Impact of microbial ecology of meat and poultry products on predictions from exposure assessment scenarios for refrigerated storage.

A novel extension of traditional growth models for exposure assessment of food-borne microbial pathogens was developed to address the complex interactions of competing microbial populations in foods. Scenarios were designed for baseline refrigeration and mild abuse of servings of chicken broiler and ground beef Our approach employed high-quality data for microbiology of foods at production, refrigerated storage temperatures, and growth kinetics of microbial populations in culture media. Simple parallel models were developed for exponential growth of multiple pathogens and the abundant and ubiquitous nonpathogenic indigenous microbiota. Monte Carlo simulations were run for unconstrained growth and growth with the density-dependent constraint based on the "Jameson effect," inhibition of pathogen growth when the indigenous microbiota reached 10(9) counts per serving. The modes for unconstrained growth of the indigenous microbiota were 10(8), 10(10), and 10(11) counts per serving for chicken broilers, and 10(7), 10(9) and 10(11) counts per serving for ground beef at respective sites for backroom, meat case, and home refrigeration. Contamination rates and likelihoods of reaching temperatures supporting growth of the pathogens in the baseline refrigeration scenario were rare events. The unconstrained exponential growth models appeared to overestimate L. monocytogenes growth maxima for the baseline refrigeration scenario by 1500-7233% (10(6)-10(7) counts/serving) when the inhibitory effects of the indigenous microbiota are ignored. The extreme tails of the distributions for the constrained models appeared to overestimate growth maxima 110% (10(4)-10(5) counts/serving) for Salmonella spp. and 108% (6 x 10(3) counts/serving) for E. coli O157:H7 relative to the extremes of the unconstrained models. The approach of incorporating parallel models for pathogens and the indigenous microbiota into exposure assessment modeling motivates the design of validation studies to test the modeling assumptions, consistent with the analytical-deliberative process of risk analysis.