Bayesian beta regression analysis of the association between Klebsiella pneumoniae resistance rates and antibiotic use: a 6-year quarterly time-series study
Year: 2026
Type: article
Study design: Observational Study
Abstract: Background This exploratory study examined the potential association between the resistance rate of Klebsiella pneumoniae and antimicrobial consumption in the preceding quarter, aiming to provide a reference for antimicrobial stewardship and the rational use of antimicrobials. Methods This retrospective study used single-center aggregated data from our institution over the past 6 years (24 quarters). Bayesian beta models were used to explore the potential association between the resistance rate of K. pneumoniae and antimicrobial consumption in the preceding quarter while controlling for temporal trends. A three-tier strategy—model comparison, robustness checks, and sensitivity analysis—was adopted to assess the direction and robustness of this association. Results A total of 3,326 K. pneumoniae isolates were identified, accounting for 13.25% of all clinical isolates. Respiratory specimens were the predominant source ( n = 1,682), indicating that the respiratory tract was a major site of infection. Association analysis showed that the trimethoprim-sulfamethoxazole resistance rate was positively associated with tigecycline DDDs: the main model outperformed the time-only baseline (ΔLOOIC = −4.47), with β = 0.0012 (95% CrI [0.00020, 0.00226]; P ( β > 0) = 98.9%); each IQR increase in tigecycline DDDs was associated with an estimated increase of 4.81 percentage points in the trimethoprim-sulfamethoxazole resistance rate. This association remained consistent under MBB-based inference, with HAC-robust standard errors, and after 5% winsorization. It also showed the same direction after adding a 2023 indicator, a 2023 step variable, or a seasonal term. After excluding 2022–2024 data, the direction did not reverse [ P ( β > 0) = 72.4%], but the reduced sample size may have limited the strength of evidence in this subsample. The ceftazidime resistance rate showed a positive association with tigecycline DDDs in the full sample ( β = 0.0017, 95% CrI [0.00007, 0.00334]), and MBB, Winsorization, and HAC-robust inference all supported this association. In the subsample excluding the 2022–2024 data, the positive association was substantially attenuated [ P ( β > 0) = 34.1%]; its temporal stability therefore warrants cautious interpretation and further validation in large independent samples. Piperacillin-tazobactam resistance showed a suggestive negative association with amoxicillin-clavulanate DDDs, whereas trimethoprim-sulfamethoxazole resistance showed a suggestive positive association with meropenem DDDs; evidence for both was limited. The remaining combinations provided no clear evidence of association. Conclusion This exploratory study suggests a potential association between the resistance rate of K. pneumoniae and antimicrobial consumption, supporting further validation in independent large-sample studies. These findings may inform antimicrobial stewardship research and suggest that agents such as tigecycline warrant inclusion in future monitoring and evaluation.
Source: Frontiers in Microbiology
Authors: Fangyuan Xia(Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Visualization, Writing – original draft), Yu Pan(Conceptualization, Formal analysis, Writing – review & editing), Xiaohao Pan(Data curation, Writing – review & editing), Wensi Gan(Data curation, Investigation, Resources, Writing – review & editing), Sucai Chen(Resources, Supervision, Validation, Writing – review & editing)
Institution: Wenzhou Hospital of Traditional Chinese Medicine
Language: English
Subfield: Infectious Diseases
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $3,801
OpenAlex ID: W7234549276
