Association of the dietary carbohydrate-to-fiber ratio with prevalent rheumatoid arthritis in NHANES 2009–2018

This study examines the cross-sectional association between the carbohydrate-to-fiber ratio (CFR), an indicator of carbohydrate quality, and prevalent rheumatoid arthritis (RA) in the National Health and Nutrition Examination Survey (NHANES). The exposure was overall CFR, calculated from 2 days of dietary recall and categorized into population-wide quartiles. RA diagnosis was the outcome. Survey-weighted logistic regression was used to evaluate the CFR-RA association, adjusting for demographic, lifestyle, dietary, and metabolic covariates. Generalized additive models, subgroup analyses, and interaction tests were performed. An exploratory survey-weighted mediation analysis evaluated SII as a hypothesized intermediate factor. The study ultimately included 19,260 participants aged ≥ 20 years with valid 2-day dietary weights, with a survey-weighted RA prevalence of 4.31%. Higher CFR was significantly associated with increased RA odds, with the highest quartile showing 67% higher odds than the lowest (P < .001). After exclusion of CFR values above the upper fence defined using the interquartile range, the model 4 GAM showed an effectively linear positive association. A sensitivity analysis retaining all CFR observations showed curvature in the sparsely populated upper tail with wide uncertainty. In the exploratory model 4 mediation analysis, the estimated indirect association through SII was small (4.77% of the total association; 95% confidence interval: -5.33% to 14.87%), and its confidence interval included the null. Higher CFR was associated with greater odds of prevalent self-reported RA. Longitudinal and genetically informed studies are needed before causal or preventive interpretations can be made.