Sunday, 16 August 2009

Nelsons - Can they be trusted not to misrepresent research... again?

I recently noticed a job advert for a Research Associate at the University of Westminster to work on a collaborative research project with Nelsons Ltd, purveyors of natural and quack medicines. It appears Nelsons needs “expert endorsement” from a university to enhance their “scientific credentials”. A bit of digging around shows us that Nelsons have previously misrepresented scientific research to suit their marketing objectives. Might they be about to do the same thing again?

Before we look at the detail, let’s start with a little background.

Westminster are well known for their rather gullible approach to quackery, and indeed have a whole department dedicated to “Chinese Medicine and Complementary Therapies”.

Nelsons are one of the UK’s largest providers of “natural medicine”, with a turnover of £28.9m in 2007 (latest figures available) and profits of £7.6m. They sell, amongst other things, homeopathic remedies and Bach flower remedies. In 2006 they were featured on Newsnight when they tried to sell an undercover reporter a homeopathic alternative to anti-malarial drugs, stating that:

“They make it so your energy doesn’t have a malaria-shaped hole in it so the malarial mosquitos won’t come along and fill that in.”

This is a company with a record of being more interested in PR and promoting products than accurate research. Last year they were involved in developing a PR toolkit for homeopaths to “reach local people in order to grow your client base, keep in touch effectively with your clients and make use of local or regional newspapers to print positive stories about homeopathy”. This was covered by the excellent Gimpy.

Nelsons are also involved in nutritional supplements and sell Spatone, “a 100% natural liquid iron supplement” through a sister company, Spatone Ltd.

The research project they have set up, which focuses on Spatone, is done through a subsidised (in this case by £74,928) government scheme called Knowledge Transfer Partnerships (KTP). The idea behind KTP is that companies and Universities can pool their commercial and academic expertise to run a commercially focused research project to their mutual benefit, as well as to the benefit of the UK economy. Each KTP project recruits a recent graduate to conduct the bulk of the research, and this is the job I saw advertised.

This research project will focus on identifying the active compounds in Spatone and comparing these to competitor products. Now, there’s nothing wrong with that, but from looking at the job description it seems that the results of the project, before it has even taken place, have been pre-determined:

“The results will help establish Nelsons products as unique product, with significant points of difference verses other supplements in the market”

But how can they already be claiming what the results of studies on their existing products will show?

The job description also indicates that the research Nelsons have conducted into their products has been focussed on quality control, rather than effectiveness:

“Nelson’s facilities have been so far mainly limited to quality control”

For a company with a turnover of nearly £30m and profits of nearly £8m you’d think they could afford to do a little more research into the efficacy of their products before releasing them!

It seems to me that the key reason that Nelsons/Spatone are conducting this study is to make their product sound all “sciencey” for marketing and PR purposes. Their “five pillars” to support growth of the company specifically include “expert endorsement” and the outcome of the project is identified by the company as being to:

“give the brand the scientific credentials necessary for such a development…”

If I was the lead academic or the research assistant working on this project I would be concerned that rigorous scientific analysis is not what the company may be looking for. One thing that is clear is that Nelsons/Spatone can’t be trusted to accurately report research into their products. They make a very big thing on their website about the fact that Spatone has been backed up by a range of scientific studies, however reading the original research papers gives a very different picture.

I contacted Nelsons to discuss this with them, initially speaking to Chris Oldham, their R&D Manager. However, following this I was referred to a member of their communications team. They claim that the requirement for “expert endorsement” is to meet regulatory requirements when selling into new territories. I’m not convinced that this is the only reason for wanting this endorsement, as the way they have misrepresented their previous research for marketing purposes leads me to think this is as much about marketing and PR as it is about meeting regulations.

“Bod”, a blogger and contributor to the forum at Badscience, has a background in biochemistry. He examined the claims made on the Spatone website and compared these with the original research. The Spatone website was then edited on the 6th August, after I contacted them. The claims we are investigating below appear as they were originally. Here’s what Bod found:

Claim: "Spatone is effective against iron deficiency in pregnancy"

The research (1) involved a sample of only 24 anaemic pregnant and 21 non-pregnant, non-anaemic women, and therefore has little statistical power. The groups allocated to different regimes were of different sizes. The research was partially funded by Spatone. It showed Spatone to be about as effective as other forms of ferrous sulphate. Why wouldn't it be?

Claim: "Spatone does not cause the side-effects usually associated with conventional iron supplements"

The study did not prove side effects to be lower for Spatone, but rather assumed it to be the case. No-one reported the side-effects, but there is no evidence they were asked about side effects.

