Source: UNIV OF MINNESOTA submitted to
SYSTEMS APPROACH TO IDENTIFYING TARGETED INTERVENTIONS FOR MINIMIZING ANTIBIOTIC RESISTANCE IN THE POULTRY PRODUCTION SYSTEM
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
TERMINATED
Funding Source
Reporting Frequency
Annual
Accession No.
1005062
Grant No.
2015-68003-22972
Project No.
MINV-63-115
Proposal No.
2014-05283
Multistate No.
(N/A)
Program Code
A4171
Project Start Date
Jan 15, 2015
Project End Date
Jan 14, 2020
Grant Year
2015
Project Director
Singer, R. S.
Recipient Organization
UNIV OF MINNESOTA
(N/A)
ST PAUL,MN 55108
Performing Department
Veterinary Biosciences
Non Technical Summary
Chicken consumption has been linked to antibiotic resistant infections in humans, primarily caused by Salmonella enterica and Escherichia coli. The elimination of antimicrobial growth promoters from poultry production might help reduce some of the resistance in these bacteria. Given that efficient production of chickens has occurred in conjunction with these antibiotics over many decades, there will likely be system-wide perturbations, such as increased necrotic enteritis, that will further change antibiotic usage patterns. This proposal will use systems approaches to investigate the ecology of resistance gene reservoirs in all stages of broiler production and identify critical control points for mitigating antimicrobial resistance. The Specific Objectives are:1. Evaluate the evidence for specific disease management approaches, including the use of antibiotics at the hatchery2. Quantify the effect of antibiotic administration in the hatchery and broiler farm on the selection of antibiotic resistance3. Associate farm-specific practices and antimicrobial resistance patterns in Salmonella4. Develop tools for predicting efficacy, cost-effectiveness and consequences of antimicrobial resistance mitigation strategies5. Deliver intervention options for reducing antimicrobial resistant Salmonella and E. coli to diverse stakeholdersOur research team believes that effective mitigation of antibiotic resistance must assess the consequences of interventions, both positive and negative, throughout the production system so that the net benefits of each plausible mitigation strategy can be evaluated and compared prior to adoption. This project will develop a software tool to aid this decision-making process.
Animal Health Component
75%
Research Effort Categories
Basic
25%
Applied
65%
Developmental
10%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
7223220117050%
7223220110030%
7223220208020%
Goals / Objectives
The long-term goals of this project are 1) to expand our understanding of the practices in the poultry production system that increase the selection of antibiotic resistant bacteria and 2) to develop multi-disciplinary extension and outreach programs within the poultry industry that deliver efficacious and cost-effective targeted interventions for reducing antibiotic resistance.Broiler production in the U.S. is an integrated system of production modules, from replacement of the breeders to the laying and hatching of eggs to the placement of the broiler chicks to processing. Strains of specific Salmonella and E. coli are known to flow downstream from one production module to the next. Pressures that select for resistance, both antibiotic and non-antibiotic, are present in every module. A successful program for mitigating resistance must address the consequences of interventions in all phases of production and must enable all stakeholders, especially producers, to evaluate the tradeoffs of implementing a specific mitigation strategy. Therefore, the Specific Objectives of this project are:Specific Objectives and Approaches:1) Evaluate the evidence for specific disease management approaches in the broiler production system. We will perform formal, rigorous systematic reviews to examine the scientific basis for efficacy and impact of hatchery antibiotic administration and for approaches to NE management and control in the absence of AGP.2) Quantify the effect of antibiotic administration in the hatchery and broiler farm on the selection of antibiotic resistance. We will use experimental trials to examine the ability of commonly used antibiotics in broiler production to select for resistant Salmonella and E. coli and overall resistance genes.3) Associate farm-specific practices, such as the use of antibiotic and non-antibiotic compounds, with farm-specific antimicrobial resistance patterns in Salmonella. We will leverage the data that we are collecting through the National Antimicrobial Resistance Monitoring System (NARMS) and our collaborations with committed industry partners to link samples from breeder flocks, hatchery and broiler growout flocks with farm practices identified in surveys of the broiler industry.4) Develop tools for predicting efficacy, cost-effectiveness and consequences of antimicrobial resistance mitigation strategies. This extension-based objective will build a dynamic simulation model of the broiler production system that estimates the expected reduction in the resistant subpopulations of Salmonella and E. coli on-farm and on the final product, enabling producers and other stakeholders to compare the efficacy and cost of different strategies for mitigating antibiotic resistance.5) Deliver intervention options for reducing antimicrobial resistant Salmonella and E. coli to broiler producers through a diversity of extension and outreach activities. We will disseminate customizable mitigation strategy programs to broiler producers and other stakeholders using our industry, consumer advocacy, and government collaborators.
Project Methods
We will conduct systematic reviews of key questions within this area. High quality systematic reviews are increasingly becoming indispensable in forming the basis of evidence-based clinical management decisions. Similarly, the systematic reviews proposed here will aid in selection of management practices that improve poultry health, minimize resistance, and ensure that antibiotics are used minimally and responsibly.We will conduct field trials to evaluate the importance of specific selection pressures on antibiotic resistance within the poultry production system.The data will be analyzed with a novel network model. This model allows us to quantify the strength that each treatment has on selecting for resistance. We will generate estimates of significant associations (including their uncertainty) between predictors such as treatment and the resistance under study.We will conduct surveys of the industry about the ways in which antibiotics are being used. We will have matched samples from the farms participating in the surveys. We will be able to assess whether early selective bottlenecks (e.g. hatchery antibiotics) produce lasting effects that persist to the end of the production cycle (i.e. at slaughter).We will build systems dynamics models to enable long-term predictions about changes in resistance following changes in production. This model will have a user-friendly interface to allow stakeholders to assess specific situations in their system.The twin objectives of any Outreach and Extension program are to increase knowledge and change behaviors of the target populations. The specific measures for evaluating these programs will be determined in consultation with the Advisory Board to ensure that they reflect meaningful changes in industry practices. Pending development of these metrics, the generalizable expected results and outcomes of the Outreach and Extension activities include:· Exposure of the results, recommendations and tools to broiler industry veterinarians;· Statistically meaningful change in awareness of the degree to which current industry practices contribute to the selection for antibiotic resistance;· Significant portion of those who participate in the programs making one or more changes in their regular practices to minimize their contribution to antibiotic resistance.

