Validating the Centre of Excellence on Longevity Self-AdMinistered (CESAM) Questionnaire: An Online Self-Reported Tool for Frailty Assessment of Older Adults

Subjective health measures are often used to assess frailty, but the validity of self-reported online tools to identify frailty remains to be established. We aimed to assess concurrent, known-groups, convergent and predictive validity of the Centre of Excellence on Longevity Self-AdMinistered (CESAM) questionnaire for frailty assessment of older adults in an outpatient setting. Cross-sectional analysis of 120 participants. Participants of age ≥65 were recruited from an outpatient geriatric clinic. Individuals who had severe neurological, cognitive, or motor deficits were excluded. We assessed concurrent validity with area under receiver operating characteristic curve (AUC) against the Frailty Index (FI) and Clinical Frailty Scale (CFS). We analyzed known-groups validity between CESAM scores with frailty status (CFS and FI), Modified Barthel Index (MBI), and modified Chinese Mini-Mental State Examination (mCMMSE) using 1-way analysis of variance. We evaluated convergent validity using correlations with MBI, the Lawton index, mCMMSE, and Geriatric Depression Scale (GDS). Associations between CESAM-identified frailty for clinician-diagnosed geriatric syndromes, and health-related quality of life (HRQoL) was analyzed using regression analysis. The CESAM questionnaire demonstrated excellent diagnostic performance for frailty using FI ≥0.25 (AUC = 0.88; 95% CI: 0.82–0.94; P < .001) and CFS ≥4 (AUC = 0.78; 95% CI: 0.68–0.88; P < .001). CESAM scores increased significantly with increasing frailty (both CFS and FI), lower MBI, and lower mCMMSE scores (all P < .001), indicating concurrent validity. The moderate-good correlation of CESAM scores with MBI (r = −0.61; P < 0.001), Lawton Index (r = −0.54; P < .001), mCMMSE (r = −0.53; P < .001) and GDS (r = 0.58; P < .001) supports convergent validity. Using a cutoff of ≥8 for frailty identification, CESAM-identified frailty was associated with cognitive impairment (OR = 3.7; 95% CI: 1.7–8.2; P = .001) depression (OR = 4.0; 95% CI: 1.7–9.6; P = .002), falls (OR = 3.1; 95% CI: 1.2–8.2; P = .021) and poorer HRQoL (β = −0.1; 95% CI: −0.2 to −0.02; P = .017). Our results support the validity of an online self-reported tool to identify frailty and geriatric syndromes in an outpatient setting, an approach that is potentially applicable for remote screening of frailty.