CONTEXT: Each year, foodborne diseases cause an estimated 48 million illnesses resulting in 128000 hospitalizations and 3000 deaths in the United States. Fast and effective outbreak investigations are needed to identify and remove contaminated food from the market to reduce the number of additional illnesses that occur. Many state and local health departments have insufficient resources to identify, respond to, and control the increasing burden of foodborne illnesses. PROGRAM: The Centers for Disease Control and Prevention (CDC) Foodborne Diseases Centers for Outbreak Response Enhancement (FoodCORE) program provides targeted resources to state and local health departments to improve completeness and timeliness of laboratory, epidemiology, and environmental health activities for foodborne disease surveillance and outbreak response. IMPLEMENTATION: In 2009, pilot FoodCORE centers were selected through a competitive application process and then implemented work plans to achieve faster and more complete surveillance and outbreak response activities in their jurisdiction. By 2019, 10 centers participated in FoodCORE: Colorado, Connecticut, Minnesota, New York City, Ohio, Oregon, South Carolina, Tennessee, Utah, and Wisconsin. EVALUATION: CDC and FoodCORE centers collaboratively developed performance metrics to evaluate the impact and effectiveness of FoodCORE activities. Centers used performance metrics to document successes, identify gaps, and set goals for their jurisdiction. CDC used performance metrics to evaluate the implementation of FoodCORE priorities and identify successful strategies to develop replicable model practices. This report provides a description of implementing the FoodCORE program during year 1 (October 2010 to September 2011) through year 9 (January 2019 to December 2019). DISCUSSION: FoodCORE centers address gaps in foodborne disease response through enhanced capacity to improve timeliness and completeness of surveillance and outbreak response activities. Strategies resulting in faster, more complete surveillance and response are documented as model practices and are shared with state and local foodborne disease programs across the country.

Current surveillance methods may not capture the full extent of COVID-19 spread in high-risk settings like food establishments. Thus, we propose a new method for surveillance that identifies COVID-19 cases among food establishment workers from news reports via web-scraping and natural language processing (NLP). First, we used web-scraping to identify a broader set of articles (n = 67,078) related to COVID-19 based on keyword mentions. In this dataset, we used an open-source NLP platform (ClarityNLP) to extract location, industry, case, and death counts automatically. These articles were vetted and validated by CDC subject matter experts (SMEs) to identify those containing COVID-19 outbreaks in food establishments. CDC and Georgia Tech Research Institute SMEs provided a human-labeled test dataset containing 388 articles to validate our algorithms. Then, to improve quality, we fine-tuned a pretrained RoBERTa instance, a bidirectional transformer language model, to classify articles containing ≥ 1 positive COVID-19 cases in food establishments. The application of RoBERTa decreased the number of articles from 67,078 to 1,112 and classified (≥ 1 positive COVID-19 cases in food establishments) articles with 88% accuracy in the human-labeled test dataset. Therefore, by automating the pipeline of web-scraping and COVID-19 case prediction using RoBERTa, we enable an efficient human in-the-loop process by which COVID-19 data could be manually collected from articles flagged by our model, thus reducing the human labor requirements. Furthermore, our approach could be used to predict and monitor locations of COVID-19 development by geography and could also be extended to other industries and news article datasets of interest. © 2021, Springer Nature Switzerland AG.