Via a post on Crooked Timber aptly titled „Ghostwriters of Science“ I got to read a science article in the guardian, and googling around found an actual journal publication by Richard Smith, a long-time editor and chief executive with BMJ (the British Medical Journal, one of the high-impact medical journals), which criticizes the same mix-up under the title „Medical Journals Are an Extension of the Marketing Arm of Pharmaceutical Companies“ (May 2005 Issue of PLoS Medicine).

The picture that presents itself is shocking. A majority of randomized controlled trials of drugs is funded by the respective companies by now, they adhere to high scientific standards and are published in high-ranking peer-reviewed journals, and they still manage to be deceptive — for instance, being four times more likely to yield favorable results than independently funded studies in one comparison. How?

From the journal article:

The companies seem to get the results they want not by fiddling the results, which would be far too crude and possibly detectable by peer review, but rather by asking the “right” questions—and there are many ways to do this [10]. Some of the methods for achieving favourable results are listed in the Sidebar, but there are many ways to hugely increase the chance of producing favourable results, and there are many hired guns who will think up new ways and stay one jump ahead of peer reviewers.

Then, various publishing strategies are available to ensure maximum exposure of positive results. Companies have resorted to trying to suppress negative studies [11,12], but this is a crude strategy—and one that should rarely be necessary if the company is asking the “right” questions. A much better strategy is to publish positive results more than once, often in supplements to journals, which are highly profitable to the publishers and shown to be of dubious quality [13,14]. Companies will usually conduct multicentre trials, and there is huge scope for publishing different results from different centres at different times in different journals. It’s also possible to combine the results from different centres in multiple combinations.

These strategies have been exposed in the cases of risperidone [15] and odansetron [16], but it’s a huge amount of work to discover how many trials are truly independent and how many are simply the same results being published more than once. And usually it’s impossible to tell from the published studies: it’s necessary to go back to the authors and get data on individual patients.

Here are the „sidebar“ tricks to „ask the right questions“:

• Conduct a trial of your drug against a treatment known to be inferior.
• Trial your drugs against too low a dose of a competitor drug.
• Conduct a trial of your drug against too high a dose of a competitor drug (making your drug seem less toxic).
• Conduct trials that are too small to show differences from competitor drugs.
• Use multiple endpoints in the trial and select for publication those that give favourable results.
• Do multicentre trials and select for publication results from centres that are favourable.
• Conduct subgroup analyses and select for publication those that are favourable.
• Present results that are most likely to impress—for example, reduction in relative rather than absolute risk.

All that makes the work very hard for peer-reviewers, as most of these manipulations will not be visible in the manuscript and, as mentioned, could only be revealed through additional data provided from the authors…

The editors‘ and publishers‘ own financial interest play an important role, too, of course:

Publishers know that pharmaceutical companies will often purchase thousands of dollars‘ worth of reprints, and the profit margin on reprints is likely to be 70%. Editors, too, know that publishing such studies is highly profitable, and editors are increasingly responsible for the budgets of their journals and for producing a profit for the owners. Many owners—including academic societies—depend on profits from their journals. An editor may thus face a frighteningly stark conflict of interest: publish a trial that will bring US$100 000 of profit or meet the end-of-year budget by firing an editor.

The majority of high-ranking articles are by now payed for by the industry:

[…] between two-thirds and three-quarters of the trials published in the major journals—Annals of Internal Medicine, JAMA, Lancet, and New England Journal of Medicine—are funded by the industry [9].

And the interest of the pharmaceutical companies is easily explained:

A large trial published in a major journal has the journal’s stamp of approval (unlike the advertising), will be distributed around the world, and may well receive global media coverage, particularly if promoted simultaneously by press releases from both the journal and the expensive public-relations firm hired by the pharmaceutical company that sponsored the trial. For a drug company, a favourable trial is worth thousands of pages of advertising, which is why a company will sometimes spend upwards of a million dollars on reprints of the trial for worldwide distribution. The doctors receiving the reprints may not read them, but they will be impressed by the name of the journal from which they come. The quality of the journal will bless the quality of the drug.

For how this is realized in practice, I highly recommend the Guardian article, but here is a decisive part about this new industry:

Drug companies exert this hold on knowledge through publication planning agencies, an obscure subsection of the pharmaceutical industry that has ballooned in size in recent years, and is now a key lever in the commercial machinery that gets drugs sold.

The planning companies are paid to implement high-impact publication strategies for specific drugs. They target the most influential academics to act as authors, draft the articles, and ensure that these include clearly-defined branding messages and appear in the most prestigious journals.

[…]

There are now at least 250 different companies engaged in the business of planning clinical publications for the pharmaceutical industry, according to the International Society for Medical Publication Professionals, which said it has over 1000 individual members.

Now, the most interesting aspect of the PLoS article is maybe its recommendations for solving that problem, particularly this radical idea:

Secondly, journals should perhaps stop publishing trials. Instead, the protocols and results should be made available on regulated Web sites. Only such a radical step, I think, will stop journals from being beholden to companies. Instead of publishing trials, journals could concentrate on critically describing them.

That would be pretty interesting work for a researcher! But he or she would still need somebody to pay for that, and it would certainly not be the industry.