Nov 21, 2007 (CIDRAP News) Sick restaurant workers, including one who vomited in the kitchen, and inadequate cleaning products contributed to a norovirus outbreak at a Michigan restaurant in 2006 that sickened at least 364 customers, the US Centers for Disease Control (CDC) reported today.
An account of the outbreak investigation, handled by the Barry-Eaton District Health Department (BEDHD), appears in the Nov 23 issue of Morbidity and Mortality Weekly Report (MMWR).
The BEDHD first learned of the outbreak on Jan 30, when several customers reported gastrointestinal illnesses after having dined at the national chain restaurant 2 days earlier. Soon after launching epidemiologic and environmental investigations, the department identified an index case, a server who was sick with vomiting on about Jan 19 but did not work while ill. The worker's sibling, who tended bar and handled some administrative duties at the restaurant, became ill with gastrointestinal symptoms 2 days later and worked on the first and second illness days.
Norovirus genogroup I (GI) was detected in stool samples from several customers and workers, the report says.
On Jan 28, a line cook who was sick before his or her shift vomited into a waste bin near the frontline work station while on duty. The same employee worked the next day while still ill with loose stools.
The health department contacted as many restaurant patrons as possible who ate at the restaurant between Jan 19 and Feb 3. Several others came forward after media reports publicized the outbreak. Investigators also interviewed the restaurant staff.
The BEDHD investigation included two studies: a descriptive study to characterize the people who became ill and an analytic study using case-control methods to determine if certain foods were linked to the illness.
Of 584 people who were interviewed, 364 met the descriptive study's case definition. Illness onset peaked on Jan 30, when the BEDHD was notified about the outbreak. Investigators found that norovirus transmission was already occurring before the cook vomited in the kitchen on Jan 28, and the illness attack rate was highest for customers who ate a few hours after the incident.
Of 32 restaurant employees who were interviewed, 17 said they had worked while they were sick, and 12 of them became ill after working on Jan 28, the day the cook vomited in the restaurant. The illness struck a greater percentage of cooks than of servers who worked that day63% versus 29%.
The analytic arm of the study, which involved 45 restaurant patrons who had been sick and 91 controls, found the antipasti platter and garlic mashed potatoes were associated with illness.
Stool samples tested by polymerase chain reaction were positive for norovirus GI; companion bacterial tests were negative.
An environmental assessment revealed problems with:
- Hand-washing methods used by employees
- Cleaning and sanitizing of food-contact and nonfood-contact surfaces
- Monitoring of food temperatures and maintenance of proper conditions for potentially hazardous foods
- Maintenance of hand-sink stations for employees.
Several interventions were launched on Jan 30, the day the BEDHD learned of the outbreak. All of the food prepared between Jan 27 and 30 was discarded, sick employees were barred from the restaurant for at least 72 hours after their symptoms subsided, and the facility was thoroughly cleaned.
However, 4 days later the department received reports of three more customers who became ill after eating at the restaurant on Feb 1.
BEDHD sanitarians discovered the restaurant had used a quaternary ammoniumbased sanitizer that was ineffective against norovirus. They directed the restaurant to disinfect the premises with bleach solution as specified in state guidelines.
In the report, the CDC emphasizes that a vomiting incident can double the norovirus attack rate in an outbreak and that the line cook's vomiting might have contributed to disease transmission. Contaminated surfaces and food likely contributed to transmission as well, the agency says.
The authors caution that some quaternary ammoniabased disinfectants claim they are effective against norovirus, but these products also contain alcohol and base their claims on laboratory studies of effectiveness against a proxy virus such as feline cailicvirus.
"In this outbreak, the restaurant's use of cleaning cloths soaked with quaternary ammoniumbased cleaning product likely was ineffective in disinfecting the restaurant," the report states.
It says the outbreak was probably linked to a single infection source, because all stool samples tested positive for one norovirus groupGI/4.
CDC. Norovirus outbreak associated with ill food-service workersMichigan, January-February 2006. MMWR 2007;56(46):1212-16 [Full text]