Version vom 4. April 2010, 14:25 Uhr

The influence of the pharmaceutical industry in medicine

George A Jelinek and Sandra L Neate*

Abstract

Pharmaceutical companies are known to be amongst the most profitable companies in the world. Proceedings of legal cases and published research provide insights into the nature of the influence of drug companies into research and publication practices relating to the drugs they manufacture, marketing disguised as ‘education’ and their influence on doctors who prescribe their drugs. The influence of drug companies extends further to sponsorship of opinion leaders promoting their drugs and groups producing clinical guidelines. More rigorous regulation of the relationship between the industry and medicine is required.

INTRODUCTION

There is widespread concern within the medical profession and the community about the conflict of interest that exists between for-profit drug companies -- with their obligations to maximise sales of their products for the benefit of shareholders -- and members of the medical profession who recommend these products to patients. This article explores the influence of the pharmaceutical industry on medical research and practice. It seeks to discover from the medical literature the extent of such conflict of interest, and the effects of that conflict on the conduct and outcomes of research and its publication, and hence clinical practice.

THE PHARMACEUTICAL INDUSTRY

The pharmaceutical industry is highly profitable. Some pharmaceutical companies are among the world's largest and most profitable companies. Marcia Angell, former Editor-in-Chief of the New England Journal of Medicine, has described the industry as being "awash with money" and as consistently being the most profitable industry since the 1980s. Her book, The Truth About the Drug Companies,1 noted that in the 2002 Fortune 500, which outlined the takings of the top 500 companies in the United States, the combined profits for the top 10 drug companies (US$35.9 billion) outweighed the combined profits of the other 490 companies (US$33.7 billion). Further, she found that the drug companies' profits comprised 17% of sales, compared with 4.6% for the other companies. These are highly profitable, powerful multinational companies, and it is worth bearing this in mind when considering their influence in medicine.

The pharmaceutical companies state that such profits are necessary, as much of the profit in the pharmaceutical industry is ploughed back into research and development because drug development is so costly. Angell laid this claim to rest, noting that the marketing budgets of these corporations far outweigh their research and development budgets, by a factor of around three-to-one. Further, the salaries of drug company executives are enormous, and disturbingly, a considerable amount of money is spent on "educating" doctors. The education of doctors by the companies that sell the products they are educating doctors about creates the potential for profound conflicts of interest.

INFLUENCE OF THE INDUSTRY ON RESEARCH FINDINGS

Drug company-sponsored research is known to be inferior to independent research in a number of ways. The association between positive findings in randomised trials and industry funding appears to have first been documented in 1986. [2] It has since been shown that industry-sponsored research is significantly more likely to reveal positive findings in cancer [3] and orthopaedics [4] journals and in randomised controlled trials published in the British Medical Journal [5] and five general interest medical journals. [6] Systematic reviews have found that drug company-sponsored research is around 3.67 to four [8] times as likely to be favourable to a company's product as independently funded research. Research has concluded that such trials may be more positive due to the biased interpretation of results. [9] Further, authors of company-sponsored research are more than five times as likely as independent authors to recommend the company's drug. [10] Researchers with industry connections have been shown to be far more likely to publish papers favourable to company products than those without such connections. In the case of calcium channel blockers, eg, 96% of authors of papers favourable to a particular drug had financial ties to the company that made the drug compared to 37% of those whose papers had been critical. [11]

INFLUENCE OF THE INDUSTRY ON RESEARCH CONDUCT, REPORTING AND DRUG APPROVAL

It is likely that the drug industry, through its control of much of the research into pharmaceuticals, is able to persuade regulatory authorities such as the Food and Drug Administration (FDA) in the United States and the Therapeutic Goods Administration (TGA) in Australia that drugs should be licensed when the evidence supporting their licensing is, in reality, less than compelling. In approving a drug for clinical use, the FDA accepts the results of clinical trials run by the drug company which produces the drug. These studies do not necessarily have to have been published in peer-reviewed journals.

In the case of antidepressants, researchers examined the 42 FDA reviews for the six most widely used antidepressants approved from 1987 to 1999.12 Interestingly, most of the 42 trials lasted six weeks, despite the fact that these agents are generally promoted and prescribed for long-term use. The studies showed that in the randomised controlled trials, placebos were in fact 80% as effective as the drugs. The difference in drugs versus placebos averaged to two points on a 62-point scale of mood. This was a statistically significant difference; whether it was clinically significant is a different matter. The researchers found that the effect was really only of clinical significance for those who were very severely depressed.13 If there was, in fact, so little additional benefit of the antidepressants over placebos, why are so many patients, most of them not severely depressed, prescribed them?