Progress 01/15/15 to 01/14/20

Outputs
Target Audience:This project focuses on three main target audiences. First, the project informs researchers in the area of antibiotic resistance about the impacts of diverse selection pressures in the poultry industry. Through presentations and publications, the scientific community will be apprised of our progress on this topic. Second, the agricultural sector, and in particular the poultry industry, is an intended target audience. Professionals within the agricultural industry make decisions that can affect antibiotic resistance, and these decisions can have nothing to do with antibiotic usage. Through presentations, this target audience was reached and the challenges of selection pressures were discussed. Third, this project must bridge the gap between the animal health and public health sectors. The intended target audience is individuals, groups and governmental agencies that pursue activities at this nexus. Changes/Problems:We encountered several challenges during this project. When we began analyzing the metagenomic data generated in this project we noticed some inconsistencies. We spent much of 2017 investigating the problems and found that there were major quality control problems at our genomics center. We have worked with the center to redo all samples and are now waiting for the results. This caused a major delay in the project. There were also problems with consistency in the MF-qPCR assay that we use to quantify antimicrobial resistance genes. We have moved our work to a lab in North Carolina to assist with this assay. Reproducibility and reliability have been greatly enhanced by making this switch. What opportunities for training and professional development has the project provided?This project involved the training of two post-docs, three graduate students, and one undergraduate student. Also there were several technicians who got significant training during the study period. One post-doc served as a teaching assistant in a course in Veterinary Epidemiology for veterinary students and in a course in Epidemiology and Biostatistics for graduate students. She delivered lectures, administered and graded assignments and exams, and mentored students. Another post-doc trained on the project learning about the metagenomics shotgun sequencing data analysis and networkanalysis. How have the results been disseminated to communities of interest?The results of the work have been presented to the scientific community at numerous conferences, as mentioned in the Outcomes section of this report. Results and lessons learned from this project are regularly presented in different formats to stakeholders within the poultry industry. Information from this project has also been provided to organizations such as Codex through the U.S. Delegation. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Objective 1 of this project is to "Evaluate the evidence for specific disease management approaches in the broiler production system. We will perform formal, rigorous systematic reviews to examine the scientific basis for efficacy and impact of hatchery antibiotic administration and for approaches to necrotic enteritis (NE) management and control in the absence of AGP." Our group, using the advice of our Advisory Board, has decided to focus our first systematic review on the issue of NE. We have published the protocol. The abstract reads: Necrotic enteritis is one of the most common and economically important bacterial diseases affecting the broiler industry. Limitations on the use of antimicrobials have highlighted the need to evaluate the efficacy of non-antibiotic alternatives and management strategies. However, the available evidence on the efficacy of non-antibiotic interventions for necrotic enteritis has not been systematically synthesized. Here we present a protocol to conduct a systematic review of the literature to assess the efficacy of non-antibiotic interventions compared to antibiotic interventions in preventing and treating necrotic enteritis cases in broiler chickens. Objective 2 of this project is to "Quantify the effect of antibiotic administration in the hatchery and broiler farm on the selection of antibiotic resistance. We will use experimental trials to examine the ability of commonly used antibiotics in broiler production to select for resistant Salmonella and E. coli and overall resistance genes." We completed a large experimental field trial and are currently analyzing the data. Two-thousand-one-hundred birds (2,100) were assigned to seven (7) treatment groups with five (5) replicate pens per treatment and 60 birds per pen. The study was described in the 2016 Progress Report. In one analysis from this study, which was presented at conferences, 47 antibiotic resistance gene quantities were assessed with a quantitative PCR microfluidic device. In the pens that used oxytetracycline, there was an accumulation of antibiotic in the litter over the flock cycle, and because the litter was reused, there was carryover of tetracycline in the litter in the high dose pens. There were no significant differences in gene quantities among the groups, although there were differences in microbial community structure. Resistance gene quantities and microbiomes did change over time as each flocked aged. Administration of antibiotics for prevention of NE in broiler chickens might not play a significant role in augmenting antibiotic resistance within the broiler litter. Objective 3 of this project is to "Associate farm-specific practices, such as the use of antibiotic and non-antibiotic compounds, with farm-specific antimicrobial resistance patterns in Salmonella. We will leverage the data that we are collecting through the National Antimicrobial Resistance Monitoring System (NARMS) and our collaborations with committed industry partners to link samples from breeder flocks, hatchery and broiler growout flocks with farm practices identified in surveys of the broiler industry." With the assistance of this funding, The current goal is to track the trends on antibiotic usage within specific broiler farms and relate these usage patterns to matched sample sets of bacteriological test results. We are now enrolling companies and complexes within these companies. Then, each quarter the contact for the company and complex will select farms to be sampled during that quarter. Each farm will only be sampled once per year (at most), and the entire sampling effort is stratified by calendar quarter to ensure that farms within participating companies are sampled throughout the year. Each quarter, a minimum of 4 farms are sampled from each enrolled complex for each participating company. Sampling is conducted on 1 house per farm. Farm identity remains anonymous to the University of Minnesota; pre-labeled sampling supplies will use a coded system to maintain farm anonymity. Sample kits are sent from the University of Minnesota, and after sampling, samples will be shipped overnight on ice packs to the University of Minnesota. Samples consist of composite litter samples. In prior years, bootsocks were also used to sample the broiler houses. However, the bootsock samples were more time-consuming for study participants to collect and often had inferior results compared to litter. Therefore, we have switched to the collection of two composite litter samples per house. These litter samples are collected as close to slaughter age as possible. The litter samples are cultured for Salmonella, E. coli and Campylobacter. DNA is also extracted from one litter sample per house. In the laboratory, each Salmonella isolate is confirmed by PCR and then serotyped. Each Campylobacter isolate is speciated (e.g. jejuni versus coli). Antibiotic resistance profiles are generated for all bacterial isolates recovered, including E. coli. Antibiotic resistance genes will be quantified from the litter samples. Antibiotic usage information will be collected from the sampled farms in real time. These data include the use of hatchery antibiotics, ionophores, and any other antibiotics used in the feed or water. Demographic data will be collected, including number of birds placed, number of birds slaughtered, and age and weight at slaughter.

