By Peel, Michael B.
Nieman Reports , Vol. 63, No. 1
Project manager Jim Morris dropped a box of Environmental Protection Agency (EPA) records next to my desk. It was December 2007, two weeks after I'd started work at The Center for Public Integrity as a computer-assisted reporter. This box contained thousands of reports of pesticide exposures involving people, pets and wildlife.
"What can we do with these?" he asked.
Not much, I thought, since what he'd handed me represented only a fraction of the EPA's pesticide exposure archive. Even this amount would have buried us in paper, making analysis all but impossible. Morris said that he'd asked for the records in an electronic format months earlier, but he'd been told that the information was available only on paper. At the time, he hadn't argued the point. But the records obviously had been printed from some sort of database, so we concluded that if we could acquire all of the data in its original electronic form, we could look for meaningful trends.
Morris, a veteran environmental reporter, first heard of the EPA's pesticide exposure data system while writing an investigative article about the health effects of the pesticide chlorpyrifos--better known by its trade name Dursban--for U.S. News & World Report in 1999. Despite assurances from Dursban's manufacturer, Dow AgroSciences, that the widely used bug-killer was perfectly safe, the EPA banned it for residential use in 2000. In his article, Morris relied on the EPA exposure reports--and the agency official then in charge of interpreting them--to tell the story of a product that showed signs of harming people and animals with alarming regularity. He'd resolved then to one day go after the entire EPA database, believing there were other dangerous products on the market that weren't being adequately regulated.
Analyzing Government Records
In January 2008, the center moved to acquire the database, called one of the "Ten Most Wanted Government Documents" by the Center for Democracy and Technology. (1) An employee in the EPA's Office of Pesticide Programs told me that these records were "kind of" available in an electronic format, but that the database was old and would be difficult to work with. The employee suggested that I tell her what I was looking for, and the EPA would conduct the analysis for me. Morris and I quickly rejected this idea and filed an FOIA request for the data. To our surprise, and to the EPA's credit, the agency responded in about two months, even though it had to remove thousands of names and other confidential information from the database.
After familiarizing ourselves with the more than 300,000 records in the database, one fact became abundantly clear: The number of incidents attributed to pyrethrins--a family of insecticides extracted from chrysanthemums--and their synthetic relatives, pyrethroids, had increased dramatically since 1998, according to reports filed with the EPA by pesticide manufacturers.
Our analysis also revealed that the number of human health problems, including severe reactions, attributed to pyrethrins and pyrethroids had increased by about 300 percent in the previous decade. Our review of more than 90,000 adverse-reaction reports found that pyrethrins and pyrethroids together accounted for more than a quarter of all fatal, "major," and "moderate" human incidents in the United States in 2007, up from 15 percent in 1998. Although the number of deaths was low--about 20 from 2003 to 2007--the number of moderate and serious incidents (more than 6,000) attributed to the group of chemicals was significantly higher than for any other class of insecticide.
The numbers surprised us. Americans, it turned out, increased their use of pyrethroid and pyrethrin insecticides after the EPA had cracked down on organophosphate pesticides such as Dursban. Toxicologists and epidemiologists told us that pyrethrins and pyrethroids are thought to be less acutely toxic than organophosphates, and the manufacturers insisted that the products are safe when properly applied. …