Meta-analysis of Escherichia coli O157 prevalence in foods of animal origin in Turkey

The present study aims to analyze the prevalence of E. coli O157 detected in foods of animal origin by meta-analysis. The prevalence of E. coli O157 detected in the different studies was combined to provide a common prevalence estimate, and heterogeneities between studies were investigated. The study material consisted of 49 studies investigating E. coli O157 prevalence in a total of 9600 food samples, including milk and dairy products, red meat and products, poultry meat and products, and cold appetizers between the years 1997-2019 in Turkey. In the meta-analysis, the Der-Simonian-Laird method was used. Meta-analyses were performed using the R 3.6.1. As a result of the meta-analysis, the common prevalence of E. coli O157 was 0.024 (0.018-0.029). As a result of the Egger’s Linear Regression Test, the study samples were found to be biased (t-value=6.092, P<0.001). To determine the source of heterogeneity between studies, sub-group and meta-regression analyses were performed in milk and dairy products, red meat and products, poultry meat and products, and ready-to-eat foods (RTEs). Accordingly, the prevalence of E. coli O157 in milk and dairy products, red meat and products, poultry meat and products, and RTEs was determined as 0.017, 0.031, 0.023, and 0.080 in Turkey, respectively. This study provides a stronger and more accurate estimation of the prevalence of E. coli O157 in foods of animal origin with the meta-analysis by eliminating inconsistencies in the effect of the sampling size of independent prevalence studies. However, in order to obtain accurate prevalence results in practice, it is necessary carefully to select the studies to be included in the analysis, to use the appropriate statistical model, and to interpret the results of the analysis correctly.