BMC Geriatrics · 6 authors, 6 centres
This summary was generated by AI from a single paper. It has not been reviewed by a clinician and is not clinical advice. Verify against the source before acting on it.
This 2-year prospective study of 482 community-dwelling older Taiwanese adults identified three cognitive frailty (CF) trajectories: robust, improvement, and progression. CF progression was significantly associated with older age, poorer balance, slower gait, and having four or more comorbidities, while CF improvement was only associated with the baseline CF state. These findings highlight modifiable factors that may help delay or reverse CF progression in older populations.
**Background:** Cognitive frailty (CF) is the simultaneous presence of physical frailty and cognitive impairment (excluding dementia). CF increases the risk of falls, dementia, hospitalizations, disability, and mortality. Two subtypes have been proposed: reversible CF (RCF; pre-frailty/frailty + subjective cognitive decline) and potentially reversible CF (PRCF; pre-frailty/frailty + mild cognitive impairment). While cross-sectional studies have identified risk factors for CF, longitudinal trajectories and their predictors remain unknown. This study aimed to identify CF trajectories over 2 years and examine factors associated with CF progression and improvement in older Taiwanese adults.
**Methods:** From August 2017 to June 2019, 832 community-dwelling adults aged ≥65 years were enrolled from outpatient clinics at Taipei Medical University Hospital. Exclusion criteria included difficulty performing basic daily tasks, communication difficulties, or major diseases (e.g., advanced cancer, major cardiovascular disease, dementia). CF was assessed at baseline and at 12- and 24-month follow-ups. Physical frailty was measured using Fried's five criteria (weight loss, weakness, exhaustion, low physical activity, slowness). Cognitive impairment was assessed using the Clinical Dementia Rating (CDR) and Mini-Mental State Examination (MMSE). RCF was defined as pre-frailty/frailty plus subjective cognitive decline (SCD); PRCF as pre-frailty/frailty plus MCI (CDR=0.5). A CF score (0–5) was assigned to each participant at each time point. Sociodemographics, lifestyle, comorbidities, depressive symptoms (Geriatric Depression Scale >5), gait characteristics (GAITRite), Tinetti balance and gait scores, activities of daily living (OARS ADL scale), and Activities-specific Balance Confidence (ABC) scale were collected. Group-based trajectory modeling identified latent CF trajectory groups. Multinomial logistic regression examined associations of explanatory variables with CF improvement and progression versus robust, with and without adjustment for baseline CF state.
**Key Results:** Of 832 participants at baseline, 320 (38.5%) had non-CF, 399 (47.9%) had RCF, and 113 (13.6%) had PRCF. 482 completed both follow-ups. Three CF trajectories were identified: robust (n=236, 28.4%), improvement (n=142, 17.1%), and progression (n=454, 54.6%). Posterior probabilities ranged 0.76–0.82. In the adjusted model (Model 1), CF progression was significantly associated with older age (OR=1.08 per year; 95% CI 1.02–1.14), lower Tinetti balance score (OR=0.72 per point; 95% CI 0.54–0.96), slower gait velocity (OR=0.98 per cm/s; 95% CI 0.97–0.99), and ≥4 comorbidities (OR=2.65; 95% CI 1.19–5.90). CF improvement was not significantly associated with any variable except baseline CF state. In the unadjusted model (Model 2), female sex, regular exercise, and depression were also significant for progression, and female sex and comorbidities for improvement, but these associations were attenuated after adjusting for baseline CF.
**Clinical Implications:** This is the first study to identify CF trajectories using group-based modeling. The findings suggest that older age, poor balance, slow gait, and multimorbidity are independent predictors of CF progression, while CF improvement appears largely driven by baseline CF status. Factors such as female sex, regular exercise, and depression may influence CF development before the study period but not its short-term trajectory. These results support targeting balance training, gait speed improvement, and comorbidity management to prevent CF progression in older adults. Limitations include a short follow-up (2 years), potential selection bias due to loss to follow-up (those who dropped out had more RCF/PRCF), and arbitrary CF scoring. Longer-term studies are needed to confirm these findings.