Multimorbidity patterns and their associations with demographic characteristics, behavioral factors, and urban–rural residence: A latent class analysis in the Thai population

Abstract

Multimorbidity, defined as the presence of two or more chronic conditions, is a growing public health concern. Understanding disease clustering and risk profiles is crucial for prevention and planning purposes. This study analyzed data from the National Health Survey in Thailand, which included 142,753 participants. Latent class analysis investigated nine non-communicable diseases, while multinomial logistic regression explored the characteristics associated with class membership. Four distinct classes were identified: low burden, hypertension–metabolic, predominant hypertension, and high multimorbidity. Advancing age revealed an increased probability of membership in all multimorbidity classes. Females exhibited a higher propensity to belong to the hypertension–metabolic and predominant hypertension classes. Urban residency was associated with an elevated probability of membership in the hypertension–metabolic and high-multimorbidity classes. The body mass index displayed significant variations across classes. Overweight and obesity were linked to the hypertension–metabolic and predominant hypertension classes, with obesity also associated with the high multimorbidity class. Underweight individuals demonstrated a reduced likelihood of membership in the hypertension–metabolic class but an increased likelihood of membership in the predominant hypertension class. Smoking was correlated with a higher probability of membership across classes, particularly among current smokers. Dietary behaviors were distinctly associated across classes. Frequent consumption of high-fat and instant foods was linked to the hypertension–metabolic class, whereas instant food consumption was also associated with the predominant hypertension class. Fruit consumption was correlated with a higher probability of membership in the hypertension–metabolic class, whereas vegetable consumption was associated with a reduced likelihood of membership in the predominant hypertension class. Moderate consumption of non-alcoholic sugary drinks was associated with a lower probability of hypertension–metabolic class membership. These findings underscore the heterogeneity of multimorbidity patterns in Thailand and delineate the demographic, behavioral, and residential profiles associated with non-communicable disease clustering.