Trial by ordeal

It is no secret that the European food industry is trailing behind its rivals in terms of research and development (R&D) spend. Figures released by the Confederation of Food and Drink Industries of the EU (CIAA) show that in 2004 European R&D spend, expressed as a proportion of industry turnover, was 0.24%. This compares to 0.35% in the US, 0.40% in Australia, and a whopping 1.21% in Japan, which is famous for its forward-thinking. But, while the pressure is on to keep up with competition, question marks hanging over the credibility of industry-sponsored research could prove to be the food sector's biggest setback.

A recent report claimed that research on soft drinks funded exclusively by the industry was four to eight times more likely to have conclusions favourable to the financial interests of the sponsoring company than independent studies.

"We had hypothesised that there would be a bias, but the magnitude was surprising," says David Ludwig, associate professor of paediatrics at Children's Hospital Boston, US, who coordinated the study.

"It's widely recognised that when a drug company pays for research, the result is likely to be in favour of the company. But the impact of food industry funding on nutrition research had never before been looked at and it's an area where conflicts of interest could affect everyone."

He claims that a study can become biased in many ways, for example if it ignores a topic likely to take a divergent view from the main, positive thrust of the research. For example, it has been recognised that saturated fat raises LDL [low-density lipoprotein - 'bad' cholesterol] levels, he explains. However, saturated fat is also known to raise levels of HDL [high-density lipoprotein - 'good' cholesterol], but to a lesser degree. So a study showing that saturated fat increases HDL levels could be accurate, yet still misleading with regard to cardiovascular disease if it ignores the effects on LDL levels, he says.

There are also more subtle types of bias. Researchers, for example, may not publish less favourable studies because they fear they will not receive further funding from food companies, claims Ludwig. "It can also become an arms race: if a manufacturer's competition is sponsoring research to come up with a favourable outcome, they too may have to do so to remain competitive."

Biased research ultimately hurts everyone: public health directly, but also the industry when questionable practices are revealed, he says. "Look at the tobacco industry - it has lost a lot of credibility by misleading people."

There are increasingly questionable relationships between food companies, academia and professional societies, which may erode the objectivity of research, he claims.

"The system will only work when each group fully understands its role. There is no issue with companies spending any amount they wish on research and development - it is when they try to publish these studies as objective that problems occur. Food manufacturers can give money to an academic centre with no strings attached, but some companies are not doing it for the sake of science, but for marketing."

It's ultimately the responsibility of governments to sponsor research into the relationship between nutrition and health and it's up to the food industry to translate this information to benefit public health, he claims. Difficulties occur when the system is reversed, he adds.

Ingredients manufacturer Orafti disagrees that processors should distance themselves from R&D. "It's not realistic for governments to sponsor all research," says marketing and communications manager Tim Van der Schraelen.

"Very often [government] projects are long term and broadly focused and we may have some nice results in a particular area. We would then fund a more specific project."

There are also circumstances where a company wishes to conduct detailed research on its own ingredients, believing that they may have additional benefits to those already proven. "Beneo Oligofructose, which can be used for fibre claims and prebiotic claims, has been shown to have an effect on diet, but we want to do a few studies before we make hard claims that the ingredient can suppress appetite," says Van der Schraelen.

"We don't do a lot of research in house. We try to shop around to find the best people," he says. Orafti has worked with Professor Abrams, of Baylor College of Medicine and the Texas Children's Hospital in Houston, US, because he is an expert in calcium and professor Glenn Gibson, an authority on prebiotics at Reading University (UK).

"We work together with a specialist - these people cannot permit themselves to be unscientific in their work," he says. "If you want to be in business long term then you have to be honest. It's frustrating when you see claims on products without science as it lets the whole industry down."

But some experts claim industry-sponsored research will be questioned regardless of whether the scientists involved have a good reputation. "Once industry supports research, it becomes tainted and people treat it as less reliable, so it's a dilemma," says Philip Calder, professor of nutritional immunology at the University of Southampton. "The argument from the government is: why should we support these things for companies to make money? But the government has a responsibility to ensure food is safe," he claims.

Ingredients supplier DKSH is well aware of the controversy surrounding industry-funded research.

"We get approached by a lot of universities asking for samples of dairy peptides, but we don't instigate research because it costs a lot of money to carry out and because of the credibility issue," says Kavan Ranasinghe, business line manager for food and ingredients. "I think you have to let the research organisation be a third party and let them have their freedom."

Julian Mellentin, executive director at the centre for Food and Health Studies, accepts that industry-sponsored research can "err towards the positive". But he claims that academics also have their own agenda: "Last year there was garbage research published on omega-3."

He refers to a study conducted by Lee Hooper, a lecturer at the school of medicine, health policy and practice at the University of East Anglia in Norwich, that claimed omega-3 may not benefit heart health. The research was later heavily criticised.

He claims that manufacturers' research is, in the main, extremely trustworthy because they have so much at stake. "There are a lot of decent people in R&D and they have their reputations to protect," he adds.

In some circumstances, published papers sponsored by industry have raised questions, admits David Richardson, a specialist consultant in nutrition and food science. "But for the most part, when studies are submitted to peer reviewed journals it's a case of trying to determine the evidence quality - were the objectives of the study properly defined? It's the same with all papers, not just those that are commercially sponsored."

In many cases you have to analyse the paper in more detail, he claims. "Other abstracts and conclusions of the paper may not reflect what the author would like you to see and, in some cases, you have to interrogate the data."

He says that food industry research is as reliable as that of other industries and he isn't surprised that much of it obtains positive results. "You have to be pragmatic if you're in industry - you're not going to waste £1M, so you're going to design an experiment where you have a reasonable chance, rather than doing something for blue-sky reasons."

He adds: "If it was all down to government research it would be a disaster - it is too politically correct and it has lost its way. To claim that research isn't credible just because it is commercially sponsored is ludicrous."