Related Experiment Video
Updated: Jan 5, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Development and Validation of a Tool to Screen for Cognitive Frailty among Community-Dwelling Elders
1Liang-Kung Chen, Center for Geriatrics and Gerontology, Taipei Veterans General Hospital, No. 201, Sec. 2, Shih-Pai Rd., Taipei 11217, Taiwan, lkchen2@vghtpe.gov.tw; Or Li-Ning Peng, Center for Geriatrics and Gerontology, Taipei Veterans General Hospital, No. 201, Sec. 2, Shih-Pai Rd., Taipei 11217, Taiwan,
Objectives:
Reciprocal age-related impairments in physical and cognitive functioning have been termed 'cognitive frailty', which is associated with adverse health outcomes and is a potential target for preventing or delaying the onset of disability in older people. However, cognitive frailty as currently defined is challenging to diagnose. To facilitate earlier diagnosis and intervention, we conducted this study to develop and validate a simple evidence-based instrument to identify community-dwelling elders at risk of cognitive frailty.
Design:
Retrospective analyses of data from the I-Lan Longitudinal Aging Study (ILAS) to develop a prediction model, and from the Longitudinal Aging Study of Taipei (LAST) for external validation.
Setting:
Community-dwelling adults from Taipei City, New Taipei City and Yilan (I-Lan) County, Taiwan.
Participants:
1271 community residents ≥65 years old, without impaired global cognition or dependency for activities of daily living/instrumental activities of daily living.
Measurements:
Demographic characteristics, anthropometric measurements, medical history, Mini-Mental State Examination, Montreal Cognitive Assessment, Functional Autonomy Measuring System, Functional Assessment Staging Test, Center for Epidemiologic Studies Depression Scale, handgrip strength, 6-metre walk speed.
Methods:
Baseline characteristics of groups with/without cognitive frailty were analyzed and factors differing significantly in univariate analysis input to binary logistic regression to develop a cognitive frailty risk (CFR) score.
Results:
The prevalence of cognitive frailty was 15.8% overall; ILAS 21.4%, LAST 8.4%. Predictors of CFR comprised: age ≥75 years; female sex; waist circumference ≥90 cm (male), ≥80 cm (female); calf circumference <33 cm (male), <32 cm (female); memory deficits; and diabetes mellitus. CFR ≥5/14 had sensitivity of 70%, specificity of 60%, and predictive accuracy of 72%.
Conclusions:
A CFR score based on simple history-taking and anthropometric measurements integrates age, sex, cardiometabolic risk, memory deficits, sarcopenia, and nutrition, with validated predictive accuracy, and could be performed easily in community settings to identify seniors with cognitive frailty for appropriate interventions.