Claim: “Spatone has an absorption rate of iron which is up to 40% - an absorption rate unmatched by iron intake from food sources or conventional iron supplements"

The small study showed 5-40% absorption of the Spatone’s iron by anaemic pregnant women. It did not compare this with conventional supplements, but instead compared it with literature values, measured under different circumstances with different subjects. As pregnant women are known to have a far higher absorption of iron than non-pregnant women, and people with low iron stores have a far higher absorption of iron than people with high stores, all the study has done is demonstrate these two points. Comparing the results obtained with those for non-pregnant non-anaemic people is invalid.

Claim: "Spatone Iron-Plus (previous trade name of Spatone) does not cause these side-effects"

This seemingly better-designed trial (2) with 102 subjects randomised and supposedly double-blinded (hopefully the placebo was as coloured and flavoured as iron containing water would be) did not show this at all. "Dyspepsia scores did not differ between the two groups" is what it said. Once again, the researchers explain why Spatone should have a lower incidence of GI side effects without actually demonstrating it in their experiment.

It does claim that compliance scores were different between the two groups, but fails to note the lack of statistical significance of the result meant that it was probably down to chance. The conclusion was that a randomised controlled trial would probably be worthwhile. No note is made of who funded the study in the paper, but we would guess Spatone played a part.

When asked to comment about this, Nelsons stated that the dyspepsia scores did not differ between the Spatone and placebo groups and that it is on this basis that the claim of no side effects is made. However, this doesn’t tell us anything about how Spatone compares to alternative products, and the trial didn’t examine this. If Nelsons wish to show that their ferrous sulphate differs in some way from the standard kind, they needed to do an experiment to show this directly, rather than playing games with the known facts by doing half-experiments.

They compare absorption of the product by pregnant anaemic women with literature values for those who we should already know will absorb less iron, instead of comparing Spatone and standard product. They compare a low dose of Spatone with placebo, then make a false comparison with a higher dose of conventional product. If they want to know whether their product has magic powers, they need to compare the same dose of Spatone and standard product, at the same dose and concentration.

Claim: "It has... been shown that spatone has an absorption of iron of up to 40%"

This study (3) only had 13 subjects initially, of whom one was eliminated as they might have had a history of iron malabsorbtion. This already causes a problem as it is bad practice to pick and choose your subjects when your sample size is so small. The study tells us that Spatone's active ingredient is ferrous sulphate, just like the standard iron tablets. The study does NOT claim that Spatone differs in any way from any other dilute solution of ferrous sulphate. The study tells us that up to 40 % of the iron in ferrous sulphate can be absorbed under certain special conditions. The study thanks Spatone for their financial contribution.

In summary, this is poor quality research on statistically insignificant, unrepresentative and in one case cherry-picked groups which give explanations for effects not shown by the research. They are funded in at least two of the three cases by Spatone Ltd.

What this research doesn’t do is show us how Spatone differs from any of their competitor’s products. Perhaps what is needed is a clinical trial of Spatone against Irn-bru, which is also a dilute solution of iron salts. Irn-bru has around 3.5mg/l of iron as ammonium ferric citrate, a very well tolerated iron salt. Using Nelson’s level of scientific robustness we can therefore conclude that one dose of Spatone is therefore slightly inferior to one litre of Irn-bru (probably the daily consumption for some in Scotland!). One would also predict slightly better scores for GI side effects. Of course Barr's are not claiming that their product is a neutraceutical. I don't think they are even allowed to claim it is made in Scotland from girders any more.

Derek Renshaw, the KTP's lead academic at Westminster was contacted about these issues and said he would have to discuss with colleagues whether and how to respond. I have heard nothing from him since.

Previous poor quality research was spun with weasel words into what look to be claims that Spatone differs from conventional supplementation. Given the stated aims of this current research project, can we really expect anything different from them this time? Is it really an appropriate use of public money to support a multi-million profit making company’s PR and marketing work?


(1) G. HALKSWORTH, L. MOSELEY, K. CARTER, M. WORWOOD (2003) Iron absorption from Spatone (a natural mineral water) for prevention of iron deficiency in pregnancy, Clinical and Laboratory Haematology, 25, 227–231

(2) D. MCKENNA, D. SPENCE, S.E. HAGGAN, E. MCCRUM, J.C. DORNAN,T.R. LAPPIN (2003) A randomized trial investigating an iron-rich natural mineral water as a prophylaxis against iron deficiency in pregnancy, Clinical and Laboratory Haematology, 25, 99–103

(3) M.WORWOOD, W.D.EVAN, R.J.VILLIS, A.K.BURNETT (1996) Iron absorption from a natural mineral water (Spatone Iron+), Clinical and Laboratory Haematology, 18, 23-27

I'd like to thank "Bod" for his help in putting this together and analysing the previous research conducted on Spatone.