Analysing foodborne illness outbreak severity in the USA, 2009–19

Abstract

Background: In the USA, one in six Americans have had a foodborne illness, resulting in approximately US$17·6 billion (2018) of losses annually attributed to medical costs, productivity losses, and economic burden due to death. Previous research has characterised these burdens with use of various methods, including disability-adjusted life-years (DALYs), medical care costs, and revenue losses of food retailers. However, these indices assess only the socioeconomic burdens of outbreaks without providing an index to compare the health-related severity of outbreaks regarding geographical spread of transmission, outbreak duration, and the morbidity and mortality of case outcomes. Methods: In this study, we proposed a data-driven index for evaluating outbreak severity using the publicly available US Centers for Disease Control and Prevention's National Outbreak Reporting System (NORS) database from Jan 1, 2009, to Dec 31, 2019. We identified 67 metrics related to the geographical extent (eg, counties or states where individuals with a foodborne illness resided and were exposed), case severity (eg, primary cases, hospital admissions, emergency room visits, and deaths), duration (eg, days of illness and days of exposure events), and their ratios. We selected 11 metrics to create the index on the basis of completeness and distributional characteristics of these metrics. We generated index scores by normalising natural log-transformed values on a 0–1 scale and averaging all metrics per outbreak. Scores ranged from 0 (lowest severity) to 1 (highest severity). Findings: NORS reported 9407 outbreaks in our 11-year study period. We found that bacterial outbreaks associated with the genera Streptococcus, Listeria, and Salmonella had the highest median severity scores, whereas Bacillus, Staphylococcus, and Clostridium had the lowest scores. Among protozoa, Cryptosporidium and Cyclospora had the lowest median scores, whereas Giardia and Toxoplasma had the highest scores. Hepatitis had the highest median score for virus-associated outbreaks, whereas Norovirus, which had the most outbreaks reported in our study period (n=3088), had the lowest score. Interpretation: Our index illustrates the usefulness of publicly reported surveillance data in establishing a data-driven approach for monitoring the severity of foodborne outbreaks regarding their spread, duration, and morbidity and mortality. Differences in outbreak severity by pathogen inform how to best allocate resources for monitoring, tracking, and preventing foodborne disease outbreaks. Being data driven, our index emphasises the importance of improved data quality, integrity, and completeness of surveillance records. Future research should strive to validate this index by use of longitudinal cohort studies. Funding: US Department of Defense, National Institute of Food and Agriculture.