This study was designed to assess the relationship between dietary insulin index (DII) and dietary insulin load (DIL) and rheumatoid arthritis (RA) risk in a case–control study. This study enrolled ninety-five newly diagnosed RA patients and 200 age- and sex-matched healthy controls. Dietary intakes were assessed using a validated 168-item semi-quantitative FFQ. DII and DIL were calculated using food insulin index values from previously published data. In the unadjusted model, individuals in the highest DIL tertile had the significantly higher odds of RA than those in the lowest tertile of the DIL scores (OR = 1·32, 95 % CI (1·15, 1·78), Pfor trend = 0·009). After adjusting for confounders, the risk of RA was 2·73 times higher for participants in the highest tertile of DIL than for those in the lowest tertile (OR = 2·73, 95 % CI (1·22, 3·95), Pfor trend < 0·001). In addition, patients in the highest DII tertile had higher risk of RA than those in the first tertile (OR = 2·22, 95 % CI (1·48, 3·95), Pfor trend = 0·008). This association persisted after adjusting for potential confounders (OR = 3·75, 95 % CI (3·18, 6·78), Pfor trend = 0·002). Our findings suggest that diets high in DII and DIL may increase the risk of developing RA, independent of other potential confounders. These findings can be verified by more research, particularly with a prospective design.