Frailty trajectories in three longitudinal studies of aging : Is the level or the rate of change more predictive of mortality?

BACKGROUND: frailty shows an upward trajectory with age, and higher levels increase the risk of mortality. However, it is less known whether the shape of frailty trajectories differs by age at death or whether the rate of change in frailty is associated with mortality. OBJECTIVES: to assess population frailty trajectories by age at death and to analyse whether the current level of the frailty index (FI) i.e. the most recent measurement or the person-specific rate of change is more predictive of mortality. METHODS: 3,689 individuals from three population-based cohorts with up to 15 repeated measurements of the Rockwood frailty index were analysed. The FI trajectories were assessed by stratifying the sample into four age-at-death groups: <70, 70-80, 80-90 and >90 years. Generalised survival models were used in the survival analysis. RESULTS: the FI trajectories by age at death showed that those who died at <70 years had a steadily increasing trajectory throughout the 40 years before death, whereas those who died at the oldest ages only accrued deficits from age ~75 onwards. Higher level of FI was independently associated with increased risk of mortality (hazard ratio 1.68, 95% confidence interval 1.47-1.91), whereas the rate of change was no longer significant after accounting for the current FI level. The effect of the FI level did not weaken with time elapsed since the last measurement. CONCLUSIONS: Frailty trajectories differ as a function of age-at-death category. The current level of FI is a stronger marker for risk stratification than the rate of change.
CC BY-NC-ND 4.0© The Author(s) 2021. Published by Oxford University Press on behalf of the British Geriatrics SocietyAddress correspondence to: Juulia Jylhävä. Tel: 358-505351712. Email: juulia.jylhava@ki.seDeclaration of Sources of Funding: National Institutesof Health (grant numbers R01 AG04563, AG10175,AG028555, AG08861-09); the MacArthur Foundation Research Network on Successful Aging; the Swedish Council for Working Life and Social Research (FAS/FORTE) (grant numbers 97:0147:1B to N.L.P., 2009-0795 toN.L.P.); the Swedish Research Council (grant numbers 825-2007-7460, 825-2009-6141, 521-2013-8689, 2015-03255, 2018-02077 to J.J., 2016-03081, 2019-01272 to S.H.); JPND/Swedish Research Council (2015-06796 to N.L.P.); FORTE (2013-2292 to N.L.P.); the Loo & Hans Osterman Foundation to S.H. and J.J.; the Foundation for Geriatric Diseases to S.H.; the Magnus Bergwall Foundation to S.H.; the Strategic Research Program in Epidemiology at Karolinska Institutet to S.H. and J.J. and the King Gustaf V’s and Queen Victoria’s Freemason Foundation to S.H. and China Scholarship Council to G.B. and X.L. The financial sponsors had no role in the design, execution, analysis andinterpretation of data, or writing of the study.