Validation and adaptation of the empirical dietary inflammatory pattern across nations: A test case

Nutrition. 2020 Nov-Dec;79-80:110843. doi: 10.1016/j.nut.2020.110843. Epub 2020 Apr 29.


Objective: The empirical dietary inflammatory pattern (EDIP) assesses the inflammatory potential of diet in the US population. The aim of this study was to assess the applicability of the EDIP to a Brazilian population.

Methods: Data from 19- to 75-y-old participants of two editions of the cross-sectional population-based Health Survey of Sao Paulo (HS-SP) were used to validate the EDIP (n = 269; HS-SP 2008), develop an EDIP adapted to a Sao Paulo population, the EDIP-SP (n = 441; HS-SP 2008), and replicate EDIP-SP results in an independent sample (n = 501; HS-SP 2015). Dietary data was assessed through two 24-h recalls and one validated food frequency questionnaire. Plasma C-reactive protein (CRP), and nine other inflammatory biomarkers were determined. EDIP was tested for its association with the 10 inflammatory biomarkers. For development of the EDIP-SP, 21 food groups and their contributions to plasma CRP levels were modeled using a stepwise multiple linear regression adjusted for age and sex.

Results: The EDIP was not associated with concentrations of inflammatory biomarkers in a Brazilian population. The components of EDIP-SP were processed meats (β = 0.27; P = 0.082), fruits and vegetables (β = -0.12; P = 0.018), and rice and beans (β = -0.27; P = 0.007). EDIP-SP significantly predicted dietary quality (β = -6.1; P < 0.001) and its inflammatory potential was replicated among men (β = 0.36; P = 0.01), but not among women (β = 0.05; P = 0.82).

Conclusion: EDIP was adapted to the Sao Paulo population. EDIP-SP, composed of high processed meat intake and low intake of fruits and vegetables, and rice and beans, constitutes an important tool to investigate dietary quality based on its inflammatory potential, in Brazilian populations.

Keywords: A posteriori; Data-driven; Dietary pattern; Empirical dietary inflammatory pattern; Hypothesis-driven; Inflammation.

MeSH terms

  • Biomarkers
  • Brazil
  • C-Reactive Protein
  • Cross-Sectional Studies
  • Diet*
  • Female
  • Humans
  • Inflammation*
  • Male


  • Biomarkers
  • C-Reactive Protein