Visual Demerit and Microbiological Evaluation of Beef Carcasses: Methodology

J Food Prot. 1993 Feb;56(2):114-119. doi: 10.4315/0362-028X-56.2.114.

Abstract

A repeatable, automated method was developed for estimating aerobic bacterial populations on surfaces of groups of beef carcasses. Ten sample cluster sites (CS) were identified by localizing visual demerits (Canadian Streamlined Inspection System) on 200 carcasses at one plant. Most probable number growth units per cm2 (MPNGU/cm2) on hydrophobic grid membrane filters (HGMF) were assessed by an automated HGMF interpreter for excision samples from the centers of these CS. Between-sample variation of more than 90% of the total log10 MPNGU/cm2 variance indicated good repeatability between HGMF of the same sample and interpretations of the same HGMF. Variance component estimates indicated that there was considerable variation in MPNGU/cm2 between carcasses and between paired adjacent samples for a CS. A statistically significant but weak association was found between the demerit scores of a CS and MPNGU at its center. The variance component estimates will be used to estimate the sample size required for future group-carcass evaluations.