ABSTRACT

The economic burden of foodborne illness has been estimated to be as high as US$90 billion annually. For policy purposes, it is often important to understand not only the overall cost of illness but also the costs associated with individual products or groups of products. In this study, I estimate the cost of foodborne illnesses from 29 pathogens associated with nongame meat and poultry products that are regulated by the U.S. Department of Agriculture. To complete this, I merge results from a food attribution model with results from an illness model and an economic burden of illness model. The food attribution model uses outbreak and expert elicitation data to attribute foods to pathogens. The illness model is a replication of the 2011 study published by the Centers for Disease Control and Prevention. The economic cost model is an updated version of previously published studies that include costs for medical care, lost productivity, loss of life, and pain and suffering. The primary attribution model, based largely on Interagency Food Safety Analytics Collaboration assumptions, estimates that meat and poultry products are vectors for 30.9% of all foodborne illnesses. This translates into 2.9 million annual illnesses, yielding economic costs of up to$20.3 billion. The costliest food-pathogen pairs include Campylobacter spp. in poultry ($6.9 billion), Salmonella spp. in chicken and pork ($2.8 and $1.9 billion, respectively), and Toxoplasma gondii in pork ($1.9 billion). Results based on alternative attribution and economic model assumptions are also presented, generating meat and poultry attribution estimates ranging from 27.1 to 36.7% and economic costs of $8.1 to$22.5 billion.

HIGHLIGHTS
• Meat and poultry are vectors for 30.9% of foodborne illnesses and 46.6% of costs.

• Beef and chicken are associated with the largest numbers of illnesses.

• Meat- and poultry-related illnesses lead to economic costs of up to US\$20.3 billion.

• Costs are highest for Campylobacter, Salmonella, and Toxoplasma gondii.

• Results from alternative modeling assumptions affect burden of illness estimates.