Background. Foodborne disease is estimated to cause 48 million illnesses annually in the US resulting in 3000 deaths [1]. Although most infections occur as sporadic cases, outbreak surveillance offers valuable insight about the foods and pathogens responsible for illnesses [2]. A total of 1632 foodborne disease outbreaks were reported during 2011-2012 [3] and recent data indicates an overall decrease in the number of outbreaks reported each year [4]. Understanding which factors contribute to the successful identification of a food vehicle in a foodborne outbreak investigation is crucial for improving outbreak response [5-10]. The purpose of this study was to describe outbreak characteristics and to determine which may be associated with the success of a foodborne outbreak investigation (i.e. one in which a food vehicle has been reported). Methods. A foodborne disease outbreak was defined as the event in which two or more people acquired similar illnesses from consuming the same food or beverage. Outbreaks occurring in FoodNet sites during 2003 through 2010 were included in the analysis. Results. Data were available for 1441 (87%) of the 1655 foodborne disease outbreaks documented in FoodNet Outbreak Supplement forms from 2003 through 2010. A food vehicle was identified in 692 of the 1441 (48%) outbreaks. Six outbreak characteristics remained statistically significant in both univariate and multivariate analyses: environmental and/or food culture collection, FDA or state agriculture involvement, outbreak size, case-control studies, and number of fecal specimens tested for norovirus. Conclusions. Less than half of foodborne outbreaks examined here resulted in a food vehicle being identified. Having more robust resources available for outbreak detection and investigation may improve likelihood of a food vehicle being identified.

Foodborne illnesses continue to have a negative impact on the nation's health, accounting annually for an estimated 76 million illnesses, 325,000 hospitalizations, and 5,000 deaths in the United States. Syndromic surveillance systems that analyze pre-diagnostic data, such as pharmaceutical sales data are being used to monitor diarrheal disease. The purpose of this study is to evaluate the usefulness of a poison control center (PCC) data collection and triage system for early detection of increases in foodborne illnesses.Data on calls to the Arizona Poison and Drug Information Center (APDIC) reporting suspected foodborne illnesses, and Pima County Health Department (PCHD) enteric illness reports were obtained for July 1, 2002 - June 30, 2007. Prediction algorithms were constructed using the first two and a half years, and validated in the remaining two and a half years. Multiple outcomes were assessed using unadjusted and adjusted raw counts, five and seven day moving averages, and exponentially weighted moving averages. Sensitivity analyses were conducted to evaluate model performance. Increases in PCHD laboratory reports of enteric illnesses were used as a proxy measure for foodborne disease outbreaks.Over the five year study period there were 1,094 APDIC calls reporting suspected foodborne illnesses, and 2,433 PCHD enteric illness cases. Seventy-five percent of cases were reported to PCHD within 23 days of symptom onset. In contrast, 62% of callers contacted APDIC within 24 hours of symptom onset. Forty percent of PCHD cases were missing symptom onset dates, which necessitated constructing and validating predictive algorithms using only those PCHD cases with known symptom onset dates.None of the prediction models performed at sensitivity levels considered acceptable by public health department standards. However, it is possible that a temporal relationship actually exists, but data quality (lack of outbreak dates, and missing symptom onset dates) may have prevented its detection. The study suggests that current surveillance by PCCs is insufficient as a univariate model for syndromic surveillance of diarrheal illness because of low caller volume reporting suspected foodborne illnesses; this can be improved. Methods were discussed to utilize PCCs for active surveillance of foodborne illnesses that are of public health significance.