https://orcid.org/0000-0001-8922-5676
Figures
Abstract
Background
Antimicrobial-resistant bacteria are a growing public health threat. In 2017 the U.S. Food and Drug Administration implemented Veterinary Feed Directive (VFD) rules changes to limit medically important antimicrobial use in food-producing animals, combating antimicrobial-resistant bacteria. The effect of the VFD rule changes on the occurrence of bacteria resistant to medically-important antimicrobials in retail meats is yet to be investigated in the U.S. This study investigates whether the VFD rule changes affected the occurrence of tetracycline-resistant and erythromycin-resistant bacteria (Salmonella, Escherichia, and Campylobacter) in retail meats in the U.S.
Methods
Multivariable mixed effect logistic regression models were used to analyze 2002–2019 retail meats surveillance data from the National Antimicrobial Resistance Monitoring System (NARMS) in the U.S. Variables included VFD rule changes, meat type, quarter of year, and raising claims. A potential association between these variables and the occurrence of tetracycline-resistant and erythromycin-resistant bacteria (Salmonella, Escherichia, and Campylobacter) in retail meats was estimated.
Results
Analysis included data regarding tetracycline-resistant Salmonella (n = 8,501), Escherichia (n = 20, 283), Campylobacter (n = 9,682), and erythromycin-resistant Campylobacter (n = 10,446) in retail meats. The odds of detecting tetracycline-resistant Escherichia (OR = 0.60), Campylobacter (OR = 0.89), and erythromycin-resistant Campylobacter (OR = 0.43) in chicken breast significantly decreased after the VFD rule changes, compared to the pre-VFD rule change period. The odds of detecting tetracycline-resistant Salmonella (0.66), Escherichia (OR = 0.56), and Campylobacter (OR = 0.33) in ground turkey also significantly decreased. However, the odds of detecting tetracycline-resistant Salmonella (OR = 1.49) in chicken breast and erythromycin-resistant Campylobacter (OR = 4.63) in ground turkey significantly increased. There was no significant change in the odds of detecting tetracycline-resistant Salmonella and Escherichia in ground beef or pork chops.
Conclusions
The implementation of VFD rule changes had a beneficial effect by reducing the occurrence of tetracycline-resistant and erythromycin-resistant bacteria in chicken and ground turkey. Ongoing surveillance of antimicrobial resistance and antimicrobial use could complement the implementation of stewardship such as VFD rule in food-producing animals in the U.S.
Citation: Sarkar S, Okafor C (2023) Effect of veterinary feed directive rule changes on tetracycline-resistant and erythromycin-resistant bacteria (Salmonella, Escherichia, and Campylobacter) in retail meats in the United States. PLoS ONE 18(8): e0289208. https://doi.org/10.1371/journal.pone.0289208
Editor: Csaba Varga, University of Illinois Urbana-Champaign College of Veterinary Medicine, UNITED STATES
Received: April 13, 2023; Accepted: July 13, 2023; Published: August 3, 2023
Copyright: © 2023 Sarkar, Okafor. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All the datasets used in this study are secondary datasets, and the reference to access the datasets are given in this article. These data sets are publicly available, and no specific rights are required to access these datasets. The data can be accessed at: https://www.fda.gov/animal-veterinary/national-antimicrobial-resistance-monitoring-system/integrated-reportssummaries (accessed on March 1, 2023).
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Animal-sourced protein consumption has increased globally over the last decade [1]. The animal-sourced protein consumed in the United States (U.S.) mostly comes from the industrial food-animal production system [2], which is characterized by large-scale, densely stocked conditions [3], and the widespread usage of antimicrobials to treat sick animals and control disease [4,5]. Before the implementation of the Veterinary Feed Directive (VFD) rule changes in the U.S., antimicrobials were commonly used in food-producing animals as preventive measures. However, since the VFD rule changes were implemented, antimicrobials can no longer be used in a prophylactic manner [4].
Antimicrobial resistance is a global health concern [6,7]. Multiple factors are associated with the emergence of antimicrobial resistance bacteria [8]. Overuse and misuse of antimicrobials are important factors associated with resistance to antimicrobial drugs [9,10]. The use of antimicrobials in food-producing animals in the U.S. varies across different classes of antimicrobials. In the U.S., specifically in 2020, tetracycline was the most frequently used antimicrobial, representing 66% (3,948,745 kg) of total use and 7% (433,394 kilograms) of macrolides in food-producing animals [11]. Antimicrobial resistance has emerged due to the selective pressure exerted by antimicrobial use in food-animal production [12–14]. The improper use of antimicrobial drugs in food-producing animal farming has increased antimicrobial resistance bacteria in animal food products [15,16]. The dose, route, duration, and class of antimicrobials (selective pressure) are important variables that may affect the development of antimicrobial resistance [17,18]. Also, antibiotic usage, whether for therapeutic, prophylactic, or growth promotion purposes, all contribute to the selection pressure leading to antibiotic-resistant development [19,20]. Reducing the misuse of antibiotics, whether for therapeutic, growth promotion, or prophylactic purposes, may strengthen the prevention and control of the development of antimicrobial resistance [18]. Biologically, reducing antimicrobial use can decrease the selection pressure for developing resistance [21]. For example, the resistance level in hospital-acquired pathogens may change rapidly within weeks or months after reforms in antimicrobial use [22]. Multifaceted interventions are needed to reduce the improper use of antimicrobials [23] and promote the judicious use of medically important antimicrobials in food-producing animals to control the development of antimicrobial resistance. Resistance to tetracycline and erythromycin were selected for the current analysis because these are medically important antibiotics [24] approved for use in food-producing animals in the U.S. [25].
The VFD rule changes are an important strategy to ensure the judicious use of medically important antimicrobials in food-producing animals in the U.S. On October 1, 2015, the U.S. Food and Drug Administration (FDA) changed the rules for the VFD. The rules were fully enforced on January 1, 2017, following the FDA’s Guidance for Industry # 2013 [4]. This VFD rule change highlighted the FDA’s concerns about developing antimicrobial resistance (AMR) pathogenic bacteria in humans and animals when medically important antimicrobial drugs are imprudently used in food-producing animal production. The changes made to the VFD rules limit the use of medically important antimicrobial drugs given to animals via feed and water, only allowing usage for treating illness. The VFD has impacted a number of medically important antimicrobial classes, such as tetracyclines, macrolides, and penicillins. However, it is important to note that these antimicrobials can still usage for therapeutic purposes through the administration via drinking water under the supervision of a licensed veterinarian [26,27]. The VFD rule changes have reduced the presence of violative sulfonamide and penicillin residues in the tissue of food animals at U.S. slaughterhouses, compared to before the VFD rule changes [28]. The effect of the VFD rule changes on the occurrence of bacteria resistant to medically important antimicrobials in retail meats is yet to be investigated in the U.S.
The National Antimicrobial Resistance Monitoring System (NARMS) was set up in 1996 as a joint initiative involving state and local public health departments, universities, the FDA, the Center for Disease Control and Prevention (CDC), and the U.S. Department of Agriculture’s (USDA) [29]. This national surveillance system has routinely collected data on the antimicrobial susceptibility of enteric bacteria from meats sold in stores since 2002 [29]. This program aims to monitor changes in the antimicrobial susceptibility of enteric bacteria in the US., with a specific component of the surveillance being devoted to retail meats.
