Abstract
OBJECTIVE: To identify latent classes based on symptom clusters and to explore the association between these distinct symptom experience subtypes and social isolation in older adults with comorbid diabetes mellitus (DM) and coronary heart disease (CHD). METHODS: A cross-sectional study was conducted among 337 older adults with DM and CHD recruited from the Department of Endocrinology and Cardiology of Nantong Sixth People's Hospital between February 2023 and October 2025. Data were collected using a general information questionnaire, the Chinese version of the Memorial Symptom Assessment Scale (MSAS), and the Lubben Social Network Scale-6 (LSNS-6). Exploratory factor analysis (EFA) was used to identify symptom clusters. Latent profile analysis (LPA) was then employed to classify patients into different symptom experience subtypes based on the symptom cluster scores. One-way ANOVA, Chi-square tests, and multiple linear regression were used to analyze the association between latent classes and social isolation. RESULTS: EFA extracted three symptom clusters (cardiopulmonary-fatigue, emotional-perceptual, and metabolic), accounting for 62.3% of the total variance. LPA identified three distinct latent classes: Class 1 "Low Burden-Balanced Pattern" (45.4%), Class 2 "Psycho-Somatic Co-dominant Pattern" (31.8%), and Class 3 "Metabolic-Physical Dominant Pattern" (22.8%). Univariate analysis revealed significant differences in social isolation scores (LSNS-6) across the three classes (F = 35.67, p < 0.001). Multiple linear regression analysis, after adjusting for confounders, indicated that compared to Class 1, both Class 2 (β = -4.82, p < 0.001) and Class 3 (β = -3.25, p < 0.001) were significantly associated with lower LSNS-6 scores, suggesting a greater degree of social isolation. CONCLUSION: The findings reveal significant heterogeneity in symptom experiences among older adults with comorbid DM and CHD, which can be categorized into distinct latent classes. The subtype characterized by a Psycho-Somatic Co-dominant Pattern shows the strongest association with social isolation. In clinical practice, early identification of this high-burden subgroup may facilitate the provision of integrated interventions that address physical, psychological, and social dimensions.