a worldwide interest in drugs targeted in these fields. The impact of sponsorship will be high as the individuals invited for this workshop, being opinion leaders in their field, are influential with the regulatory agencies. Currently, we are seeking major sponsors to pledge support of U.S. $5,000 and $10,000. These major sponsors will be given the opportunity to nominate participants to represent industry’s interest and to participate actively in the conference.
CME ought to be the ultimate test of medicine’s professionalism.43 What could be more central than physicians educating other physicians to improve the quality of care? Yet doctors hope to obtain something of value without paying the full price for it, and a variety of commercial predators take full advantage of these hopes to line their own pockets.43
The drug industry supplies one-third of the American Psychiatric Association’s budget, and when interviewed, a spokesperson said that without this funding, instead of meeting at the Philadelphia Convention Center, the members would be sitting in the basement at the YMCA.43 The reporter cleverly asked: ‘And what, exactly, is wrong with meeting in the basement at the YMCA?’ Apart from this, the psychiatrists are rich enough to pay for themselves.
What has received little attention so far is the fact that by buying most knowledgeable experts in the field, the drug industry also corrupts the peer review system. Journal editors look to experts to tell them whether a submitted research study has been done well, and experts on industry payroll may tell them it is, even when that’s not the case. Many experts have shares in companies and know perfectly well what it means for the company to have a trial published in one of our most prestigious journals.
Industry liaisons may also mean that doctors fail to notify the regulators when a drug-related death is suspected, e.g. if they ask the company before they submit the report. Some doctors prefer to send their reports to the companies instead of to the regulators because of their close relations with them, and the FDA and the EMA have found many cases where companies failed to sent them on even when the patients had died.44,45
References
1 Danish National Board of Health. [List of permissions for physicians and dentists]. Available online at: http://ext.laegemiddelstyrelsen.dk/tilladelselaegertandlaeger/tilladelse_laeger_tandlaeger_full_soeg.asp?vis=hele (accessed November 2010).
2 Gagnon M-A. Corporate influence over clinical research: considering the alternatives. Rev Prescrire. 2012; 32: 311–14.
3 Light DW, Lexchin JR. Pharmaceutical research and development: what do we get for all that money? BMJ. 2012; 344: e4348.
4 Derry CJ, Derry S, Moore RA. Sumatriptan (oral route of administration) for acute migraine attacks in adults. Cochrane Database Syst Rev. 2012; 2: CD008615.
5 Tfelt-Hansen PC. Unpublished clinical trials with sumatriptan. Lancet. 2009: 374: 1501–2.
6 Tfelt-Hansen P, Hauchildt Juhl H. [Treatment of migraine with triptans – a commented foreign health technology assessment]. Copenhagen: Sundhedsstyrelsen; 2008.
7 Spielmans GI, Parry PI. From evidence-based medicine to marketing-based medicine: evidence from internal industry documents. Bioethical Inquiry. 2010. DOI 10.1007/s11673-010-9208-8.
8 Grill M. Kranke Geschäfte: wie die Pharmaindustrie uns manipuliert. Hamburg: Rowohlt Verlag; 2007.
9 Sox HC, Rennie D. Seeding trials: just say ‘no’. Ann Intern Med. 2008; 149: 279–80.
10 Harris G. As doctors write prescriptions, drug company writes a check. New York Times. 2004 June 27.
11 Andersen M, Kragstrup J, Søndergaard J. How conducting a clinical trial affects physicians’ guideline adherence and drug preferences. JAMA. 2006; 295: 2759–64.
12 Psaty BM, Rennie D. Clinical trial investigators and their prescribing patterns: another dimension to the relationship between physician investigators and the pharmaceutical industry. JAMA. 2006; 295: 2787–90.
13 Nielsen HL. Linking Healthcare: an inquiry into the changing performances of web-based technology for asthma monitoring [PhD dissertation]. Copenhagen Business School, Department of Organization and Industrial Sociology; 2005.
