Association Between Dietary and Circulating Folate Markers and Kidney Stones in American Adults: A Cross-Sectional NHANES Analysis

Folate intake and circulating folate markers had no significant association with kidney stone disease in 9,208 US adults after adjusting for confounders.

Journal: Medicine | Published: 2026-05-29 | Type: Cross-sectional analysis | PMID: 42216366 Authors: Safargar Mohammad, Basiri Abbas, Kord-Varkaneh Hamed (Urology and Nephrology Research Center, Shahid Labbafi Nejad Hospital, Iran; Tabriz University of Medical Sciences) Funding/COI: Funding not reported. Authors declare no conflicts of interest.

Summary

Folate participates in one-carbon metabolism, which overlaps with oxalate and homocysteine pathways implicated in kidney stone formation — so a connection seemed plausible enough to investigate. Using NHANES 2017–2020 data on 9,208 adults, researchers tested whether dietary folate intake or four circulating folate biomarkers predicted kidney stone disease prevalence. After adjusting for sociodemographic, lifestyle, and clinical variables, the association disappeared entirely across all measures.

Claims

• Weighted prevalence of kidney stone disease in the sample was 9.4%
• In unadjusted models, higher RBC folate and tetrahydrofolate (THF) concentrations were associated with increased odds of KSD
• After full adjustment (Model 3), dietary folate showed no significant association: Q4 vs Q1 OR 0.94 (95% CI: 0.57–1.55)
• No significant associations found for any circulating biomarker: total serum folate, 5-methyltetrahydrofolate, THF, or RBC folate
• No linear or nonlinear dose-response trend detected for any folate measure

Study Quality

NHANES is a large, nationally representative survey with validated dietary recall methodology and laboratory-measured biomarkers — a legitimate data source for cross-sectional questions. Survey-weighted multivariable logistic regression with quartile analysis and dose-response modeling is appropriate for the design. Crucially, the unadjusted associations (higher RBC folate and THF linked to more kidney stones) evaporated after confounding adjustment, which is the correct finding to report — confounders matter here.

The irreducible problem is cross-sectional design: kidney stone history is self-reported and retrospective, dietary recall captures one to two days of intake and is notoriously noisy, and temporality cannot be established. Whether folate status preceded stone formation or resulted from dietary changes after diagnosis is unknowable from this data. The authors adjusted for common confounders but cannot rule out residual confounding from unmeasured variables like fluid intake, gut oxalate absorption, or genetic polymorphisms in folate metabolism (MTHFR variants, for instance).

Red Flags

• Cross-sectional design: causality cannot be inferred in either direction
• Kidney stone history is self-reported, not verified by imaging or metabolic workup — misclassification is likely
• 24-hour dietary recall is a weak instrument for habitual nutrient intake; one or two days of recall does not represent long-term folate exposure
• Funding source not reported — cannot assess whether pharmaceutical or supplement industry had any role
• All three authors are affiliated with Iranian institutions; NHANES is a US dataset; no explanation of institutional access or IRB context provided
• Stone type (calcium oxalate, uric acid, struvite, cystine) not distinguished — mechanisms differ substantially and pooling all stone types may obscure subgroup effects

Strengths

• Large sample (n = 9,208) with nationally representative weighting
• Multiple folate biomarkers assessed simultaneously, including RBC folate and individual metabolites (5-MTHF, THF) rather than dietary intake alone
• Rigorous survey-weighted analysis respects NHANES complex sampling design
• Tested both linear and nonlinear dose-response relationships, not just quartile comparisons
• Null result reported clearly, without spin — the abstract conclusion matches the data

Verdict

A null finding, reported honestly. Folate likely does not belong on any serious list of kidney stone risk factors in the general US adult population. The unadjusted signal for RBC folate and THF is almost certainly confounded — wealthier, supplement-taking adults have higher folate levels and also different dietary and clinical profiles. Once you control for that, nothing remains. The cross-sectional design and self-reported stone history are real limitations, but the sample is large enough and the biomarker panel comprehensive enough that a substantial effect would have survived adjustment. This paper closes a speculative question more than it opens a new one.