Multistate Outbreak of Norovirus Gastroenteritis Among Attendees at a Family Reunion --- Grant County, West Virginia, October 2006
On October 17, 2006, the West Virginia Department of Health and Human Resources (WVDHHR) was notified of an outbreak of acute gastroenteritis, characterized by vomiting and diarrhea, among attendees at a family reunion. The outbreak initially was reported by a group of attendees to their local health department in Garrett County, Maryland. The same day, the information was relayed to the Grant County Health Department in West Virginia and subsequently to WVDHHR. The reunion was held on October 14 at a private residence in Grant County, West Virginia, and the 53 identified attendees included residents from Florida, Maryland, New York, Pennsylvania, Virginia, and West Virginia. This report describes a collaborative, multijurisdictional epidemiologic investigation using a cohort study and laboratory analyses to determine the source of infection and appropriate control measures. The results indicated that a combination of person-to-person and foodborne transmission of two strains of norovirus, likely introduced by persons from two different states and subsequently at least two food items, was the probable cause of these illnesses, highlighting the challenge of investigating and controlling norovirus outbreaks. During periods of peak norovirus activity, public health officials should emphasize the importance of appropriate handwashing and the exclusion of ill persons from social gatherings.
In collaboration with state and local health departments, interviews were conducted with 11 reunion attendees to help generate hypotheses and develop a list of attendees and foods served. A questionnaire was then developed to conduct a cohort study involving all reunion attendees. Questions addressed illness onset, symptoms, attendance at prereunion gatherings, consumption of specific food items, contact with ill persons, and onset of symptoms among nonattendees. Questionnaires were administered by telephone and in person by state and local health department staff members from West Virginia and Maryland in coordination with health departments from the other attendee jurisdictions in Florida, New York, Pennsylvania, and Virginia.
An attendee case was defined as two or more episodes of nonbloody diarrhea (i.e., two or more loose stools in a 24-hour period) or vomiting within a single 24-hour period on or after October 7, 2006, in a person who attended the reunion. A nonattendee case was defined as acute illness characterized by vomiting or diarrhea with onset after 12 a.m. on October 18 in persons who did not attend the reunion but who had direct contact (i.e., within 3 feet) with attendees after the reunion.
The list of reunion attendees included 53 persons, of whom 48 (91%) were interviewed. Of those interviewed, 28 (58%) had illness that met the attendee case definition. In addition, four cases were identified among nonattendees, all of whom were household contacts of attendees. Symptoms reported by the 28 ill attendees included diarrhea (96%), vomiting (75%), abdominal cramps (71%), nausea (61%), headache (54%), chills (36%), body aches (32%), fever (not specified) (21%), and fatigue or malaise (18%). Nineteen (68%) of the 28 ill attendees were female, and six (21%) were aged <10 years. Six (21%) of the patients sought medical care. For the 25 patients who reported both date of illness onset and date of recovery, the median duration of illness was 54 hours (range: 6--135 hours). Twenty-one of the 28 attendee cases occurred during October 14--16 (Figure).
The 1-day reunion began at 11 a.m. on October 14. Persons with illness onset after 8 p.m. on October 14 through 12 a.m. on October 18 were included in the cohort study, as were persons who attended but did not become ill (Figure). Persons with illness onset either before the reunion or after 12 a.m. on October 18 were excluded. Incubation periods were calculated by subtracting the date and time of the first possible exposure from the date and time of illness onset. The first possible exposure was defined as either the time the person arrived at the reunion or the time the person arrived at a prereunion gathering where previously ill persons were present. Nine of the 48 interviewed attendees were excluded from the cohort study because they did not meet the defined illness-onset criteria. Three had illness onset >72 hours after the reunion. Six attendees had illness onset either before the reunion or within 6 hours after the reunion began and might have introduced the illness into the reunion; four of these six were immediate family members from New York who had traveled to the reunion together, including a child who was ill with vomiting and diarrhea during the reunion, and the other two were West Virginia residents who had no contact with each other or the family from New York immediately before the reunion.
Of the 39 attendees included in the cohort study, 19 met the case definition and illness-onset criteria, and 20 did not become ill. The median incubation period for the 19 cases was 36 hours (range: 20--61 hours). Of 31 food items served at the reunion (Table 1), two items were identified as significant risk factors for developing illness (p<0.05, by two-tailed Mantel-Haenszel chi-square test) and were eaten by the majority of ill persons: scalloped potatoes (relative risk [RR] = 2.8, 95% confidence interval [CI] = 1.1--6.9) and chicken (RR = 2.2, CI = 1.0--4.8). Both food items were eaten at the reunion by persons who were ill before the reunion, which might have provided an opportunity for these persons to contaminate the food at the event. The chicken was purchased at a store by the family from New York, whose four members had been ill before the reunion, which provided another opportunity for the food to be contaminated. The scalloped potatoes were brought by persons from West Virginia who were not ill before the reunion. Consumption of the chocolate cheese ball also was statistically associated with illness (p = 0.04), but the item was only eaten by seven persons. In addition, six of the seven attendees who ate the chocolate cheese ball also ate both the chicken and scalloped potatoes; all seven ate the chicken. Self-reported direct contact with ill persons at the reunion, including with the symptomatic child, also was a significant risk factor for developing illness (RR = 2.3, CI = 1.0--5.1). Attendance at prereunion gatherings at either home A or home B was not associated with illness. Reunion attendees were provided information on appropriate hand hygiene and the potential for viral shedding and secondary transmission up to 2 weeks after symptoms resolved.
