Journal or Publishing Institution: BioScience
Date of Publication: 01/22/2015
Study: http://bioscience.oxfordjournals.org/content/early/2015/01/22/biosci.biu229.extract
Author(s): Boone, M.D., and Rohr, J.R.
Article Type: Journal Publication
Abstract:
Forbes indicates that we have missed the “real” problem of risk assessment in Boone and colleagues (2014)—the failure of standard tests to reveal the actual environmental risks. We agree that there are many issues to be addressed with risk assessment (e.g., Forbes and Calow 2013) and that risk assessments will fall short if we fail to incorporate or account for ecological complexity. Although Forbes indicates that we considered industry-funded work to be “inherently biased and therefore not [to] be trusted” as a result of its conflicts of interest (COIs), this is not the case, and it misrepresents a COI. Having a COI does not depend on underlying motives; rather, it is amoral, neither moral nor immoral—an objective state occurring when personal interests could benefit from professional decisions or judgments. A COI is, consequently, inherent in research directly conducted or funded by industry for assessments or regulatory purposes. Therefore, COIs should be managed. We acknowledge that there are other COIs (e.g., the desire for prestige) that can influence scientists, but financial COIs are the most insidious and have been documented to influence the outcome of results (e.g., Bekelman et al. 2003). Although financial COIs are most commonly managed through disclosure in the acknowledgement section of manuscripts, they could be further managed through a third-party funding agency that separates an industry from study analysis and interpretation, and eliminates restrictions on publication, regardless of outcome. For the public to have confidence in the scientific process, they need assurance that scientific outcomes are not influenced by personal financial gains. For this reason, failure to address and manage issues of financial COIs is a structural flaw in risk assessment. Addressing COIs still allows for “all of the key stakeholder[s]” to come together, but we need to explicitly acknowledge that some have a greater vested interest in the outcome and manage that risk.
Whereas issues related to COIs are of great concern, the potential for exclusion of most of the available data in US Environmental Protection Agency (USEPA) risk assessments was our central criticism of the process. On the basis of a single industry-funded study, USEPA concluded that “exposure to atrazine at concentrations ranging from 0.01 to 100 [milligrams per liter] had no effect on Xenopus laevis development (which included survival, growth, metamorphosis, and sexual development)” (p. 60) and that the “level of concern for effects on aquatic plant communities… was lower than the atrazine concentration observed to produce significant direct or indirect effects on invertebrates, fish, and amphibians” (USEPA 2012, p. 97), which would eliminate further assessments of atrazine’s impacts on amphibians despite significant effects at these concentrations in other studies. Fixing other issues of risk assessment, like incorporation of ecological data, will not improve the process if most of the independent research is omitted, whereas studies with inadequately managed COIs are included. If we do not repair the foundation of regulatory risk assessment, the best standard tests, ecological studies, and ecological models may currently not even be used to inform policy and regulatory decisions—the antithesis of evidence-based decision making.
Keywords: risk assessment, conflicts of interest, Environmental Protection Agency
Citation:
Boone, M.D., and Rohr, J.R. (2015). The trouble with risk assessment lies at the foundation. BioScience, 65(3), 227-228.
Category:
- Regulatory issues
Record ID: 234
