Objective: SARC-F is a self-reported questionnaire to screen for an increased risk of sarcopenia. Since it requires self-assessment, it is not appropriate for use in patients with impaired judgment or severe communication problems. Whether it can be reliably administered to proxies on behalf of patients is an issue that needs to be clarified. We aimed to study the reliability of SARC-F by proxy and examine the agreement between patient and proxy responses.

Methods: This is a cross-sectional study conducted between September 2019 and October 2021. Patients were recruited from 2 settings: outpatients and nursing home residents. Proxies were relatives/caregivers in community-dwelling setting and nurses in nursing home. We transformed SARC-F to SARC-F by proxy and studied its reliability with interrater and test–retest reliability analyses in the first phase. In the second phase, we examined the concordance between patient and proxy responses in total and item by item.

Results: Total sample size was 279 (172 patients and 107 proxies). Community-dwelling older adults made up 58.1% of the older adult population. Median age of older adults was 72 (60-93), and 44.8% were female. SARC-F by proxy showed an excellent interrater and test–retest reliability, with intraclass correlation coefficients of 0.91 and 0.90, respectively (P < .001). It also demonstrated a high level of internal consistency, with a Cronbach’s alpha value of 0.82. The total scores of SARC-F by patient and SARC-F by proxy showed a moderate correlation (r = 0.635; P < .001). The fourth item demonstrated the highest, and the fifth item showed the lowest correlation (r values = 0.591 and 0.443, respectively).

Conclusion: According to our study, SARC-F by proxy can be reliably administered to proxies on behalf of older adults when conditions that prevent reliable judgment or communication exist. Further validity studies of SARC-F by proxy are needed to verify whether it will work well in identifying sarcopenia cases in older adults.

Keywords: Geriatric assessment, older adults, patient-reported outcome measures, proxy, sarcopenia