RESUMO
BACKGROUND: Frailty has emerged as one of the major risk factors of loss of autonomy and it can be reverted through early and appropriate interventions. A wide range of available frailty screening tools are administered, mainly in clinical settings. However, few frailty instruments are self-administered. OBJECTIVES: The aim of this study was to determine the diagnostic test accuracy of a modified self-administered questionnaire derived from the Study of Osteoporotic Fractures (SOF) index against the Fried frailty phenotype in identifying frailty. DESIGN: Observational, multicenter, diagnostic test accuracy study. PARTICIPANTS: Participants aged 70 and over, living at home or in community-dwelling (n=5134) in two centers in France were contacted. MEASUREMENTS: Participants were mailed self-administered questionnaires derived from the SOF index. Responders who accepted the home evaluation were assessed by trained nurses, blinded to results of the questionnaire, using the Fried frailty phenotype as the reference method. RESULTS: The questionnaire was sent to 5134 participants, of which 1878 (36.6%) met inclusion criteria and returned the questionnaire. Fried frailty assessments were obtained in 691 (35.4%) participants. A total of 639 subjects had a complete evaluation on both the self-administered questionnaire and the Fried phenotype. Mean age was 78.9 (standard deviation [SD]: 5.95) years and 359 (56.2%) participants were women. According to the questionnaire, 159 (24.9%) subjects were considered frail, 172 (26.9%) pre-frail, and 308 (48.2) robust. With the home evaluation, Fried frailty phenotype results were respectively, 114 (17.8%), 295 (46.2%) and 230 (36%). The self-administered questionnaire presented a sensitivity of 66.6% (95% CI: 57.2-75.2) and a specificity of 84.2% (95% CI: 80.8-87.2). CONCLUSIONS: A self-administered questionnaire can be used in elders and represents an opportunity for empowering them in the management of their health in the context of frailty.
