Whole fruit intake and 10-year ASCVD risk: mediation analysis in NHANES
Abstract
Background: Higher fruit intake is associated with better cardiovascular health, but mechanisms remain unclear and may differ for whole fruit versus fruit juice. We evaluated the association between usual whole-fruit intake and predicted 10-year ASCVD risk and explored potential dietary and biochemical mediators.
Methods: We analyzed 12,771 adults aged ≥40 years from NHANES 2007–2018 without prior cardiovascular disease. Usual whole-fruit intake (excluding juice) was estimated using the NCI multivariate method. The outcome was logit-transformed 10-year ASCVD risk based on the Pooled Cohort Equations. A high-dimensional mediation framework was applied to identify a latent mediator from multiple dietary and biomarker variables using machine learning (ridge regression, support vector regression, gradient boosting, and deep learning). Effects were estimated in a test set with bootstrapping and multiple imputations.
Results: Fruit consumers differed from non-consumers in socio-demographic and lifestyle characteristics, including lower smoking prevalence and obesity. Higher whole-fruit intake was associated with lower ASCVD risk. An increase of 80 g/day (one portion) corresponded to a total effect of − 0.037 (95% CI: −0.059, −0.015; p = 0.001), with consistent direct effects across models (−0.033 to − 0.037; p ≤ 0.002). Indirect effects were small and model-dependent; only gradient boosting detected a significant mediated effect (−0.003; p = 0.019), accounting for 8% of the total effect. Key contributors included eGFR, total energy intake, sodium, cured meat, and legumes/soy.
Conclusions: Higher whole-fruit intake was associated with lower predicted ASCVD risk, largely independent of the evaluated mediators. Findings suggest that whole-fruit intake reflects broader dietary and metabolic patterns rather than a single causal pathway.
Key messages: • Higher whole-fruit intake is associated with lower predicted 10-year cardiovascular risk in US adults, supporting current dietary recommendations.
• The association is only weakly mediated, suggesting fruit intake reflects broader healthy dietary and metabolic patterns rather than a single pathway.
Related
- Association between ultra-processed food intake and osteoarthritis prevalence among U.S. adults : A cross-sectional study using NHANES data
- Association Between Dietary Vitamin Intake and Short Sleep Duration Among Adults in the United States: A Cross-Sectional Analysis of 2021-2023 National Health and Nutrition Examination Survey (NHANES) Data
- Self-reported fruit and vegetable intake obtained through a brief telephone interview among users of the Academia da Saúde Program: a validation study, Belo Horizonte, Minas Gerais, Brazil, 2016-2017.
- Impact of the Food4Moms produce prescription program on readiness for healthy eating, fruit and vegetable intake, and food security
- Machine learning–driven identification of whole-food dietary patterns associated with anxiety and rheumatoid arthritis comorbidity patterns: evidence from the NHANES and the USDA MyPyramid Equivalents database
BibTeX
@article{llane2026whole,
title = {{Whole fruit intake and 10-year ASCVD risk: mediation analysis in NHANES}},
author = {A Llane and D Zampieri and H Ocagli and C Dente and F Fortuna and G Lorenzoni and D Gregori},
journal = {The European Journal of Public Health},
year = {2026},
month = oct,
doi = {10.1093/eurpub/ckag169.1335},
url = {https://doi.org/10.1093/eurpub/ckag169.1335},
note = {CC BY-NC 4.0},
}
RIS
TY - JOUR
TI - Whole fruit intake and 10-year ASCVD risk: mediation analysis in NHANES
AU - A Llane
AU - D Zampieri
AU - H Ocagli
AU - C Dente
AU - F Fortuna
AU - G Lorenzoni
AU - D Gregori
PY - 2026
DA - 2026/10/01
JO - The European Journal of Public Health
DO - 10.1093/eurpub/ckag169.1335
UR - https://doi.org/10.1093/eurpub/ckag169.1335
AB - Background: Higher fruit intake is associated with better cardiovascular health, but mechanisms remain unclear and may differ for whole fruit versus fruit juice. We evaluated the association between usual whole-fruit intake and predicted 10-year ASCVD risk and explored potential dietary and biochemical mediators. Methods: We analyzed 12,771 adults aged ≥40 years from NHANES 2007–2018 without prior cardiovascular disease. Usual whole-fruit intake (excluding juice) was estimated using the NCI multivariate method. The outcome was logit-transformed 10-year ASCVD risk based on the Pooled Cohort Equations. A high-dimensional mediation framework was applied to identify a latent mediator from multiple dietary and biomarker variables using machine learning (ridge regression, support vector regression, gradient boosting, and deep learning). Effects were estimated in a test set with bootstrapping and multiple imputations. Results: Fruit consumers differed from non-consumers in socio-demographic and lifestyle characteristics, including lower smoking prevalence and obesity. Higher whole-fruit intake was associated with lower ASCVD risk. An increase of 80 g/day (one portion) corresponded to a total effect of − 0.037 (95% CI: −0.059, −0.015; p = 0.001), with consistent direct effects across models (−0.033 to − 0.037; p ≤ 0.002). Indirect effects were small and model-dependent; only gradient boosting detected a significant mediated effect (−0.003; p = 0.019), accounting for 8% of the total effect. Key contributors included eGFR, total energy intake, sodium, cured meat, and legumes/soy. Conclusions: Higher whole-fruit intake was associated with lower predicted ASCVD risk, largely independent of the evaluated mediators. Findings suggest that whole-fruit intake reflects broader dietary and metabolic patterns rather than a single causal pathway. Key messages: • Higher whole-fruit intake is associated with lower predicted 10-year cardiovascular risk in US adults, supporting current dietary recommendations. • The association is only weakly mediated, suggesting fruit intake reflects broader healthy dietary and metabolic patterns rather than a single pathway.
SN - 1101-1262
SN - 1464-360X
ER -