Meat and meat products can be an important source of exposure to antimicrobial-resistant bacteria in humans [30–33]. Numerous studies have indicated the existence of antimicrobial-resistant bacteria found in meats sold at retail stores globally [31,34–39]. Antimicrobial-resistant bacteria originating from meats can be transmitted to humans through direct contact with meats or the consumption of meat products [40,41]. In cases where humans contract antimicrobial-resistant bacteria, the resulting infection may be challenging to manage or lead to unfavorable clinical consequences [42,43]. Foodborne illness associated with enteric bacteria is a significant public health issue in the U.S. Approximately 1 in 6 Americans suffer annually from a foodborne illness, resulting in almost 48 million cases, 128,000 hospitalizations, and 3000 deaths in the U.S. [44]. Three food-associated enteric bacteria, namely Salmonella, Escherichia, and Campylobacter, are responsible for approximately 60% of these illnesses and hospitalizations in the U.S. [44,45]. These bacteria also infect food-producing animals, making them a suitable focus for the current study.
Several studies have been conducted on the prevalence of antimicrobial-resistant bacteria in retail meats in the U.S. Most of these studies aimed to characterize and assess the profiles of antimicrobial-resistant bacteria isolates in retail meats, using cross-sectional surveys conducted in the U.S. between 1998 and 2018 [32,46–48]. A recent study utilized a subset of NARMS data covering the period from 2008 to 2017 and compared Salmonella prevalence and antibiotic susceptibility profiles in retail poultry meats with and without antibiotic-related claims [49]. Another study explored the associations between meat production methods and AMR and bacterial contamination of retail meats, using NARMS data covering 2012 to 2017 [50]. However, recent studies utilizing NARMS data did not evaluate the potential impact of VFD rule changes on the risk of antimicrobial-resistant bacteria in retail meats in the U.S. Therefore, our study’s objective was to investigate the effect of the 2017 VFD rule changes on the occurrence of tetracycline and erythromycin-resistant bacteria (Salmonella, Escherichia, and Campylobacter) in the U.S. using the NARMS dataset. The findings of this study will provide evidence of the magnitude of the impact that VFD rule changes have had on the risk of AMR bacteria isolates in retail meats in the U.S.
Materials and methods
Data source and retrieval
On March 1, 2023, the retail meats surveillance dataset from 2002 to 2019 was downloaded from the NARMS, which is publicly available data [51]. Under this surveillance system, chicken breast, ground beef, ground turkey, and pork chops are collected from grocery stores and tested for bacterial contamination and their resistance to antimicrobial drugs [51]. Subsequently, the obtained data was transferred from Microsoft Excel to STATA 17.1 software (Stata Corporation, College Station, TX, USA) for data validation and analysis.
Data preparation and variables
The primary variable of interest utilized in selecting the final dataset for analysis was the presence of the minimum inhibitory concentration (MIC) values of tetracycline and erythromycin. The other selected variables for analysis in each dataset included month, year, state, meat type, raising claims (regarding how the animals were raised and whether they were given antibiotics), and type genera of bacteria found (Salmonella, Escherichia, and Campylobacter). The bacteria categories analyzed in this study include Salmonella, Escherichia, and Campylobacter, which encompass the aggregation of their respective serotypes. The “raising claims” variable has numerous text-form responses, which was further categorized into three groups (conventional, without any antibiotics, and not specified). Responses that included words such as “organic”, or “no antibiotic” or “antibiotic-free” (without any antibiotics) were classified together as “without any antibiotics” categories [52]. Likewise, responses without specific antibiotic-free claims such as “all-natural”, “use antibiotic responsibly”, “grass-fed” were grouped together as “conventional”. Similarly, responses containing the words “none” or “not specified” were grouped together as "not specified". After analyzing the data, we found that there was no difference between the conventional and not specified categories. As a result, we merged the not specified category with the conventional category. The final raising claims were either ‘without any antibiotic’ or ‘conventional’. The “year” variable was treated as an individual year variable. Also, we collapsed “year” variable into a categorical variable called “years of sampling”. This categorization was based on three years: ‘2002–2004’, ‘2005–2007’, ‘2008–2010’, ‘2011–2013’, ‘2014–2016’, and ‘2017–2019’. Additionally, we also created a dichotomous variable called “VFD rule change,” which collapse the “year” variable into two categories: “before VFD rule change (2002–2016)” and “after VFD rule change (2017–2019).” The “VFD rule changes” is a two-level categorical variable, with ‘after VFD rule changes’ meaning the retail meats were sampled and tested between 2017 and 2019, and ‘before VFD rule changes’ meaning the retail meats were sampled and tested between 2002 and 2016.
The "month" variable was categorized into four quarters: Quarter 1 (January, February, and March), Quarter 2 (April, May, and June), Quarter 3 (July, August, and September), and Quarter 4 (October, November, and December).
The outcome variables were the presence of tetracycline-resistant Salmonella, Escherichia, and Campylobacter isolates in retail meats and the presence of erythromycin- resistant Campylobacter isolates in retail meats. Tetracycline-resistant Salmonella and Escherichia isolates in retail meat were categorized based on the MIC breakpoint values of tetracycline (≥16 μg/ml). Similarly, the tetracycline-resistant Campylobacter in retail meats was categorized based on the MIC values of tetracycline (≥4 μg/ml). Erythromycin-resistant Campylobacter in retail meats was defined based on the MIC breakpoint values of erythromycin (≥8 μg/ml). The breakpoints were based on the 2021 NARMS Interpretive Criteria for Susceptibility Testing [53].
Statistical analysis
Frequencies and percentages were used to summarize categorical predictor variables. Four mixed-effect logistic regression models were built, using a random intercept to control for state-level clustering for the tetracycline-resistant Salmonella, Escherichia, and Campylobacter isolates and erythromycin-resistant Campylobacter isolates in retail meats in this study.
For each model-building process, two steps were involved. In the first step, a univariable mixed effect logistic regression model using the state as a random intercept was fitted to assess the unadjusted associations between potential predictor variables and the outcome variable. We conducted the univariable analysis to investigate the association of the “year”, “years of sampling”, “VFD rule change”, meat type, quarter of year, and raising claims with the resistant outcome.
A relaxed p-value ≤ 0.2 [54–57] was used to identify predictors that were chosen for further examination in the multivariable mixed-effect logistic regression models in step two. To prevent the inclusion of collinear variables in the multivariable models, the pairwise collinearity of these variables was examined using Spearman’s rank correlation coefficient. If the correlation coefficient between two variables is ≥0.6 [58], only the variable with the highest odds ratio, the fewest missing observations in the initial screening, and biological plausibility would be included in the multivariable model. Categorical variables with more than two levels of categories were analyzed to evaluate pairwise differences using the Tukey-Kramer adjustment for multiple comparisons. Temporal graphs were generated in excel to visualize the temporal patterns of resistant outcomes by years of sampling. However, for the variables measuring similar characteristics (e.g., year, year of sampling, and VFD rule change), only one of the variables was used in model building, determined by the results of the univariable analysis.
