Screening Frailty Among Community-Dwelling Older Adults: A Cross-Sectional Study Comparing the Edmonton Frail Scale and the Clinical-Functional Vulnerability Index-20

Background/Objectives: The Edmonton Frail Scale (EFS) and the Clinical-Functional Vulnerability Index-20 (IVCF-20) are multidimensional screening instruments that differ in the domains assessed, scoring systems, and classification thresholds, which may lead to different frailty-related classifications. Although both instruments are used to identify vulnerability in older adults, the extent to which their classifications agree remains insufficiently characterized. This study compared EFS and IVCF-20 classifications among community-dwelling older adults by examining category distribution, cross-classification, agreement, and correlation between total scores. Methods: This cross-sectional observational study included 110 community-dwelling adults aged ≥60 years recruited by convenience sampling from a public community center in Rio Verde, Goiás, Brazil. Participants completed the EFS and IVCF-20 during the same assessment session. Agreement across the original three ordinal categories was evaluated using linear-weighted Cohen’s kappa, and the relationship between total scores was assessed using Spearman’s rank correlation coefficient. Results: The EFS classified 43.6% of participants as non-frail, 31.8% as pre-frail, and 24.5% as frail, whereas the IVCF-20 classified 43.6% as robust, 37.3% as potentially frail, and 19.1% as frail. Exact agreement was observed in 65.5% of participants (95% CI: 56.2–73.7%), and linear-weighted Cohen’s kappa indicated moderate agreement (κ = 0.543; 95% CI: 0.415–0.670; p < 0.001). Most discordant classifications occurred between adjacent categories. Total scores were strongly positively correlated (Spearman’s ρ = 0.745; p < 0.001). Conclusions: The EFS and IVCF-20 showed a strong correlation between total scores but only moderate agreement between frailty-related classifications. Although the instruments capture related dimensions of vulnerability, their classifications should not be considered interchangeable.