Cardiovascular
Does an inflammatory diet raise stroke risk?
Metadata + PaperFren explanation · cc by · source: Europe PMC
Higher dietary inflammatory index scores associated with greater stroke odds in NHANES adults.
Study at a glance
- Design
- Cross-sectional — NHANES 1999–2018; DII from 24-hour recalls vs self-reported stroke
- N
- N=44019 · US NHANES participants after exclusions
- Population
- US adults in NHANES 1999–2018 with dietary recall and stroke data
- Outcome
- Odds of self-reported stroke by dietary inflammatory index quartile
Structured fields used in claim comparison tables when every cited study has a complete layer.
Key findings
Stroke prevalence was 3.38%. Higher DII quartiles had rising stroke odds (Q4 OR 1.87 vs Q1). Association was nonlinear; LASSO-selected dietary factors yielded a nomogram with AUC 79.8%.
Methodology
Cross-sectional analysis of 44,019 US NHANES participants (1999–2018) linking dietary inflammatory index (DII) from 24-hour recalls to self-reported stroke, with weighted logistic and spline models.
Limitations
Cross-sectional design cannot establish causation; stroke and diet were measured at one time point and DII depends on recall accuracy.
How this study connects
Role on claims
Each row is a claim on a concept or method page where this paper supports, challenges, or qualifies the statement. Roles are hand-checked — not a model guess.
A larger weight-adjusted waist index associates with higher stroke odds in US survey data.
In US NHANES adults (2011–2020), higher weight-adjusted waist index was associated with higher odds of self-reported stroke after multivariable adjustment (about 25% higher odds per 1-unit WWI; highest vs lowest quartile OR about 1.62).
Scope note — different exposure — dietary inflammatory load, not WWI
Limits the claim's scope: a different population, assay, or outcome.
A more inflammatory diet score associates with higher stroke odds in related survey data.
In a related NHANES analysis, higher dietary inflammatory index quartiles were associated with higher stroke odds (highest vs lowest quartile OR about 1.87), with a nonlinear pattern and a LASSO-based nomogram AUC near 80%.
Evidence for the claim as stated.
These markers flag observational risk — they are not proven treatment targets.
In the prospective 45 and Up Study, current smoking was associated with elevated risks across many CVD subtypes including stroke, with especially large relative risks for peripheral arterial disease; former smokers showed intermediate risks, consistent with risk reduction after quitting.
Scope note — observational design — cross-sectional self-reported stroke odds
Limits the claim's scope: a different population, assay, or outcome.
WWI and DII both associate with stroke in NHANES, but they measure different exposures (central adiposity vs dietary inflammatory load). Neither analysis falsifies the other; treating them as a head-to-head disagreement confuses two questions that happen to share a survey outcome.
Evidence for the claim as stated.
NHANES stroke associations are cross-sectional and self-reported; 45 and Up smoking estimates are prospective event risks. A stronger causal reading is more defensible for the smoking cohort design than for the NHANES odds ratios—without requiring the NHANES associations to be “wrong.”
Evidence for the claim as stated.
Open questions
Tensions this paper is part of
From concept pages' “where studies disagree.” Disagreement means the same question; scope means different assays, populations, or outcomes.
WWI and DII both associate with stroke in NHANES, but they measure different exposures (central adiposity vs dietary inflammatory load). Neither analysis falsifies the other; treating them as a head-to-head disagreement confuses two questions that happen to share a survey outcome.
NHANES stroke associations are cross-sectional and self-reported; 45 and Up smoking estimates are prospective event risks. A stronger causal reading is more defensible for the smoking cohort design than for the NHANES odds ratios—without requiring the NHANES associations to be “wrong.”
Related papers in this topic
Same topic cluster — not a recommendation engine.