Household Size and Age-Modified Patterns of Cardiometabolic Biomarkers and Lifestyle Behaviors: KNHANES 2015–2024

Background Household structure is a salient social determinant that may shape cardiometabolic risk through daily routines, resource access, and health-related behaviors. However, the extant evidence remains limited in scope, particularly regarding whether associations between household size and cardiometabolic health differ across the lifespan. We examined age-dependent associations between household size and cardiometabolic outcomes using nationally representative survey data, integrating components of metabolic syndrome, blood biomarkers, and daily health behaviors. Methods A repeated cross-sectional analysis of nationally representative health and nutrition survey data was conducted, with data collected annually from 2015 to 2024. The study population comprised adults aged ≥ 19 years who provided complete information on household size, cardiometabolic outcomes, health behaviors, and covariates. Household size (FN) was categorized (e.g., 1, 2, 3, 4, ≥ 5 members), and age group (AG) was modeled categorically. To this end, survey-weighted regression models were employed to estimate associations of FN with metabolic syndrome components (including waist circumference and blood pressure), blood biomarkers (liver enzymes, renal function, hematologic and endocrine-related markers), and behaviors (physical activity domains, sedentary time, and sleep duration), testing FN×AG interactions. The models were adjusted for key sociodemographic and health-related covariates, and complex sampling was accounted for (weights, strata, and primary sampling units). Results Across various domains, the associations between household size and cardiometabolic indicators were frequently age-dependent rather than uniform. Among the components of metabolic syndrome, waist circumference and systolic blood pressure exhibited evidence of household-size associations, in conjunction with pronounced age effects, manifesting distinct FN×AG interaction patterns (waist circumference: p_FN = 0.003; p_int < 0.001; systolic blood pressure: p_FN = 0.022; p_int < 0.001). The data revealed interaction-dominant patterns in several additional components, including fasting glucose, triglycerides, and HDL-C. This finding highlights the heterogeneity observed across different life stages. Furthermore, the investigation revealed that biomarkers showed age-dependent correlations with household size, including interaction signals for liver enzymes (AST, p_int = 0.019; ALT, p_int = 0.003) and renal biomarkers (creatinine, p_int = 0.014). The behavioral findings yielded actionable candidates for pathways, demonstrating consistent associations between sedentary behavior and both FN and age, with a pronounced interaction (p_FN = 0.002; p_int < 0.001). The study's findings indicated significant domain-specific heterogeneity in physical activity outcomes, with notable interactions observed across specific domains and intensity levels, such as recreational vigorous activity and total activity. However, these interactions were not consistently observed across all domains or intensity levels, including occupational moderate or vigorous activity and recreational moderate activity. The present models did not demonstrate any significant FN×AG interactions for sleep duration. Conclusions Household size is not merely a demographic descriptor; rather, it is a contextual factor associated with cardiometabolic health in an age-dependent manner. The consistent interaction patterns observed across metabolic risk markers, biomarkers, and behaviors—particularly sedentary time—suggest that prevention and surveillance strategies may benefit from incorporating household structure as a pragmatic stratification marker and tailoring interventions to age-specific living contexts.