In coordination with state and local health departments, reunion attendees were encouraged to submit stool or vomitus samples to their respective local health departments. Stool specimens were submitted by 13 ill reunion attendees from Pennsylvania, Maryland, New York, and West Virginia, and the specimens were then submitted to the respective state laboratories for analysis.* No vomitus samples were analyzed. Norovirus reverse transcription--polymerase chain reaction (RT--PCR), genotype sequencing analyses, and enteric bacterial cultures were performed by the Maryland, New York, and Pennsylvania state laboratories. Initial genogroup assignment was made by differential probe binding. Results were compiled and compared to identify specific etiologic agents involved in the outbreak. No environmental samples were collected.
Of the 13 stool specimens submitted (six from Pennsylvania residents, three from Maryland, three from New York, one from West Virginia, and none from Florida or Virginia), 12 (92.3%) tested positive for norovirus genogroup II by RT--PCR (Table 2). Using genetic sequencing of the RT--PCR products from norovirus region B and comparison with GenBank,§ the closest match for the strain detected was identified by each state laboratory. The same strain (Hu/GII-4/Chester/2006/UK) was identified in the two sequenced norovirus-positive specimens from Maryland, the two positive specimens from New York, and the one positive specimen from West Virginia. A second strain (Hu/NLV/Oxford/B6S6/2003/UK) was identified in all six positive specimens from Pennsylvania. No differences in exposures between persons infected with the two different strains could be identified. No other etiologies (e.g., bacterial) were identified.
Reported by: S Glasscock, Grant County Health Dept; J Welch, Preston County Health Dept; J Dailer, W Elmer, Randolph County Health Dept; K Kline, Pendleton County Health Dept; D Bixler, MD, M del Rosario, MD, M Myers, West Virginia Dept of Health and Human Resources. T Buckel, E Cvetnick, Garrett County Health Dept; R Myers, PhD, Maryland Dept of Health and Mental Hygiene. A Johnson, Chautauqua County Health Dept; B Rosen, PhD, New York State Dept of Health. B Perry, Pennsylvania Dept of Health. A Hall, DVM, EIS Officer, CDC.
Noroviruses are the most common cause of gastroenteritis in the United States, with an estimated 23 million cases occurring annually (1--3). The average incubation period for norovirus is 24--48 hours, and clinical disease is characterized by acute onset of vomiting, nonbloody diarrhea, or both, lasting 12--60 hours (4). In clinical studies, approximately two thirds of persons infected with norovirus experienced symptoms of disease (5). The primary route of transmission for noroviruses is fecal-oral, including consumption of fecally contaminated food or water, direct person-to-person contact, and contaminated objects or environments (4,5). Airborne transmission via vomitus droplets also can occur (4,5). During outbreaks, primary cases often result from exposure to a fecally contaminated food item, object, or environment, whereas secondary cases result from person-to-person transmission (6). Noroviruses and norovirus infections have numerous characteristics that facilitate their spread during outbreaks, including the low dose required for infection; prolonged, asymptomatic shedding that can occur in infected persons; environmental stability of the virus; and lack of lasting immunity in persons who have been infected previously (4). Molecular epidemiologic techniques have identified substantial strain diversity, and epidemic strains of norovirus might be more virulent or more environmentally persistent than nonepidemic strains (7).
This outbreak highlights the challenges of investigating and controlling norovirus outbreaks, including multiple modes of transmission. The findings of this investigation, including the detection of two different norovirus strains in patients, suggest that illness was independently introduced into the reunion by several sources (i.e., persons from New York and from West Virginia). Food items might have been contaminated by persons who were ill when they attended the reunion. Infection likely was propagated through a combination of person-to person contact and foodborne transmission; transmission through contaminated fomites cannot be ruled out. Laboratory evidence confirmed that at least two different norovirus strains were circulating among attendees. The convergence of two virus strains in a single outbreak coincided with a period of high norovirus activity in the region. During October--December 2006, a total of 20 other outbreaks of acute gastroenteritis in West Virginia were reported to WVDHHR, representing a sevenfold increase in the number reported during the same period in 2005.
Prevention and control of norovirus outbreaks, especially during periods of increased norovirus circulation, should emphasize standard infection-control practices, including the exclusion of ill caregivers and food handlers from work settings and exercising adequate hand hygiene (8). Persons who have had gastroenteritis recently should pay attention to washing their hands after toileting and should not prepare food. Food items that might have been contaminated by persons with gastroenteritis should be discarded. As demonstrated by this outbreak, collaboration among multiple state and local health departments often is required for prompt public health investigations of norovirus outbreaks, which can be complicated by multiple sources, viral strains, and routes of this highly transmissible infection.
The findings in this report are based, in part, on contributions by Tucker County Health Dept; C Clark, S Arrington, MA, L Clay, S Comstock, S Hill, T Shwe, MPH, S Stowers, S Wilson, MPH, West Virginia Dept of Health and Human Resources; A Weltman, MD, Pennsylvania Dept of Health; K St George and DNA Sequencing Core, New York State Dept of Health; Washington County Health Dept, Maryland; Virginia Dept of Health; Florida Dept of Health; D Bensyl, PhD, Office of Workforce and Career Development, T Chen, MD, G Mirchandani, PhD, and J Schaffzin, MD, EIS officers, CDC.
* Three of the four New York family members who were ill before the reunion submitted stool samples; neither of the two persons from West Virginia who were ill submitted a stool sample. The specimen from one West Virginia resident was analyzed at the Maryland state laboratory because of assay availability.
RT--PCR primers targeted region B of the viral genome, which includes the polymerase gene commonly used for genetic classification.
§ Genetic sequence database maintained by the National Institutes of Health (http://www.ncbi.nlm.nih.gov/Genbank/index.html).
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: 7/11/2007