When researchers receive money from pharmaceutical companies, their trials are more likely to produce favorable results, a review of almost 200 published studies has found. Although the influence of money on trials has long been suspected, quantitative evidence has been scarce. The review also has a lesson for cynics who think corporate involvement means all drug trials are suspect, as many studies were also apparently conducted without bias.
To see how much difference funding makes, Oregon Health and Science University student Rosa Ahn picked a random sample of 195 drug trials whose results were published in 2013, and investigated the independence of the researchers involved. Attempts to examine industry influence have been done before, but most have not differentiated between direct funding for a study and other financial connections to researchers. Ahn looked at each independently.
Two-thirds of trials had one or more principal investigators who had received money from the pharmaceutical industry, although this could be as small as getting travel expenses paid or as large as owning stock in the company whose drug was being tested. Five percent held a patent over the drug they were investigating.
More than three-quarters of trials with positive results had a principal investigator with financial connections to the industry, while just 49 percent of those that produced negative results had such conflicts of interest.
Ahn's findings have been published in the British Medical Journal. The paper notes that there are other possible explanations for the raw data besides researcher bias. For example, it's possible that financial connections are more common at a particular study phase, which might in turn be one where positive outcomes are more common.
However, Ahn and her co-authors controlled for a range of factors they thought might have influenced the data and the results barely budged – the presence of a financial tie halved the chance of a negative result being reported.
The situation could even be worse than Ahn's results suggest, since not all financial influence is disclosed. The paper's authors made efforts to uncover undisclosed financial connections and found 34 examples where a principal investigator failed to disclose money they had received, but it is likely additional cases were missed.
Intriguingly, while the investigators' financial ties made a difference, actual funding from industry didn't, at least in this sample. Trials that were publicly funded were as likely to produce positive results as those where a company put up the cash.
In 2002, 15.5 percent of scientists involved in drug trials admitted in a survey to changing the design, or even the results, of their study to meet pressure from their paymasters.
In a linked editorial, Dr Andreas Lundh of the University of Southern Denmark suggests that insufficient blinding, where participants have a good idea of who is on the placebo and who is taking the trial drug, could be distorting results. “Perhaps authors with financial ties are more willing to accept such practices because of commercial pressure,” Lundh speculates. “However, the empirical evidence for this explanation is lacking.”
Lundh proposes that peer-reviewed journals should require increased transparency about scientists' financial ties before publication, something some journals already do.