Purpose: To assess the performance of alternate method against the reference methods for detection of Salmonella and E. coliO157: H7 in food and environmental samples.
Methods: The Salmonella method evaluation (alternate vs. USDA-FSIS-MLG-4.08) was performed on raw ground beef (25 g and 375 g), raw chicken breast, raw fish, creamy peanut butter, vanilla ice cream, dry pet food (all 25-g samples) and stainless steel (sponges). Alternate method for E. coli O157:H7 was compared to the USDA-FSIS- MLG 5.08 for ground beef (25 g and 375 g) and FDA-BAM for spinach (200 g). All results were compared using the probability of detection (POD) following AOAC guidelines.
Results: For samples with low contamination levels (n=20/ matrix), the difference in POD for confirmed Salmonella alternate and reference methods (dPODC values along with 95% confidence interval in parenthesis) were -0.15 (-0.41,0.14) for 25-g beef samples and -0.15 (-0.41,0.15) for 375-g beef samples; 0.15 (-0.15, 0.41) for chicken breast; 0.05 (-0.24, 0.33) for fish; 0.15 (-0.13, 0.40) for peanut butter; 0.10 (-0.19, 0.37) for ice cream; 0.15 (-0.15, 0.41) for pet food; and 0.05 (-0.23, 0.33) for steel surface samples. For E. coli O157:H7, dPODC values were 0.15 (-0.11, 0.39) for 25 g beef; -0.10 (-0.37, 0.20) for 375-g beef; and 0.15 (-0.13, 0.40) for 200-g spinach samples. The confidence interval of all above dPODC values as well as for all un-inoculated and high contamination level samples (n=5 each/ matrix) contained zero, showing no statistical difference between the compared methods.
Significance: No performance difference was found between the alternate and reference methods. The alternate method employs a simplified workflow and offers a significant time saving in comparison to the reference methods.