Author(s): SUNGJAE, Y., HAN, Y., CHOI, Y., KIM, Y., Institution: SEOUNL NATIONAL UNIVERSITY, Country: KOREA, SOUTH, Abstract-ID: 938

INTRODUCTION:
Large-scale meta-analyses confirm that sedentary time (ST) independently predicts all-cause mortality, yet whether this risk differs by body mass index (BMI) and age has not been formally tested. Prior individual participant data meta-analyses reported broadly consistent ST-mortality associations across subgroups but did not test for multiplicative interaction. Moreover, existing evidence derives predominantly from Western cohorts with minimal East Asian representation. South Korea-where workers log the longest hours in the OECD and the population is aging faster than any high-income country-provides a relevant setting to examine this question. We therefore conducted a formal interaction analysis of the ST-mortality association by BMI and age in a nationally representative East Asian cohort.
METHODS:
We analyzed 23,823 adults (>=19 years) from the Korea National Health and Nutrition Examination Survey 2014-2018 linked to national death records through December 2022 (692 deaths; mean follow-up 5.3 years). Moderate-to-vigorous physical activity (MVPA) and ST were assessed via the Global Physical Activity Questionnaire. MVPA was categorized as Inactive (0 min/week), Insufficiently active (1-149 min/week), and Active (>=150 min/week); ST as <8, 8 to <12, and >=12 h/day. Survey-weighted Cox models estimated hazard ratios (HRs) adjusting for age, sex, BMI, smoking, alcohol, income, education, marital status, hypertension, and diabetes, with mutual MVPA-ST adjustment. Dose-response curves were modeled via restricted cubic splines. Multiplicative interactions were tested with cross-product terms.
RESULTS:
Active adults had 36% lower mortality versus Inactive (HR 0.64, 95% CI 0.51-0.78); ST <8 versus >=12 h/day was associated with 30% lower risk (HR 0.70, 0.54-0.89). A significant ST x BMI interaction was observed (P=0.024): reduced ST was associated with 42% lower mortality among normal-weight adults (HR 0.58, 0.43-0.78) but not among those with BMI >=25 (HR 1.13, 0.72-1.77). The ST x age interaction approached significance (P=0.074), with a 49% risk reduction among adults >=65 years (HR 0.51, 0.39-0.68) versus no significant association in younger adults (HR 0.99, 0.56-1.73). Dose-response analysis confirmed a non-linear L-shaped MVPA curve (P-nonlinearity=0.006) and a linear ST gradient (P=0.849). Joint analysis showed meeting MVPA guidelines attenuated ST-associated risk across all sitting levels. In contrast to ST, MVPA showed no effect modification by BMI or age.
CONCLUSION:
Our findings indicate that the ST-mortality association is modified by BMI and age: sedentary time reduction was strongly associated with lower mortality among normal-weight and older adults but not among those with elevated BMI, who may require integrated approaches addressing sedentary behavior alongside weight management. These results suggest that reducing ST may not benefit all adults equally, warranting age- and BMI-specific guidance.