NARMS conducts research to develop and test methods that improve the scientific basis of NARMS data. Current work at FDA to help advance the objectives of the 2021-2025 NARMS Strategic Plan falls into general categories:
On this page:
Targeted Surveys of Additional Commodities
NARMS scientists perform targeted surveys to explore antimicrobial resistance in commodities and bacteria that are not included in routine NARMS data collection. These activities help broaden our understanding of resistance patterns in the food chain and inform decisions on whether and how to expand the scope of NARMS monitoring. Recent pilot surveys are listed below.
- Retail veal (2018-2021): A survey of 538 veal samples from 7 states was completed by FDA in 2020 and the manuscript was published. We anticipate that veal testing will occur on a periodic basis (e.g., every 3-4 years) depending on competing priorities and resources.
- Sheep, lamb, goats, and catfish (2020-2022): Surveys of sheep, lamb, goats, and catfish at slaughter were conducted by FSIS from 2020-2022. Data generated from these surveys will be used determine if periodic sampling of sheep, lamb, goat and catfish at slaughter is warranted.
- Chicken giblets (2021-2022): Analysis of US foodborne outbreaks reported to FSIS between 2000 and 2016 show that chicken giblets were increasingly associated with Campylobacter and Salmonella infections in humans (Lanier et al. 2018). Since little is known about the antimicrobial resistance in pathogens from chicken giblets, and their similarity to the same bacteria from whole meats, the product was selected by FDA for a study starting in 2021. Genomic sequencing and antibiotic susceptibility testing of the isolates is ongoing.
- Animal Feed (2005-2011): FDA published a manuscript on the analysis of 1,025 animal food samples (647 pet food and 378 animal feed) collected during 2005–2011 for two indicator organisms (Escherichia coli and Enterococcus spp.). These findings help establish a historic baseline for the prevalence and antimicrobial resistance of these indicator organisms among U.S. animal food products. FDA is also in the process of analyzing historical samples tested by FDA’s Office of Regulatory Affairs for antimicrobial resistant Salmonella in animal food and human food.
- Other bacteria: FDA initiated a retail seafood survey in 2021 that included testing for Aeromonas and Vibrio. As a result of this work, AMR monitoring of salmon, shrimp, and tilapia was made a routine part of NARMS in 2022. In addition, USDA and FDA have been examining enteric bacteria other than E. coli to capture a broader view of resistance in the Enterobacterales (Citrobacter, Enterobacter, Klebsiella, etc.), many of which also can colonize humans. Vet-LIRN has continued to expand its surveillance activities, with the additions of Campylobacter spp. and Enterobacter cloacae in 2021, two important pathogens with One Health importance for humans and animals. We are also leveraging the FDA Vet-LIRN network of laboratories to perform resistance surveillance of aquaculture pathogens and are planning work to help establish interpretive criteria for these pathogens both to aid the veterinarian and to improve reporting on resistance trends.
Genomics and Metagenomics Research
Goal 2 of the NARMS Strategic Plan outlines plans to take advantage of genomic and metagenomic methods to provide comprehensive microbiological data at the DNA sequence level. In 2015, NARMS agencies began routinely using whole genome sequencing (WGS). This technology represents a major advancement in microbial monitoring by providing definitive information on the relatedness of bacteria from different sources and their carriage of specific genes, including antimicrobial resistance genes. Genomic sequences from NARMS are uploaded to the NCBI Pathogen Detection web portal on a weekly basis for access by all stakeholders.
Metagenomics is the study of the total genetic content of a complex biological sample that is assessed by direct sequencing. For resistance monitoring, metagenomic sequencing reveals the overall complement of resistance genes (the “resistome”) present in a sample, including genes from unknown bacteria, those that are not easy to cultivate (e.g., anaerobes), and those for which no culture methods exist.
The use of metagenomics to understand the dynamics of microbial ecosystems is an area of intense study and immense promise for public health. It provides a much more complete picture of the taxonomic structure and resistance profile of a given source and can be used to evaluate the impact of different conditions on microbial populations. Information about abundance and diversity of AMR genes in the microbiome are likely to come increasingly from metagenomic studies and the integration of metagenomic and WGS data. In NARMS, metagenomics is being used to assess the resistome of water, meats, animals, and food.
Molecular Genetic Studies
Molecular genetics studies are used for many purposes. They can uncover the role of known genes in antimicrobial resistant bacteria and find new ones. Other studies include genome-wide association studies (GWAS) to reveal linkages of important traits (e.g., virulence, heavy metal and disinfectant resistance) with AMR resistance, modeling of plasmid distribution to understand resistance gene origin, spread, and comparisons of the evolutionary relationships of NARMS bacteria with those found in other countries.
Epidemiological and Statistical Research
NARMS scientists conduct epidemiological and statistical research to understand the sources of resistance, to identify emerging resistance trends, to help understand conditions associated with resistance in different environments, and to measure the possible impact of interventions. NARMS is currently examining whether different packaging types are associated with AMR, how temporal trends vary by commodity type, whether seasonal resistance patterns occur in samples from different animals, and the difference in genetic determinants in E. coli from meat and poultry with different antibiotic usage claims. This work helps us understand the drivers of resistance and may help identify control points where antimicrobial resistance could be better mitigated.
The large datasets generated by genomic sequencing have shifted the expertise needed in surveillance towards bioinformatics and machine learning and increased the need for high-performance computing power and computer programming expertise. While machine learning in NARMS is in the early phases of use, it has already proven very powerful for analyzing genome-wide variations associated with antimicrobial susceptibility. NARMS scientists are applying machine learning to analyze metagenomics data sets and to determine the origin of resistant strains, among other work.