PERSPECTIVE Chemical Mixtures: An Unsolvable Riddle?

It is difficult to overstate the complexity of assessing risks from chemical mixtures. For every valid reason to assess risks from mixtures, there appears an equally valid question as to whether it is possible to do so in a scientifically rigorous and relevant manner. Because so few data exist for mixtures, current mixture assessment methods must rely on untested assumptions and simplifications. That the accuracy of risk estimates improve with the number of chemicals assessed together as mixtures is a valid assumption only if assessment methods for mixtures are better than those based on individual chemicals. On the other hand, arbitrarily truncating a mixture assessment to make it manageable may lead to irrelevant risk estimates. Ideally, mixture assessments should be as broad as necessary to improve accuracy and reduce uncertainty over assessments that only use toxicity data for single chemicals. Further broadening the scope may be ill advised because of the tendency to increase rather than decrease uncertainty. Risk assessment methods that seek to be comprehensive at the expense of increased uncertainty can hardly be viewed as improvements. It would be prudent to verify that uncertainty can be reduced before burdening the risk assessment process with more complexity.