In the second step, a multivariable mixed-effect logistic regression model (PROC GLIMMIX, SAS version 9.4) was built using a manual backward elimination method. A full model was first constructed by including all the screened variables and non-correlated as fixed factors, and subsequently, the state was fitted as a random effect and state as a random effect. We included the VFD rule change in every full multivariable model, regardless of the p-value obtained from the univariable analysis, as it was our primary exposure of interest for the analysis. Non-significant variables were removed through a manual backward elimination process. However, if the removal of a non-significant variable resulted in a substantial change (more than 20%) in the coefficients of any remaining variables in the model, it was considered a potential confounder and was retained in the final model [59]. Each final multivariable model was checked for possible multicollinearity using variance inflation factor (VIF) to avoid modeling issues associated with multicollinearity. If the VIF value exceeded 10, it indicated the of presence of multicollinearity [60]. The relevant pairwise significant interaction term was assessed in the final multivariable models [61]. For instance, the two-way interaction between VFD rule changes and the meat type was evaluated. For all final multivariable model results were presented as an odds ratio (OR) with a corresponding 95% confidence interval (CI), and p-values. A p-value ≤0.05 was considered statistically significant [62]. The overall assessment of the final multivariable model was done using the Akaike’s Information Criterion (AIC) [63]. The model with the lowest AIC values was considered as the best-fitting model.
Results
The original dataset contained 342,041 records from 2002 to 2019. However, only 67,731 (19.8%) and 39,720 (11.6%) of these records had MIC values for the target antibiotics, tetracycline and erythromycin, respectively. Data regarding tetracycline-resistant Salmonella (N = 8,501), Escherichia (N = 20,283), Campylobacter (N = 9,682), and erythromycin-resistant Campylobacter (N = 10,446) from records of retail meats were included in the statistical analysis.
Univariable mixed effect logistic regression results
Meat type, quarter of year, raising claims, and year were significantly associated with the occurrence of tetracycline-resistant Salmonella in retail meats (Table 1). Additionally, the estimated odds ratios from the years of sampling did not demonstrate a distinct linear pattern across all comparisons for tetracycline-resistant Salmonella in retail meats (S1 Table in S1 File). Furthermore, no linear trend was observed in the proportion of tetracycline-resistant Salmonella in retail meats based on our graphical analysis (S1 Fig in S1 File). Similarly, meat type, quarter of year, raising claims, year, and the VFD rule changes were significantly associated with tetracycline-resistant Escherichia in retail meats (Table 2). Additionally, the estimated odds ratios from the years of sampling did not demonstrate a distinct linear pattern across all comparison’s tetracycline-resistant Escherichia in retail meats (S2 Table in S1 File). Furthermore, no linear trend was observed the proportion of tetracycline-resistant Escherichia in retail meats based on our graphical analysis (S2 Fig in S1 File). Additionally, meat type, quarter, year, and VFD rule changes were significantly associated with tetracycline-resistant Campylobacter in retail meats (Table 3). However, there was no statistically significant association between raising claims and quarter year for tetracycline-resistant Campylobacter in retail meats (Table 3). In addition, the estimated odds ratios from the years of sampling did not demonstrate a distinct linear pattern across all comparison’s tetracycline-resistant Campylobacter in retail meats (S3 Table in S1 File). Furthermore, no linear trend was observed in the proportion of tetracycline-resistant Campylobacter in retail meats based on our graphical analysis (S3 Fig in S1 File). Lastly, meat type, year, and VFD rule changes were significantly associated with erythromycin-resistant Campylobacter in retail meats (Table 4). In addition, the estimated odds ratios from the years of sampling did not demonstrate a distinct linear pattern across all comparison’s erythromycin-resistant Campylobacter in retail meats (S4 Table in S1 File). Furthermore, no linear trend was observed in the proportion of erythromycin-resistant Campylobacter in retail meats based on our graphical analysis (S4 Fig in S1 File).
Multivariable mixed effect logistic regression model results
The final model was fitted for tetracycline-resistant Salmonella, which included 8,501 records of retail meats (Table 5). Variables significantly associated with detecting tetracycline-resistant Salmonella—controlling for other variables—was meat type. No multicollinearity issue was found in the final model. However, there was a significant interaction between the VFD rule changes and meat type after controlling for all other variables in the model. The significant interaction between the VFD rule changes and meat type implies that the effect of the VFD rule changes on the odds of detecting tetracycline-resistant Salmonella was not the same across the meat types. For example, the odds of detecting tetracycline-resistant Salmonella were decreased by 44% in ground turkey following implementation of the VFD rule changes compared to ground turkey in the period prior to implementation (OR = 0.66, 95% CI: 0.56, 0.77; p<0.0001) (Table 5). In contrast, the odds of detecting tetracycline-resistant Salmonella were increased by 49% in the chicken breast following implementation of the VFD rule changes, compared to chicken breast in the period prior to implementation (OR = 1.49; 95% CI: 1.31, 1.69; p<0.0001) (Table 5).
The final model was fitted for tetracycline-resistant Escherichia, which included 20,283 records of retail meat (Table 6). Variables significantly associated with detecting tetracycline-resistant Escherichia—controlling for other variables—were the VFD rule changes, meat type, and sampling quarter. No multicollinearity issue was found in the final model. However, there was a significant interaction between the VFD rule changes and meat type after controlling for all other variables in the model. The significant interaction between the VFD rule changes and meat type implies that the effect of the VFD rule changes on the odds of detecting tetracycline-resistant Escherichia was not the same across the meat types. For example, the odds of detecting tetracycline-resistant Escherichia were decreased by 40% in the chicken breast following the implementation of the VFD rule changes compared to chicken breast in the period prior to implementation (OR = 0.60, 95% CI: 0.50, 0.72; p<0.0001) (Table 6). Likewise, the odds of detecting tetracycline-resistant Escherichia were decreased by 44% in ground turkey following implementation of the VFD rule changes compared to ground turkey in the period prior to implementation (OR = 0.56, 95% CI: 0.48, 0.65; p<0.0001) (Table 6). After the VFD rule changes, there was no significant difference in the odds of detecting tetracycline-resistant Escherichia in ground beef and pork chop compared to before the VFD rule changes (Table 6). Considering quarter of year, the odds of detecting tetracycline-resistant Escherichia in Quarter 1 and Quarter 2 were 22% (OR = 1.22, 95% CI: 1.09, 1.36); p<0.0001) and 12% (OR = 1.12, 95% CI: 1.00, 1.26, p = 0.008) higher, compared to Quarter 3 (Table 6).
The final model was fitted for tetracycline-resistant Campylobacter, which included 9,682 records of retail meats (Table 7). Variables significantly associated with detecting tetracycline-resistant Campylobacter—controlling for other variables—were the VFD rule changes and meat type. No multicollinearity issue was found in the final model. However, there was a significant interaction between the VFD rule changes and meat type after controlling for all other variables in the model. The significant interaction between the VFD rule changes and meat type implies that the effect of the VFD rule changes on the odds of detecting tetracycline-resistant Campylobacter was not the same across the meat types. For example, the odds of detecting tetracycline-resistant Campylobacter were decreased by 11% in the chicken breast following implementation of the VFD rule changes, compared to chicken breast in the period prior to implementation (OR = 0.89, 95% CI: 0.79, 0.99; p = 0.0370) (Table 6). Likewise, the odds of detecting tetracycline-resistant Campylobacter were decreased by 67% in ground turkey following implementation of the VFD rule changes, compared to ground turkey in the period prior to implementation (OR = 0.33, 95% CI: 0.17, 0.66; p = 0.0017) (Table 7).
