In a letter to Veneman, Sen. Tom Harkin (D-Iowa) said he had "fundamental questions" about whether the department's testing plan for mad cow disease that is supposed to begin June 1 "is properly designed according to scientific and statistical principles."
Harkin noted that after a single case of mad cow disease was discovered in Washington State in December 2003, an international panel of experts on the disease recommended that USDA increase testing. The department announced it would move from testing a total of 20,000 animals per year to testing between 201,000 to 268,000 cattle over 30 months of age, considered high-risk, and 20,000 younger cattle, considered a lower risk. The goal is to detect BSE at prevalence as low as 1 in 10 million.
"The plan seems to be dictated primarily by how many cattle USDA wants to test, rather than by the number that would have to be tested, using statistical methods, to reach accurate and reliable conclusions," Harkin said.
He also charged that USDA "seems to have assembled an ad hoc mixture of somewhat increased testing, but less than the expert panel recommended, along with principles from the Harvard Risk Assessment."
Some scientists and veterinarians have suggested 20,000 may be too few to test to yield any statistically valid conclusions regarding the existence or prevalence of BSE in such a sub-population of apparently healthy cattle, Harkin said. "Previously, you and other USDA officials repeatedly stated that 40,000 tests would be 47 times the international standard, but that was plainly not true," Harkin said.
The Iowa senator also said USDA needs to come up with a clearer definition of so-called downer cows. USDA officials said the BSE-infected cow in Washington State was a downer, meaning nonambulatory, but workers at the plant have said the cow was able to stand.
Harkin's letter followed substantial criticism of the USDA effort on mad cow disease at the Consumer Federation of America's annual conference Friday on food safety.