Proactive Use of Supply Chain Data in Foodborne Illness Outbreak Investigation
Director, National Center for Food Protection and Defense, Assistant Professor, Veterinary Population Medicine, University of Minnesota
Unfortunately, traditional epidemiological foodborne outbreak investigations are generally forensic. They allow us to know what went wrong so that preventive controls can be put in place for the future and we know where to assign blame for the outbreak. However, these investigations do not allow us to intervene and help those who will become ill because, with episodic contamination, the majority of the contaminated food is usually consumed before the epidemiologic investigation has identified the vehicle. Traditional epidemiological investigations are only really a mitigation strategy (i.e., interventional) for systemic contamination events in which there is low-level contamination over an extended period of time. Part of the challenge of the traditional epidemiologic approach is that we need outbreaks to be recognized before epidemiologists can carry out a case control study (i.e., identifying possible causal factors by comparing ill individuals to nonsick individuals). However, our primary detection system is currently the emergency room. When the investigation begins, the epidemiologist has to do extensive interviews to find out which foods to consider in the case control study. The more foods included, the longer the study takes — an inherent conflict. There is an opportunity to dramatically simplify these investigations by utilizing private sector supply chain data, but this requires strong public-private partnerships. In essence, by comparing the illness pattern with the specific product distribution patterns, one could identify which products are possible contamination vehicles. Using that data in a meaningful way, however, requires near real-time analysis of vast amounts of information. New approaches on how data from competitors could be combined without compromising proprietary business information or exposing companies to additional regulatory risk must be identified.