A group of ecotoxicologists claim that the US Environmental Protection Agency's evaluations of pesticide safety are inadequate and lead to bias.

Writing in BioScience, Michelle Boone of Miami University and colleagues worry note that most pesticide toxicity tests used in risk assessments are conducted by pesticide manufacturers themselves, which the authors believe can result in untenable conflicts of interest. This has been longstanding policy since the beginning of the EPA; manufacturers are required to prove product safety according to EPA guidelines, to prevent taxpayers from having to foot the bill to discover if a product is harmful.

Then they say the EPA is too strict, because it uses stringent criteria for valid studies and claims that lack data or valid methodology are not used.

The environmentalists are specifically interested in the popular herbicide atrazine, which is undergoing re-registration again and had two special Scientific Assessment Panels called based on mainstream news media publication of studies that the EPA disqualified in its panels. The EPA looked at over 6,000 studies regarding atrazine and only one received its highest rating, because it was duplicated in two countries and created according to EPA guidelines. The maker of atrazine, Syngenta, was forced to pay for the studies, which leads the authors to believe the EPA is not doing its job.

They say if lower quality studies had been included atrazine may not have been approved - but the same argument could be made about vaccines and autism or GMO corn versus organic. Instead, criteria for studies are created and those do not meet it are excluded.

They then say that there is over-reliance on laboratory studies.

They think that the scientific method and proving harm is a flaw in the EPA. "Presumption of innocence" is inappropriate for the evaluation of potentially harmful substances, the authors say, and say that "the risk assessment process can and should be improved so that decisions are made with the best available data with an evidence-based approach."

The EPA has not commented on these claims. They produce detailed papers about every re-registration and they are publicly available.  But the authors contend that the EPA cannot be objective and want another layer between the EPA and the products they monitor. Loosely-controlled field studies with multiple variables are not currently considered high quality by the EPA and they also suggest that the assessment process should be made more transparent - though the EPA was created to be protected from exactly the kind of political pressure that lobbying scientists brings.

Environmentalists have good company on that issue - global warming skeptics also think the EPA is not transparent enough and would like to have their advocates involved in decisions about emissions.