Causal Mediation Analysis of Foodborne Salmonella Outbreaks in the United States: Serotypes and Food Vehicles

Abstract

Various Salmonella serotypes have caused numerous foodborne outbreaks associated with food vehicles in different categories. This study provides evidence on the occurrence and inter-relations between Salmonella serotypes and the number of deaths mediated by the number of illnesses and hospitalizations. Confirmed foodborne outbreaks of Salmonella serotypes (n = 2868) that occurred between 1998 and 2021 were obtained from the Centers for Disease Control and Prevention National Outbreak Reporting System. Causal mediation analysis was performed based on 500 bootstrap samples. The serotypes and the Interagency Food Safety Analytics Collaboration (IFSAC) food categories as confounding effects were considered as categorical variables. A total of 106 single Salmonella serotypes were associated with foodborne outbreaks. Foodborne outbreaks caused by Salmonella serotypes resulted in 81,996 illnesses, 11,018 hospitalizations, and 115 deaths between 1998 and 2021 in the United States. The serotypes Enteritidis (815 outbreaks, 28.42%), Typhimurium (359 outbreaks, 12.52%), and Newport (220 outbreaks, 7.67%) accounted for almost half of Salmonella-linked outbreaks. Poultry products, chickens, eggs, and turkey, were the leading IFSAC food categories, accounting for 14.02% of total outbreaks and 10.44% of total deaths. Certain serotypes had a significant effect on illness, hospitalization, and death counts. Two serotypes, Heidelberg and Saintpaul, and fruits as the food vehicle in IFSAC categories had a significant direct effect on the number of illnesses, hospitalizations, and deaths as outcomes of Salmonella outbreaks (p <= 0.05). There was strong evidence that illness and hospitalization counts played a key role in the pathway from serotype to death counts on foodborne outbreaks caused by Salmonella based on causal mediation analysis. The findings of this study can help outbreak investigations and lead to prevention and control measures by providing insightful information about the frequencies of Salmonella serotypes and the associated food vehicles causing foodborne diseases.