Publications

  • Type: Journal Articles Status: Published Year Published: 2020 Citation: Galarneau, K.D., R.S. Singer, and R.W. Wills. A system dynamics model for disease management in poultry production. Poultry Science. 99:5547-5559. doi: 10.1016/j.psj.2020.08.011.
  • Type: Conference Papers and Presentations Status: Other Year Published: 2019 Citation: Assessing the Effects of Disease Prevention Antibiotics on Antimicrobial Resistance in Broiler Litter. Randall S. Singer. 5th International Symposium on the Environmental Dimension of Antibiotic Resistance. June 2019, Hong Kong.
  • Type: Journal Articles Status: Other Year Published: 2020 Citation: Bueno, Irene; Smith, Emily; Hwang, Haejin; Nault, Andre J.; Valeris-Chacin, Robert; Singer, Randall S.. (2020). Efficacy of antibiotic and non-antibiotic interventions in preventing and treating necrotic enteritis in broiler chickens: a protocol for a systematic review. Retrieved from the University of Minnesota Digital Conservancy, https://hdl.handle.net/11299/216460.
  • Type: Conference Papers and Presentations Status: Other Year Published: 2018 Citation: Pen trial to evaluate effects of necrotic enteritis disease prevention antibiotics on antimicrobial resistance in broiler litter. R. Singer, B. Wass, T. Johnson, C. Hofacre, E. Lamont, D. Dopfer, R. Wills, The 1th International Symposium of Veterinary Epidemiology and Economics (ISVEE), November 2018, Thailand.


