Purpose: The goal was to assess whether the microbial load of packing facility surfaces are associated with the microbial load of produce.
Methods: From November 2000 to December 2003, 487 matched produce (14 types) and equipment surfaces (6 production steps) were sampled from 8 packing facilities near the U.S. and Mexico border and enumerated for aerobic plate counts (APC), generic E. coli, enterococci, and coliforms. Bivariate correlations were assessed by Spearman’s rho and adjusted associations were assessed by multilevel mixed linear regression models.
Results: In general, the microbial loads on produce and equipment surfaces changed across production steps but did not consistently trend in any one direction. Equipment surface and produce microbial loads were correlated, but correlations varied from none to high depending on the equipment surface. For example, significant correlations (P<0.01) included: APC (rho=0.386) and enterococci (rho =0.562) with the harvest bin, E. coli (rho =0.372) and enterococci (rho =0.355) with the merry go round, enterococci (rho =0.679) with rinse cycle equipment, APC (rho =0.542) with the conveyer belt, and for all indicators with the packing box (rho=0.310-0.657). To identify and compare equipment surfaces at elevated risk for contaminating produce, a model was constructed that controlled for crop type, produce-swab group, and sampling location. There were significant positive associations between the log concentration of enterococci on produce and the harvest bin (beta=0.259, P<0.01) and the rinse cycle (beta=0.010, P=0.01), and between the log concentration of all indicators on produce and the packing box (beta = 0.155-0.500, all P<0.01).
Significance: Statistically significant associations between microbial loads on packing facility surfaces and fresh produce confirm the importance of packing facility sanitation to protect produce quality and safety.