Deploying the Model as a Surveillance and Decision-Support Tool for U.S. Food Safety

Abstract

This study has examined the deployment of a model as a surveillance and decision-support tool for U.S. food safety, addressing the problem that food safety systems generate large volumes of surveillance, laboratory, inspection, recall, complaint, and supply-chain data, but these data are not always integrated into timely and actionable decision-support outputs. The purpose of the study has been to determine whether model deployment capability, data integration quality, predictive surveillance readiness, decision-support usability, and organizational preparedness improve food safety surveillance effectiveness and food safety decision-making performance. The study has followed a quantitative, cross-sectional, case-based design within the U.S. food safety surveillance and enterprise decision-support context. Data have been collected from 246 valid respondents, representing an 82.0% valid response rate from 300 distributed questionnaires, including food safety and quality assurance officers, public health monitoring personnel, regulatory compliance and inspection personnel, laboratory staff, supply-chain safety personnel, and data or risk-analysis personnel. The analysis plan has included descriptive statistics, reliability analysis, correlation analysis, regression modeling, mediation analysis, a Model-Readiness Index, and a Surveillance-to-Decision Conversion Assessment. The findings have shown high agreement across the major variables, with data integration quality recording the highest mean score of 4.18, followed by food safety surveillance effectiveness at 4.14, predictive surveillance readiness at 4.11, food safety decision-making performance at 4.07, model deployment capability at 4.05, decision-support usability at 3.98, and organizational preparedness at 3.92. Reliability has been strong, with Cronbach’s alpha values ranging from .82 to .91 and overall instrument reliability at .94. Regression results have shown that readiness variables explained 63.0% of the variance in surveillance effectiveness, while surveillance effectiveness and readiness variables explained 67.0% of the variance in decision-making performance. Surveillance effectiveness has also partially mediated the relationship between model deployment capability and decision-making performance, with an indirect effect of .19. The Model-Readiness Index has been 4.05, and the Surveillance-to-Decision Conversion Rate has been 98.31%. These findings have implied that U.S. food safety organizations should strengthen data integration, predictive readiness, usability, and organizational preparedness to convert surveillance signals into faster, evidence-based food safety decisions.