Progress 01/15/17 to 01/14/18

Outputs
Target Audience:This project focuses on three main target audiences. First, the project informs researchers in the area of antibiotic resistance about the impacts of diverse selection pressures in the poultry industry. Through presentations and publications, the scientific community will be apprised of our progress on this topic. Second, the agricultural sector, and in particular the poultry industry, is an intended target audience. Professionals within the agricultural industry make decisions that can affect antibiotic resistance, and these decisions can have nothing to do with antibiotic usage. Through presentations, this target audience was reached and the challenges of selection pressures were discussed. Third, this project must bridge the gap between the animal health and public health sectors. The intended target audience is individuals, groups and governmental agencies that pursue activities at this nexus. Specific audiences that were targeted included the scientific community attending the Conference for Research Workers in Animal Disease in Chicago (Dec 2016, Dec 2017),the Food Research Institute at UW-Madison during the spring conference (2016), and the American Association of Avian Pathologists (July 2018). Changes/Problems:When we began analyzing the metagenomic data generated in this project we noticed some inconsistencies. We spent much of 2017 investigating the problems and found that there were major quality control problems at our genomics center. We have worked with the center to redo all samples and are now waiting for the results. This caused a major delay in the project. What opportunities for training and professional development has the project provided?Year 3 of this project involved the training of two post-docs, three graduate students, and one undergraduate student. Also there were several technicians who got significant training during the study period. One post-doc served as a teaching assistant in a course in Veterinary Epidemiology for veterinary students and in a course in Epidemiology and Biostatistics for graduate students. She delivered lectures, administered and graded assignments and exams, and mentored students. Another post-doc trained on the project learning about the metagenomics shotgun sequencing data analysis and network analysis. How have the results been disseminated to communities of interest?The results of the work have been presented to the scientific community at numerous conferences, as mentioned in the Outcomes section of this report. What do you plan to do during the next reporting period to accomplish the goals?During the next reporting period we will proceed with the Systematic Review in Objective 1. We will complete the analysis of the microbiome and metagenome data from the experimental pen trial, once the quality control problems have been fixed at our genomics center. Metagenomics pipelines for quantification ofassociations between treatment cohorts and shotgun sequencing data will be applied to the new experimental data from theproject when they become available. Network analysis approaches are the focus of this analysis. Detailed mathematicalmodels will be developed that analyze the data from Objectives 2 and 3 and the literature. The system dynamics model ofbroiler production from chick delivery to harvest will be expanded to incorporate the dynamics of antimicrobial resistancedevelopment and transfer. The model will also be expanded to incorporated additional production information and economicsof production and health. Information from other objectives of the research project will also be integrated into the systemsdynamics model to further refine the model. The model will be used in outreach programs and made available to producersand other stakeholders enable them to compare the efficacy and cost of different strategies for mitigating antibiotic resistance.The extension program will be implemented with significant outreach to industry.

Impacts
What was accomplished under these goals? Objective 1 of this project is to "Evaluate the evidence for specific disease management approaches in the broiler production system. We will perform formal, rigorous systematic reviews to examine the scientific basis for efficacy and impact of hatchery antibiotic administration and for approaches to necrotic enteritis (NE) management and control in the absence of AGP." Our group, using the advice of our Advisory Board, has decided to focus our first systematic review on the issue of NE. We are assembling our team and are beginning to draft our search strings to conduct this systematic review. Objective 2 of this project is to "Quantify the effect of antibiotic administration in the hatchery and broiler farm on the selection of antibiotic resistance. We will use experimental trials to examine the ability of commonly used antibiotics in broiler production to select for resistant Salmonella and E. coli and overall resistance genes." We completed a large experimental field trial and are currently analyzing the data. Two-thousand-one-hundred birds (2,100) were assigned to seven (7) treatment groups with five (5) replicate pens per treatment and 60 birds per pen. The study was described in the 2016 Progress Report. The metagenomic data that was analyzed in 2016 had to be completely rerun due to quality control problems at our genomics center. Data are not yet available for analysis. Metagenomics pipelines for quantification of associations between treatment cohorts and shotgun sequencing data have beenoptimized and will be applied to the new experimental data from the project as soon as they become available. Networkanalysis approaches are the focus of this analysis. Causal and differential network analysis is being applied to existing datasets. In addition, microbiome analysis is being applied to 16S microbiome data sets and their associated treatment and riskfactors for disease outcomes. These skills are important for the analysis of the data from the current project. Objective 3 of this project is to "Associate farm-specific practices, such as the use of antibiotic and non-antibiotic compounds, with farm-specific antimicrobial resistance patterns in Salmonella. We will leverage the data that we are collecting through the National Antimicrobial Resistance Monitoring System (NARMS) and our collaborations with committed industry partners to link samples from breeder flocks, hatchery and broiler growout flocks with farm practices identified in surveys of the broiler industry." During 2016, our on-farm NARMS program collected data from over 150 farms sampled every flock cycle. Over 10 of the largest broiler producing companies were represented. Salmonella and Campylobacter were cultured from all samples. Antibiotic use information was collected from every farm. These data are being analyzed separately as part of the NARMS program, but our causal network modeling approaches developed in Objective 2 will also be applied to these data. Objective 4 of this project is to "Develop tools for predicting efficacy, cost-effectiveness and consequences of antimicrobial resistance mitigation strategies. This extension-based objective will build a dynamic simulation model of the broiler production system that estimates the expected reduction in the resistant subpopulations of Salmonella and E. coli on-farm and on the final product, enabling producers and other stakeholders to compare the efficacy and cost of different strategies for mitigating antibiotic resistance." A system dynamics model of broiler production from chick delivery to harvest was developed using Vensim modeling software. The model follows flocks of birds during their grow-out period within the broiler house. Parameters that can be modified include delivery size, down time, grow-out period, and litter age at removal. In addition, the model simulates spread of infectious agents in the broiler house over time. Sub-models incorporate susceptible-infected-recovered (SIR) models to follow transmission dynamics of representative organisms such as Salmonella, necrotic enteritis, and coccidiosis in broiler production. Disease severity triggers antimicrobial use in the model. The efficacy of each selected antimicrobial is adjustable for each pathogen. Disease parameters that can be modified include morbidity, mortality, disease start date, number initially infected, mortality threshold for antimicrobial use, and disease specific antimicrobial efficacy. Simulation of the model demonstrates the dynamics of susceptible, infected, recovered, dead, and total harvested birds in response to changing infectivity, mortality rates, environmental factors, and antimicrobial use over multiple grow-out flocks. The model can be modified to accommodate other disease agents and interventions. It will be expanded to simulate the dynamics of antimicrobial resistance concurrently with disease and animal production dynamics and antimicrobial use. Objective 5 of this project is to "Deliver intervention options for reducing antimicrobial resistant Salmonella and E. coli to broiler producers through a diversity of extension and outreach activities. We will disseminate customizable mitigation strategy programs to broiler producers and other stakeholders using our industry, consumer advocacy, and government collaborators." To date we have given numerous presentations to the poultry industry regarding antibiotic stewardship, antimicrobial resistance, and current knowledge about measures to mitigate resistance in the poultry production system.

Publications

  • Type: Conference Papers and Presentations Status: Other Year Published: 2017 Citation: "Antibiotic Resistance in Poultry Production", Randall Singer, International Poultry Council, Banff, Canada, September 2017
  • Type: Conference Papers and Presentations Status: Other Year Published: 2017 Citation: "Empleo de antimicrobianos y su resistencia en la producción avícola de los USA", Randall Singer, XXV Congreso Latinoamericano de Avicultura, Guadalajara, Mexico, September 2017
  • Type: Conference Papers and Presentations Status: Other Year Published: 2017 Citation: "The Changing Paradigm of Antimicrobial Use in Veterinary Medicine", Randall Singer, The National Academies of Sciences - Combating Antimicrobial Resistance: A One Health Approach to a Global Threat ⿿ A Workshop, Washington, D.C., June 2017


Progress 01/15/16 to 01/14/17

Outputs
Target Audience:This project focuses on three main target audiences. First, the project informs researchers in the area of antibiotic resistance about the impacts of diverse selection pressures in the poultry industry. Through presentations and publications, the scientific community will be apprised of our progress on this topic. Second, the agricultural sector, and in particular the poultry industry, is an intended target audience. Professionals within the agricultural industry make decisions that can affect antibiotic resistance, and these decisions can have nothing to do with antibiotic usage. Through presentations, this target audience was reached and the challenges of selection pressures were discussed. Third, this project must bridge the gap between the animal health and public health sectors. The intended target audience is individuals, groups and governmental agencies that pursue activities at this nexus. Specific audiences that were targeted included the scientific community attending the Conference for Research Workers in Animal Disease in Chicago (Dec 2016)and the Food Research Institute at UW-Madison during thespring conference (2016). Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Year 2 of this project involved the training of two post-docs, three graduate students, and one undergraduate student. Also there were several technicians who got significant training during the study period. One post-doc served as a teaching assistant in a course in Veterinary Epidemiology for veterinary students and in a course in Epidemiology and Biostatistics for graduate students. She delivered lectures, administered and graded assignments and exams, and mentored students. She gave an oral presentation on her work at a national scientific meeting, the 2016 Conference for Workers in Animal Disease. Another post-doc trained on the project learning about the metagenomics shotgun sequencing data analysis and network analysis. He trained a summer scholar student who presented her results about microbiome data analysis during the CRWAD meeting in Chicago in December of 2016. How have the results been disseminated to communities of interest?The results of the work have been presented to the scientific community at numerous conferences, as mentioned in the Outcomes section of this report. What do you plan to do during the next reporting period to accomplish the goals?During the next reporting period we will proceed with the Systematic Review in Objective 1. We will complete the analysis of the microbiome and metagenome data from the experimental pen trial. Metagenomics pipelines for quantification of associations between treatment cohorts and shotgun sequencing data will be applied to the new experimental data from the project when they become available. Network analysis approaches are the focus of this analysis. Detailed mathematical models will be developed that analyze the data from Objectives 2 and 3 and the literature. The system dynamics model of broiler production from chick delivery to harvest will be expanded to incorporate the dynamics of antimicrobial resistance development and transfer. The model will also be expanded to incorporated additional production information and economics of production and health. Information from other objectives of the research project will also be integrated into the systems dynamics model to further refine the model. The model will be used in outreach programs and made available to producers and other stakeholders enable them to compare the efficacy and cost of different strategies for mitigating antibiotic resistance. The extension program will be implemented with significant outreach to industry.

