Impact of multidomain lifestyle intervention on dynamics of cognitive frailty: post hoc analysis of the FINGER trial

BACKGROUND: Cognitive frailty (CF), a condition with physical frailty and mild cognitive impairment (MCI) without dementia, is potentially reversible and linked to adverse outcomes. We aimed to investigate the impact of a multidomain lifestyle intervention on temporal dynamics of CF in older adults at risk of dementia. METHODS: In the Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER), 1259 participants, aged 60-77, were randomized to a 2-year multidomain lifestyle intervention or standard health advice. Frailty was defined by the modified Fried phenotype, and MCI by the lowest quintile in the neuropsychological test battery z score. Having pre-frailty/frailty and MCI was classified as CF. Transition probabilities and predictiveness of CF by 4 different baseline groups (healthy, MCI, pre-frail/frail, CF) were examined using multinomial logistic regression. RESULTS: At baseline, 219 participants (18%) had CF. The risk for developing CF at 2 years was higher in the control group (risk ratio [RR], 1.88, p = .003). The intervention effect was not modified by baseline CF (p = .493). Reversal from CF to no-CF group was more likely in the intervention group, and progression to or persisting with CF was more likely in the control group. Compared with healthy participants (n = 401) at baseline, the MCI group (n = 244) had an RR of 5.10, pre-frail/frail (n = 336) of 3.06, and CF of 30.61 for having CF at 2 years, with no difference between MCI and pre-frail/frail groups (p = .116). CONCLUSIONS: The 2-year multidomain lifestyle intervention was effective in preventing and reversing CF. Participants with MCI or pre-frailty/frailty were both at increased risk for CF compared with healthy participants.