Multisite Outbreak of Norovirus Associated with a Franchise Restaurant --- Kent County, Michigan, May 2005
The majority of cases of foodborne gastroenteritis in the United States are caused by noroviruses (1). This report summarizes an investigation by the Kent County Health Department (KCHD) in Michigan into three norovirus outbreaks and a cluster of community cases that were associated with a national submarine sandwich franchise restaurant during May 3--9, 2005. The investigation identified a potential source, a food handler who had returned to work within a few hours of having symptoms of gastrointestinal illness while he was still excreting norovirus in his stools. To prevent norovirus outbreaks, food service workers should be educated regarding norovirus transmission and control. In 2005, new guidelines for state health departments regarding norovirus containment were published by the Food and Drug Administration (FDA) (2); guidelines for local health departments in Michigan were issued by the state's Department of Community Health and Department of Agriculture (3). The new guidelines for Michigan recommend that food service workers with suspected norovirus not return to work until they are asymptomatic for 48--72 hours.
Outbreak 1. On May 5, the KCHD was notified of a gastroenteritis outbreak among employees who had attended a school staff luncheon on May 3. Staff members were served a party-sized submarine sandwich catered by a national franchise restaurant. A case was defined as illness in a person who ate the suspect meal during May 3--8 and became ill 8--56 hours later with vomiting or diarrhea and two of the following: fever (documented), abdominal cramps, or nausea. A total of 23 (80%) of 29 school staff members reported illness. Among the 23, predominant symptoms were diarrhea (87%) and vomiting (74%). Of the six stool specimens collected, all tested positive for norovirus by polymerase chain reaction (PCR). A retrospective cohort study was conducted, and exposures to 26 food items were analyzed. Twenty-two of 23 ill persons reported eating lettuce; however, no specific food item was significantly associated with illness.
Outbreak 2. On May 6, KCHD was notified of a gastroenteritis outbreak at a publishing company staff luncheon that had occurred on May 5. Party-sized submarine sandwiches were served by the same restaurant that catered the luncheon in outbreak 1. Among 95 persons who could be interviewed and who ate the suspect meal, 55 (58%) had become ill. Predominant symptoms were diarrhea (94%) and vomiting (83%). Because the entire cohort of exposed persons could not be interviewed, a case-control study was conducted, and exposures to 16 food items were analyzed. Results indicated that eating lettuce was significantly associated with illness (odds ratio [OR] = 11.24; 95% confidence interval [CI] = 1.30--95.2). Fifty-three of 54 ill persons who responded to the question reported eating lettuce. Two other food items were significantly associated with illness: jalapeno peppers (OR = 3.45; CI = 1.04--11.40) and onions (OR = 3.09; CI = 1.27--7.80). Fifteen of 52 ill persons for whom data were available reported eating jalapenos, and 21 of 50 reported eating onions. Of two stool specimens that were tested during this outbreak, one was positive by PCR for norovirus. The owner of the restaurant was contacted, and a log of other catered events was requested.
Outbreak 3. On May 6, KCHD learned of another outbreak through inquiries to groups identified in the same restaurant's catered event log. A social service organization that held a luncheon on May 4 reported that employees became sick after eating a party-sized submarine sandwich. Of 18 persons who attended the luncheon, nine (50%) became ill and met the case definition. Predominant symptoms were vomiting and diarrhea (both 78%). A cohort study was conducted, but no specific food item was significantly associated with illness. Norovirus was detected by PCR in both of the two stool specimens tested.
Community cases. Reports from the community identified an additional 28 persons with illness onset dates of May 4--9, including 25 (90%) who reported being ill after eating sandwiches from the same franchise restaurant that catered the events in the three outbreaks. Major symptoms included diarrhea (92%) and vomiting (80%). A case-control study was conducted, but no specific food item was associated with illness. All three stool specimens tested were positive by PCR for norovirus.
Epidemiologic and Laboratory Investigation
The three outbreaks and community cases all appeared to have been linked to sandwiches prepared by the same restaurant (Figure). Sequence analysis was performed on 21 stool specimens from the three outbreaks. Results demonstrated 100% sequence homology for the 21 specimens.
