Burden of disease, disability-adjusted life years and frailty prevalence

Abstract

Background Burden of disease (BoD) using disability-adjusted life years (DALY) is a useful summary measure of population health and estimates are provided for Ireland annually. We hypothesized that BoD may be used as a predictor of frailty prevalence. Aim To examine the correlation between frailty measured by the accumulation of deficits (frailty index, FI) and Fried frailty phenotype (FFP) classifications and BoD, in an Irish context. Design Cross-sectional secondary analysis. Methods Data were obtained from waves two and three of The Survey of Health, Ageing and Retirement in Europe for Irish adults aged ≥65 in 2007. Frailty was defined by a 70-item FI and the FFP. Years lived with disability (YLD), years of life lost (YLL) and DALY were calculated using adapted equations from the World Health Organization and, where possible, disability weights, sequelae and durations as in the Global BoD (GBD) project (2016). Results Of 1035 participants, 442 were ≥65 years. Mean DALY were significantly higher in those identified as frail (FI: 3.31, P < 0.0001, n = 406; FFP: 2.46, P = 0.005, n = 319). For the FI, stronger correlation was found for DALY (r = 0.5431, P < 0.0001) than for age (r = 0.275, P < 0.0001). Controlling for confounders, DALY were an independent predictor of frailty when measured with the FI (OR 1.17, 95% CI: 1.10-1.24) but not with the FFP (OR 1.079, 95%% CI 1.00-1.17). Conclusions Frailty correlates significantly with DALY, and more so with the FI than the FFP, reaffirming that these measures are different constructs. GBD data could represent a predictor of population-level frailty estimates, facilitating improved comparisons.