14 Jackson T. Are you being duped? BMJ. 2001; 322: 1312.
15 Kassirer JP. On the Take: how medicine’s complicity with big business can endanger your health. Oxford: Oxford University Press; 2005.
16 Abramson J. Overdo$ed America. New York: HarperCollins; 2004.
17 Henry D, Doran E, Kerridge I, et al. Ties that bind: multiple relationships between clinical researchers and the pharmaceutical industry. Arch Intern Med. 2005; 165: 2493–6.
18 Boseley S. Junket time in Munich for the medical profession – and it’s all on the drug firms. The Guardian. 2004 Oct 5.
19 Abelson R. Whistle-blower suit says device maker generously rewards doctors. New York Times. 2006 Jan 24.
20 Thompson M, Heneghan C. BMJ open data campaign: time to move the debate forward. BMJ 2012; 345: 25.
21 Godlee F. Open letter to Roche about oseltamivir trial data. BMJ. 2012; 345: e7305.
22 Moore J. Medical device payments to doctors draw scrutiny. Star Tribune. 2008 Sept 8.
23 Lenzer J. Doctor’s group files legal charges against nine French doctors over competing interests. BMJ. 2009; 338: 1408.
24 Moynihan R. Key opinion leaders, independent experts of drug representatives in disguise? BMJ. 2008; 336: 1402–3.
25 Steinman MA, Bero LA, Chren MM, et al. Narrative review: the promotion of gabapentin: an analysis of internal industry documents. Ann Intern Med. 2006; 145: 284–93.
26 [Guidelines for the requirement of physicians and dentists to get permission to be connected to a drug company]. Copenhagen: Sundhedsstyrelsen; 2011 June 28.
27 Boseley S. Scandal of scientists who take money for papers ghostwritten by drug companies. The Guardian. 2002 Feb 7.
28 Elliott C. Pharma goes to the laundry: public relations and the business of medical education. Hastings Cent Rep. 2004; 34: 18–23.
29 Brownlee S. Overtreated: why too much medicine is making us sicker and poorer. New York: Bloomsbury; 2007.
30 Bowman MA, Pearle DL. Changes in drug prescribing patterns related to commercial company funding of continuing medical education. J Contin Educ Health Prof. 1988; 8: 13–20.
31 Bowman MA. The impact of drug company funding on the content of continuing medical education. Möbius. 1986; 6: 66–9.
32 Moynihan R. Doctors’ education: the invisible influence of drug company sponsorship. BMJ. 2008; 336: 416–17.
33 Burton B, Rowell A. Disease Mongering. SpinWatch. 2003. Available online at: www.spinwatch.org/component/content/article/47-pharma-industry/29-disease-mongering (accessed 11 November 2012).
34 Key Opinion Leaders Europe. Conference announcement. SMI. 2009 June 15–16.
35 Can I buy you a dinner? Pharmaceutical companies increasingly use doctors’ talks as sales pitches. 2005 Aug. Available online at: www.worstpills.org (accessed August 2005).
36 Wilmshurst P. Academia and industry. Lancet. 2000; 356: 338–44.
37 Carlat D. Dr drug rep. New York Times. 2007 Nov 25.
38 Fugh-Berman A, Ahari S. Following the script: how drug reps make friends and influence doctors. PLoS Med. 2007; 4: e150.
39 Blum A, Solberg E, Wolinsky H. The Surgeon General’s report on smoking and health 40 years later: still wandering in the desert. Lancet. 2004; 363: 97–8.
40 Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women’s Health Initiative randomized controlled trial. JAMA. 2002; 288: 321–33.
41 Avorn J. Powerful Medicines: the benefits, risks, and costs of prescription drugs. N
ew York: Vintage Books; 2005.
