Psychometric analysis of the Evaluation of Multidimensional Functioning and Risks in Aging (EMFRA) scale: reliability, structural validity, and concurrent validity.

Background: EMFRA is a brief scale developed to assess physical, cognitive, emotional, and social functioning in older adults. This study examined its initial psychometric properties. Methods: A cross-sectional study was conducted with 201 older adults aged ≥65 years without a diagnosis of dementia, recruited from the community, day-care and nursing centres from Madrid. Dimensional structure was examined through bootstrap parallel analysis (PA), bootstrap exploratory graph analysis (EGA), and factor analysis (FA), followed by progressive item refinement. Internal consistency, test-retest reliability in a subsample (n = 22), SEM, MDC95, floor and ceiling effects, and associations with SPPB, MoCA, Barthel Index, and HADS were assessed. Results: The 21-item version did not reproduce the four hypothesized domains, yielding a three-dimensional solution with limited stability. A nine-item version comprised two factors, Autonomy and Psychological Distress, although FA fit was limited (TLI = 0.80; RMSEA = 0.13). A low hierarchical omega (0.409) did not support a total score, whereas factor reliability was adequate (omega = 0.840 and 0.785). Test-retest ICCs were 0.82 and 0.78, with wide 95% CIs. No floor or ceiling effects were observed. Autonomy correlated with SPPB (r = 0.751), MoCA (r = 0.697), and Barthel (r = 0.56), and Psychological Distress with HADS (r = 0.625). Conclusions: The nine-item EMFRA shows preliminary evidence of a two-factor structure whose factors should be scored separately. Its temporal reliability and measurement error remain uncertain, and external validation is required before clinical implementation.