The final model was fitted for erythromycin-resistant Campylobacter, which included 10,446 records of retail meats (Table 8). Variables significantly associated with detecting erythromycin-resistant Campylobacter—controlling for other variables—was meat type. No multicollinearity issue was found in the final model. However, there was a significant interaction between the VFD rule changes and meat type after controlling for all other variables in the model. The significant interaction between the VFD rule changes and meat type implies that the effect of the VFD rule changes on the odds of detecting erythromycin-resistant Campylobacter were not the same across the meat types. For example, the odds of detecting erythromycin-resistant Campylobacter were decreased by 57% in the chicken breast following implementation of the VFD rule changes, compared to chicken breast in the period prior to implementation (OR = 0.43, 95% CI: 0.29, 0.64; p<0.0001). In contrast, the odds of detecting erythromycin-resistant Campylobacter were 4.63 times higher in ground turkey following implementation of the VFD rule changes compared to ground turkey in the period prior to implementation (OR = 4.63; 95% CI: 1.48, 14.44; p = 0.0084) (Table 8).
Discussion
The VFD rules changes were implemented in 2017 to reduce the use of medically important antibiotics in food-producing animals. These rule changes are crucial in decreasing the amount of medically important antibiotics used and, consequently, reducing the development of antibiotic-resistant bacteria in food-producing animals and their products in the U.S. This current investigation sought to determine the association between VFD rule changes and the occurrence of tetracycline and erythromycin-resistant bacteria isolated from retail meats obtained under the NARMS in the U.S., after controlling for other variables. After controlling for other variables in the multivariable models, there was a significant interaction between the VFD rule changes and meat type on the occurrence of tetracycline-resistant Salmonella in retail meats (Table 5). Additionally, after controlling for other variables in the multivariable models, there was a significant interaction between the VFD rule changes and meat type on the occurrence of tetracycline-resistant Escherichia in retail meats (Table 6). Likewise, after controlling for other variables in the multivariable models, there was a significant interaction between the VFD rule changes and meat type on the occurrence of tetracycline-resistant Campylobacter in retail meats (Table 7). Moreover, after controlling for other variables in the multivariable models, there was a significant interaction between the VFD rule changes and meat type on the occurrence of erythromycin-resistant Campylobacter in retail meats (Table 8). The significant interaction between the VFD rule changes and meat type implies that the effect of the VFD rule changes on the odds of detecting antibiotic-resistant bacteria were not the same across the meat types.
The findings of this study reveal that the implementation of VFD rule changes in the U.S. resulted in a significant reduction in the odds of detecting tetracycline-resistant Escherichia, Campylobacter, and erythromycin-resistant Campylobacter isolated from chicken breast. Also, our data revealed significant reduction in the odds of detecting tetracycline-resistant Salmonella, Escherichia, and Campylobacter isolated from ground turkey. To our knowledge, this report is the first to provide quantitative evidence of the association between changes in VFD rules changes and the odds of detecting tetracycline-resistant Escherichia, Campylobacter, and erythromycin-resistant Campylobacter isolated from chicken breast, as well as tetracycline-resistant Salmonella, Escherichia, and Campylobacter in ground turkey in the U.S.
Multiple factors could explain the decreased odds of detecting tetracycline-resistant Escherichia (40%) and Campylobacter (11%) isolated from chicken breast compared to the period before implementation. One is the VFD rule changes potentially decreased the use of tetracycline in chicken production, leading to a decrease in the occurrence of tetracycline-resistant bacteria in chicken breast in the U.S. Reduced use of tetracycline may have reduced the selective pressure for the emergence of tetracycline-resistant bacteria in food animals, including chickens [13,15]. The U.S. FDA’s recent report indicates that tetracycline sales decreased in chicken production after the VFD rule changes were implemented [11]. Recent data on on-farm antimicrobial use in the U.S. broiler chicken industry shows a substantial decline in the utilization of medically important in-feed and water-soluble tetracycline since 2017 [64]. Similarly, data on on-farm antimicrobial use in U.S. turkey production demonstrates a substantial reduction in the usage of medically important in-feed and water-soluble tetracycline in 2017 [65]. The reduction of antimicrobial use in food-producing animals is associated with a reduction in the occurrence of antibiotic-resistant bacteria originating from food animals [15]. An observational study conducted by Stuart B. Levy et al. [66] demonstrated that six months after the removal of tetracycline-supplemented feed from a chicken farm, there was a lower frequency of tetracycline-resistant Escherichia isolates compared to before the removal of the tetracycline-supplemented feed in the farm. There is also evidence that a reduction in antimicrobial use is associated with reduced occurrence of antimicrobial-resistant bacteria isolates from humans [21,56]. In addition, chicken production type and management practices could be associated with decreased odds of detecting tetracycline-resistant Escherichia and Campylobacter. For instance, extensive and comprehensive cleaning and hygiene practices in chicken farms could reduce the reservoirs of resistant bacteria [67] that may have occurred on dirty floors and fomites within the chicken houses [68]. Also, improved feeding systems, such as providing safe and antibiotic-free feed and water in chicken production, could reduce the risk of the occurrence of resistant bacteria [67] in chicken meats. Furthermore, the organic food industry is rapidly growing in the U.S., with organic meat and poultry sales increasing by 2.5% and 4.7%, respectively, in 2021 [69]. Transitioning from conventional chicken production to organic chicken production, which involves the elimination of antibiotic use, can significantly reduce the occurrence of antibiotic-resistant bacteria in chicken meat in the U.S. There is evidence that the prevalence of antibiotic-resistant Enterococcus significantly decreased in poultry farms in the U.S. which transitioned from conventional farms to organic farming practices [70].
Another explanation could be that the VFD rule changes may have improved veterinary-client-patient relationships, resulting in better oversight of antimicrobial usage in chicken production by licensed veterinarians. For instance, following the implementation of VFD rule changes in the U.S., administering medically important antimicrobials to food animals via feed or drinking water has been brought under licensed veterinarian supervision [71]. Additionally, it is possible that the VFD rule changes led to increased awareness and adoption of responsible antimicrobial use practices among both chicken producers and veterinarians, resulting in more judicious use of antimicrobials to help mitigate the development of resistance to antimicrobials [72,73]. Also, it is possible that the reduction in antimicrobial-resistant bacteria from retail chicken meats observed after the implementation of VFD rule changes may be due to factors not directly related to the rule changes—such as natural variations in resistance patterns in meat-associated bacterial population over time or improved hygiene practices during meat processing at slaughterhouses, in storing, or handling. Overall, this study demonstrates that the VFD rule changes were beneficial in reducing the occurrence of tetracycline-resistant Escherichia and Campylobacter in chicken breasts in the U.S.
However, the multivariable logistic regression analysis showed that the odds of detecting tetracycline-resistant Salmonella were significantly increased in the chicken breast following the VFD rule changes compared to the pre-VFD rule changes period. The findings of our study is consistent with previous research that has reported a higher level of tetracycline-resistant Salmonella in chicken breast compared to ground turkey in California in 2018 [57]. The exact cause of this outcome remains unknown. However, it is possible that the composition of Salmonella serotypes within chicken production may vary over time [74], or there could be other unidentified factors that explain our study’s findings. Control of tetracycline-resistant Salmonella in chicken meat is difficult as chickens can serve as perpetual vectors and reservoirs without exhibiting symptoms [57]. Tetracycline-resistant Salmonella can contaminate the chicken farm environment and potentially spread the bacteria by infected chicken and contaminated eggs [74]. It is important to conduct further investigations to explore potential risk factors associated with an increased likelihood of tetracycline-resistant Salmonella in chicken meat.
