Assessment tools for social frailty in older adults based on COSMIN guidelines: A systematic review

The ageing population faces social isolation and a lack of resources, placing older adults at risk of social frailty. Characterised by diminished social engagement, lack of social resources, and prolonged isolation, social frailty impacts outcomes such as depression, cognitive decline, and mortality. However, the instruments used to assess social frailty in older adults have not been thoroughly evaluated for their measurement properties and methodological quality.

This systematic review aims to assess the measurement properties and methodological quality of social frailty assessment instruments for older adults using the COSMIN methodology for systematic reviews of Patient-Reported Outcome Measures version 2.0. A modified Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach was used to grade evidence and guide recommendations.

A systematic search was conducted in PubMed, Embase, Web of Science Core Collection, the Cochrane Library, China Biomedical Literature Database, China National Knowledge Infrastructure, Wanfang Data, and VIP Database from inception to 28 February 2026. Studies published in English or Chinese that evaluated instruments measuring social frailty in older adults were eligible. Measurement properties were assessed using the COSMIN Risk of Bias checklist and the COSMIN criteria for good measurement properties. The quality of evidence for each instrument was graded using the modified GRADE approach.

Twenty-one study reports covering 13 social frailty assessment instrument families were included. Among reflective and mixed instruments, structural validity and internal consistency were the most frequently reported measurement properties, whereas among formative instruments, hypotheses testing for construct validity was reported most often. The social life and socioeconomic deprivation subscales of the Chinese version of the Social Frailty Index met category A recommendation criteria. The Persian version of the Social Frailty Scale was classified as category C. The remaining reflective and mixed instrument versions and independently scored subscales, as well as all formative instruments, were classified as category B. Important evidence gaps remained in cross-cultural validity or measurement invariance, measurement error, and responsiveness.

Evidence for social frailty assessment instruments in older adults remains uneven across instrument types and measurement properties. The strongest current support was found for the social life and socioeconomic deprivation subscales of the Chinese version of the Social Frailty Index. However, important gaps remain in the evaluation of cross-cultural validity or measurement invariance, measurement error, and responsiveness. Further validation and refinement of available instruments are needed to support more robust instrument selection in research and practice.

PROSPERO CRD420251107875.