An official website of the United States government.
<OL> <LI> Complete a quantitative review of salmonella which includes best available data on the prevalence and concentration of salmonella at all points along the pork production chain. <LI> Develop simulation and algebraic models of the chain, incorporating all available salmonella data and input-output relationships. <LI>Develop methods for integrating economic aspects related to alternative interventions into the risk framework and test the model with available data on prices and costs. <LI>Quantify how changes in the microbial load at various points in the chain would change Salmonella illness risks in the US population by computer simulations using user-friendly virtual engineering interface. <LI>Compare the costs of various interventions with the value of expected benefits (measured in terms of the value associated with changes in Salmonella-related illnesses.)
NON-TECHNICAL SUMMARY: This research will specifically address the question of how changes at various points in the pork production process affect the predicted number of human salmonellosis cases attributable to pork and the costs of alternative control interventions. These resulting change estimates (quantitative mitigation elasticity) and relative costs will provide meat industry management and food safety policy makers a tool to evaluate and determine priority areas for intervention that are indeed "risk-based" and cost effective. There is a shortage of data and information about how much of the human salmonellosis burden can be attributed to different food and non-food sources. Various estimates have suggested that about 9 to 15% of human cases are due to pork. (Hald et al. 2004). It is important that systems of salmonella typing and monitoring be further developed to improve this attribution. However, exact attribution is not critical, to answer the above risk management questions. The critical element needed is some estimation of how much human risk (e.g. cases or illness days per year) changes in response to changes in various salmonella impacting interventions (mitigation elasticity). (Ebel and Schlosser, 2000). With a model that quantitates these mitigation elasticities, the most cost effective, from a human risk standpoint, interventions can be identified.
<P>
APPROACH: These objectives will be accomplished by meta analysis of available data and development of two models. The proportion of human salmonella illness attributable to pork will be estimated by reconciliation of the forward and backward calculation models. The forward-calculation (farm-to-fork) will start with on-farm prevalence estimates and process them through slaughter, handling, etc. into a dose-response model that would predict the expected number of human cases. The resulting model will be programmed, by virtual engineering, into a user-friendly visual format with an intuitive input panel and graphic outputs. The backward calculation (clinic-to-farm) model will use the currently reported number of human salmonella cases and adjust back to an upper bound of those that could be a due to pork. The economic aspects will be developed to coordinate with the structure of the risk model and market organization, but focus on economic relationships and linkages between farm and processor. The retail sector (and links to the processor) will be through relatively limited mechanisms of price and cost transmission. The relative impact on human illness and the cost of various interventions at different points along the chain will be evaluated. This analysis will provide an estimate of the investment required for each case of illness prevented. It will give decision makers some priorities about where along the chain to most efficiently apply limited resources.