The Nonlinear Association Between Catheterisation Duration and Urinary Tract Infection in Patients With Geriatric Hip Fracture: A Retrospective Study

ABSTRACTAimTo rigorously examine the association between catheterisation duration and urinary tract infections among geriatric patients with hip fractures.DesignRetrospective cohort study.MethodsWe analysed data from 872 patients aged 60 years and older, treated at a tertiary care hospital between January 2022 and May 2023. To address missing data, we employed multiple imputations using chained equations (50 iterations). A comprehensive set of statistical methods, including multivariate logistic regression, generalised additive models, and smooth curve fitting techniques, was applied to investigate variable relationships. Additionally, we used a two‐piecewise binary logistic regression model to further elucidate nonlinear associations.ResultsThe cohort included 296 males and 576 females, with an overall urinary tract infection prevalence of 25.3%. Multivariate smooth spline analysis revealed a nonlinear relationship between catheterisation duration and urinary tract infection. This finding suggested that the risk of urinary tract infection increased by 2.8% for every 10‐h increase with an indwelling duration < 106 h. However, the risk no longer significantly increased with an indwelling duration of ≥ 106 h. Sensitivity and interaction analyses confirmed the robustness of these results.ConclusionsThis study identified an inverse L‐shaped relationship between catheterisation duration and urinary tract infection. Given the potential for confounding, further investigations are necessary to validate these findings and explore their implications for clinical practice.Implications for the Profession and Patient CareNurses should prioritise early catheter removal protocols to mitigate infection risk, coupled with enhanced surveillance within the first 106 h after insertion.Reporting MethodAdhered to STROBE guidelines for observational studies.Patient or Public ContributionNo public Contribution. Patients contributed through data collected from the Hospital Information System, which was used for analysis.