A Scoping Review of Existing Models of Antimicrobial Resistance and Evidence to Parameterize a One Health Model in a Swedish Food System Context. (March 2022)

• Record Type: Journal Article
• Authors: Cousins, M.; Parmley, J.; Neiterman, E.; Greer, A.; Lambraki, I.; Carson, C.; Wernli, D.; Jørgensen, P. Søgaard; Majowicz, S.
• Is Part Of: International journal of infectious diseases. Volume 116(2022)Supplement
• Page Start: S7
• Publication Date: 2022-03
• Subjects: Communicable diseases -- Periodicals; Communicable Diseases -- Periodicals; Communicable diseases; Periodicals; Electronic journals; 616.9
• DOI: 10.1016/j.ijid.2021.12.017
• Languages: English
• ISSNs: 1201-9712

Abstract

Abstract : Purpose: Antimicrobial resistance (AMR) is a global One Health issue. Simulation modelling is used to understand AMR in many contexts but there is lack of integration of models across sectors. The purpose of our research is to explore whether it is possible to quantify an existing model consisting of 92 compartments deemed important to the emergence and transmission of AMR in different parts of the One Health spectrum relevant to the food system of a high-income country (Sweden). Methods & Materials: We performed a literature scan to address 2 research objectives: (1) to identify current models of AMR representing different parts of the One Health system, and (2) to identify existing evidence to inform such models in a high-income (specifically Swedish) context. Searches were conducted in Google, Google Scholar, and Pubmed and articles that fit pre-determined criteria were read, models were categorized by type in an excel database. Results: Of the 140 simulation models identified (Objective 1), the majority of the models were deterministic (48%), compartmental (69%), and limited to one sector (57%). The majority of the AMR-specific models were human-focussed (54%) and were either at a macroscopic-level (transfers of patients between hospitals and community members; 22%) or microscopic (pathogen) level (within-host models of gene transfer and AMR emergence; 27%), with very few multi-level models including both mechanisms (2%). A total of 376 data sources were included to inform the models (Objective 2). These sources provided quantitative and qualitative data that addressed 79 of the 92 compartments from peer-reviewed (138 citations) and grey literature (72 web pages, 72 reports, and 94 other sources). Major data gaps included the environmental sector, wildlife, and nodes that are broad, ill-defined, or abstract in nature (e.g. human experience and knowledge). Conclusion: The number of models that address AMR beyond one context or sector is limited. Existing models addressing different parts of the system have the potential to be integrated together into a more holistic model. Mixed methods may be useful to incorporate a variety of data sources to fill current gaps.