for the scientific decision-making process is not the conflict itself but the impact it may have on the context of a situation. Michaels agreed that there may be certain situations for which context comes into play. For example, a government advisory committee meeting for which a vote is expected should not be composed of people with conflicts of interest, as the credibility of the process may be questioned. Some participants noted that conflict of interest can derail the scientific process and needs to be resolved.


Further discussion focused on how to ensure openness and a systematic structure in the environmental health decision-making process. Goldman proposed that it may be time for the field to develop a code of ethics similar to that used in the legal profession, since there is no current agreed-on roadmap to ensure that biases and points of view are noted. In the legal profession’s code of ethics, once a conflict is identified, lawyers recuse themselves from the situation; this is looked on favorably as a way to avoid conflict and bias. Hattis explained an effort to do this in the community of risk analysts that took the form of a set of “ideals” (Hattis, 2000). On this point, Michaels argued that while codes of ethical conduct can be beneficial in certain professions, when it comes to decision making, those with financial conflicts of interest should not be in a decision-making position, regardless of a code. Ultimately, one participant stated, the facts matter, and when looking at conflict, whether from a legal or scientific perspective, facts are what should drive the decision-making process.


The discussion concluded with input from the speakers and the audience as to the future direction of scientific decision making. Numerous suggestions were offered as a path to making overall improvements in the current decision-making process. The list below does not constitute recommendations of the group, but rather captures the range of ideas that people would like to see explored in future discussions. These include

  • Not necessarily instituting a standard for how one actually weighs the evidence, but rather providing a rationale for the inclusion and exclusion of material studied in order to simply show why something should or should not be studied.

  • Tailoring the approach to decision making to eliminate the “one size fits all” risk assessment and incorporate context.

  • Discussing regulatory agency decisions to explain why agencies are regulating some substances and not others. This could eliminate the presumption of innocence in the current decision-making process.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement