Abstract
This study examined associations between fluid intake characteristics, consumption patterns, hydration status, and MASLD among community-dwelling older adults, exploring whether hydration status shows statistical associations consistent with an indirect pathway in cross-sectional analyses. This cross-sectional study enrolled 876 community-dwelling adults aged ≥ 65 years recruited from community health centers. Fluid intake was assessed using 3-day beverage records, including two weekdays and one weekend day. First-morning urine osmolality was used as the overnight hydration indicator. MASLD was diagnosed based on ultrasound-detected hepatic steatosis and metabolic dysfunction criteria. Principal component analysis (PCA) was used to identify beverage patterns. Bootstrap-based exploratory analyses with 5000 resamples were performed as secondary descriptive decompositions to examine whether first-morning urine osmolality showed statistical associations consistent with an indirect association in cross-sectional data. These analyses are descriptive and hypothesis-generating only, and do not imply causality or temporal ordering. Mean total fluid intake was 1385 ± 355 mL/day, with plain water comprising 85.2% (1180 ± 285 mL/day) of consumption. Overall MASLD prevalence was 36.1% (316/876). In the fully adjusted model (adjusted for age, sex, education, physical activity, total energy intake, smoking, DII, red meat, vegetables, whole grains, fruit, season, and diuretic use), lower fluid intake ( 800 mOsm/kg) was also associated with higher odds (OR 2.55, 95% CI 1.89-3.43, P < 0.001). PCA identified three beverage patterns that cumulatively explained 49.9% of the total variance. In fully adjusted models, the Sweet Beverage Preference pattern was positively associated with MASLD (OR 2.00, 95% CI 1.52-2.63, P < 0.001); after additional adjustment for total added sugar intake, this OR attenuated to 1.76 (95% CI 1.31-2.36). The Basic Hydration pattern was inversely associated with MASLD (OR 0.67, 95% CI 0.51-0.89, P = 0.005). The Basic Hydration pattern was inversely associated with MASLD (OR 0.67, 95% CI 0.51-0.89, P = 0.005). The fully adjusted model showed good calibration (Hosmer-Lemeshow P = 0.420) and acceptable discrimination (AUC = 0.782). The Sweet Beverage Preference pattern should be interpreted primarily as a marker of broader unhealthy dietary behavior and cardiometabolic risk rather than as evidence that minimal SSB intake directly increases MASLD risk. In older adults with plain water-dominant reported beverage intake, lower reported beverage intake, higher first-morning urine osmolality, and a Sweet Beverage Preference pattern were associated with MASLD. The Sweet Beverage Preference pattern should be interpreted primarily as a marker of broader dietary and metabolic risk rather than as a beverage-specific risk signal. These findings should be interpreted as exploratory cross-sectional associations and may reflect broader cardiometabolic and dietary-pattern differences rather than direct causal effects.