Association Between Frailty and Incident Urinary Incontinence in Older Adults: An Updated Systematic Review and Meta-Analysis

Abstract ID
4337
Authors' names
Afra Wasama Islam1, Harsahaj Singh Wilkhoo1
Author's provenances
1Faculty of Medicine, Tbilisi State Medical University, Tbilisi, Georgia
Abstract category
Abstract sub-category
Conditions

Abstract

Background

Frailty and urinary incontinence (UI) are common in older adults, contributing to functional decline, reduced quality of life, and increased healthcare use. The role of frailty in predicting new-onset UI remains unclear, as prior reviews focused on prevalence and cross-sectional data. We conducted an updated systematic review and meta-analysis of longitudinal studies to quantify the association between frailty and incident UI.

Methods

We searched MEDLINE, EMBASE, CINAHL, and Cochrane Library from January 2015 to December 2025 for studies examining frailty and incident UI in adults ≥60 years. Inclusion criteria were adults ≥60 years, living in nursing home settings, baseline frailty assessed with validated tools (Fried phenotype, Frailty Index), incident UI as a separate outcome, and follow-up ≥12 months. Exclusion criteria included cross-sectional studies, pre-existing UI, non-human studies, and abstracts without full text. Two reviewers screened 2,134 abstracts and 142 full texts. Data extracted included participant characteristics, frailty definition, UI type, follow-up, and effect estimates. Risk of bias was assessed using the Newcastle-Ottawa Scale (7/9). Pooled relative risks (RR) with 95% confidence intervals (CI) were calculated using random-effects models, with heterogeneity quantified by I²; subgroup analyses were performed by frailty phenotype and UI type.  

Results

Twelve longitudinal studies (n=18,750 participants) met the inclusion criteria. Frailty at baseline was associated with a significantly increased risk of incident UI (pooled RR 1.65, 95% CI 1.35–2.01; I²=58%). Subgroup analyses revealed stronger associations for physical frailty phenotypes (RR 1.82, 95% CI 1.42–2.33) and for urge incontinence compared to stress or mixed UI. Sensitivity analyses confirmed robustness of results, and publication bias was low.

Conclusion

Frailty is a strong predictor of incident UI in older adults. These findings support integrating continence risk assessment into routine frailty evaluation within comprehensive geriatric assessment and highlight opportunities for early, preventive interventions targeting vulnerable older adults.

Comments

Sounds like a really thorough meta-analysis - lovely to have a good evidence base for what we include in our CGA, particularly when our demographic of patients sometimes struggle to make it in to research populations!

Submitted by charlotte.haye… on

Permalink

Thank you, that’s very kind. We completely agree, one of the key aims was to strengthen the evidence base for CGA, especially given how often older, frailer patients are underrepresented in research. Hopefully, these findings help support more proactive continence assessment within routine frailty evaluation.

Submitted by islam1048@gmail.com on

In reply to by charlotte.haye…

Permalink

Interesting work. Could you clarify which Frailty Index was used (e.g. Rockwood deficit accumulation FI, electronic frailty index, or alternative)?

Submitted by kathryn.fisher… on

Permalink

Thank you for that important question. Across the included studies, frailty was assessed using validated but heterogeneous tools. The most commonly used approaches were the Fried frailty phenotype and the Rockwood deficit accumulation Frailty Index. A smaller number of studies used modified or study-specific frailty indices based on deficit accumulation principles. 

We did not restrict inclusion to a single Frailty Index, as our aim was to capture the full breadth of longitudinal evidence. However, we accounted for this variability through subgroup analyses, where we found that physical frailty phenotypes showed a stronger association with incident urinary incontinence. 

This variability in frailty measurement likely contributes in part to the observed heterogeneity, but also enhances the external validity of our findings across different clinical settings.