Impacts
What was accomplished under these goals? Objective 1 of this project is to "Evaluate the evidence for specific disease management approaches in the broiler production system. We will perform formal, rigorous systematic reviews to examine the scientific basis for efficacy and impact of hatchery antibiotic administration and for approaches to necrotic enteritis (NE) management and control in the absence of AGP." Our group, using the advice of our Advisory Board, has decided to focus our first systematic review on the issue of NE. We are assembling our team and are beginning to draft our search strings to conduct this systematic review. Objective 2 of this project is to "Quantify the effect of antibiotic administration in the hatchery and broiler farm on the selection of antibiotic resistance. We will use experimental trials to examine the ability of commonly used antibiotics in broiler production to select for resistant Salmonella and E. coli and overall resistance genes." We completed a large experimental field trial and are currently analyzing the data. Two-thousand-one-hundred birds (2,100) were assigned to seven (7) treatment groups with five (5) replicate pens per treatment and 60 birds per pen. The pen facility was divided into five (5) blocks with each block containing each of the seven (7) treatment groups. Treatment groups were assigned to pens using randomized complete block. The study began when birds are placed (day-of-hatch; DOT 0), at which time birds were allocated to experimental pens. Only healthy birds were selected. On DOT 0, group body weights were recorded by pen. No birds were replaced during the course of the study. Chicks were raised in 5 x 10 feet floor pens (stocking density of 0.83 feet-squared per bird) with reused litter obtained from a previous broiler type study, in a solid-sided barn with dirt floors (as used in conventional U.S. broiler house environments), and under ambient humidity. Litter was not replaced or amended during the course of this study. Feed and water was available ad libitum throughout the trial. Each pen contained one tube feeder and three nipple drinkers (60 bird to feeder/20 drinker ratio). Treatment groups were the following: T1 Narasin (70g/ton); T2 Narasin (70g/ton)/Bacitracin (50g/ton); T3 Narasin (70g/ton)/Bambermycin (2g/ton); T4 Narasin (70g/ton)/Oxytetracycline (100g/ton); T5 Narasin (70g/ton)/Oxytetracycline (400g/ton); T6 Narasin (70g/ton)/Virginiamycin (20g/ton); T7 No Narasin/No Antibiotic. Treatments were applied via the feed from Days 0 through 28 of each flock. Birds were sacrificed at Day 35 and there was a 7 day downtime between flocks. Three total flocks were placed in each pen, each for a 35-day flock cycle. Litter samples were collected from all pens on DOT 0, 7, 14, 21, 28, and 35. A Whirlpak bag was filled with a composite sample of litter. Bags were labeled to identify pen number and sample date. Samples were shipped overnight on ice to Dr. Randall Singer, University of Minnesota. Litter samples were cultured for Salmonella and E. coli. E. coli cultures were performed using a spiral plater to get enumeration data, and each sample was cultured on multiple plates, including agar with antibiotics to get specific resistance phenotype quantitative data. Results are currently being analyzed. DNA was extracted from all litter samples. These DNA aliquots are currently being analyzed for specific resistance gene quantities. All samples have been assessed for their microbiome and complete metagenome. These results are being analyzed currently. Litter samples were also prepped for antibiotic metabolite testing, and the tetracycline assay has been completed. Results are being analyzed. Metagenomics pipelines for quantification of associations between treatment cohorts and shotgun sequencing data have been optimized and will be applied to the new experimental data from the project as soon as they become available. Network analysis approaches are the focus of this analysis. Causal and differential network analysis is being applied to existing data sets. In addition, microbiome analysis is being applied to 16S microbiome data sets and their associated treatment and risk factors for disease outcomes. These skills are important for the analysis of the data from the current project. Objective 3 of this project is to "Associate farm-specific practices, such as the use of antibiotic and non-antibiotic compounds, with farm-specific antimicrobial resistance patterns in Salmonella. We will leverage the data that we are collecting through the National Antimicrobial Resistance Monitoring System (NARMS) and our collaborations with committed industry partners to link samples from breeder flocks, hatchery and broiler growout flocks with farm practices identified in surveys of the broiler industry." During 2016, our on-farm NARMS program collected data from over 150 farms sampled every flock cycle. Over 10 of the largest broiler producing companies were represented. Salmonella and Campylobacter were cultured from all samples. Antibiotic use information was collected from every farm. These data are being analyzed separately as part of the NARMS program, but our causal network modeling approaches developed in Objective 2 will also be applied to these data. Objective 4 of this project is to "Develop tools for predicting efficacy, cost-effectiveness and consequences of antimicrobial resistance mitigation strategies. This extension-based objective will build a dynamic simulation model of the broiler production system that estimates the expected reduction in the resistant subpopulations of Salmonella and E. coli on-farm and on the final product, enabling producers and other stakeholders to compare the efficacy and cost of different strategies for mitigating antibiotic resistance." A system dynamics model of broiler production from chick delivery to harvest was developed using Vensim modeling software. The model follows flocks of birds during their grow-out period within the broiler house. Parameters that can be modified include delivery size, down time, grow-out period, and litter age at removal. In addition, the model simulates spread of infectious agents in the broiler house over time. Sub-models incorporate susceptible-infected-recovered (SIR) models to follow transmission dynamics of representative organisms such as Salmonella, necrotic enteritis, and coccidiosis in broiler production. Disease severity triggers antimicrobial use in the model. The efficacy of each selected antimicrobial is adjustable for each pathogen. Disease parameters that can be modified include morbidity, mortality, disease start date, number initially infected, mortality threshold for antimicrobial use, and disease specific antimicrobial efficacy. Simulation of the model demonstrates the dynamics of susceptible, infected, recovered, dead, and total harvested birds in response to changing infectivity, mortality rates, environmental factors, and antimicrobial use over multiple grow-out flocks. The model can be modified to accommodate other disease agents and interventions. It will be expanded to simulate the dynamics of antimicrobial resistance concurrently with disease and animal production dynamics and antimicrobial use. Objective 5 of this project is to "Deliver intervention options for reducing antimicrobial resistant Salmonella and E. coli to broiler producers through a diversity of extension and outreach activities. We will disseminate customizable mitigation strategy programs to broiler producers and other stakeholders using our industry, consumer advocacy, and government collaborators." To date we have given numerous presentations to the poultry industry regarding antibiotic stewardship, antimicrobial resistance, and current knowledge about measures to mitigate resistance in the poultry production system.

