United States Department of Agriculture, Agricultural Research Service, Northeast Area, Eastern Regional Research Center, Chemical Residue and Predictive Microbiology Research Unit, University of Maryland Eastern Shore Worksite, Room 2111, Center for Food Science and Technology, Princess Anne, MD 21853, USA.
J Food Prot. 2023 Feb;86(2):100036. doi: 10.1016/j.jfp.2022.100036. Epub 2023 Jan 2.
The Poultry Food Assess Risk Model (PFARM) project was initiated in 1995 to develop data collection and modeling methods for simulating the risk of salmonellosis from poultry food produced by individual production chains. In the present study, the Initial Contamination (IC) step of PFARM for Salmonella and chicken gizzards (CG) was conducted as a case study. Salmonella prevalence (Pr), number (N), and serotype/zoonotic potential (ZP) data (n = 100) for one sample size (56 g) of CG were collected at meal preparation (MP), and then Monte Carlo simulation (MCS) was used to obtain data for other sample sizes (112, 168, 224, 280 g). The PFARM was developed in Excel and was simulated with @Risk. Data were simulated using a moving window of 60 samples to determine how Salmonella Pr, N, and ZP changed over time in the production chain. The ability of Salmonella to survive, grow, and spread in the production chain and food, and then cause disease in humans was ZP, which was based on U. S. Centers for Disease Control and Prevention data for salmonellosis. Of 100 CG samples tested, 35 were contaminated with Salmonella with N from 0 to 0.809 (median) to 2.788 log per 56 g. Salmonella serotype Pr per 56 g was 16% for Kentucky (ZP = 1.1), 9% for Infantis (ZP = 4.4), 6% for Enteritidis (ZP = 5.0), 3% for Typhimurium (ZP = 4.9), and 1% for Thompson (ZP = 3.7). Results from MCS indicated that Salmonella Pr, N, and ZP among portions of CG at MP changed (P ≤ 0.05) over time in the production chain. Notably, the main serotype changed from Kentucky (low ZP) to Infantis (high ZP). However, the pattern of change for Salmonella Pr, N, and ZP differed over time in the production chain and by the statistic used to characterize it. Thus, a performance standard (PS) based on Salmonella Pr, N, or ZP at testing or MP will likely not be a good indicator of poultry food safety or risk of salmonellosis.
家禽食品评估风险模型 (PFARM) 项目于 1995 年启动，旨在开发数据收集和建模方法，以模拟来自个体生产链的家禽食品中沙门氏菌病的风险。在本研究中，PFARM 的初始污染 (IC) 步骤针对沙门氏菌和鸡肫 (CG) 进行了案例研究。在膳食准备 (MP) 时收集了 1 个 CG 样本大小 (56 g) 的沙门氏菌流行率 (Pr)、数量 (N) 和血清型/人畜共患病潜力 (ZP) 数据 (n = 100)，然后使用蒙特卡罗模拟 (MCS) 获得其他样本大小 (112、168、224、280 g) 的数据。PFARM 是在 Excel 中开发的，并使用 @Risk 进行模拟。使用 60 个样本的移动窗口来模拟数据，以确定沙门氏菌 Pr、N 和 ZP 在生产链中的时间变化。沙门氏菌在生产链和食品中的存活、生长和传播能力，然后导致人类疾病的能力是 ZP，这是基于美国疾病控制与预防中心的数据对于沙门氏菌病。在测试的 100 个 CG 样本中，有 35 个样本被沙门氏菌污染，N 从 0 到 0.809(中位数)到 2.788 log/56 g。56 g 时的每克沙门氏菌血清型流行率为 16%为肯塔基州 (ZP = 1.1)、9%为婴儿型 (ZP = 4.4)、6%为肠炎型 (ZP = 5.0)、3%为伤寒型 (ZP = 4.9)和 1%为汤普森型 (ZP = 3.7)。MCS 的结果表明，MP 时 CG 各部分的沙门氏菌 Pr、N 和 ZP 随时间在生产链中发生变化 (P ≤ 0.05)。值得注意的是，主要血清型从肯塔基州 (低 ZP) 变为婴儿型 (高 ZP)。然而，沙门氏菌 Pr、N 和 ZP 的变化模式随时间在生产链中以及用于描述其变化的统计数据而有所不同。因此，基于测试或 MP 时的沙门氏菌 Pr、N 或 ZP 的性能标准 (PS) 不太可能成为家禽食品安全或沙门氏菌病风险的良好指标。
