This is another it what will be a long – too long – series of outbreak investigations where we have represented consumers in what I hope will be a cautionary tale, and a learning experience, for manufacturers of food.
I have told restaurant and grocery store associations many, many times that they are only as safe as are their inputs.
2000 Taco Bell Green Onion Outbreak – In early December 2000, Lake County Health Department (LCHD) learned of seven hepatitis A cases, including five hospitalizations, in Lake and neighboring Sumter Counties in a two-week span. During the previous two years, the total number of known hepatitis A cases in Lake County was twenty-two. Recognizing the possibility of an outbreak, LCHD notified the Florida Department of Health.
LCHD began its investigation immediately. Assisting in the investigation were individuals from the Florida Bureau of Epidemiology and the Florida Bureau of Environmental Epidemiology, Food and Waterborne Disease. The investigation commenced with a case definition:
A primary case of hepatitis A was defined as a positive hepatitis A IgM antibody test in a Lake or Sumter county resident who had elevated liver enzyme tests or an acute onset of jaundice or abdominal pain, onset of illness between November 10 and December 16, 2000, and no other explanation for the elevated liver tests or abdominal pain.
Known “cases” were interviewed using the CDC viral hepatitis case record form. Demographic and epidemiologic information, including onset dates and sources for virus acquisition or spreading, were collected. Also, hypotheses-generating interviews were conducted with several “cases,” and the information revealed in these interviews spawned a questionnaire that was eventually answered by all case-patients.
In an effort to locate unknown “cases,” the investigation team contacted area health care facilities and labs, known visitors to Lake County during the outbreak period, and known “case” family members and acquaintances who had experienced similar symptoms. Investigators also contacted the business acquaintances and contacts of those known cases that worked in the food and childcare industries.
After the first twenty-one cases were identified, two case-control studies were done. The first aimed to determine the source of the outbreak. For this study, “controls were defined as adults who did not report a history of hepatitis A in the past, or symptoms of hepatitis [in] November or December 2000, [and] who resided on the same street as a case.” Individuals who fit the definition were interviewed over the phone.
This first study revealed a strong association between cases and the Taco Bell Mexican restaurant in Fruitland Park. Consequently, serologic testing was done on all Taco Bell employees who had worked during the exposure period. Other than the individual who was a known case, the employees tested negative.
The second case-control study was conducted to identify the Taco Bell food item(s) that were, or had been, contaminated. Six meal items and eight ingredients were significantly associated with illness. Of the meal items, only two were eaten by a majority of cases. And of the eight ingredients, green onions bore the strongest statistical association. Further analysis revealed that the green onions were the most likely vehicle for transmission.
While the Lake County investigation was ongoing, the LCHD learned from the CDC that hepatitis A outbreak investigations were also underway in Russell County, Kentucky and Clark County, Nevada. Taco Bell green onions would soon be implicated in these outbreaks as well.
LCHD and other investigators ultimately identified twenty-three people who met the case definition. Illness onset for these cases was between November 21 and December 11, 2000. In total, fifteen cases (65%) required hospitalization due to the severity of their symptoms.
The CDC extracted viral RNA from the sera of twelve cases for molecular sequencing. Sequencing in eleven of the samples matched exactly, and sequencing of the twelfth varied by one base pair over a 250 base pair gene segment. The CDC then compared the matching Lake County samples to four serum samples from the Kentucky outbreak and one from the Nevada outbreak. Sequencing studies revealed a 100% sequence homology among all the samples, and, again, the twelfth Florida sample varied from all other samples by one base pair.
These studies, together with the epidemiological, environmental, and laboratory investigations, convinced the LCHD that the 2000 Florida hepatitis A outbreak occurred at the Fruitland Park Taco Bell. See Outbreak Report at 10. The LCHD further concluded that “[a]lthough most foodborne outbreaks of hepatitis A are due to food contaminated by an infected food preparer, we believe the ingredients were contaminated prior to arrival at the outlet in this outbreak. . . . The most likely contaminated ingredient is green onion.”
2003 Chi-Chi’s Green Onion Outbreak – In late October of 2003, Pennsylvania health officials learned of a potential hepatitis A outbreak from emergency room doctors treating patients in Beaver County. The Beaver County Health Department (BCHD) and Pennsylvania Department of Health (PDOH) began investigating the apparent outbreak, and learned through interviews that all case patients had eaten at the Chi Chi’s restaurant at the Beaver Valley Mall in the weeks before becoming ill.
On November 3, PDOH issued a hepatitis A advisory, encouraging anyone who had eaten at the Beaver Valley Mall Chi-Chi’s restaurant within the past 14 days to receive an Immune globulin (Ig) shot to prevent becoming ill with the hepatitis A virus. Ig is only effective in preventing infection with hepatitis A if it is administered within 14 days of exposure to the virus. PDOH scheduled Ig immunization clinics at several locations over the following days.