Our research shows that there was no significant change in the likelihood of detecting tetracycline-resistant Salmonella and Escherichia in ground beef and pork chops after the VFD rule changes were implemented in 2017, compared to before the changes. Our study’s results align with recent research that found no significant change in the prevalence of tetracycline-resistant Escherichia in cecal samples of swine at slaughter in the U.S. between 2013 and 2019 [75]. There is no systematic collection of medically important antibiotic consumption data at the farm-level in the U.S. [71]. Such data could provide important information to explain the observed findings. However, the FDA’s medically important antibiotic drug and species-specific sales data between 2016 and 2019 could serve as a proxy for antibiotic consumption data at the farm-level in the U.S. [11]. The data shows that tetracycline sales were highest in 2016 for both cattle (2,840,519 kilograms) and swine (2,520,680 kilograms) production. Tetracycline sales sharply decreased for cattle (1,560,542 kilograms) and swine (1,579,145 kilograms) production in 2017 when the VFD rule changes were implemented. However, tetracycline sales increased in both cattle and swine production in 2018 and 2019 in the U.S. [11]. This pattern of tetracycline sales suggests that there was no consistent reduction in tetracycline consumption in cattle and swine production following the implementation of the VFD rule changes measures in the U.S. Furthermore, the increased usage of this drug for therapeutic purposes could also contribute to the rising sales of tetracycline following the implementation of the VFD rule changes. It is important to conduct further investigations to explore potential factors associated with the increased usage of tetracycline in cattle and swine production just one year after implementing the VFD rule changes.
The current study also shows that there were lower odds of detecting erythromycin-resistant Campylobacter in chicken breast (57%), following the VFD rule changes implementation period, compared to the period prior to implementation. These observed lower odds of detecting erythromycin-resistant Campylobacter in chicken breast may be due to declines in the frequency of usage of erythromycin following the implementation of VFD rule changes. This decrease in usage could result in lower selective pressure [72,73]. The U.S. FDA antimicrobials sold data indicates that the sale of macrolides, including erythromycin, sharply decreased in chicken production following the 2017 VFD rule changes [11]. Alternative explanations may include improved farming management and biosecurity practices, resulting in a decreased occurrence of bacterial disease (respiratory disease, infectious sinusitis, air sacculitis), and, thus, a reduction of the need for macrolide (erythromycin, tylosin) usage in chicken production. There is evidence from other studies that reducing or restricting antimicrobial use in food-producing animals reduces the occurrence of antimicrobial-resistant bacteria by decreasing selective pressure [72,73] on the bacteria originating from the food animals [15,76]. Likewise, a low prevalence of ciprofloxacin-resistant Campylobacter in Louisiana retail chickens was observed after the enrofloxacin ban [77]. In human studies, evidence of low use of azithromycin (macrolide) during summer months was found to be associated with decreased prevalence of resistant pneumococci among children [18,78]. The findings of our work show that implementing the VFD rule changes has had a significant effect on the occurrence of erythromycin-resistant Campylobacter in chicken breast in the U.S.
However, the odds of detecting erythromycin-resistant Campylobacter were significantly increased in ground turkey following the implementation of the VFD rule changes compared to the pre-VFD rule changes period. There was no systematic collection of medically important antibiotic consumption data at the turkey farm level in the U.S. [71]. Such data could provide important information to explain this finding. However, the FDA’s medically important antibiotic drug and species-specific sales data between 2016 and 2019 could serve as a proxy for antibiotic consumption data at the turkey farm-level in the U.S. The data shows a sharp increase in sales of macrolide, including erythromycin, in 2017 (1307 kilograms) when the VFD rule changes were implemented [11]. An increasing trend in macrolide sales was observed in 2018 (1653 kilograms) and 2019 (1944 kilograms) in the U.S. [11]. This sales pattern suggests an increasing consumption of macrolide, including erythromycin/tylosin, in turkey production following the implementation of the VFD rule changes in the U.S., which can create selective pressure in developing erythromycin-resistant bacteria. Previous research identified erythromycin-resistant Campylobacter in commercial turkey farms and at slaughterhouses in Ohio, USA [79]. There is evidence of an association between antimicrobial usage and antimicrobial-resistant bacteria isolated from fecal samples in turkey farms in European countries [80]. Other possible reasons may be the presence of poor hygiene, inadequate flock health management, and intensive farming conditions at turkey farms. These factors can contribute to a higher prevalence of infectious diseases such as respiratory disease/infectious sinusitis/air sacculitis in turkeys. These conditions often require treatment with antimicrobial drugs from the macrolide group. The usage of antibiotics can increase the selection pressure for erythromycin-resistant bacteria among turkeys [81]. As a result, there is an increased likelihood of erythromycin-resistant bacteria-infected turkeys entering slaughterhouse establishments. Additionally, suppose turkeys are exposed to feed or water contaminated with erythromycin-resistant bacteria or genes. In that case, they can become carriers of these resistant bacteria. Furthermore, improper sanitary and cleanliness of slaughterhouses and fecal contamination during slaughtering processing [82,83] can result in resistant-bacteria cross-contamination in turkey meat. Evidence of improper scalding or scalding processes without scalding water temperature control was significantly associated with bacteria contamination of chicken meat in slaughterhouses [84]. The U.S. is the world’s largest producer of turkey and turkey meat. In 2019, approximately 229 million turkeys were raised in the U.S., resulting in a total live weight production of 7.4 billion pounds. It is important to conduct further investigations at the turkey farm-level to explore potential factors associated with the emergence of erythromycin-resistant Campylobacter within U.S. turkey production following the implementation of the VFD rule changes.
While this study provides valuable insights into the reduction of tetracycline-resistant Escherichia, Campylobacter, and erythromycin-resistant Campylobacter in chicken breast, as well as tetracycline-resistant Salmonella, Escherichia, and Campylobacter in ground turkey, it is important to note that these findings may not apply to all chicken and turkey retail meats sold in the U.S. This limitation arises from the sampling strategy employed during the study period (2002–2019), which involved convenience sampling [85]. The utilization of convenience sampling could potentially introduce sample selection bias.
To analyze the occurrence of antibiotic-resistant bacteria in retail meats, we initially examined both the individual year and three-year trend using a univariable mixed effect logistic regression model. Additionally, we visually assessed the linear trend of the proportion of antibiotic-resistant bacteria in retail meats at three-year intervals. However, the estimated odds ratios from both the individual year and three-year wise analyses did not demonstrate a consistent linear pattern across all comparisons (S1, S2, S3 and S4 Tables in S1 File). Furthermore, no linear trend of the proportion of antibiotic-resistant bacteria in retail meats was observed based on our graphical analysis (S1, S2, S3 and S4 Figs in S1 File). These non-linear relationships suggest that the impact of individual years or three-year intervals on antibiotic-resistant bacteria in retail meats vary across different points of comparison. Considering these observations, we chose to analyze the variable “VFD rule change” as a reasonable approach. We categorized data into two groups: “before VFD rule change (2002–2016)” and “after VFD rule change (2017–2019)”. Our research question aimed to investigate significant differences in antibiotic-resistant bacteria in retail meats, and the “VFD rule change” variable allowed us to examine the overall effect of the period after the implementation of VFD rule changes compared to the reference period (pre-VFD period: 2002–2016). Furthermore, collapsing these years into binary variable increased the sample size of the category, thereby improving the statistical power of our analysis. Additionally, we employed a mixed-effect logistic regression model to address the clustering effect of state-level during the analysis to avoid biased estimates [86].
