BACKGROUND: Approximately eight of 2000 Salmonella infections reported annually in Georgia are serotype Thompson. During April 2016, the Georgia Public Health Laboratory (GPHL) reported a cluster of 12 S. Thompson isolates. An outbreak investigation was initiated (Outbreak 1). During October 2016, three illnesses (one confirmed salmonellosis) were associated with Restaurant A. Another outbreak investigation was initiated (Outbreak 2).
METHODS: Epidemiologists attempted to interview all Georgia S. Thompson case-patients reported from April 2016 forward using standard Salmonella and restaurant questionnaires. Outbreak 1 included a case-control study to associate Restaurant A with illness. Cohort studies were performed for both outbreaks. Cobb County Environmental Health (EH) performed environmental assessments, implemented control measures, and collected environmental and employee specimens during both investigations. Food samples were collected for Outbreak 2. Monthly testing of all employee and environmental samples began following Outbreak 2. GPHL performed culture, serotyping, and pulsed-field gel electrophoresis (PFGE) on clinical and food specimens and isolates; whole genome sequencing (WGS) is ongoing. GPHL and the Georgia Department of Agriculture Laboratory tested environmental samples.
RESULTS: Outbreak 1 included 51 cases, 11 hospitalizations, and no deaths. Outbreak 2 included 47 cases, six hospitalizations, and no deaths. Salmonellosis was associated with Restaurant A in the Outbreak 1 case-control study (p<0.05). Chicken consumption was epidemiologically associated with illness in Outbreak 2 (p<0.05). During Outbreak 1, EH documented improper cooling procedures, inadequate sanitization records, and preparation surfaces used for both raw meat and vegetables. The facility underwent functional and structural changes. The facility closed during Outbreak 2. New standard operating procedures were implemented. Salmonella-positive employees were excluded from work pending two negative stool cultures. PFGE patterns from Outbreaks 1 and 2 were distinct. WGS results are pending. Environmental samples were negative for Salmonella. One food specimen (raw chicken) was S. Thompson-positive. Three of 12 and eight of 13 employees were S. Thompson-positive in Outbreaks 1 and 2, respectively. Employee A tested positive during both outbreaks. Employee B was intermittently positive during and after Outbreak 2. Employee C wasn’t tested during Outbreak 1 and persistently tested positive over 2 months after Outbreak 2.
CONCLUSIONS: In 2016, two Salmonella Thompson outbreaks associated with Restaurant A resulted in almost 100 illnesses. While a food vehicle was identified, we cannot rule out a role for food worker carriage as a contributing factor. Non-traditional control measures, including employee and environmental testing for 3 months or more may be needed for persistent foodborne outbreaks.