Salmonella Infections Among Employees of a Turkey Hatchery in Minnesota

BACKGROUND:  In October 2014, the Minnesota Department of Health (MDH) identified a case with a Salmonella I 4,5,12:i:- (MDH pulsed-field gel electrophoresis [PFGE] pattern designation TM918) infection who reported working at a turkey hatchery before illness onset. An investigation was initiated when the case reported other ill employees, and two additional S. I 4,5,12:i:- TM918 cases who reported working at the hatchery were identified through routine surveillance.

METHODS:  A case-control study design was used. A case was defined as a hatchery employee who had a laboratory-confirmed Salmonella I 4, 5, 12: i:-, MDH PFGE pattern TM918 infection, or who reported fever and diarrhea (≥3 loose stools in a 24-hour period) or diarrhea of ≥3 days duration with illness onset since July 1. A control was defined as a hatchery employee who did not report illness. Cases and controls were interviewed about employment length, illness history, job tasks, personal protective equipment (PPE) usage, handwashing practices, and PPE training. Odds ratios and 95% confidence intervals were calculated using logistic regression for binomial variables. T-tests were performed for continuous variables. Stool samples from consenting cases were submitted to MDH for Salmonella testing.

RESULTS:  Forty-four (83%) of 53 hatchery personnel were interviewed; 12 (27%) interviewees met the case definition. Four were confirmed to have S. I 4,5,12:i:- TM918 infections.  In univariate analyses, egg setting (odds ratio [OR], 7.8; 95% confidence interval [CI], 1.45-41.8; p=0.02) and pulling the hatch (OR, 13.5; 95% CI, 2.74-66.5; p=0.001) were significantly associated with illness. In the multivariate model, only pulling the hatch was associated with illness (adjusted [aOR], 7.0; 95% CI, 1.02-48.1; p=0.048). Cases who pulled the hatch were more likely to have worked at the hatchery for less than 6 months. There were no significant associations between illness and any other job tasks, reported usage of PPE, or handwashing. The mean age of cases was 35.5 years (range, 24-56 years), and the mean age of controls was 44.8 years (range, 22-62 years) (p=0.04). Cases had worked at the hatchery for a mean of 1.1 years (range, 1 month-7 years), compared to a mean of 5.1 years for controls (range, 1 month-34 years) (p=0.006).

CONCLUSIONS:  The source of this salmonellosis outbreak was handling turkey poults or exposure to environments contaminated by poults. The hatchery took immediate action to increase biosecurity, improve PPE compliance and infection control procedures, and perform a trace-back investigation to breeder flocks supplying the hatchery.