Background/objectives: The physical phenotype of frailty, described by Fried et al., shows significant overlap with sarcopenia. EWGSOP2 recommends the SARC-F questionnaire to screen for sarcopenia. Considering common features between both conditions, we aimed to investigate whether the SARC-F questionnaire could also be a valid and reliable tool to screen or evaluate frailty.

Design: Retrospective, cross-sectional.

Setting: Istanbul University Istanbul Faculty of Medicine.

Participants: A total of 447 older adults (70.7% female, mean age: 74.5±6.6 years).

Measurements: Frailty was assessed by the modified Fried scale. SARC-F questionnaire was performed by all participants. We used a receiver operating characteristics curve to obtain SARC-F cut-off values to detect frailty, and calculated the area under the curve and 95% confidence interval.

Results: There were 93 (20.8%) older adults with frailty according to the modified Fried scale. SARC-F cut-off ≥1 had 91.4% sensitivity and 44.9% specificity. SARC-F cut-off ≥2 presented the best balance between sensitivity and specificity (sensitivity: 74.1% vs. specificity: 73.7%) to identify frailty (area under curve: 0.807; 95% confidence interval: 0.76-0.84, p<0.001). SARC-F ≥4 had high specificity of 92.6% with a sensitivity of 46.2%.

Conclusion: We suggest that SARC-F ≥1 point can be used to screen for frailty with high sensitivity, and SARC-F ≥4 can be used to diagnose frailty with high specificity. SARC-F may be used to evaluate frailty in usual geriatric practice.