(P17-016-25) Trends in Extreme Sodium Intake: Analysis of NHANES Data From 2003 to 2018

Objectives: While global dietary sodium analyses typically focus on average intake estimates that provide a broad overview of dietary patterns, less attention is given to the temporal trends of intake extremes across populations. Yet, examining extremes is crucial to identifying at-risk groups, comprehending dietary behaviors fully, and designing targeted interventions to lessen health disparities and enhance outcomes. We investigated 16-year trends in high sodium intake in US adults and individuals with chronic kidney and cardiovascular diseases (CKD and CVD), using data from the National Health and Nutrition Examination Survey (NHANES) from 2003 to 2018.

Methods: We constructed sodium consumption distribution curves for the general population, stratified by sex, age, and race, and for subpopulations with CVD (hypertension, coronary heart disease, heart attacks, stroke) and CKD. We first estimated percentile sequences from the minimum to the maximum intake {min-P5-P10-P25-P50-P75-P90-P95-max} and coefficients of skewness (CS) and kurtosis (CK). Then, a machine learning model was applied to detect decadal shifts in consumption patterns and the prevalence of intake exceeding the Chronic Disease Risk Reduction threshold of 2.3 g/day.

Results: Our preliminary analysis of the 2003–2004 (n=8,213) and 2017–2018 (n=6,491) NHANES cycles revealed significant shifts in sodium intake patterns. In the 2003-2004 cycle, the mean (± SD) sodium intake, percentile sequence, CS and CK were: 3.1±1.8 g/day; {0.0-0.7-1.2-1.9-2.8-3.9-5.4-6.4-15.1} g/day, 1.2 and 2.7, respectively. In contrast, the 2017–2018 cycle recorded a shift toward higher mean intake of 3.2±1.9 g/day; the percentile sequence {0.2-0.9-1.3-1.9-2.8-4.0-5.5-6.6-20.8} g/day, and increased CS=1.7 and CK=6.7. The shifts in both upper and lower percentiles, as well as the extremes of the intake distributions suggest notable and significant changes in dietary sodium intake patterns over time (Kolmogorov–Smirnov test, p < 0.0001). Among individuals with CVD and CKD, these trends were more pronounced, with mean intakes shifting from 2.9±1.3g/day to 3.2±2.1 g/day.

Conclusions: Our trend analysis demonstrates significant shifts in sodium consumption patterns and needs for targeted nutritional interventions sensitive to the evolving dietary habits and health conditions.

Funding Sources: None