42 Clark J. A hot flush for big pharma. BMJ. 2003; 327: 400.
43 Brody H. Hooked: ethics, the medical profession, and the pharmaceutical industry. Lanham: Rowman & Littlefield; 2008.
44 Petersen M. Our Daily Meds. New York: Sarah Crichton Books; 2008.
45 Wise J. European drug agency criticises Roche for failing to report adverse reactions and patient deaths. BMJ. 2012; 344: e4344.
9
Hard sell
The drug industry is unique in that it can make exploitation appear a noble purpose.
Former medical director of Squibb at a US Senate hearing1
Clinical trials are marketing in disguise
Whatever the industry does, whatever it calls it, and whatever it says about its noble motives, it all boils down to one thing: selling drugs.
This is done very effectively by tightly controlling the flow and type of information about its drugs, both in scientific papers and in marketing. Its clinical trials are rarely research in the true sense of this word (see Chapter 5); it is marketing disguised as research. The trials are often flawed by design, additional flaws are introduced during data analysis, and the misleading results are spun to make sure that whatever an honest trial might have shown, the trial concludes something that is useful for boosting sales.2,3,4,5,6,7,8
My thesis showed that what the industry publishes just cannot be true. I identified 196 double-blind trials where a new nonsteroidal anti-inflammatory drug (NSAID) had been compared with a control NSAID in patients with rheumatoid arthritis.2 This is a highly variable disease, which makes it difficult to find differences between two similar drugs, but despite this, the trials were microscopically small, with a median sample size of only 27 patients in each group.3 One would therefore expect virtually all statistically significant differences in effect to have arisen by chance, i.e. 5% of the differences would be statistically significant, 2.5% in favour of the new drug and 2.5% in favour of the control drug.
However, 14% of the differences, or three times as many as expected, were statistically significant, and in 73 trials, all differences favoured the new drug compared with only 8 trials that favoured the control drug.3 It was rarely possible to check the statistical analyses, but I found 12 trials where the claimed statistically significant differences were not statistically significant and 5 trials where I strongly suspected it. In all 17 cases, the false significant findings favoured the new drug. The results for side effects were even more striking. In all 39 trials with a significant difference in side effects, this difference favoured the new drug.
Thus, the new NSAIDs appeared to be considerably better than the old ones. And the spin in the conclusion or abstract was even more formidable. In 81 cases, the biased conclusion favoured the new drug, and in only one case did the conclusion favour the control drug (P = 3.4 ∙ 10–23).
However, when I looked at the data (presented as means in the papers), the impressive superiority of new drugs disappeared. The most common outcome was grip strength, and there was no difference between new drugs and control drugs, on average.9
I reasoned that the most important outcome when two NSAIDs are being compared is which drug the patients prefer in crossover trials where they try both drugs in random order. The patients are surely the best judges for weighing a certain degree of pain relief against the side effects of the drugs. Most trials had used indomethacin as comparator, an old drug marketed in 1963, which, according to industry myths and flawed trials, had many side effects. However, in indomethacin crossover trials, the patients preferred indomethacin about equally often as the new NSAID (see Figure 9.1).10 The figure shows that the larger the trial, the less the variation in the difference between the two drugs. This is expected from statistical theory. When we randomise rather few patients, there will sometimes be more patients with a good prognosis in the indomethacin group than in the other group, and sometimes there will be more patients with a poor prognosis. When we randomise many patients, the groups will be very similar and the result therefore more precise than in a small trial. We expect the results from many similar trials to lie within a symmetric funnel and this is also the case, apart from two results that are so outlying that fraud must be suspected.
Figure 9.1 Difference in proportion of patients preferring a new NSAID and proportion preferring indomethacin (therapeutic gain) in 34 cross-over trials. Arrows mark two outlying, likely fraudulent, trials
Two of 32 trials is a high fraud rate (6%), but when I showed the graph to a colleague from industry, he laughed and said that everybody knew that about 5% of trials were fraudulent, i.e. more or less made up. Data fabrication was so widespread in the drug industry that there was slang for it: ‘dry labelling’ or ‘graphiting’ in the United States and ‘making’ in Japan.11
When I defended my thesis in 1990, the two examiners felt that NSAIDs were a particularly bad area because there was so much competition on the market. My results were too shocking for them to fully realise what they meant. Since then, however, we have seen similar problems in all therapeutic areas that have been thoroughly investigated.
