A Russian roulette of food poisoning in American states

By THOMAS HARGROVE

More than 50,000 people got sick or died from something they ate in a hidden epidemic that went undiagnosed by the nation's public health departments over a five-year period.

Americans play a sort of food-poisoning Russian roulette depending on where they live, an investigation by Scripps Howard News Service found. Slovenly restaurants, disease-infested food-processing plants and other sources of infectious illness go undetected all over the country, but much more frequently in some states than others.

Scripps studied 6,374 food-related disease outbreaks reported by every state to the federal Centers for Disease Control and Prevention from Jan. 1, 2000, through Dec. 31, 2004. The causes of nearly two-thirds of the outbreaks in that period were officially listed as "unknown."

The findings translate into an alarming potential for tragedy. If health officials are unable to connect illness to food, victims who might eat from the same poisoned source cannot be warned. If food is known as the culprit, but the specific disease lurking within is not diagnosed, the victims may get even sicker or die without proper treatment.

The poor track record of so many state labs also raises chilling questions about their ability to spot or deal with a food-borne terrorist attack.

Families of children who got sick during the five-year period in the study tell heart-rending stories of heroic efforts they made to convince the medical establishment they were victims of food illness.

"My daughter's death would have been listed just as a 'stroke' and swept under the rug," said Todd Nelson, a Continental Airlines pilot and father of a 19-month-old girl who died of E. coli. "But I wanted to know what my daughter really died of. And I wanted somebody to blame."

The Nelson family believes Ana Leigh Nelson ate infected hamburger meat from a popular Minnesota restaurant in 2002. The family demanded further private tests that confirmed a rare strain of E. coli and then demanded that the medical examiner change her death certificate to correctly report death from complications of food poisoning.

"We sort of fell through the cracks," Nelson said.

[inline:1=click to enlarge map]

The study found that Kentucky, Oklahoma and Nebraska are virtually blind to outbreaks of food sickness, rarely detecting that scattered illnesses have common food causes.

In Alabama, Florida and New Jersey, the cause of food poisoning is almost never found, even when it is known that dozens or hundreds of people became violently ill or died from something they ate, according to the Scripps study.

The CDC defines an "outbreak" as two or more people who got sick or died after eating the same food. State and local epidemiologists are diagnosing an average of just 36 percent of the nation's reported outbreaks even though some outbreaks have hundreds of victims.

Alabama was the worst in the nation, diagnosing only 5 percent of its reported outbreaks, the study found. "It's a real struggle. We've never identified a virus at the state level," said Alabama State Epidemiologist John Lofgren.

After learning of the study's findings, Kentucky officials ordered changes to their disease-reporting system. "We really hadn't been categorizing food- and waterborne outbreaks," admitted Kentucky Epidemiologist Kraig Humbaugh.

During the five-year period studied, Florida reported only seven people sickened by E. coli outbreaks, a suspiciously low number for a state of its size. Nationwide, at least 3,349 people contracted E. coli in food-poisoning outbreaks.

Eight-year-old Chris Ware of St. Johns County, Fla., nearly died from massive kidney failure and bloody diarrhea in 2004 in what his mother, Cindy Ware, believes was an undiagnosed E. coli outbreak. He developed a sometimes-fatal complication from E. coli called Hemolytic Uremic Syndrome (HUS).

"Mommy, I'm going to die," Chris said at one point during his three-week ordeal in the intensive-care unit.

Chris survived after undergoing days of emergency dialysis, although his kidneys must be monitored for the rest of his life.

Ware said Chris' doctors expressed surprise that four children in her area contracted HUS during this time.

"One of them asked me: 'What's going on in St. Johns County?' " Ware said bitterly. "Nobody ever contacted us from the health department, and I called them twice."

The study found that health departments are more likely to make a diagnosis when a very large number of people get sick. They failed to determine the cause in 31 percent of the outbreaks that sickened 50 people or more. But the failure rate increases rapidly with smaller groups.

Fifty-three percent of outbreaks affecting 10 to 49 people went undiagnosed, while 75 percent of outbreaks that sickened nine or fewer people were listed as "unknown" causes.

Several state and local epidemiologists said large outbreaks give them more chances to isolate the exact disease involved. More victims mean a better chance of obtaining blood, stool and urine samples that can be tested for specific pathogens.

But epidemiologists admit that failures to diagnose food illness are common, even when the only suspect for outbreaks of a widespread intestinal disease is food. The Scripps study found that the disease went undiagnosed in 4,054 of the 6,374 reported outbreaks. Those unknown causes sickened or killed 50,968 people.

"We did what we could do," said Lisa Dallmeyer, epidemiologist for Peoria, Ill., after extensive local and federal lab tests failed to discover why 95 public-school children started vomiting after eating lunches served in December 2005 and the following January.

Dallmeyer said it "doesn't surprise me" when told Illinois is diagnosing the cause of only 27 percent of its outbreaks.
Every year, an estimated 5,000 Americans die from food-based diseases like Salmonella, E. coli, Shigellosis and Campylobacter. Another 325,000 people are hospitalized. The CDC estimates that food-based sickness probably afflicts 76 million Americans annually.

Although the Scripps study found that the quality of the nation's network of public health departments varies alarmingly, there were some bright spots.

Wisconsin, Minnesota and Hawaii do a good job of diagnosing disease outbreaks.

Wisconsin came out on top in the study by diagnosing the cause of 90 percent of its food-poisoning cases. Wisconsin also was the first state to detect and report September's deadly E. coli outbreak from infected raw spinach grown in California and shipped nationwide. The outbreak killed at least three people and sickened at least 199 others.

But the study found little to celebrate overall since most outbreaks go undiagnosed.
Federal officials and public health experts agreed with the findings and conclusions of the Scripps study.

"Our surveillance systems were designed to ring a bell when there is a problem. Are they perfect? Absolutely not. Could they be better? Absolutely yes," said spokesman Tom Skinner at the CDC's Atlanta headquarters after reviewing some of the study's findings. "We've already come a long way, but certainly, we can do better than this."
Skinner offered no explanation when asked why the CDC didn't warn underperforming states and local health departments.

"The CDC, like most government agencies, is pretty conservative. Why would they want to rock the boat?" said Ewen Todd, director of the Food Safety Policy Center at Michigan State University. "It takes someone who is independent to say: 'This is crazy.' "
Todd agreed that the quality of public health is erratic in the United States. He said state health programs are especially poor in the South.

"Our laboratories are pretty good. But, overall, the whole public health system is not working very well," Todd said. "There are no national standards for the surveillance and reporting of food illnesses."

Other experts said they believe the quality of state labs also varies alarmingly.
"Lab capacity is a serious issue for many reasons," said Jeffrey Levi, executive director of the Trust for America's Health in Washington, which studies trends in public health. "The federal government should set minimum standards and expectations and provide the resources necessary to assure this capacity."

Kentuckian Lindsay Ronay would agree.
When Ronay got married in 1998, at least 30 of her guests were put on antibiotics after a Louisville hospital lab reported they were suffering an outbreak of Campylobacter, a bacteria often associated with raw poultry.

Yet that year Kentucky reported only 35 people had suffered from a food outbreak _ none of them from the Ronay wedding.
"We had more people than that get sick just at my wedding!" Ronay said. "Two people were hospitalized, one from complications from diabetes and the other from dehydration. I've never been so sick in my life.

"We didn't think this was properly investigated at all," she said.

(Scripps Howard News Service reporters Sruthi Kunnel and Lee Bowman contributed to this story. For more stories visit scrippsnews.com.)