The Best Research Money Can Buy

Imagine you are a university administrator and someone approaches you with an offer of $15 million for your research budget. You would leap at it, because that amount of money can fund a lot of resources. I worked in a very sophisticated laboratory once that cost $8 million to build on an empty block of land, so even if you were to create the facilities from scratch there would still be money left over for running costs.

An offer of this amount has been made to at least two Australian universities over recent months, plus another two who are rumoured but not yet confirmed. The first three rejected the offer, which was finally accepted by La Trobe University. You might wonder why anyone would refuse this, but it was because the donation was tied to doing specific research into the products of a single company,  the vitamin manufacturer Swisse. Swisse have come in for some attention from the ACCC recently for the wording used in their advertisements, but as all distributors of vitamins at least suggest some amorphous benefits from using their products this isn't enough reason to refuse to take their money.

The matter received publicity when Dr Ken Harvey resigned from his position at La Trobe because of the Swisse donation. Dr Harvey is a well-known critic of medical quackery, and was sued by a distributor of a make-you-thin product for pointing out that there was no such thing. I respect Dr Harvey for acting according to his principles, but I will leave discussion of his motivation and his arguments for it to his (and my) friends at Friends of Science in Medicine. (You can see the FSM response here.)

Had Swisse simply turned up at the university with money there probably would not have been any outcry, but because the donation was solely to be spent on researching the efficacy of Swisse branded products three serious issues were raised. These were transparency, conflict of interest, and accuracy of current claims.

Because the donation was made publicly with the donor clearly identified it might seem that transparency is not a problem, but as the research program was to span several years the link between the donor and any published results might have been gradually forgotten. Media reports of scientific breakthroughs rarely extend to mentioning that little paragraph in italics at the bottom of papers: "This research was partly funded by a donation from …". This is not always the fault of the journalists, because they often work from media releases from research institutions without necessarily checking back to the published journal articles. The danger becomes even bigger when the media releases come from the donor company itself. I'm not suggesting deliberate deception in either case, but PR people usually aren't scientists and journalists have to work to deadlines, often having to produce a story before the work has been published in the medical journals. "Research to be published in a coming edition of the Journal of Reproducibility suggests that XXX can …".

Conflict of interest should be obvious, and with the best intentions it is always going to be assumed that money influences the results of sponsored research. It taints everything. With the amount of money involved it would be difficult for the university to do equivalent research on products from other manufacturers for comparison unless another donor could be found, and other possible benefactors might be reluctant to donate if there is a perception that the answers have already been bought. Things might even get worse, with researchers paid for by say, Blackmore's, competing with colleagues down the corridor funded by Swisse, with both groups trying to show that their funder's products are superior.

The third issue is perhaps the most important, because it is not obvious to the buying public.

There are two levels of approval in Australia for products making medical or health claims – Registered and Listed. All Swisse products sold are in the Listed class, defined by the Therapeutic Goods Administration as "Lower risk medicines containing pre-approved, low-risk ingredients and that make limited claims" and "Listed medicines are assessed by the TGA for quality and safety but not efficacy. … It is a requirement under the Act that sponsors hold information to substantiate all of their product's claims".

Note the final sentence. If Swisse already have evidence of efficacy, why are they willing to pay millions to find it again? If they don't have the evidence, how could they have honestly applied for listing of the products?

We already know that vitamins and supplements help people with diagnosed deficiencies, but might even be harmful for others. No amount of sponsorship can change this reality, but it and the halo effect of being associated with a respected university can change audience perception. And that is the real problem with the Swisse donation to La Trobe.

This article was published as the Naked Skeptic column in the April 2014 edition of Australasian Science