Sex-Specific Dietary Inflammation and Metabolic Syndrome Across Three Nations: Cross-National IPD Harmonisation with Observational and Supportive Genetic Evidence for the Inflammatory Pathway (Diet-Attributable Component Bounded at OR=1.001-1.013)
TL;DR
A critical finding is the construct validity failure of the 20-item DII in non-Western dietary contexts, underscoring the need for population-specific inflammation metrics.
Abstract
Abstract Background: The dietary inflammatory index (DII) has been associated with metabolic syndrome (MetS), but evidence for sex-specific effects with causal support and cross-national validation remains limited. Methods: We harmonised individual-participant data from 56,475 adults across US NHANES, Korean KNHANES, and Mexican ENSANUT using survey-weighted sex-stratified logistic regression. Three complementary approaches assessed the inflammatory pathway: CRP-based Mendelian randomization (primary: 84 female-matched instruments; sensitivity: 56-SNP sex-overlap subset), exploratory direct diet MR, and proxy gap decomposition. Country-specific population attributable fractions evaluated cross- cultural DII construct validity. Results: Observational analyses revealed a robust female-specific DII-MetS association (OR=1.15, P<0.0001; Q5 vs Q1 OR=1.56) across three countries, with null associations in men (sex interaction P<0.0001; RERI P>0.05). CRP-based MR provided conservative support: the primary 84-SNP set yielded an attenuated estimate (IVW OR=1.03, P=0.31), while the pleiotropy-robust MR-RAPS method indicated a modest signal (OR=1.03-1.05, P<=0.01). Exploratory direct diet MR showed directionally consistent effects (female OR=0.91, P=0.10). Country-specific PAFs exposed a construct validity gradient: US women 20.2% versus negative PAFs in Korea (-8.0%) and Mexico (-33.1%). Formal triangulation, noting partial evidence-line dependency, yielded Fisher combined P=4.4e-8. Conclusions: Pro-inflammatory diet shows a consistent female-specific association with MetS across three nations. CRP-based MR provides conservative genetic support for the inflammatory pathway hypothesis (MR-RAPS OR=1.03-1.05). A critical finding is the construct validity failure of the 20-item DII in non-Western dietary contexts, underscoring the need for population-specific inflammation metrics. The multiplicative sex interaction was significant while the additive interaction was not, suggesting population-wide dietary guidance may capture the potential benefit. Trials are needed to establish causality. Keywords: Dietary Inflammatory Index; Metabolic syndrome; Mendelian randomization; Sex differences; Cross-national; Construct validity