Lipid-lowering drugs is another example of a highly competitive market. In head-to-head statin trials there is often no blinding, no concealment of treatment allocation (which means that the randomisation could have been violated), poor follow-up and no intention-to-treat analysis (where the fate of all randomised patients are accounted for, also those who drop out).12 Funding from the test drug company rather than the comparator drug company was associated with more favourable results (odds ratio 20) and more favourable conclusions (odds ratio 35). This is not surprising considering that head-to-head statin trials are not fairly designed, as the compared doses in most of the trials are not equivalent.13 Further, there are no trials of good quality that have compared different statins for clinically relevant outcomes such as coronary disease events. In contrast, no less than 29 placebo-controlled trials have reported on such events, which suggest that many of the trials were unethical, as patients on placebo were denied an effective drug.
Looking at it from a helicopter perspective, a Cochrane review that included 48 papers that in total comprised thousands of individual trials found that industry-sponsored studies more often had favourable efficacy results, favourable harms results and favourable conclusions for the drug or medical device of interest, compared with non-industry sponsored studies.14
The industry’s many tricks make the impossible possible. It is very rare that the title of a paper tells you everything you need to know, but here is an example:15
‘Why olanzapine beats risperidone, risperidone beats quetiapine, and quetiapine beats olanzapine: an exploratory analysis of head-to-head comparison studies of second-generation antipsychotics.’ In a mathematical sense, this shouldn’t be possible. If A is higher than B, and B is higher than C, then C cannot be higher than A.
Ghostwriting
The misleading information in original research papers is subsequently propagated in scores of ghostwritten reviews and other secondary articles. Ghostwriting is very harmful to public health, as it misleads doctors about the benefits and harms of drugs.16 It is also fraud, as doctors are deceived deliberately. The very purpose of not informing the readers about who wrote the paper is to make it appear as if it came from disinterested, prominent academics and not from a corporate sponsor.
Ghostwritten papers are subsequently cited in promotional materials and in other ghost-papers, as if they provide independent verification that the drug is effective and safe and better than other drugs. Thus, marketing people produce ghost-papers that are used by the same marketing people, a perfect incestuous way of fooling unsuspecting doctors into believing what they think their own leaders have written.
If deceit wasn’t intended, we would expect the company to tell us who the writer was, make it clear that the writer was paid and commissioned to do the work, and publish the paper in that writer’s name. Instead, companies go to great length
s to find academics that cover up the scam with their names and omit any mentioning of the medical writer’s contribution, even in the acknowledgement. The academics get paid for their non-work and may receive a letter offering them tens of thousands of dollars simply to add their names to a review they have never seen that praises the company’s new drug.17
Ghostwriting corrupts the trust that is so essential for scientific communication. It’s looks like a win–win situation for the doctors and the company that share an interest in not telling anyone about the arrangement, but lawsuits have made it possible for everyone to get a glimpse of the dirty business. I shall first describe how common the practice is.
A study of papers on the antidepressant drug sertraline (Zoloft, Pfizer) showed that in a 3-year period, 55 papers had been written by a medical writing agency, Current Medical Directions, whereas only 41 papers had been written by other people.18 Only two of the 55 papers acknowledged writing support from people not listed as authors, and all results were favourable for Pfizer.
In 2007, the International Society of Medical Planning Professionals included a workshop at its annual meeting where a consultant warned of the dangers of regulators seeing publication plans: ‘If they looked at a publication strategy that, I don’t know, had, “We’re going to put out 80 papers this year on one drug, all off-label. Fifty of those will be review articles where we’ll pay someone to write about off-label use …”’19