Modeling the survival kinetics of Salmonella in tree nuts for use in risk assessment
Modeling the survival kinetics of Salmonella in tree nuts for use in risk assessment
复制标题
DOI：
10.1016/j.ijfoodmicro.2016.03.014
复制
发表时间：
2016-06-16
影响因子：
5.4
通讯作者：
Van Doren, Jane
中科院分区：
文献类型：
--
作者：
Farakos, Sofia M. Santillana;Pouillot, Regis;Van Doren, Jane
沙门氏菌已被证明可以在坚果中存活很长时间。这种存活能力及其变异性是坚果中沙门氏菌风险评估的关键因素。本研究的目的是开发一个数学模型，预测在环境储存温度下，考虑变异性和不确定性分别树坚果沙门氏菌的生存，可以很容易地纳入风险评估模型。沙门氏菌在生杏仁、山核桃、开心果和核桃上的存活数据来自同行评议文献。选择Weibull模型作为基线模型，并将各种固定效应和混合效应模型拟合至数据。通过统计分析测试确定的最佳模型，然后用于开发分层贝叶斯模型。坚果中的沙门氏菌在21摄氏度至24摄氏度的温度下缓慢下降。在文献中报道的树坚果研究中观察到生存率的高度变异性。统计分析结果表明，最适用的模型是一个混合效应模型，其中包括一个固定的和随机的变化5每树坚果（这是第一个日志（10）减少所需的时间）和一个固定的变化rho每树坚果（参数，定义曲线的形状）。沙门氏菌在开心果上的估计存活率较高（6），其次是山核桃，杏仁和核桃。使用从贝叶斯推断获得的后验分布来估计每种坚果存活率log（10）下降水平的变异性以及这些估计的不确定性。这些建模的不确定性和变异性分布的估计，可用于获得一个完整的暴露评估树坚果沙门氏菌时，包括时间-温度参数的存储和消费数据。本研究中提出的统计方法可应用于任何旨在开发预测模型的研究，这些模型将在概率暴露评估或定量微生物风险评估中实施。由Elsevier B. V.出版，这是CC BY-NC-ND许可下的开放获取文章。
Salmonella has been shown to survive in tree nuts over long periods of time. This survival capacity and its variability are key elements for risk assessment of Salmonella in tree nuts. The aim of-this study was to develop a mathematical model to predict survival of Salmonella in tree nuts at ambient storage temperatures that considers variability and uncertainty separately and can easily be incorporated into a risk assessment model. Data on Salmonella survival on raw almonds, pecans, pistachios and walnuts were collected from the peer reviewed literature. The Weibull model was chosen as the baseline model and various fixed effect and mixed effect models were fit to the data. The best model identified through statistical analysis testing was then used to develop a hierarchical Bayesian model. Salmonella in tree nuts showed slow declines at temperatures ranging from 21 degrees C to 24 degrees C. A high degree of variability in survival was observed across tree nut studies reported in the literature. Statistical analysis results indicated that the best applicable model was a mixed effect model that included a fixed and random variation of 5 per tree nut (which is the time it takes for the first log(10) reduction) and a fixed variation of rho per tree nut (parameter which defines the shape of the curve). Higher estimated survival rates (6) were obtained for Salmonella on pistachios, followed in decreasing order by pecans, almonds and walnuts. The posterior distributions obtained from Bayesian inference were used to estimate the variability in the log(10) decrease levels in survival for each tree nut, and the uncertainty of these estimates. These modeled uncertainty and variability distributions of the estimates can be used to obtain a complete exposure assessment of Salmonella in tree nuts when including time-temperature parameters for storage and consumption data. The statistical approach presented in this study may be applied to any studies that aim to develop predictive models to be implemented in a probabilistic exposure assessment or a quantitative microbial risk assessment. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license.