By November 7, PDOH had identified 130 people who had contracted hepatitis A as part of the outbreak. The number had grown to 240 cases by November 11, and kept climbing. By November 14, three people had died due to liver failure caused by hepatitis A, and the number of ill people had risen to 500.
PDOH, the Centers for Disease Control and Prevention (CDC) and the Food and Drug Administration (FDA), conducted an epidemiological study of the outbreak, and determined that green onions imported from Mexico were the source of the outbreak. The FDA issued a statement dated December 9, 2003, affirming that this outbreak was associated with eating raw or undercooked green onions.
Ultimately, over 650 confirmed cases of hepatitis A, both primary and secondary, were linked to consumption of green onions at the Beaver Valley Mall Chi-Chi’s. The victims included at least 13 employees of the restaurant, and numerous residents of six other states. Four people died as a consequence of their hepatitis A illness. In addition, more than 9,000 people who had eaten at the restaurant during the period of potential exposure, or who had been exposed to ill Chi-Chi’s customers, obtained immune globulin shots to prevent hepatitis A infection.
2005 Soleil Lettuce Outbreak – In October 2005, epidemiologists at the Los Angeles County Department of Health Services – Public Health (“Public Health”) became aware of an increase in the number of reported cases of people infected with the hepatitis A virus (“HAV”). Notably, the number of reported cases in October 2005 was four times the number of cases reported in October 2004. At first, cases appeared to be occurring largely among the County’s homeless population, and efforts to identify potential sources of HAV infection centered on food served at charitable food kitchens. Investigators had difficulty locating the largely transient group of persons for interview and, thus, made little progress in identifying a potential source of exposure to the virus.
In early November, Public Health officials learned that several cast and crewmembers from the movie set, “The Good German,” were confirmed with HAV. Preliminary interviews revealed that the only day that all six ill individuals had eaten food on the set was October 3. Interviews revealed that, on October 3, all six cases had eaten food provided by each of the two food purveyors, Silver Grill Location Catering Inc. (“Silver Grill”) and Jeff Winn, craft service for the movie. Twelve more cast and crewmembers from the movie were ultimately identified to be infected with HAV, and all had eaten on location on October 3. Public Health launched a case-control study of movie workers to identify risk factors for HAV infection and determine if all had become ill from the same source.
A case-patient was defined as a person, who had eaten food provided on the set of The Good German on October 3, had then experienced symptoms of acute hepatitis, and had a positive HAV IgM test. Control subjects had eaten on the set on the same day and had not had clinical symptoms, a past diagnosis or vaccination for hepatitis A, or a recent history of receiving immune globulin. Using a standardized questionnaire, cases and controls were asked about potential exposures to known risk factors for an HAV infection, including contact with young children, travel, shellfish consumption, and number of sexual contacts. A list of 65 food items available to the cast and crew on October 3 was obtained from Silver Grill Catering and Jeff Winn.
A total of 116 Good German cast and crew who were on set on October 3 was interviewed; forty were excluded from the analysis because of previous HAV infection or vaccination. Eighteen individuals met the case definition. The median age of case-patients was forty-one years and 78% were male. The median incubation period was twenty-five days (range nineteen to thirty-seven days). One person was hospitalized and none died. No common exposures other than eating on the set of the Good German were identified.
Data analysis showed that none of the food items from craft services was associated with illness. Investigators consequently turned their attention to food items prepared and served by Silver Grill Catering. Data analysis showed that case-patients were more likely to have eaten baby mixed green lettuce (adjusted odds ratio [aOR]: 4.81; exact 95% confidence interval [CI] = 1.3-19.8) or jerk chicken (aOR: 4.43; CI=1.2-18.3) than their non-ill counterparts. The jerk chicken had no raw ingredients and was only handled with tongs making it an unlikely vehicle for the virus unless cross contamination occurred after cooking. But, because all Silver Grill Catering food-handlers tested negative for acute hepatitis A, contamination by an ill food handler did not occur. Investigators concluded that lettuce was the likely source of the outbreak.
Public Health Environmental Health staff and the California Department of Health Services’ Food and Drug Branch conducted a traceback and identified the distributor of produce to Silver Grill Catering as Soleil Produce, a Los Angeles distributor. There was no environmental investigation at any farm that might have supplied the lettuce.
On January 6, 2006, Public Health issued an outbreak investigation report of the Hepatitis A outbreak that occurred among members of the set of The Good German. Citing statistical evidence and known likelihood of produce contamination with HAV, investigators concluded that the salad provided by Silver Grill Catering most likely caused at least eighteen confirmed cases of hepatitis A virus infection among Good German cast and crew.
Fresh produce contaminated during cultivation, harvesting, processing, and distribution has also been a source of hepatitis A. In 1997, frozen strawberries were the source of a hepatitis A outbreak in five states. Other produce, such as blueberries, has been associated with hepatitis A outbreaks in the U.S. as well as other developed countries. See, www.about-hepatitis.com.