Should risk assessment be more transparent?
The idea of conducting a risk assessment is a relatively recent development. The impetus stems from serious toxic incidents such as the use of ethylene glycol in the 1930s as a solvent for sulphanilamide, a poorly water-soluble antibacterial drug or the introduction of thalidomide in 1959. Ethylene glycol was responsible for many cases of kidney failure in patients, and this disaster was critical not only for the development of toxicology as a science but also for the establishment of government agencies to control the marketing of pharmaceuticals, e.g. the U.S. Food and Drug Administration. The thalidomide catastrophe was the instigator of drug registration practices in Europe.
The Delaney amendment written into law in the U.S. (1958) stated that no chemical found to be carcinogenic in laboratory animals or humans could be added to food. At that time, it meant that the analytic detection level was the limit, but chemistry developed rapidly, and soon small amounts of carcinogenic chemicals could be detected in all foodstuffs. Attitudes towards pesticides changed after the appearance of Rachel Carson’s Silent Spring (1962) and toxicology developed by leaps and bounds. In addition, Bruce Ames’s comment in the 1970s that one molecule could cause cancer especially stimulated the evaluation of carcinogenicity of chemicals.
Risk assessment developed for registration purposes is not transparent
At different times in different countries, a few groups of chemicals were recognized as groups that cannot be marketed without performing toxicity studies and they had to be approved by national authorities. Typically, the three groups about which there was concern were pharmaceuticals, food additives and pesticides; pharmaceuticals and food additives, because they are foreign chemicals deliberately given to people, and pesticides, since they are, by definition, toxic compounds.
Most of the risk assessment procedures were developed for these purposes, and it also meant that the manufacturer was obliged to provide the authorities with convincing data about the safety of their product. These studies were often performed by commercial laboratories commissioned and paid by the producer of the chemical. This system means that the manufacturers are very protective of the data, and would be most unhappy if it were publicly available, because then their competitors would gain an unfair advantage, i.e. having access to work performed and paid by others. This has lead to different principles being applied here from other scientific activities; most of the toxicity data are seen only by a few government officials and it is not publicly available in the scientific literature.
It is worth considering if it is in anyone’s interests that data crucial for human health should be a closely guarded secret. Should it not be open to scientific criticism, comments by other scientists who would be able to make their own interpretation of the results, and also freely available to those who suffer, as well as simply being able to evaluate if the data are reliable? One of the excellent points in the REACH-program of the European Union is that entrepreneurs are encouraged to buy the data from those who have already conducted studies, rather than repeating them for themselves. In fact, there could be a similar system as with patents: the first firm doing the toxicity studies would make them public through an official channel, and subsequent users of the data would need to buy a license from the original firm to use the data for their own product.
Could also risk assessment be open?
Another perhaps even more dramatic novel departure from the previously somewhat secretive toxicity studies and risk assessment is the idea of performing the whole risk assessment in an open forum comparable to the Wikipedia, open encyclopedia. This would represent also an attempt to combine science and values in a single discussion forum aiming at achieving a synthesis that would be acceptable to all parties. Like Wikipedia, it would be based on web-based information storage and workspaces building on the idea of mass collaboration. The goal of one such attempt, Opasnet, is specifically to provide information for supporting decision making by means of systems analysis.
The fundamental concept of open risk assessment is that a large number of people working together towards a common aim can achieve not only much more but reach their goal also much more rapidly than individual researchers, however bright and productive they are. This also would make it possible to avoid mistakes resulting from preconceived ideas that prevent the free exchange of novel concepts.
It is in the interests of the society and the public that all research used for risk assessment and also the risk assessment process itself are as reliable and as open as possible. Human health should not be based on secret documents. In addition, risk assessment itself could benefit from openness and novel ideas, especially input from individuals outside the groups of professional people normally performing it.
Notes and references
- See the chapter "Why are experts usually so sceptical?"
- See the chapter "If this kills insects, what effect is it having on me?"
One level up: Here a risk, there a risk, everywhere risks, risks!
Next chapter: Is environmental science just another religion?