CropLife America (CLA) is surprised and disturbed to learn the U.S. Environmental Protection Agency (EPA) has a plan to convene a Science Advisory Panel (SAP) next month to consider the use of flawed epidemiology studies in pesticide regulation. This breaks with long-established Agency standards for sound scientific risk assessment and fails to address concerns expressed in previous SAPs convened by the agency.
CLA calls on EPA to postpone this SAP until: 1) all data generated by the epidemiology studies have been received and appropriately evaluated in the same way EPA’s policies require the review and evaluation of all other data used in regulatory decision making; and 2) the agency’s intended use of epidemiology studies has been presented transparently and subjected to public review. CLA is concerned that EPA has not been able to fully review all of the collected human epidemiology data because the authors of the studies in question have declined to provide the underlying data despite repeated Agency requests. EPA currently bases its health and safety standards for pesticide regulation on robust studies following EPA-approved protocols. Exposures in these studies are known, effects are documented, human health impacts are determined, results can be replicated, and the underlying data are available for EPA evaluation.
EPA’s long-standing practice has been to use epidemiology studies, not to set regulatory standards, but as a basis for requiring additional regulatory guideline studies and, potentially, further investigation. This practice reflects a recognition that epidemiology studies, by their observational nature, are not appropriate for use in regulatory decision-making for pesticides.
As noted in EPA’s own guidance:
“When animal and epidemiologic data do not provide a consistent toxicological picture of a particular pesticide, more weight would likely be given to those studies with robust study design and availability of replication or confirmatory data.” – Draft Framework for Incorporating Human Epidemiologic & Incident Data in Health Risk Assessment,” 2010, Office of Pesticide Programs, U.S. Environmental Protection Agency. Washington, DC, page 31.
Thus according to EPA policy, when data conflicts and decisions must be made, higher quality data trumps data of lesser quality. Other data may form a basis for additional investigation, but it cannot not be accorded greater weight than high-quality guideline studies specifically designed for regulatory use. To do so would result in serious damage to the scientific credibility of EPA risk assessments.