As Donald Rumsfeld taught us, how you handle and communicate what you don’t know is just as important as dealing with what you do know.
A new paper recently published by the scientific journal Environment International offers several different ways to help better address the uncertainty conundrum when it comes to sharing the results of chemical assessments. The paper, “Approaches for describing and communicating overall uncertainty in toxicity characterizations: U.S. Environmental Protection Agency’s Integrated Risk Information System (IRIS) as a case study,” offers five approaches to help people better understand how assumptions and uncertainty factors were used to establish toxicity values.
The recommendations are the result of a great deal of thoughtful input that stemmed from a workshop hosted by ACC’s Center for Advancing Risk Assessment Science and Policy (ARASP) to discuss and explore better methods for presenting 
uncertainty and risk information in federal chemical hazard assessment programs. The workshop benefited from the input of a wide-range of participants including more than 60 experts in toxicology, risk assessment, risk communication, exposure assessment, and hazard characterization drawn from academia, government, industry, and non-governmental organizations. Several of the proposed approaches also reflect the input from another workshop that was held during the annual meeting of the Society of Toxicology.
While IRIS was used as a case study, the approaches can be applied to the findings in similar databases, including the Agency for Toxic Substances and Disease Registry minimal risk levels and the European Chemicals Agency derived no-effect levels, and could also be adopted or adapted by other agencies and programs when evaluating chemicals. All of these programs could benefit from improving how uncertainty is quantified and communicated by risk assessors. More transparency will surely increase the utility of information generated by chemical assessors and boost the public’s confidence in the results.