The findings of this study suggest that the implementation of VFD rule changes had a positive impact on reducing the presence of tetracycline and erythromycin-resistant bacteria (Escherichia and Campylobacter) in retail meats, specifically chicken breast and ground turkey. The implementation of VFD rule changes has likely contributed to a decrease in antibiotic-resistant bacteria isolates found in retail meats by reducing the usage of antibiotics in food animals. These findings highlight the importance of continued efforts to promote the judicious use of medically important antimicrobials in food-producing animals to combat the emergence and spreading of antimicrobial-resistant bacteria. Further research is necessary covering a more extended period after the implementation of VFD rule changes could more comprehensively evaluate the effectiveness of the VFD rule changes on the occurrence of bacteria resistant to medically-important antimicrobials in retail meats in the U.S. Additionally, it is crucial to assess the impact of the implementation of the VFD rule changes on human health in the U.S.
Supporting information
S1 File. Contains all the supporting tables and figures.
https://doi.org/10.1371/journal.pone.0289208.s001
(DOCX)
Acknowledgments
The authors are grateful to the National Antimicrobial Resistance Monitoring System (NARMS) as the data source. We thank the Comparative and Experimental Medicine Program of University of Tennessee, Knoxville, for supporting this study. Also, we would like to thank Jessica Vidlund for checking the English language expression in the manuscript.
References
- 1.
OECD, Food, Nations AOotU. OECD-FAO Agricultural Outlook 2021–20302021.
- 2.
USDA. Census of Agriculture: United States Department of Agriculture; 2017 [March 7, 2023]. Available from: https://www.nass.usda.gov/Publications/AgCensus/2017/Full_Report/Volume_1,_Chapter_1_US/.
- 3. Casey JA, Kim BF, Larsen J, Price LB, Nachman KE. Industrial food animal production and community health. Current environmental health reports. 2015;2:259–71. pmid:26231503
- 4.
FDA. Guidance for Industry #2013 [March 7, 2023]. Available from: https://www.fda.gov/regulatory-information/search-fda-guidance-documents/cvm-gfi-213-new-animal-drugs-and-new-animal-drug-combination-products-administered-or-medicated-feed.
- 5.
FDA. Potential Approach for Defining Durations of Use for Medically Important Antimicrobial Drugs Intended for Use In or On Feed: A Concept Paper 2017 [March 7, 2023]. Available from: https://www.fda.gov/media/144927/download.
- 6. Karp BE, Tate H, Plumblee JR, Dessai U, Whichard JM, Thacker EL, et al. National antimicrobial resistance monitoring system: two decades of advancing public health through integrated surveillance of antimicrobial resistance. Foodborne pathogens and disease. 2017;14(10):545–57. pmid:28792800
- 7. Murray CJ, Ikuta KS, Sharara F, Swetschinski L, Aguilar GR, Gray A, et al. Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. The Lancet. 2022;399(10325):629–55.
- 8. Chokshi A, Sifri Z, Cennimo D, Horng H. Global contributors to antibiotic resistance. Journal of global infectious diseases. 2019;11(1):36. pmid:30814834
- 9. Holmes AH, Moore LS, Sundsfjord A, Steinbakk M, Regmi S, Karkey A, et al. Understanding the mechanisms and drivers of antimicrobial resistance. The Lancet. 2016;387(10014):176–87. pmid:26603922
- 10. van den Bogaard AE, Stobberingh EE. Epidemiology of resistance to antibiotics: links between animals and humans. International journal of antimicrobial agents. 2000;14(4):327–35.
- 11.
FDA. Antimicrobials Sold or Distributed for Use in Food-Producing Animals 2021 [March 7, 2023]. Available from: https://www.fda.gov/media/163739/download.
- 12. Schwarz S, Kehrenberg C, Walsh T. Use of antimicrobial agents in veterinary medicine and food animal production. International journal of antimicrobial agents. 2001;17(6):431–7. pmid:11397611
- 13. Laxminarayan R, Duse A, Wattal C, Zaidi AK, Wertheim HF, Sumpradit N, et al. Antibiotic resistance—the need for global solutions. The Lancet infectious diseases. 2013;13(12):1057–98. pmid:24252483
- 14. Rantala M, Huovinen P, Hölsö K, Lilas A, Kaartinen L. Survey of condition‐based prescribing of antimicrobial drugs for dogs at a veterinary teaching hospital. Veterinary Record. 2004;155(9):259–62. pmid:15461362
- 15. Tang KL, Caffrey NP, Nóbrega DB, Cork SC, Ronksley PE, Barkema HW, et al. Restricting the use of antibiotics in food-producing animals and its associations with antibiotic resistance in food-producing animals and human beings: a systematic review and meta-analysis. The Lancet Planetary Health. 2017;1(8):e316–e27. pmid:29387833
- 16. Fontana C, Patrone V, Lopez CM, Morelli L, Rebecchi A. Incidence of Tetracycline and Erythromycin Resistance in Meat-Associated Bacteria: Impact of Different Livestock Management Strategies. Microorganisms. 2021;9(10):2111. pmid:34683432
- 17. Rice LB. Antimicrobial resistance in gram-positive bacteria. American journal of infection control. 2006;34(5):S11–S9.
- 18.
Friedman CR, Whitney CG. It’s time for a change in practice: reducing antibiotic use can alter antibiotic resistance. The University of Chicago Press; 2008. p. 1082–3.
- 19. Singer RS, Finch R, Wegener HC, Bywater R, Walters J, Lipsitch M. Antibiotic resistance—the interplay between antibiotic use in animals and human beings. The Lancet infectious diseases. 2003;3(1):47–51. pmid:12505035
- 20. Group WSW. Antimicrobial resistance. Bulletin of the World Health Organization. 1983;61(3):383. pmid:6603914
- 21. Smith RA, M’ikanatha NM, Read AF. Antibiotic resistance: a primer and call to action. Health communication. 2015;30(3):309–14. pmid:25121990
- 22. Lipsitch M, Samore MH. Antimicrobial use and antimicrobial resistance: a population perspective. Emerging infectious diseases. 2002;8(4):347. pmid:11971765
- 23. Prestinaci F, Pezzotti P, Pantosti A. Antimicrobial resistance: a global multifaceted phenomenon. Pathogens and global health. 2015;109(7):309–18. pmid:26343252
- 24.
World, Health Organization. Critically important antimicrobials for human medicine– 5th rev. Geneva: World Health Organization; 2017.
- 25.
FDA. Antimicrobial Use and Resistance in Animal Agriculture in the United States, 2016–2019 2022 [March 1, 2023]. Available from: https://www.fda.gov/media/159544/download.
- 26. Okello E, Williams DR, ElAshmawy WR, Adams J, Pereira RV, Lehenbauer TW, et al. Survey on antimicrobial drug use practices in California preweaned dairy calves. Frontiers in Veterinary Science. 2021;8:636670. pmid:33969034
- 27.
FDA. Veterinary Feed Directive 2015 [March 8, 2023]. Available from: https://www.federalregister.gov/documents/2015/06/03/2015-13393/veterinary-feed-directive.
