Objective: To identify distinct multimorbidity patterns among hospitalized elderly patients using latent class analysis (LCA) and to examine the association between these patterns and frailty status.

Methods: A cross-sectional study was conducted among patients aged ≥65 years admitted to a tertiary hospital between January 2024 and December 2025. Demographic data, chronic disease diagnoses, and frailty assessments using the Clinical Frailty Scale (CFS) were collected. LCA was applied to identify multimorbidity subgroups based on 15 common chronic conditions. Multivariate logistic regression was used to evaluate associations between latent classes and frailty, adjusting for age, sex, body mass index, and polypharmacy.

Results: A total of 586 patients were included (mean age 73.8±6.9 years; 52.7% male). LCA identified four distinct multimorbidity patterns: cardiometabolic class (31.6%), musculoskeletal-respiratory class (24.3%), neuropsychiatric-metabolic class (23.5%), and complex multisystem class (20.6%). Overall frailty prevalence was 41.2%. After multivariable adjustment, the complex multisystem class showed the highest frailty risk (OR=4.16, 95% CI: 2.73–6.34, P<0.001), followed by the neuropsychiatric-metabolic class (OR=2.43, 95% CI: 1.58–3.74, P<0.001), compared with the cardiometabolic class. Conclusion: Hospitalized elderly patients exhibit heterogeneous multimorbidity patterns. Complex multisystem and neuropsychiatric-metabolic clusters confer substantially elevated frailty risk, highlighting the need for pattern-based, individualized geriatric interventions.

Authors:
Dan Xiang, Chongqing University, China

About the Presenter(s)
Xiang Dan, female, Bachelor's degree, Senior Nurse, Comprehensive Geriatric Assessor, Cardiovascular Specialist Nurse, graduated from Wuhan University of Traditional Chinese Medicine, and completed advanced training in Cardiovascular Medicine at the

See this presentation on the full schedule – On Demand Schedule

Bookmark