Publications

  • Type: Conference Papers and Presentations Status: Other Year Published: 2016 Citation: "Antibiotic Usage in Poultry: Assessing the Effects on Antibiotic Resistance and Human Health", Randall Singer, International Production and Processing Expo, Atlanta, GA, January 2016
  • Type: Conference Papers and Presentations Status: Other Year Published: 2016 Citation: "System dynamics model of broiler production as a tool for studying disease and intervention" K.D. Galarneau, R. S. Singer, R.W. Wills, 2016 Conference for Research Workers in Animal Disease, Chicago, IL
  • Type: Conference Papers and Presentations Status: Other Year Published: 2016 Citation: "Uso de antibi�ticos y resistencia antimicrobiana: una relaci�n compleja", Randall Singer, ANECA - the Mexican national association of poultry science specialists, Ixtapa, Mexico, April 2016
  • Type: Conference Papers and Presentations Status: Other Year Published: 2016 Citation: "Managing Antibiotic Use and Resistance in the Poultry Production System", Randall Singer, Institute of Medicine, Washington, DC, April 2016
  • Type: Conference Papers and Presentations Status: Other Year Published: 2016 Citation: "Mitigating Antimicrobial Resistance in Agricultural Systems", Randall Singer, American Society for Microbiology, Boston, MA, June 2016
  • Type: Conference Papers and Presentations Status: Other Year Published: 2016 Citation: "Antibiotic Resistance Development", Randall Singer, Poultry Science Association, New Orleans, LA, July 2016
  • Type: Conference Papers and Presentations Status: Other Year Published: 2016 Citation: "El uso de antibi�ticos en aves de corral: Evaluaci�n de los efectos sobre la resistencia a los antibi�ticos y la salud humana", Randall Singer, AMEVEA, XV Seminario Internacional de Patolog�a y Producci�n Av�cola, Santiago, Chile, October 2016
  • Type: Conference Papers and Presentations Status: Other Year Published: 2016 Citation: "The diversity and abundance of bacterial taxa and antimicrobial resistance factors in cecal shotgun metagenomes chicken treated with anticoccidial drugs and growth promotors", Dorte Dopfer, Food Research Institute (FRI), University of Wisconsin, Madison, WI, May 2016


Progress 01/15/15 to 01/14/16

Outputs
Target Audience:This project focuses on three main target audiences. First, the project informs researchers in the area of antibiotic resistance about the impacts of diverse selection pressures in the poultry industry. Through presentations and publications, the scientific community will be apprised of our progress on this topic. Second, the agricultural sector, and in particular the poultry industry, is an intended target audience. Professionals within the agricultural industry make decisions that can affect antibiotic resistance, and these decisions can have nothing to do with antibiotic usage. Through presentations, this target audience was reached and the challenges of selection pressures were discussed. Third, this project must bridge the gap between the animal health and public health sectors. The intended target audience is individuals, groups and governmental agencies that pursue activities at this nexus. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Year 1 of this project involved the training of two post-docs, three graduate students, and one undergraduate student. Also there were several technicians who got significant training during the study period. There will be a significant increase in training and development in Year 2 of this project. How have the results been disseminated to communities of interest?The studies in this project have only recently begun. Currently there is nothing to report regarding results. Year 2 will begin the dissemination of results to communities of interest. What do you plan to do during the next reporting period to accomplish the goals?During the next reporting period we will proceed with the Systematic Review in Objective 1. We will analyze the samples collected in the first experimental trial of Objective 2. We will conduct the second trial of Objective 2. We will be conducting the study of Objective 3. Detailed mathematical models will be developed that analyze the data from Objectives 2 and 3 and the literature. The system dynamics models of Objective 4 will be developed and applied to antimicrobial resistance in the poultry production system. The extension program will be implemented with significant outreach to industry.