- 28. Sarkar S, Okafor CC. Effect of Changes in Veterinary Feed Directive Regulations on Violative Antibiotic Residues in the Tissue of Food Animals from the Inspector-Generated Sampling in the United States. Microorganisms. 2022;10(10):2031. pmid:36296306
- 29.
FDA. The National Antimicrobial Resistance Monitoring System in United States [March 8, 2023]. Available from: https://www.fda.gov/animal-veterinary/antimicrobial-resistance/national-antimicrobial-resistance-monitoring-system.
- 30. Adzitey F. Incidence and antimicrobial susceptibility of Escherichia coli isolated from beef (meat muscle, liver and kidney) samples in Wa Abattoir, Ghana. Cogent Food & Agriculture. 2020;6(1):1718269.
- 31. Ahmed AM, Shimamoto T. Molecular analysis of multidrug resistance in Shiga toxin-producing Escherichia coli O157: H7 isolated from meat and dairy products. International journal of food microbiology. 2015;193:68–73. pmid:25462925
- 32. Lestari SI, Han F, Wang F, Ge B. Prevalence and antimicrobial resistance of Salmonella serovars in conventional and organic chickens from Louisiana retail stores. Journal of food protection. 2009;72(6):1165–72. pmid:19610326
- 33. Marshall BM, Levy SB. Food animals and antimicrobials: impacts on human health. Clinical microbiology reviews. 2011;24(4):718–33. pmid:21976606
- 34. Dogruer Y, Telli N, Telli AE, Güner A. Presence and antibiotic susceptibility of Listeria monocytogenes in retail meat and meat products. International Journal of Biological Research. 2015;3(2):76–81.
- 35. Chajęcka‐Wierzchowska W, Zadernowska A, Łaniewska‐Trokenheim Ł. Diversity of antibiotic resistance genes in Enterococcus strains isolated from ready‐to‐eat meat products. Journal of food science. 2016;81(11):M2799–M807. pmid:27780292
- 36. Jaja IF, Bhembe NL, Green E, Oguttu J, Muchenje V. Molecular characterisation of antibiotic-resistant Salmonella enterica isolates recovered from meat in South Africa. Acta Tropica. 2019;190:129–36. pmid:30408462
- 37. Wang W, Chen J, Shao X, Huang P, Zha J, Ye Y. Occurrence and antimicrobial resistance of Salmonella isolated from retail meats in Anhui, China. Food science & nutrition. 2021;9(9):4701–10. pmid:34531984
- 38. Thai TH, Hirai T, Lan NT, Yamaguchi R. Antibiotic resistance profiles of Salmonella serovars isolated from retail pork and chicken meat in North Vietnam. International journal of food microbiology. 2012;156(2):147–51. pmid:22497836
- 39. Sallam KI, Mohammed MA, Hassan MA, Tamura T. Prevalence, molecular identification and antimicrobial resistance profile of Salmonella serovars isolated from retail beef products in Mansoura, Egypt. Food control. 2014;38:209–14.
- 40. Donabedian S, Thal L, Hershberger E, Perri M, Chow J, Bartlett P, et al. Molecular characterization of gentamicin-resistant enterococci in the United States: evidence of spread from animals to humans through food. Journal of Clinical Microbiology. 2003;41(3):1109–13. pmid:12624037
- 41. Van Loo I, Huijsdens X, Tiemersma E, De Neeling A, van de Sande-Bruinsma N, Beaujean D, et al. Emergence of methicillin-resistant Staphylococcus aureus of animal origin in humans. 2007.
- 42. Angulo FJ, Mølbak K. Human health consequences of antimicrobial drug—resistant salmonella and other foodborne pathogens. Clinical infectious diseases. 2005;41(11):1613–20. pmid:16267734
- 43. Varma JK, Mølbak K, Barrett TJ, Beebe JL, Jones TF, Rabatsky-Ehr T, et al. Antimicrobial-resistant nontyphoidal Salmonella is associated with excess bloodstream infections and hospitalizations. The Journal of infectious diseases. 2005;191(4):554–61. pmid:15655779
- 44.
CDC. Estimates of Foodborne Illness in the United States [March 6, 2023]. Available from: https://www.cdc.gov/foodborneburden/2011-foodborne-estimates.html.
- 45. Scallan E, Hoekstra RM, Angulo FJ, Tauxe RV, Widdowson M-A, Roy SL, et al. Foodborne illness acquired in the United States—major pathogens. Emerg Infect Dis. 2011;17(1):7–15. pmid:21192848
- 46. Schmidt JW, Vikram A, Doster E, Thomas K, Weinroth MD, Parker J, et al. Antimicrobial resistance in US retail ground beef with and without label claims regarding antibiotic use. Journal of Food Protection. 2021;84(5):827–42.
- 47. Vikram A, Miller E, Arthur TM, Bosilevac JM, Wheeler TL, Schmidt JW. Similar levels of antimicrobial resistance in US food service ground beef products with and without a “raised without antibiotics” claim. Journal of food protection. 2018;81(12):2007–18.
- 48. White DG, Zhao S, Sudler R, Ayers S, Friedman S, Chen S, et al. The isolation of antibiotic-resistant Salmonella from retail ground meats. New England journal of medicine. 2001;345(16):1147–54. pmid:11642230
- 49. Yin X, M’ikanatha NM, Nyirabahizi E, McDermott PF, Tate H. Antimicrobial resistance in non-Typhoidal Salmonella from retail poultry meat by antibiotic usage-related production claims–United States, 2008–2017. International Journal of Food Microbiology. 2021;342:109044. pmid:33529874
- 50. Innes GK, Nachman KE, Abraham AG, Casey JA, Patton AN, Price LB, et al. Contamination of retail meat samples with multidrug-resistant organisms in relation to organic and conventional production and processing: a cross-sectional analysis of data from the United States National Antimicrobial Resistance Monitoring System, 2012–2017. Environmental health perspectives. 2021;129(5):057004. pmid:33978452
- 51.
FDA. The National Antimicrobial Resistance Monitoring System (NARMS), Integrated Report/Summaries [March 1, 2023]. Available from: https://www.fda.gov/animal-veterinary/national-antimicrobial-resistance-monitoring-system/integrated-reportssummaries.
- 52. Forman J, Silverstein J, Nutrition Co, Health CoE, Bhatia JJ, Abrams SA, et al. Organic foods: health and environmental advantages and disadvantages. Pediatrics. 2012;130(5):e1406–e15. pmid:23090335
- 53.
FDA. 2021 NARMS Interpretive Criteria for Susceptibility Testing [March 1, 2023]. Available from: https://www.fda.gov/media/108180/download.
- 54. Fonseca M, Heider LC, Stryhn H, McClure JT, Léger D, Rizzo D, et al. Intramammary and systemic use of antimicrobials and their association with resistance in generic Escherichia coli recovered from fecal samples from Canadian dairy herds: A cross-sectional study. Preventive Veterinary Medicine. 2023:105948. pmid:37263090
- 55. Fossen JD, Campbell JR, Gow SP, Erickson N, Waldner CL. Antimicrobial Use in Canadian Cow–Calf Herds. Veterinary Sciences. 2023;10(5):366. pmid:37235449
- 56. Lee KY, Atwill ER, Pitesky M, Huang A, Lavelle K, Rickard M, et al. Antimicrobial resistance profiles of non-typhoidal Salmonella from retail meat products in California, 2018. Frontiers in Microbiology. 2022;13:76. pmid:35369434
- 57. Shrestha RD, Agunos A, Gow SP, Deckert AE, Varga C. Associations between antimicrobial resistance in fecal Escherichia coli isolates and antimicrobial use in Canadian turkey flocks. Frontiers in Microbiology. 2022;13. pmid:35966666
- 58. Sova A, LeBlanc S, McBride B, DeVries T. Associations between herd-level feeding management practices, feed sorting, and milk production in freestall dairy farms. Journal of Dairy Science. 2013;96(7):4759–70. pmid:23660144
- 59.
