Abstract 17275: Evaluating the Convergent and Discriminant Validity of Four Frailty Instruments in Adults With Heart Failure: Results From the Frame-HF Study

Abstract

Introduction: Frailty assessment has become increasingly common in those with heart failure. The most frequently used frailty instrument is the Frailty Phenotype (FP). The validity of this instrument in those with heart failure is yet to be determined. Aim: To examine the convergent and discriminant validity of four frailty instruments: i) the FP ii) a Questionnaire-only version of FP [q-FP]; iii) St. Vincent’s Frailty [SVF]; and iv) the Frailty Instrument for Primary Care of the Survey of Health, Ageing and Retirement in Europe [SHARE-FI] for assessing the frailty status of adults with heart failure. Methods: Using a cross-sectional survey design, individuals aged ≥ 18 years, diagnosed with heart failure were recruited from the inpatient cardiology ward and outpatient heart failure clinic of a metropolitan academic hospital. In addition to assessing all items of the frailty instruments, other data collected included: a) NYHA classification; b) quality of life and health status using the EQ-5D-5L; c) physical status using the Australian-modified Karnofsky Performance Scale (AKPS); and d) 10-item Depression in Medically Ill (DMI) measure. Results: A total of 131 participants were recruited and included in the analyses, there were wide variations in frailty status across the four frailty instruments (Figure 1). Inter-correlations were highest between SVF and SHARE-FI instruments ( r = 0.62). The NYHA classification were correlated with SVF ( r = 0.46) and SHARE-FI ( r = 0.42). Similarly, all EQ-5D-5L dimensions were correlated with both SVF and SHARE-FI but not with FP or q-FP. The SVF differentiated between high and low AKPS scores (χ 2 = 11.70, df : 2, p = 0.003) and DMI scores (χ 2 = 7.28, df : 2, p = 0.026). Similarly, the SHARE-FI also had good discriminant validity using AKPS scores (χ 2 = 9.25, df : 2, p = 0.010) and DMI scores (χ 2 = 14.32, df : 2, p = 0.001). Conclusions: The SVF and SHARE-FI demonstrated good convergent and discriminant validity. Both instruments provide a valid alternative to the FP for the assessment of frailty in those with heart failure. Figure 1: Classification of frailty categories: FP, Questionnaire-only FP, SVF and SHARE-FI