Subtypes of physical frailty: Latent class analysis and associations with clinical characteristics and outcomes

Li Kuo Liu, Chao Yu Guo, Wei Ju Lee, Liang Yu Chen, An Chun Hwang, Ming Hsien Lin, Li Ning Peng, Liang Kung Chen*, Kung Yee Liang

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

55 Scopus citations


Frailty is a well-recognized geriatric syndrome with various definitions and conceptual frameworks. This study aimed to use latent class analysis to discover potential subtypes of pre-frail and frail older people. Data from the I-Lan Longitudinal Aging Study (ILAS), a community-based cohort study was used for analysis. Latent class analysis was applied to characterize classes or subgroups with different frailty phenotypes among ILAS participants targeting older adults aged 65 and above, capable of completing a 6-meter walk, without severe major or life threatening diseases, and not institutionalized. Latent class analysis identified three distinct subgroups with different frailty phenotypes: Non-mobility-type (weight loss and exhaustion), mobility-type frailty (slowness and weakness), and low physical activity. Comparing these groups with the robust group, people with mobility-type frailty had poorer body composition, worse bone health, poorer cognitive function, lower survival (hazard ratio: 6.82, p = 0.019), and poorer overall health outcomes (hazard ratio: 1.67, p = 0.040). People in the non-mobility-type group had poorer bone health and more metabolic serum abnormalities. In conclusion, mobility-type frailty was a better predictor of adverse outcomes. However, further investigation is needed to evaluate how these phenotypic subgroups may help in predicting prognosis or in developing interventions.

Original languageEnglish
Article number46417
JournalScientific reports
StatePublished - 11 Apr 2017


Dive into the research topics of 'Subtypes of physical frailty: Latent class analysis and associations with clinical characteristics and outcomes'. Together they form a unique fingerprint.

Cite this