September 23, 2026
How does the choice of frailty assessment scale affect prevalence estimates in elderly IBD patients, and which factors are consistently associated with frailty regardless of the scale used?
Frailty is a useful framework by which to characterize elderly patients with IBD, and may help to inform decisions about monitoring and therapy
Carballo-Folgoso L, Duque Alcorta JM, Varela P, et al. The Prevalence of Frailty in Elderly Inflammatory Bowel Disease Patients Varies Depending on the Scale Used. Aliment Pharmacol Ther. 2026. Epub ahead of print. https://doi.org/10.1111/apt.70895
The multicentre Astur-Frailty-IBD study enrolled 612 consecutive inflammatory bowel disease (IBD) outpatients aged 60 or older from three hospitals in Asturias, Spain, between March and June 2024. Frailty was prospectively assessed using four validated scales: the Fried Frailty Phenotype (FFP), the FRAIL scale, the Clinical Frailty Scale (CFS) and the SHARE-FI instrument. Clinical and demographic data, comorbidity indices – including the Charlson Comorbidity Index (CCI) and the Cumulative Illness Rating Scale-Geriatric (CIRS-G) – and IBD characteristics were collected through face-to-face individual interviews. Agreement between scales was evaluated and factors independently associated with frailty were identified by multivariable logistic regression for each scale. Over 56% of patients had Crohn’s disease and the median age was 68.8 years.
Frailty prevalence varied nearly fourfold depending on the scale: 4.4% (FRAIL), 4.6% (CFS), 12.4% (SHARE-FI) and 16.8% (FFP). Overall agreement across all four scales was only “fair” (Fleiss’ Kappa 0.25; 95% Confidence Interval [CI], 0.21–0.29). Pairwise, SHARE-FI and FFP showed the highest agreement (k: 0.45) and FFP and CFS the least (k: 0.22). Against FFP as a reference, all other scales had high specificity (92–95%) but moderate-to-low sensitivity: 62% for FRAIL, 53.5% for SHARE-FI and just 33.6% for CFS. CFS identified the fewest frail patients despite being the most commonly used scale in clinical practice. Frail patients identified by FFP were older than non-frail patients (median 74.9 vs. 67.1 years), more often female (62.1% vs. 40.1%), more likely to have clinical or biochemical active disease and had higher comorbidity scores (median CIRS-G 8 vs. 4; p<0.001).
On multivariate analysis, polypharmacy (defined as five or more medications, present in 49.7% of the cohort) and CIRS-G ≥ 3 were the only two factors independently associated with higher risk of frailty across all four scales, with an OR of 2.09–2.38 and 1.75–2.03, respectively. Higher age, female sex, clinical IBD activity and current corticosteroid use were associated with frailty on three of the four scales. Conversely, good family support and longer disease duration were each independently associated with a lower risk of frailty on three scales. Notably, higher CCI scores were not associated with frailty on any scale, while CIRS-G (which evaluates severity across 14 body systems) was consistently associated with frailty across all four scales.
Details
Study Design: Prospective cohort study
Funding: No specific funding received
Allocation: Not applicable
Setting: Multicenter
Level of Evidence: 2b