Association between total dietary sugar intake and gallstones in Americans: a study based on machine learning

Objective: This study aimed to investigate the association between dietary total sugar intake and gallstone risk in the U.S. adult population. Methods: We conducted a cross-sectional analysis using data from 8975 eligible participants in the National Health and Nutrition Examination Survey (NHANES) 2017-2023. Dietary total sugar intake (g/day) was assessed via two 24-hour dietary recalls. Gallstones status comes from self-reported information. Multivariable logistic regression, restricted cubic splines (RCS) regression model, and subgroup analysis were employed to evaluate associations. Machine learning including XGBoost algorithm with SHapley Additive exPlanations (SHAP) analysis was used to further explore the potential correlation. Results: Each 100g/day increase in total sugar intake was associated with a 41% higher gallstone risk after adjusting for all covariates (OR = 1.41, 95%CI:1.20–1.65, P < 0.001). A linear dose-response relationship was observed in RCS regression model ( P for linear = 0.130). Subgroup analyses showed consistent associations across populations ( P for interaction > 0.05). XGBoost performed well on the test set (AUC = 0.896). SHAP analysis confirmed sugar intake as the sixth top predictor, with revealing age, gender, and BMI as stronger determinants. In addition, partial dependency plot revealed high sugar intake increased the risk of gallstones. Conclusion: Higher dietary sugar intake is significantly associated with increased gallstone risk, independent of traditional metabolic factors. These findings highlight sugar reduction as a potential preventive strategy, warranting further longitudinal and mechanistic studies.