Cardiometabolic multimorbidity phenotypes and their associations with functional and cognitive outcomes in older Chilean adults: findings from the Chilean national health survey

Abstract Background Multimorbidity is highly prevalent among older adults and is associated with functional decline, disability, cognitive impairment, and increased healthcare burden. However, traditional disease-count approaches may fail to capture the heterogeneity of multimorbidity patterns. This study aimed to identify cardiometabolic multimorbidity phenotypes and examine their associations with functional and cognitive outcomes among older Chilean adults. Methods A cross-sectional analysis was conducted using data from 2,031 community-dwelling adults aged ≥ 60 years from the Chilean National Health Survey 2016–2017. Six objectively measured cardiometabolic conditions were included: diabetes, hypertension, metabolic syndrome, elevated total cholesterol, elevated triglycerides, and obesity. K-means cluster analysis was applied to identify multimorbidity phenotypes, and the number of clusters was selected combining elbow, silhouette, and clinical interpretability criteria. Functional outcomes included self-rated health assessed using the EQ-5D visual analogue scale (EQ-VAS), mobility limitations, difficulty performing usual activities, and pain/discomfort. Cognitive performance was assessed using the abbreviated Mini-Mental State Examination. Adjusted linear regression models controlling for age and sex were performed. Results Overall, 53.5% of participants presented multimorbidity (≥ 2 conditions). Four multimorbidity phenotypes were identified: Low-burden (30.9%), Obesity-predominant (25.1%), Hypertension-dominant (22.1%), and Metabolic + obesity (21.9%). Significant between-group differences were observed for self-rated health (ANOVA p = 0.022), cognitive performance ( p = 0.0003), mobility limitations ( p = 0.005), difficulty performing usual activities ( p = 0.0008), and pain/discomfort ( p = 0.003). The phenotype structure was robust to the exclusion of metabolic syndrome in a sensitivity analysis (Adjusted Rand Index = 0.719). In adjusted analyses with the Low-burden phenotype as reference, the Hypertension-dominant phenotype was independently associated with greater pain/discomfort (β = 0.14, 95% CI 0.06 to 0.21; FDR-adjusted p = 0.011) and the Metabolic + obesity phenotype with higher Mini-Mental scores (β = 0.56, 95% CI 0.17 to 0.95; FDR-adjusted p = 0.039); an association between the Hypertension-dominant phenotype and lower self-rated health (β = -4.25, 95% CI -8.00 to -0.50) did not withstand correction for multiple comparisons. Conclusions Distinct cardiometabolic multimorbidity phenotypes were differentially associated with functional and cognitive outcomes among older Chilean adults. These findings support the importance of phenotype-based approaches to improve risk stratification and promote more personalized strategies in geriatric care and healthy aging.