The food industry and government regulators have focused for years on finding the most virulent strain of E. coli bacteria, which has killed hundreds of people and sickened thousands every year.
But they don't regularly test for six less common E. coli strains that can cause illnesses equally as serious. Most recently, two dozen illnesses in four states were tied this spring to bagged romaine lettuce contaminated by an uncommon E. coli strain that can be difficult to detect.
Industry officials said tests aren't available to do widespread monitoring of these other strains, but food safety advocates have begun pushing the government to step up surveillance after several outbreaks.
They're motivated by what has happened to people such as Shiloh Johnson, who two years ago picked at a roll, fried chicken, sunflower seeds and olives from a restaurant buffet. Within days, the 10-year-old was hooked to a ventilator in an Oklahoma hospital, one of 341 victims of an E. coli outbreak. She remained hospitalized for six weeks.
Investigators tied the outbreak to one of the six less common E. coli strains, and her mother, Belinda Johnson, has endorsed a petition that includes their story in urging the U.S. Department of Agriculture to test for additional E. coli strains.
The agency is reviewing the petition, filed by a Seattle law firm that represents Shiloh Johnson and is known for food-illness lawsuits.
"I was so shocked then. I thought that everything was tested for," Belinda Johnson said. "I want there to be a safe food supply. I don't want any other kids or anyone to have to go through this."
Hundreds of strains of E. coli live in the intestines of cattle and other animals and can get into the food supply. Many don't cause illness. For those that do, symptoms often go unreported or undiagnosed because people don't realize it was food that made them ill.
The food industry screens for the most prevalent strain, O157:H7, which belongs to a class of E. coli that produces a sickening toxin and causes an estimated 73,000 illnesses each year. Symptoms include bloody diarrhea, dehydration and, in severe cases, kidney failure. It is the only strain the USDA considers an adulterant in meat, requiring regular screening and recalls.
Six other E. coli strains that also produce the toxin account for the majority of non-O157 E. coli cases — estimated at 30,000 illnesses in the U.S. each year, according to the Centers for Disease Control and Prevention. But just 5 percent of public health laboratories nationally test for these strains, so there is no reliable way to know whether the number of illnesses is increasing.
The CDC recommended last year that labs test for other potentially dangerous strains when they test for E. coli O157 during an outbreak of illness.
There have been no known outbreaks of the other six strains in meat, but the Seattle law firm that specializes in food-related illnesses, headed by attorney Bill Marler, has petitioned the USDA to list them as adulterants in meat.
Marler said he hopes his call for more screening of meat will prompt other industries, such as produce, to follow suit.
Many food producers have balked at such a move, questioning the feasibility of eliminating all toxin-producing strains from products.
Dr. Patricia Griffin, head of the CDC's food-borne illness epidemiology section, said she's sympathetic to their argument because not all strains of E. coli appear dangerous.
"The problem is it's a little slippery to say which ones cause human illness. We don't have it defined yet," she said. "We know those six, and we know a few others, but the others are still in a gray zone."
Illness from eating tainted meat can be avoided by cooking it thoroughly and using a meat thermometer to confirm it has reached a temperature of at least 160 degrees.
Marler hired a private Seattle area lab, IEH Laboratories, to sample ground beef nationally for these strains to determine their prevalence. So far, about 1 percent of samples have been tainted and could have potentially caused illness.
IEH Laboratories President Mansour Samadpour said his lab has worked with some in the produce industry to monitor for these strains in the past three years, though there is no required testing. As for meat, Samadpour argued the USDA needs to set standards for testing.
"There has not been demand from the industry, but if the demand is created, I'm sure a lot of kit makers will be happy to make tests available," he said.
At least one food safety group and U.S. Sen. Kirsten Gillibrand, D-N.Y., support the effort.
"This has become an issue that is too important and too urgent to ignore any longer," Gillibrand wrote in a letter to Agriculture Secretary Tom Vilsack.
The USDA has worked for three years on a screening test that could be widely used. So far, the agency has developed a test that is reliable on four of the six strains, said Dr. David Goldman, assistant administrator for the Office of Public Health Science for the USDA's Food Safety and Inspection Service.
"We are hopeful that in the next six months or so we would have a screen that would reliably find all six," he said.
The meat industry also wants to eliminate pathogens from meat, but James H. Hodges, executive vice president of the American Meat Institute, says banning six more E. coli strains won't solve the problem.
"The food safety strategies in place in plants today are far more effective in enhancing food safety than outlawing a pathogen that nature presents us," Hodges said.
Michael Doyle, director of the University of Georgia's Center for Food Safety, said requiring beef to be free of all E. coli would be economically unfeasible.
"You would probably be condemning a lot of ground beef and meat that would otherwise be considered to be safe because it's not contaminated with a harmful strain," he said.