Impacts
What was accomplished under these goals? In Objective 2, we have the goal to "Quantify the effect of prevention antibiotic administration in the hatchery and broiler farm on the selection of antibiotic resistance." From August through December 2015 we conducted the first experimental trial of this objective. Two-thousand-one-hundred birds (2,100) were assigned to seven (7) treatment groups with five (5) replicate pens per treatment and 60 birds per pen. The pen facility was divided into five (5) blocks with each block containing each of the seven (7) treatment groups. Treatment groups were assigned to pens using randomized complete block. The study began when birds are placed (day-of-hatch; DOT 0), at which time birds were allocated to experimental pens. Only healthy birds were selected. On DOT 0, group body weights were recorded by pen. No birds were replaced during the course of the study. Chicks were raised in 5 x 10 feet floor pens (stocking density of 0.83 feet-squared per bird) with reused litter obtained from a previous broiler type study, in a solid-sided barn with dirt floors (as used in conventional U.S. broiler house environments), and under ambient humidity. Litter was not replaced or amended during the course of this study. Feed and water was available ad libitum throughout the trial. Each pen contained one tube feeder and three nipple drinkers (60 bird to feeder/20 drinker ratio). Treatment groups were the following: T1 Narasin (70g/ton); T2 Narasin (70g/ton)/Bacitracin (50g/ton); T3 Narasin (70g/ton)/Bambermycin (2g/ton); T4 Narasin (70g/ton)/Oxytetracycline (100g/ton); T5 Narasin (70g/ton)/Oxytetracycline (400g/ton); T6 Narasin (70g/ton)/Virginiamycin (20g/ton); T7 No Narasin/No Antibiotic. Treatments were applied via the feed from Days 0 through 28 of each flock. Birds were sacrificed at Day 35 and there was a 7 day downtime between flocks. Three total flocks were placed in each pen, each for a 35-day flock cycle. In the first flock cycle, ten (10) chicks per pen were tagged, color-coded (for identification), and orally dosed (gavaged) with a cocktail of three (3) Salmonella Kentucky and two (2) Escherichia coli isolates on DOT 0. These isolates were marked and were meant to serve as potential donors and recipients of antibiotic resistance plasmids. Litter samples were collected from all pens on DOT 0, 7, 14, 21, 28, and 35. A Whirlpak bag was filled with a composite sample of litter. Bags were labeled to identify pen number and sample date. Samples were shipped overnight on ice to Dr. Randall Singer, University of Minnesota. Litter samples were cultured for Salmonella and E. coli. E. coli cultures were performed using a spiral plater to get enumeration data, and each sample was cultured on multiple plates, including agar with antibiotics to get specific resistance phenotype quantitative data. Results are currently being analyzed. DNA was extracted from all litter samples. These DNA aliquots are currently being analyzed for specific resistance gene quantities, microbiome and metagenome. Litter samples were also prepped for antibiotic metabolite testing, which is currently underway. Members of the Advisory Board have convened two teleconferences, on June 8 and October 2, 2015. Members currently include: Dane Bernard, Keystone Foods; Sarah Brew, Faegre-Baker-Daniels; John Glisson, U.S. Poultry & Egg Association; David Goldman, USDA FSIS; Barbara Mahon, CDC; Barb Masters, OFW Law; Pat McDermott, FDA-CVM; Ashley Peterson, National Chicken Council; Ron Phillips, Animal Health Institute; Mike Robach, Cargill, Inc.; and Caroline Smith DeWaal, (formerly with CSPI). Shaun Kennedy, Co-PI on this project led the two calls. Dr. Singer presented research information to the Advisory Board and then left the call so that the Advisory Board could meet independently from the researchers and speak candidly and anonymously. After the discussion of research objectives, the Advisory Board made the recommendation to focus the first systematic review on necrotic enteritis and antibiotic use. The research team agreed with this recommendation. The Advisory Board made the recommendation to include a treatment group in the experimental trial of Objective 2 of No Antibiotic / No ionophore. The research team agreed with this recommendation but included a coccidia vaccine for this group because of concerns that the morbidity would be too high without any coccidia prevention. Based on the recommendation of the Advisory Board, the focus of the first Systematic Review of Objective 1 is related to necrotic enteritis and antibiotic use. The team is currently being formed to begin the design and review process. In Objective 3, we are enrolling the companies and farms that will be included in the longitudinal investigation. This work will largely commence in Year 2. The data analysis related to causal network modeling is being led by Dr. Dopfer and her post-doc. They are currently using pre-existing microbiome data from studies on antibiotic administration to develop and refine longitudinal network analyses. These approaches will be applied to the emergence of antimicrobial resistance factors in poultry in Year 2. Dr. Wills has begun the development of the system dynamics modeling of the poultry production system as described in Objective 4. Major progress will be reported in Year 2.

Publications

  • Type: Conference Papers and Presentations Status: Other Year Published: 2015 Citation: "Making Sense of the Not-So-Simple Relationship between Antibiotic Use and Antibiotic Resistance", Randall Singer, National Meeting on Poultry Health, Processing ,and Live Production, Ocean City, MD, October 2015
  • Type: Conference Papers and Presentations Status: Other Year Published: 2015 Citation: "A One Health Approach to Antibiotic Resistance", Randall Singer, National Meeting of the American Society of Agronomy, the Crop Science Society of America, and the Soil Science Society of America, Minneapolis, MN, November 2015