The Dietary Inflammatory Index (DII) is a model that attempts to predict the inflammatory potential of a diet. It assigns points to various components of the diet, giving positive scores to proposed inflammatory components (e.g. trans fat, saturated fat, and calories) and negative points to proposed anti-inflammatory components (e.g. omega-3s, green/black tea, fiber, and vitamins C, E, and D). The final score, in theory, offers a general estimate of the inflammatory or anti-inflammatory potential of the diet.

The DII was designed using data from thousands of scientific publications,^[1] but some of the research it includes and how it incorporates this research is likely not ideal. In short, when designing the DII, the investigators included non-human studies (i.e., animal studies and cell culture experiments), did not weight studies differently according to their sample sizes, and only considered whether changes in inflammatory markers were statistically significant, but not if they were clinically important. These issues could make the DII a less than reliable model.

That said, observational studies have found that people with a higher DII score do tend to have higher levels of certain inflammatory markers,^[2][3][4][5] albeit with some inconsistency when broken down by individual markers.^[6][7] However, it’s not entirely clear which of the 45 dietary components included in the DII are actually important determinants of its predictive ability and which are less relevant.