Meta-analysis on the effect of interventions used in cattle processing plants to reduce Escherichia coli contamination

Cattle coming from feedlots to slaughter often harbor pathogenic E. coli that can contaminate final meat products. As a result, reducing pathogenic contamination during processing is a main priority. Unfortunately, food safety specialists face challenges when trying to determine optimal intervention strategies from published literature. Plant intervention literature results and methods vary significantly, making it difficult to implement interventions with any degree of certainty in their effectiveness. To create a more robust understanding of plant intervention effectiveness, a formal systematic literature review and meta-analysis was conducted on popular intervention methods. Effect size or intervention effectiveness was measured as raw log reduction, and modeled using study characteristics, such as intervention type, temperature of application, initial microbial concentration, etc. Least-squares means were calculated for intervention effectiveness separately on hide and on carcass surfaces. Heterogeneity between studies (I-2) was assessed and factors influencing intervention effectiveness were identified. Least-squares mean reductions (log CFU/cm(2)) on carcass surfaces (n = 249) were 1.44 [95% CI: 0.73-2.15] for acetic acid, 2.07 [1.48-2.65] for lactic acid, 3.09 [2.46-3.73] for steam vacuum, and 1.90 [1.33-2.47] for water wash. On hide surfaces (n = 47), least-squares mean reductions were 221 [1.36-3.05] for acetic acid, 3.02 [2.16-3.88] for lactic acid, 3.66 [2.60-4.72] for sodium hydroxide, and 0.08[-0.94-1.11] for water wash. Meta-regressions showed that initial microbial concentrations and timing of extra water washes were the most important predictors of intervention effectiveness. Unexplained variation remained high in carcass, hide, and lactic add meta-regressions, suggesting that other significant moderators are yet to be identified. The results will allow plant managers and risk assessors to evaluate plant interventions, variation, and factors more effectively.This material is based upon work that was supported by the National Institute of Food and Agriculture, U.S. Department of Agriculture, under award number 2012-68003-30155 (STEC CAP grant) and by Virginia Tech's “Water INTERface” Interdisciplinary Graduate Education Program. Also, Dr. Gonzales-Barron wishes to acknowledge the financial support provided by the Portuguese Foundation for Science and Technology (FCT) through the award of a five-year Investigator Fellowship (IF) in the mode of Development Grants (IF/00570).