Investigators learned that a food handler employed by the restaurant had vomiting and diarrhea on May 2. The food handler believed he had acquired illness from his child, who had vomited on May 1. The child's illness was traced to an ill cousin who had been exposed to norovirus at a child care center. The food handler's vomiting ended by the early morning of May 3, and he returned to work at the restaurant later that morning. A stool specimen from the food handler was collected on May 10 and tested positive by PCR for norovirus on May 16. Sequence analysis was performed on strains from the food handler and eight customers who had illness consistent with the case definition. All nine matched the strains identified in the previous outbreaks.
An environmental health inspection on May 6 revealed that the restaurant had been thoroughly cleaned on May 5 in advance of a visit by corporate supervisors on May 6. During the inspection, investigators learned that the food preparation sink was being used for hand washing; a copy of the Michigan guidelines for disinfection of norovirus (3) was provided to the restaurant owner. The restaurant again was cleaned on May 7. On May 10, after the health department received new complaints of illness from the public, the health department recommended that the restaurant be closed temporarily, and the restaurant owner complied.
On May 16, a meeting was held with members of the health department staff and the food handler who had tested positive for norovirus. Because of the association with eating lettuce in outbreak 2, questions were asked about lettuce-handling procedures. Investigators learned that lettuce was sliced each morning by the food handler who had been ill. In addition, heads of lettuce were washed in the same sink in which employees washed their hands; the sink was not sanitized before and after the lettuce was washed. On May 17, a professional cleaning company was hired to clean the restaurant, and it was reopened the next day. No further cases of illness were reported as of May 18.
Reported by: J Kettlehut Payne, MPH, M Hall, MD, M Lutzke, MPH, C Armstrong, J King, Kent County Health Dept, Grand Rapids, Michigan.
The outbreak investigation described in this report underscores the challenges associated with preventing norovirus transmission. Small restaurants might have difficulty operating when an employee is absent and might not be able to afford paying leave for illness. However, employees who become ill and continue to work can place the public's health at risk.
The results of these investigations suggest that the illness of one food handler might have been linked to the illnesses of at least 100 persons in multiple settings. Illnesses at a publishing company, school, social service group, and among members of the public resulted in closure of a warehouse, employee absences, pay for substitute teachers, loss of wages, and loss of revenue to the restaurant during a week-long closure.
These outbreaks demonstrate a general lack of education regarding norovirus. Restaurant owners in Michigan are required to review the state food code; however, no free training was offered to them to interpret this lengthy legal document. To facilitate education, online training programs for food handlers and managers are being developed by Kent County public health sanitarians.
Previously, no work exclusion requirements in the Michigan food code specifically targeted norovirus. An employee was required to report Salmonella, Shigella, Escherichia coli 0157:H7, and hepatitis A infections or illnesses such as diarrhea, vomiting, fever, jaundice, or sore throat with fever. Employees were required to provide written medical documentation that they were free from the four specified infectious agents via stool testing (3); no such requirement existed for norovirus. However, the 2005 FDA Food Code included norovirus containment recommendations for states requiring food service employees to be excluded from work if symptomatic with vomiting or diarrhea and, if they have been ill with suspected norovirus, not to return to work until they have been asymptomatic for 24 hours (2). In one study, viral shedding began approximately 15 hours after exposure to norovirus and peaked 25--72 hours after exposure (4). However, in this outbreak, one specimen tested positive for norovirus 14 days after exposure. The new Michigan guidelines, issued in 2005, recommend that employees who have been ill with suspected norovirus not return to work for 48--72 hours after symptoms have ended.
Specific education on norovirus containment for food handlers should be provided, even when not documented in a state food code. The Michigan guidelines offer recommendations for cleaning and disinfection that are based on recent studies and go beyond previous guidelines (5), indicating the concentration of bleach required for cleaning various porous and nonporous surfaces. The Michigan guidelines also supply a list of other effective and ineffective disinfectants. The restaurant described in this report was not cleaned appropriately until after it had been closed for nearly 1 week. Whether containment of norovirus resulted from the final restaurant cleaning or whether the virus was no longer viable could not be determined and offers an opportunity for future study.
This report is based, in part, on data provided by D Kraker, A Kramer, S Green, D Smith, Kent County Health Dept; and JP Massey, PhD, Michigan Dept of Community Health.
Disclaimer All MMWR HTML versions of articles are electronic conversions from ASCII text into HTML. This conversion may have resulted in character translation or format errors in the HTML version. Users should not rely on this HTML document, but are referred to the electronic PDF version and/or the original MMWR paper copy for the official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.**Questions or messages regarding errors in formatting should be addressed to email@example.com.
Date last reviewed: 4/13/2006