Purpose: To develop the probabilistic and numerical data for use in a quantitative predictive risk assessment model (NorOPTIMAL) designed to simulate the spread of human norovirus contamination and disease in LCTFs, with a focus on food and surfaces.
Methods: Key model inputs were defined for the NorOPTIMAL model, and those terms were searched in relevant electronic databases (e.g., PubMed, IngentaConnect, norocorelit.com) to identify peer-reviewed literature. Final input values were derived using data from the literature review, research conducted by the NoroCORE food safety initiative, and expert judgment.
Results: Input values and distributional data were developed for five categories of information in NorOPTIMAL, including 1) Agents (e.g., residents, health care workers, kitchen staff) and their behaviors (e.g., hand washing, food washing compliance); 2) Hazard (e.g., dose-response, food washing removal); 3) Symptoms (e.g., number of vomiting episodes, shedding rates); 4) Timeline of Activities (e.g., food preparation, cleaning events), and 5) Cost of Interventions (e.g., cost of equipment, training). For example, sanitation/disinfection compliance of food utensils varies between 0 to 91%, average 61%. Transfer of norovirus between an agent and food serving is 0 to 12% for dry conditions, and 20 to 70% for wet conditions. Finally, washing food reduces norovirus by 0.39 to 1.26 log, average 1.01 log.
Significance: This work provides a unique compilation of data describing agents and the microenvironment that determine the contamination, spread, and disease burden for human norovirus in the LTCF setting, including food preparation. These data will be used to populate the model so as to evaluate the efficacy of candidate intervention strategies and inform development of effective food safety plans.