A predictive model for survival of Escherichia coli O157:H7 and generic E. coli in soil amended with animal manure

Authors: PANG, HAO - Us Food & Drug Administration (FDA); MOKHTARI, AMIR - Us Food & Drug Administration (FDA); CHEN, YUHUAN - Us Food & Drug Administration (FDA); ORYANG, DAVID - Us Food & Drug Administration (FDA); INGRAM, DAVID - Us Food & Drug Administration (FDA); Sharma, Manan; Millner, Patricia; VAN DOREN, JANE - Us Food & Drug Administration (FDA)

Submitted to: Risk Analysis
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/30/2020
Publication Date: 5/7/2020
Citation: Pang, H., Mokhtari, A., Chen, Y., Oryang, D., Ingram, D.T., Sharma, M., Millner, P.D., Van Doren, J.M. 2020. A predictive model for survival of Escherichia coli O157:H7 and generic E. coli in soil amended with animal manure. Risk Analysis. https://doi.org/10.1111/risa.13491.
DOI: https://doi.org/10.1111/risa.13491

Interpretive Summary: Predictive statistical models may be useful in determining several practices that can improve pre-harvest food safety of fruits and vegetables. The use of untreated soil amendments like manure can contain bacterial pathogens which can be introduced to fruits and vegetables to cause foodborne illness. One route to minimize foodborne illness is to determine an appropriate interval between the application of untreated soil amendments to soil and harvest of fruits and vegetables long enough to ensure that the pathogens contamination will be reduced. The predictive statistical model used shows that E. coli populations can be predicted using data that USDA ARS collected from 12 seasonal field trials over four years. The amendment type, soil moisture content, and the number of rainy days were identified as influential factors impacting E. coli levels in amended soils. This work also shows that sufficient environmental and agricultural data is needed to make accurate and meaningful predictions regarding the survival of pathogens in manure-amended soils. This work will benefit growers of fresh fruits and vegetables by providing more tools and information to comply with upcoming rules in the Food Safety Modernization Act (FSMA) related to produce safety.

Technical Abstract: The aim of this study was to develop a predictive model to describe the relationship between a variety of agricultural and environmental variables and changes in populations of enteric bacteria in soil amended with Biological Soil Amendments of Animal Origin (BSAAO) under dynamic conditions. We developed and validated a Random Forest model using data from a longitudinal field study investigating the survival of Escherichia coli O157:H7 and generic E. coli in soils amended with dairy manure, horse manure, or poultry litter. Amendment type, soil moisture content, and number of rainy days since the previous sampling day were identified as the most influential agricultural and environmental variables impacting the concentration of E. coli O157:H7 and generic E. coli in amended soils. The Random Forest model was able to capture the complex non-linear relationships and predict fluctuations in E. coli concentrations over time under different agricultural and environmental conditions. Our model also accurately characterized the variability of E. coli concentration in amended soil under certain conditions by providing upper and lower prediction bounds. With the ability to provide quantitative estimates of population of E. coli in amended soils, our predictive survival model can be further incorporated into a risk assessment model for understanding the risks associated with application of untreated BSAAO. The modeling approach reported here also can be used to predict the survival of other enteric bacteria, such as Salmonella, in amended soils, if similar data elements are available.