Polish validation of the Simpler Modified Fried Frailty Scale for frailty screening in older inpatients

This study aimed to translate and culturally adapt the Simpler Modified Fried Frailty Scale (SFS) into Polish and evaluate its psychometric performance against reference measures, with a focus on its potential as a brief frailty screening tool in inpatient settings.

The study was conducted in a geriatric inpatient setting in Poland and included 100 hospitalized adults aged ≥65 years who were able to ambulate and complete frailty assessments. Participants completed the SFS, the performance-based Modified Fried Frailty Scale (MFS), and the FRAIL scale. We assessed internal consistency, construct validity, item-level discrimination, diagnostic accuracy with positive predictive value (PPV), negative predictive value (NPV), likelihood ratio (LR).

Internal consistency was acceptable for a brief multidomain screener (α=0.61). Convergent correlations were strong with MFS and FRAIL (ρ=0.64 each; p<0.001). Agreement was fair-moderate (κ=0.38-0.46). Using the ≥3 cut-point, sensitivity was 70.8% (95% Confidence Interval (CI) 48.9-87.4), specificity 80.3% (69.5-88.5), PPV 53.1%, NPV 89.7%; LR⁺ 3.59, LR⁻ 0.36. Discrimination for MFS-defined frailty was good (the area under the curve =0.81, 95%CI 0.71-0.92). Items contributing most to discrimination were exhaustion/fatigue (Odds ratio (OR) =20.5), low activity (OR=8.1), and perceived slowness (OR=5.3), supporting the construct validity of self-reported phenotypic domains. .

The Polish SFS showed good discrimination against MFS-defined frailty, acceptable internal consistency, and a high negative predictive value in this inpatient sample. These findings support its potential as a brief frailty screening tool, although further reliability testing and external validation across clinical settings are warranted.