Scientific Reports (Feb 2024)

Healthy eating index-2015 and its association with the prevalence of stroke among US adults

  • Xiao-Fei Wu,
  • Fei Yin,
  • Gui-Jie Wang,
  • Ye Lu,
  • Rong-Fei Jin,
  • Dong-Lin Jin

DOI
https://doi.org/10.1038/s41598-024-54087-9
Journal volume & issue
Vol. 14, no. 1
pp. 1 – 12

Abstract

Read online

Abstract This study aims to investigate the relationship between the healthy eating index (HEI) and the prevalence of stroke within a diverse United States population. Employing a cross-sectional design, we utilized data sourced from the National Health and Nutrition Examination Survey (NHANES). Dietary information was collected from participants and HEI scores were computed. NHANES employed stratified multistage probability sampling, with subsequent weighted analysis following NHANES analytical guidelines. Thorough comparisons were made regarding the baseline characteristics of individuals with and without stroke. Weighted multivariable logistic regression analysis and restricted cubic spline (RCS) methods were employed to ascertain the association between stroke risk and HEI, with LASSO regression utilized to identify dietary factors most closely linked to stroke risk. Additionally, we constructed a nomogram model incorporating key dietary factors and assessed its discriminatory capability using the receiver operating characteristic (ROC) curve. Our study encompassed 43,978 participants, representing an estimated 201 million U.S. residents. Participants with a history of stroke exhibited lower HEI scores than their non-stroke counterparts. Logistic regression analysis demonstrated a robust association between lower HEI scores and stroke, even after adjusting for confounding variables. RCS analysis indicated a nonlinear negative correlation between HEI and stroke risk. Furthermore, detailed subgroup analysis revealed a significant gender-based disparity in the impact of dietary quality on stroke risk, with females potentially benefiting more from dietary quality improvements. Sensitivity analysis using unweighted logistic regression yielded results consistent with our primary analysis. The nomogram model, based on key dietary factors identified through LASSO regression, demonstrated favorable discriminatory power, with an area under the curve (AUC) of 79.3% (95% CI 78.4–81.2%). Our findings suggest that higher HEI scores are inversely related to the risk of stroke, with potential greater benefits for women through dietary quality enhancement. These results underscore the importance of improving dietary quality for enhanced stroke prevention and treatment.

Keywords