Utility of the Fatigue,Resistance, Ambulation,Illness, and Loss of weight Scale in Older Patients with Cardiovascular Disease

ObjectivesTo compare the Fried criteria for frailty diagnosis with the Frailty Screening Index (FSI) and the fatigue, resistance, ambulation, illness, and loss of weight (FRAIL) scale in older patients with cardiovascular disease (CVD).DesignWe conducted a retrospective 1-year follow-up cohort study of adult inpatients who participated in a cardiac rehabilitation program between June 2016 and September 2018.Setting and ParticipantsWe included 1472 Japanese patients age 65 years and older with CVD. After excluding 765 patients with incomplete frailty measurements, 707 patients were included in the analysis.MethodsFrailty and physical function were measured before hospital discharge according to each of the 3 definitions. Outcomes were all-cause mortality and physical dysfunction.ResultsThe prevalence of frailty according to the Fried criteria, the FRAIL scale, and the FSI was 213 (30.1%), 181 (25.6%), and 186 (26.3%), respectively. The FSI and the FRAIL scale showed moderate agreement with the Fried criteria [vs FSI: K = 0.52, 95% confidence interval (CI): 0.45–0.59; vs FRAIL scale: K = 0.45, 95% CI: 0.37–0.52; all P < .001]. We found a significant correlation between all-cause mortality and frailty assessed by all of the definitions, even after multivariate adjustment [FSI: hazard ratio (HR): 2.43, 95% CI: 1.30–4.58, P = .006; FRAIL scale: HR: 2.32, 95% CI: 1.21–4.45, P = .011; Fried criteria: HR: 1.99, 95% CI: 1.04–3.82, P = .038). However, the prediction accuracy of the FRAIL scale was higher than that of the FSI and comparable to that of the Fried criteria for physical dysfunction.Conclusions and ImplicationsThe FSI and the FRAIL scale showed moderate agreement with the Fried criteria regarding frailty diagnostic performance and had comparable prognostic value. However, only the FRAIL scale was as accurate as the Fried criteria in screening for physical dysfunction.