To Fear and to Fund

I write my column To Fear and to Fund in TCS Daily on 13 February 2003.

The case of the mad potato crisp disease: the making of a food scare

Most invitations to press conferences are dull. But one sent out last April by the Swedish National Food Administration hinted that a cancer-causing chemical had been found in a wide range of foods. The levels of this mystery chemical were high, and the new results were expected to affect risk evaluation, food production and eating habits internationally. Just the invitation itself made headlines. What was the threat? How many were at risk? What to do?

The press conference was crowded with journalists, covered live on television and mentioned internationally. The Food Administration revealed that it, together with researchers at Stockholm University, had found the possible carcinogen acrylamide in many common baked, roasted, and fried foods, especially those containing potatoes. It was everywhere! There was no escape! Sales of potato chips plummeted. The World Health Organization (WHO) and the Food and Agricultural Organization (FAO) of the United Nations expressed concern, and food administrations worldwide began to feverishly sample their local food for acrylamide. Journalists interviewed anxious consumers and vague, worried cancer researchers.

Things gradually cooled down to a steady beat of research groups and investigative teams expressing concern and demanding more funding. Governments began to suggest limits to acrylamide content in food, granting funding and promising speedy action. By September the origin of the acrylamide was starting to become clear (it is caused by a variant of the browning reaction that makes bread brown and much of the flavour of cooked meat). The snack industry reported it had decreased levels by 75 per cent by using less heat during baking.

Finally in January 2003: deliverance! In headlines suited for V-Day or the Apollo landing the newspapers proclaimed: "Crisps do not cause cancer!" According to the newspaper Aftonbladet they might even decrease the risk. Everybody was finally safe. The mad potato crisp disease would not kill us.

This is a story of a food scare and how it was created.

The discovery of acrylamide in food was to a large extent a serendipitous research accident that came to the attention of the Food Administration a long time before the announcement. The discovery of acrylamide in hamburgers was mentioned in the April 2001 newsletter of the Food Administration, but not widely reported. The press invitation one year later was on the other hand quite sensational, and the press conference was held long before the results were reviewed and published.

Two days after the invitation agricultural minister Margareta Winberg gave 1.5 million Swedish crowns to the Food Administration, which just happened to be in economic difficulties. In the next budget the raise was even larger. The Administration traded science and caution against money and visibility.

The Food Administration is supposed to find risks, and is rewarded when it succeeds in finding new dangers, especially widespread public dangers. But it is a practical way of getting more money for a department - announce a food scare widely. If they are right, they are heroes. If they are wrong, well, better safe than sorry.

Obviously, any possible risk is worth investigating, especially if it seems to be widespread. And keeping quiet about a risk discovery has a dubious morality and a serious danger of backfiring, as was seen in the BSE scandal in England. But the way risks are presented to the public matters. Presenting risks in a manner that is bound to worry people or lead to easy misinterpretation does very real damage - not just to sales but also to the perceptions of the public.

People can handle uncertainty, but when the information they get is biased or changes rapidly there is little to base an opinion on. The press releases played on the fear of the unknown and uncertain, helpfully suggesting solutions "just in case" that seemed to imply that one really should change, which in turn implied that there was a real threat. It is a weak chain of logic, but a very human way of reasoning.

How affected were people by the alarm? There was little doubt about the concern people expressed. But a poll just a week after the press conference revealed that two-thirds of the interviewed would not change their eating habits. A month after the alarm crisp consumption was back at normal. In this case the extremely loud warnings almost backfired. Perhaps the food scare went too far: there was almost nothing left to eat, and popular and well-known "healthy" foods like bread, cereals and cauliflower au gratin were on the "bad" list. Maybe one could stop eating crisps, but if one would get cancer from the rest there was not much point. So people simply gave up on trying to deal with it. The constant media overuse of risk claims has made the public indifferent. When a real and significant risk is discovered it will not stand out from the noise.

On the other hand, the study in British Journal of Cancer that caused the triumphant headlines did merely show that there was no detectable link between food with acrylamide and certain cancers. It did not show that there was no general risk from acrylamide. But in the media it was announced as a vindication of all food - now it was safe. The risks from eating chips have not suddenly vanished; it is just that one possible risk vanished.

The public is confused: one day crisps are deadly, the next day fine. The result is a general distrust of science, since it appears to be changing its position every day. Meanwhile the real risk of eating fat food with empty calories that makes you thirsty for a soft drink or beer remains, but more mediagenic risks push it back. Slight updates of risk estimates or that unhealthy food still is unhealthy are not news. It is news when somebody announces that they have discovered something.

There is a symbiosis between researchers, administration and journalists. A memorable result can make a scientific career - especially if it becomes important in the public debate or the basis of significant decisions. An administration needs issues to administrate and recognition that it deserves funding. Journalists need interesting things to report. They all benefit each other by boosting new risks.

The problem is of course that publication by press conference is not good scientific publishing and may in the long run hurt both the career of the researcher (anybody remember cold fusion?) and science itself, as results that are dramatic or fit politics become emphasized at the expense of well-tested but less interesting research.

It also forces administrations to act quickly, and often on insufficient information. This in turn makes bad policy. But bad policies are not easily changed once they have been loudly announced; there is too much prestige invested in them and this makes administrators defend them far beyond what they are worth. The Food Administration, for example, largely discounts the BJC study as being too specific and having too few samples to imply any change in organisational policy. Meanwhile it also discounts far larger studies of the osteoporosis risks of fortifying milk with vitamin A and the potential benefits of starting folate fortification instead - each policy appears to defend itself against change by ignoring new results.

The recipe for a food scare is preliminary science added to an administration in need of recognition, garnished with news-hungry journalists. It is driven just as much by self-interest as genuine fear and concern.

But don’t worry. There is a solution, and Brussels has decided it for you. Instead of relying just on national Food Administrations there should be a federal administration, the European Food Safety Authority. Instead of having hasty, national alarms about possible risks there will be international panels carefully evaluating them and making proper press releases. One wonders whether such a scheme would act swiftly if a real danger did appear - and if it would remain immune to the same temptation to boost its budget, influence and visibility as the Swedish Food Administration lost against.

The case of the mad potato crisp disease might be over, but there is no doubt that bold investigators will find many more things to fear - and fund.