Dohoo I., Martin S., Stryhn H., 2009. Veterinary epidemiologic research. VER. Inc., Charlottetown, PE, Canada.
- 60. Okafor CC, Pearl DL, Blois SL, Lefebvre SL, Yang M, Lund EM, et al. Factors associated with hematuric struvite crystalluria in cats. Journal of feline medicine and surgery. 2019;21(10):922–30. pmid:30392432
- 61. Committee ES. Statistical significance and biological relevance. EFSA Journal. 2011;9(9):2372.
- 62. Bajus A, Renaud D, Goetz H, Steele M, Kelton D, Proudfoot K, et al. The effect of transportation duration on lying behavior in young surplus dairy calves. Journal of Dairy Science. pmid:37268586
- 63. Akaike H. Likelihood of a model and information criteria. Journal of econometrics. 1981;16(1):3–14.
- 64. Singer RS, Schrag NF, Ricke I, Apley MD. Antimicrobial usage in broiler chicken production in the United States, 2013–2021. Frontiers in Veterinary Science. 2023;10. pmid:37138912
- 65. Singer RS, Porter LJ, Schrag NF, Davies PR, Apley MD, Bjork K. Estimates of on‐farm antimicrobial usage in turkey production in the United States, 2013–2017. Zoonoses and Public Health. 2020;67:36–50. pmid:33201604
- 66. Levy SB, FitzGerald GB, Macone AB. Changes in intestinal flora of farm personnel after introduction of a tetracycline-supplemented feed on a farm. New England Journal of Medicine. 1976;295(11):583–8. pmid:950974
- 67. Jimenez CEP, Keestra S, Tandon P, Cumming O, Pickering AJ, Moodley A, et al. Biosecurity and water, sanitation, and hygiene (WASH) interventions in animal agricultural settings for reducing infection burden, antibiotic use, and antibiotic resistance: a One Health systematic review. The Lancet Planetary Health. 2023;7(5):e418–e34. pmid:37164518
- 68. Adiguzel MC, Sigirci BD, Celik B, Kahraman BB, Metiner K, Ikiz S, et al. Phenotypic and genotypic examination of antimicrobial resistance in thermophilic Campylobacter species isolated from poultry in Turkey. Journal of Veterinary Research. 2018;62(4):463. pmid:30729203
- 69. Ager E, Carvalho T, Silva E, Ricke S, Hite J. Global trends in antimicrobial resistance on organic and conventional farms. bioRxiv. 2023:2023.04. 07.536071.
- 70. Sapkota AR, Hulet RM, Zhang G, McDermott P, Kinney EL, Schwab KJ, et al. Lower prevalence of antibiotic-resistant enterococci on US conventional poultry farms that transitioned to organic practices. Environmental health perspectives. 2011;119(11):1622–8.
- 71. Wallinga D, Smit LA, Davis MF, Casey JA, Nachman KE. A review of the effectiveness of current US policies on antimicrobial use in meat and poultry production. Current environmental health reports. 2022;9(2):339–54. pmid:35477845
- 72. Levin BR. Minimizing potential resistance: a population dynamics view. Clinical infectious diseases. 2001;33(Supplement_3):S161–S9. pmid:11524714
- 73. Lipsitch M. The rise and fall of antimicrobial resistance. Trends in microbiology. 2001;9(9):438–44. pmid:11553456
- 74. Jajere S. A review of Salmonella enterica with particular focus on the pathogenicity and virulence factors, host specificity and antimicrobial resistance including multidrug resistance. Vet World 12: 504–521. Salmonella enterica. 2019. pmid:31190705
- 75. Sodagari HR, Varga C. Evaluating Antimicrobial Resistance Trends in Commensal Escherichia coli Isolated from Cecal Samples of Swine at Slaughter in the United States, 2013–2019. Microorganisms. 2023;11(4):1033. pmid:37110456
- 76. McEwen SA, Fedorka-Cray PJ. Antimicrobial use and resistance in animals. Clinical infectious diseases. 2002;34(Supplement_3):S93–S106. pmid:11988879
- 77. Han F, Lestari SI, Pu S, Ge B. Prevalence and antimicrobial resistance among Campylobacter spp. in Louisiana retail chickens after the enrofloxacin ban. Foodborne pathogens and disease. 2009;6(2):163–71. pmid:19099357
- 78. Dagan R, Barkai G, Givon-Lavi N, Sharf AZ, Vardy D, Cohen T, et al. Seasonality of antibiotic-resistant Streptococcus pneumoniae that causes acute otitis media: a clue for an antibiotic-restriction policy? The Journal of infectious diseases. 2008;197(8):1094–102. pmid:18419528
- 79. Kashoma IP, Kumar A, Sanad YM, Gebreyes W, Kazwala RR, Garabed R, et al. Phenotypic and genotypic diversity of thermophilic Campylobacter spp. in commercial turkey flocks: a longitudinal study. Foodborne pathogens and disease. 2014;11(11):850–60. pmid:25184688
- 80. Horie M, Yang D, Joosten P, Munk P, Wadepohl K, Chauvin C, et al. Risk factors for antimicrobial resistance in turkey farms: a cross-sectional study in three European countries. Antibiotics. 2021;10(7):820. pmid:34356741
- 81. Chuppava B, Keller B, El-Wahab AA, Meißner J, Kietzmann M, Visscher C. Resistance of Escherichia coli in turkeys after therapeutic or environmental exposition with enrofloxacin depending on flooring. International journal of environmental research and public health. 2018;15(9):1993. pmid:30216998
- 82. Mensah GI, Adjei VY, Vicar EK, Atsu PS, Blavo DL, Johnson SAM, et al. Safety of Retailed Poultry: Analysis of Antibiotic Resistance in Escherichia coli From Raw Chicken and Poultry Fecal Matter From Selected Farms and Retail Outlets in Accra, Ghana. Microbiology Insights. 2022;15:11786361221093278. pmid:35521227
- 83. Xu Z, Wang M, Zhou C, Gu G, Liang J, Hou X, et al. Prevalence and antimicrobial resistance of retail-meat-borne Salmonella in southern China during the years 2009–2016: The diversity of contamination and the resistance evolution of multidrug-resistant isolates. International Journal of Food Microbiology. 2020;333:108790. pmid:32693316
- 84. Klaharn K, Pichpol D, Meeyam T, Harintharanon T, Lohaanukul P, Punyapornwithaya V. Bacterial contamination of chicken meat in slaughterhouses and the associated risk factors: A nationwide study in Thailand. Plos one. 2022;17(6):e0269416. pmid:35675365
- 85.
FDA. National Antimicrobial Resistance Monitoring System (NARMS). NARMS Methods. [March 1, 2023]. Available from: https://www.fda.gov/media/101741/download.
- 86. Yang J, Zaitlen NA, Goddard ME, Visscher PM, Price AL. Advantages and pitfalls in the application of mixed-model association methods. Nature genetics. 2014;46(2):100–6. pmid:24473328