Enhancing antimicrobial resistance surveillance and research: a systematic scoping review on the possibilities, yield and methods of data linkage studies
Series / Report no.
Open Access
Type
Article
Language
en
Date
2025-03-29
Research Projects
Organizational Units
Journal Issue
Title
Enhancing antimicrobial resistance surveillance and research: a systematic scoping review on the possibilities, yield and methods of data linkage studies
Translated Title
Published in
Antimicrob Resist Infect Control 2025; 14(1):25
Abstract
Surveillance data on antimicrobial resistance (AMR) determinants such as antibiotic use, prevalence of AMR, antimicrobial stewardship, and infection prevention and control are mostly analysed and reported separately, although they are inextricably linked to each other. The impact of surveillance and research can be enhanced by linking these data. This systematic scoping review aims to summarize the studies that link AMR data and evaluate whether they yield new results, implications, or recommendations for practice.
For this review, data linkage is defined as the process of linking records, from at least two independent data sources on either (I) at least two AMR determinants or (II) one AMR determinant and one or more general population characteristics. Data linkage should be performed on the level of a certain entity which, in the context of this review, can encompass persons, healthcare institutes, geographical regions or countries. A systematic literature search was performed on February 7th 2025 in Embase.com, PubMed and Scopus to identify AMR data linkage studies.
Forty-eight articles were included in our review. Most data linkage studies used two data sources, and most studies were published in the last 5 years (n = 23 in 2020-2024). A predominance of studies linked data on geographical location, and thirteen studies linked data on individual patient level. Findings demonstrate that the majority of studies (43/48) had added value and provided recommendations for clinical practice and future policies or had suggestions for further research and surveillance. Additionally, data linkage studies appeared to be suitable for hypothesis generating. Several limitations were identified. Most studies had ecological designs, which are prone to ecological fallacy and unobserved confounding, making it hard to establish causality.
This systematic scoping review showed that AMR data linkage studies are increasingly performed. They have potential to gain a more comprehensive understanding of AMR dynamics by generating hypotheses, assisting in optimisation of surveillance, and interpretation of data in the context of guideline/policy development. To increase the added value of data linkage, more studies should be performed to improve knowledge on methodological approaches, data access, data management, and governance issues.
Not applicable.
For this review, data linkage is defined as the process of linking records, from at least two independent data sources on either (I) at least two AMR determinants or (II) one AMR determinant and one or more general population characteristics. Data linkage should be performed on the level of a certain entity which, in the context of this review, can encompass persons, healthcare institutes, geographical regions or countries. A systematic literature search was performed on February 7th 2025 in Embase.com, PubMed and Scopus to identify AMR data linkage studies.
Forty-eight articles were included in our review. Most data linkage studies used two data sources, and most studies were published in the last 5 years (n = 23 in 2020-2024). A predominance of studies linked data on geographical location, and thirteen studies linked data on individual patient level. Findings demonstrate that the majority of studies (43/48) had added value and provided recommendations for clinical practice and future policies or had suggestions for further research and surveillance. Additionally, data linkage studies appeared to be suitable for hypothesis generating. Several limitations were identified. Most studies had ecological designs, which are prone to ecological fallacy and unobserved confounding, making it hard to establish causality.
This systematic scoping review showed that AMR data linkage studies are increasingly performed. They have potential to gain a more comprehensive understanding of AMR dynamics by generating hypotheses, assisting in optimisation of surveillance, and interpretation of data in the context of guideline/policy development. To increase the added value of data linkage, more studies should be performed to improve knowledge on methodological approaches, data access, data management, and governance issues.
Not applicable.