A cross sectional comparison of older people’s self-perceived frailty and their Electronic Frailty Index score

Abstract ID
3803
Authors' names
V Barber-Fleming1; A Anand1,2, H Wilkinson1, G Mead3
Author's provenances
1. Advanced Care Research Centre, University of Edinburgh; 2. Institute for Neuroscience and Cardiovascular Research, University of Edinburgh; 3. Usher Institute, University of Edinburgh
Abstract category
Abstract sub-category
Conditions

Abstract

Introduction

Small, qualitative studies suggest discrepancies between older adults’ measured and self-perceived frailty. Any mismatch will have implications for frailty interventions and advanced care planning. We therefore, aimed to report the relationship between older adults’ self-perceived frailty and the Electronic Frailty Index (eFI), an objective screening tool measure of frailty, in a large, unselected cohort of older people.

Method

One thousand people aged ≥ 70 years, randomly selected from a single GP practice, were sent a survey, asking them to rate their own frailty using self-report measures (an ordinal scale, a binary scale, Self-rated health (SRH) and PRISMA-7. We analysed a) agreement between self-perceived frailty (ordinal scale) and eFI categorised frailty (weighted Kappa and Gwet’s second order agreement co-efficient [AC2]), b) discrimination of each self-report measure for eFI defined frailty (threshold ≥ 0.12) and c) logistic regressions exploring predictors of self-perceived frailty (binary scale).

Results

375 people were included in the analysis (median age 76, 51% female). Agreement was ‘fair’ between self-perceived frailty and eFI using linear weighted kappa (k = 0.25) and quadratic weighted kappa (k = 0.37). Agreement was higher with linear and quadratic weighted AC2 (k = 0.65 and 0.81 respectively). As eFI severity increased, agreement with self-perceived frailty decreased.

Self-perceived frailty was poor at discriminating frailty as categorised by eFI. PRISMA-7 outperformed the other measures. SRH performed least well. The optimal eFI cut point for discriminating self-perceived frailty was 0.17.

Multivariate regression showed that increasing age (OR 1.10, 95% confidence intervals [CI] 1.02-1.18) and depression (OR 1.51, 95% CI 1.31-1.74) were the only significant predictors of self-perceived frailty. Sex, anxiety, eFI category and deprivation were not significant.

Conclusions

The mismatch between self-perceived and e FI-categorised frailty, especially in those categorised as severely frail, may have implications for frailty management and advanced care planning.