The Pictorial Fit-Frail Scale—Malay version (PFFS-M): reliability and validity testing in Malaysian primary care

Abstract Background This study investigated the reliability and convergent validity of the PFFS-Malay version (PFFS-M) among patients (with varying educational levels), caregivers, and health care professionals (HCPs). PFFS-M cutoffs for frailty severity were developed. Methods This is a cross-sectional study from 4 primary care clinics where 240 patients aged >60 years and their caregivers were enrolled. Patients were assigned to a nurse or a health care assistant (HCA) for 2 separate PFFS-M assessments administered by HCPs of the same profession, as well as by a doctor during the first visit (inter-rater reliability). Patients were also administered the Self-Assessed Report of Personal Capacity & Healthy Ageing (SEARCH) tool, a 40-item frailty index, by a research officer. The correlation between patients’ PFFS-M scores and SEARCH tool scores determined convergent validity. Patients returned 1 week later for PFFS-M reassessment by the same HCPs (test–retest reliability). Caregivers completed the PFFS-M for the patient at both clinic visits. Classification cut-points for the PFFS-M were derived against frailty categories defined through the SEARCH tool. Results The inter-rater (intraclass correlation coefficient [ICC] = 0.92 [95% CI, 0.90–0.93)] and test–retest (ICC = 0.94 [95% CI, 0.92–0.95]) reliability between all raters was excellent, including by patients’ education levels. The convergent validity was moderate (r = 0.637, p < 0.001), including for varying educational background. PFFS-M categories were identified as: 0–3, no frailty; 4–5, at risk of frailty; 6–8, mild frailty; 9–12, moderate frailty; and >13, severe frailty. Conclusion PFFS-M is a reliable and valid tool with frailty severity scores now established for use of this tool in primary care clinics.