While the FDA assesses all these studies in making its determination of whether to license the drugs, many of the studies are not subsequently published. In fact, a substantial bias has been shown to exist in what is published and what is not. Selective reporting may mean that clinicians have an unrealistic view of the efficacy of such drugs. To take the antidepressants again, researchers showed that of 74 studies registered with the FDA to license antidepressants, 38 had positive results; of these, 37 were published.14 Thirty-six had negative or questionable results; of these, 22 were not published, 11 were published in a way that made them appear positive, and only three were published accurately according to the researchers. Overall, the published literature suggested that 94% of studies were positive, whereas FDA literature showed that 51% were positive. Meta-analysis of published literature versus FDA literature showed a one-third greater effect size in published data. Selective reporting can clearly inflate our estimate of the benefit of these drugs.

The drug industry also has a significant influence on research conduct and reporting. In a national Australian survey of 823 medical specialists, 12.3% reported that industry staff wrote first drafts of their papers for publication.15 Further, 6.7% reported delayed publication, 5.1% non-publication of key negative findings, and 2.2% concealment of results. Overall, 71 respondents (8.6%) experienced at least one event that could represent a breach of research integrity.

While doctors who may be sceptical about drug company marketing usually trust the peer-reviewed medical literature,16 this trust may sometimes be misplaced, because of drug company manipulation of publishing practices. Research reporting has been profoundly affected by drug company sponsorship in several well-publicised cases. In the case of Merck's Vioxx (rofecoxib), researchers have shown that many articles written in-house by the company's own researchers or by commercial medical writing firms were published with "guest authors" who were academic investigators recruited by the company.17 Further, it has been reported that the company obscured the mortality risks associated with the drug in published trials despite being aware of them from in-house reports.18

In Australian legal proceedings it has been revealed that Merck funded leading medical publisher Elsevier to produce a publication that appeared to be, but was not, a peer-reviewed medical journal (the Australasian Journal of Bone and Joint Medicine) promoting the benefits of Vioxx,19 further eroding the ability of clinicians to differentiate marketing from science.

THE RELATIONSHIP BETWEEN DRUG COMPANIES AND DOCTORS

Drug companies reward doctors who undertake research for them. In the Australian survey of 823 medical specialists previously discussed, 41% of these doctors reported involvement in industrysponsored research in the previous year.20 These doctors were around three and a half times more likely to have been offered industry-sponsored items over $A500, and nearly five and a half times more likely than others to have been offered support for attending international conferences. For doctors who were members of advisory boards, this latter figure rose to seven fold, and nine fold for paid consultants to industry.

A variety of inducements is offered to doctors. The same Australian survey showed that:

• 96% had been offered food;
• 94% had been offered items for the office;
• 75 to 84% had been invited to product launches, symposia or "educational events";
• 52% had received offers of travel to conferences; and
• 50% had been offered personal gifts, journals or textbooks.21

The authors reported that 66 to 79% of offers were accepted.

These problems are widespread. A United States study reported that 94% of physicians reported a relationship with the industry, with 83% accepting food and 78% free samples.22 Further, 35% had received money as reimbursements, with 28% receiving direct payments for lectures, etc. The survey demonstrated that marketing was focused on opinion leaders likely to influence prescribing by other doctors. Cardiologists were twice as likely to receive inducements as other physicians.

This conflict of interest is widespread throughout the profession. An example of the extent was reported in a study examining the drug company associations of the committee members of the Committee on Safety of Medicines in the United Kingdom.23 This committee advises the United Kingdom regulatory agency on new drug approvals. Of 29 committee members, 23 had financial conflicts of interest. Thirteen of the members had associations with five companies, four with 10 companies, and three with 20 companies.

It may surprise that even clinical guidelines, widely thought to be objectively and independently produced, have on occasion been shown to be sponsored by the company producing a drug advocated in the guidelines,24 authors to have received speakers' fees and travel assistance from that drug company25 and dissemination of the guidelines to have been sponsored by the drug company.26

The extent of this conflict of interest can be further appreciated when one considers that the New England Journal of Medicine, widely considered to have the most stringent policy of general medical journals for restricting and declaring potential conflicts of interest of authors, was forced in 2002 to reverse a 12-year policy of precluding anyone with financial ties to industry from writing editorials or reviews, because it could not find enough authors without ties.27

THE INFLUENCE OF CONFLICT OF INTEREST ON PRESCRIBING

It may be argued that this widespread conflict of interest is not a problem if it does not actually change doctors' prescribing behaviour. If pharmaceutical industry involvement does not translate into changed prescribing practice, one wonders why drug companies would waste their resources on such activities. Social science research shows that even small insignificant gifts influence the recipient.28 While doctors generally deny that they are influenced,29 physicians attending pharmaceutical events are more likely to use the product, even without scientific evidence.30 It has been reported that promotional activities lead to increased prescribing, non-rational prescribing and acceptance of commercial rather than scientific views.31