Anecdotal evidence and aspartame
As a scientist, I see some sources of evidence as being more powerful than others. For example, published evidence from a double blind, cross-over study, done to best practice and challenged through peer review carries more weight than anecdotal evidence.
Nonetheless, all evidence is important and none is discounted when the Agency carries out a risk assessment, though it may be weighted according to its source. I have been interested to learn that the National Council of Women of Great Britain discussed the use of anecdotal evidence at its Annual Conference in October this year and adopted a motion to 'urge HMG to acknowledge the importance of anecdotal evidence in decision-making and put in place a system to seek, listen to, record, challenge and if proven act upon anecdotal evidence.
Having reproducible evidence and challenging through peer review are fundamental to the scientific process, but the Agency is not only about science. When we develop advice and make policies -the risk management part of the business - the Agency takes into account a far wider evidence base (individual liberty, regulatory constraints, economic and social consequences and consumers’ appetite for risk). This second stage is distinct from the scientific process of advocacy and challenge that generates the risk assessment; it's an iterative, consultative process which leads to our Board reaching and making a judgement accountably and in public.
I see anecdotal evidence as a trigger for more rigorous scientific studies which will provide reproducible evidence for an independent analysis of the risks.
The Agency has recently been asked how it is going to respond to the anecdotal evidence of adverse reactions to the artificial sweetener aspartame.
At a recent meeting of the European Food Safety Authority (EFSA) Advisory Forum, I proposed that EFSA should lead a research project to address anecdotal concerns of consumers about the effects of the artificial sweetener aspartame. The idea is that the study would compare reported symptoms and biochemical parameters in a double-blind cross over design in self-selected groups who report that they are sensitive to aspartame compared with control groups. EFSA responded positively to this suggestion and will arrange a meeting with key member states to discuss this further.
I believe that the Agency needs to be clearer about the relative merits of different types of evidence used in the risk assessment process and so I have asked Professor Colin Blakemore, the Chair of the Agency's new General Advisory Committee on Science to consider this in 2008.
