Product Biomonitoring and Responsible Reporting

In this issue of Environmental Health Perspectives (EHP), Schecter et al. (2011) report levels of poly brominated diphenyl ether (PBDE) flame retardants in butter purchased from retail stores. Although the investigators found prevalent levels of PBDEs in butter, they report that one sample had inordinately high levels of octa-, nona-, and deca-BDE con-geners, likely from its highly contaminated wrapping paper. The safety of PBDEs has come into question (DiGangi et al. 2010). Chemical companies have voluntarily phased out penta-and octa-BDEs in the United States and have agreed to do so for deca-BDE production, and some states have banned PBDE use in consumer products such as mattresses and electronics (U.S. Environmental Protection Agency 2007, 2010). The presence of these chemicals in food products has not been addressed explicitly in legislation or regulations in the United States. Given the lack of clear regulatory guidance and the question of risk from high levels of PBDEs in food products, it is legitimate to ask: should investigators in these types of " product biomonitoring " studies, when confronted with startling results from any potentially risky chemical, report their results to the manufacturer or to any regulatory agency prior to publication? Should the investigators identify specific brand name products and their manufacturers in their manuscript? More important, would the public want to know such results? In a study of clinical trial subjects in brain imaging studies, Kirschen et al. (2006) found that at least 91% of the subjects, depending on the setting, wanted incidental findings to be disclosed to them, regardless of their clinical significance. In a workshop at which investigators discussed incidental findings, Illes (2006) found that most of the participants believed that research protocols should provide for dis closure of incidental findings. In environmental exposure studies there is no clear consensus on reporting individual data (Morello-Frosch et al. 2009), but Brody et al. (2007) found that an over whelming majority of subjects (97%) were interested in knowing their personal results, regardless of the limited or absence of scientific information related to the health risks of that data. But the study by Schecter et al. (2011) is not a clinical trial, nor does the study include human subjects with measure ments of internal exposures. On the other hand, the butter obtained for this study would otherwise have been purchased and consumed; additional lots of butter, similarly contaminated, may be awaiting purchase or may have already …