It has been suggested that inflammation is involved in Alzheimer disease (AD) pathogenesis. The aim of this study is to evaluate the association between inflammatory aspects of diet and incident AD risk. About 2258 nondemented elderly (age ≥ 65) in New York who provided dietary information at baseline were followed-up prospectively for AD development. We examined the composite total Inflammation Factor Rating (tIFR), as a measure of inflammatory impact of foods, in relation with (i) serum level of high-sensitivity C-reactive protein (hsCRP) and (ii) risk of incident AD using Cox proportional hazard model. The tIFR was not associated with serum hsCRP level. After an average of 4.0 years of follow-up, 262 participants developed incident AD. The tIFR was not associated with AD risk: compared with the lowest tertile of tIFR (most proinflammatory), hazard ratios (95% confidence interval) for the highest tertile (most anti-inflammatory) was 0.97 (0.69-1.35) (P-for-trend=0.71), in the adjusted model. We conclude that tIFR might not be a biologically relevant measure of the inflammatory impact of the diet. In addition, although it remains possible that tIFR might be related with some other aspects of inflammation not captured by hsCRP, lack of association with AD risk suggests its limited clinical utility.