Abstract
Background: Sarcopenia, a prevalent geriatric syndrome with multifactorial origins, is strongly influenced by malnutrition alongside immobility and chronic illness and contributes substantially to falls, disability, and mortality. The SARC-F (Strength, Assistance with walking, Rise from a chair, Climb stairs, and Falls) questionnaire is widely used to screen for probable sarcopenia; however, the conventional cut-off of ≥4 has yielded insufficient sensitivity across studies.
Objective: To evaluate how SARC-F scores correspond to muscle strength and physical performance tests endorsed by the EWGSOP2 algorithm and to determine the most suitable cut-off for screening probable sarcopenia in older adults.
Methods: This cross-sectional study included community-dwelling adults aged ≥65 years who attended a tertiary geriatric outpatient clinic (January 2022–May 2024). Patients with active malignancy, non-ambulatory status, or missing performance data were excluded. Diagnostic accuracy of SARC-F thresholds was assessed against handgrip strength (HGS), the sit-to-stand test (STST), 4-meter gait speed, and the Timed Up and Go (TUG) test using receiver operating characteristic analysis.
Results: A total of 1,355 participants (mean age 74.0 ± 6.0 years; 64.9% female) were analysed. Prevalence of SARC-F ≥4 was 22.2%. For HGS and STST, a cut-off of ≥2 yielded the most favourable discrimination (AUC=0.705 and 0.735, respectively). Gait speed showed the best accuracy at ≥3 (AUC=0.788), while TUG demonstrated the highest performance at ≥4 (AUC=0.881).
Conclusion: SARC-F performance varies across muscle strength and functional domains. A threshold of ≥2 may enhance early detection of sarcopenia in community-dwelling older adults, while higher cut-offs better reflect advanced functional impairment.