Professional medical journals are full of advertisements for treatments. A typical issue of a journal- such as the British Medical Journal (BMJ) or the Lancet, two of the world’s most prestigious, based right here in the United Kingdom- consists of news, opinion pieces, peer-reviewed scientific research, and ads for medications and devices. Many may be surprised to learn that these journals often read more like magazines, with full-page and even centrefold adverts for the latest, greatest drugs.
It has long been known that the information in these adverts may be misleading. Drug companies often use various tricks in them, such as reporting ‘relative’ rather than ‘absolute’ effect sizes, which can make treatments seem more effective, often by an order of magnitude. While peer-reviewed research isn’t perfect- one only needs to read recent articles about how pharmaceutical companies continue to be allowed to withhold data, thereby making drugs seem better than they are- it is at least subject to some degree of independent review, usually by two or more scientists. The quality of information available to doctors and policymakers is an important issue in the UK right now, given the scathing conclusions of the recent House of Commons public accounts committee report on access to clinical trial information. For example, it has become increasingly clear that the decision to spend £424 million on Tamiflu, an antiviral medicine used for influenza treatment, was based on misleading and biased information, and that the drug may not be effective.
Despite the increasing popularity of online journal content, print medical journal readership continues to rise, and they remain an important source of information for practitioners. One way of improving the quality of information presented in them might be to eliminate medical journal adverts entirely. But publishers say adverts are a vital source of revenue, and without them, journals would cease to be financially viable.
To investigate whether this claim is true, our international research team from several universities in the UK, Canada, and the United States (University of Oxford, the University of Toronto, York University and other institutions) went through print editions of six of the world’s most prestigious medical journals, going back to the 1970’s. We calculated the volume of each journal taken up by adverts, opinion pieces, job listings, and peer-reviewed research. Using the price guide listed by each journal, we estimated their advertising revenue. We then calculated how much it would cost, per subscriber, to remove adverts entirely. What we found surprised us was and was completely contrary to what journal editors and publishers say in public.
Take the BMJ, for example. This journal has been at the absolute forefront of medical research, publishing groundbreaking articles debunking the connection between the MMR vaccination and autism. It was also one of the first to question the evidence base supporting the use of Tamiflu- that same flawed evidence base used by the government to spend £424 million on a drug that may not work. We determined the cost to eliminate adverts in the BMJ would be £13 per subscriber per year. Considering the salary of an NHS general practitioner is between £54,319 to £81,969 (consultants make a basic salary of between £75,249 and £101,451) this represents a tiny fraction of salary even for the lowest paid doctors in the UK. We found similar figures in other countries: it would cost each subscriber to Canada’s largest journal, the Canadian Medical Association Journal, £20 per year to eliminate adverts entirely. (In case you are wondering, the average Canadian doctor makes £194,000 per year.)
Journal publishers say without this income, they may be forced to shut their doors and stop publishing research on medical breakthroughs and new treatments. It is clear, however, that this is far from the truth. Adverts could be completely removed from print journals for less than the cost of a couple of issues of Hello magazine. I certainly think this is a price worth paying to improve the quality of medical information provided to doctors around the world.
Here’s a link to the paper, and a link to some media coverage.
Braden O'Neill 
Hi Braden,
Good research question.
But, as we’ve discussed elsewhere, the data on the BMJ are completely wrong because the journal’s UK circulation largely comprises BMA members, not subscribers, and BMJ has no control over or income from BMA subs. And you greatly underestimated the BMJ’s ad revenue because you examined only one of our print editions – our GP and CR editions have exactly the same editorial content but have different, ie additional, advertisements.
Best to disregard the data on the BMJ – and I’m saying that from a methodological viewpoint, not a defensive one!
Trish Groves
Head of research, BMJ
Thanks for an interesting post. Are you arguing against advertising per se? The science publishing aspect as I read it doesn’t appear to be central to this thesis: e.g. could equally apply to radio or roadside billboard placement.
(Disclosure: I am a journals publisher with Elsevier.)
Thanks for your interest in the paper! My perspective is that whether to remove ads is a discussion we need to have in the context of scientific publishing. We need to know how much it would cost in order to be able to have the discussion. I think adverts in medical journals are a special case, not only in relation to advertising more generally but also in relation to other scientific publishing: they provide information on the efficacy or safety of a treatment, just as the edited content often does. But whereas edited content is subject to various checks and balances, there is fairly limited regulation on what goes into the adverts.
Personally I would argue for enhanced regulation on what goes into pharma ads, rather than removing them entirely. I see so many ads that report ‘relative’ effect sizes (as opposed to absolute) for surrogate outcomes. I don’t really buy the argument that adverts are a useful source of information for clinicians: if the information is that important, it should be included in the edited content in some way, and it should be reported with absolute effect estimates on hard outcomes. A larger question, though, is what is the role of advertising in medical publishing as it transitions more and more online, as well as with a shift to open access. Revenue streams in scientific publishing are in flux at the moment, and I believe now is the time to discuss these issues.
Thanks for replying Braden. Your comments do seem to concern the legitimacy of advertising broadly: whether or not adverts appear in med journals seems to be secondary.
I’m not sure I understand the rationale behind the idea: “One way of improving the quality of information presented in them might be to eliminate medical journal adverts entirely”. How does that follow?
A converse scenario may actually be, Journal X receives $ from advertising + invests it in editorial processes, thereby improving article quality.