Antimicrobial Resistance Trends in Escherichia coli, Klebsiella pneumoniae, and Staphylococcus aureus: A Multinational WHO GLASS Analysis
Background. Antimicrobial resistance (AMR) is a leading global health threat, and the WHO Global Antimicrobial Resistance and Use Surveillance System (GLASS) is the principal mechanism for multinational AMR surveillance. Countries participate voluntarily and report differing pathogen-antibiotic combinations in different years, so the reported data form an unbalanced panel whose composition changes over time. Objectives. To describe and quantify trends in reported resistance for Escherichia coli, Klebsiella pneumoniae and Staphylococcus aureus using GLASS country-reported data (2018–2023), and to determine how far those trends depend on which countries reported in each year. Methods. Retrospective ecological secondary analysis of WHO XMART country-reported GLASS AMR indicators for 2018–2023. The original extraction on 17 October 2026 was reproduced using a new retrieval on 31 October 2026. Eight clinically prioritised pathogen–antibiotic–specimen combinations were evaluated using binomial mixed-effects models with country-level and observation-level random intercepts. Balanced-panel, inclusion-threshold, country-exclusion and model-specification sensitivity analyses were undertaken, together with Holm adjustment across the eight primary temporal tests and exploratory overdispersion-aware regional analyses. Results. The eight sentinel combinations comprised 2,294 country–year strata from 65 reporting countries and 17,132,532 interpretable antimicrobial susceptibility test (AST) results; the broader descriptive subset contained 8,824 strata from 66 countries. Three combinations showed increasing reported resistance after accounting for extra-binomial heterogeneity and Holm adjustment: K. pneumoniae meropenem (annual OR 1.1453; 95% CI 1.1060–1.1861; adjusted P = 2.34 × 10⁻¹³), E. coli meropenem (OR 1.1860; 95% CI 1.1226–1.2529; adjusted P = 8.14 × 10⁻⁹), and K. pneumoniae ciprofloxacin (OR 1.0684; 95% CI 1.0430–1.0946; adjusted P = 4.58 × 10⁻⁷). All three retained direction and significance across seven alternative scenarios. The other five combinations had no statistically significant primary temporal association. Exploratory overdispersion-aware regional interaction tests were significant in three combinations. For E. coli ciprofloxacin resistance in blood, the all-reporter pooled change was −1.8 percentage points and the balanced-panel change was +7.4 percentage points, yielding a descriptive participation contrast of −9.2 percentage points. Conclusions. Among reporting countries, reported meropenem resistance in E. coli and K. pneumoniae and ciprofloxacin resistance in K. pneumoniae increased between 2018 and 2023. These associations were robust to sensitivity and multiplicity analyses, but they do not establish national population-level resistance trends or treatment outcomes. Changes in reporting-country composition materially affected crude comparisons, highlighting the need to report denominator, participation and model uncertainty when interpreting multinational surveillance data.
Authors
- Farheen Tariq
- Sania Zehra
- Shehzad Khalil (ORCID: https://orcid.org/0009-0005-2844-2518)
- Gulwareena Yousafzai
- Mustafeez Bashar
Institutions
- University of Sindh (PK)
- The University of Agriculture, Peshawar (PK)
- BRAC University (BD)
- University of Malakand (PK)
Publication Details
- Journal
- Link Medical Journal
- Published
- 2026-11-15
- DOI
- https://doi.org/10.61919/w1v79n33
- Primary Topic
- Antibiotic Use and Resistance
- Type
- article
- Field-Weighted Citation Impact
- 0.00