Several molecular epidemiological studies have been conducted in recent years to evaluate a possible association between the interferon regulatory factor 5 (IRF5) rs2004640 polymorphism and rheumatoid arthritis risk in diverse populations. However, the results remain conflicting rather than conclusive. Our aim was to assess associations of IRF5 gene polymorphisms with rheumatoid arthritis risk. Meta-analysis was performed on six published case-control studies (from eight countries) that included 4,818 cases of rheumatoid arthritis and 4,316 controls. The rs2004640-T allele was associated with a significantly increased risk of rheumatoid arthritis when the dominant genetic model was applied (T/T + T/G versus G/G: P = 0.003, OR = 1.14, 95% CI 1.05-1.25). Upon stratified analysis by ethnicity, the rs2004640 polymorphism was associated with an increased rheumatoid arthritis risk in Caucasians when the homozygotic contrast model was employed(T/T versus G/G: P = 0.03, OR = 1.25, 95% CI 1.02-1.53) and this was also the case when the dominant genetic model was used (T/T + T/G versus G/G: P = 0.04, OR = 1.20, 95% CI 1.01-1.42), whereas, in Asian populations, only the dominant genetic model was associated with an increased rheumatoid arthritis risk (T/T + T/G versus G/G: P = 0.02, OR = 1.14, 95% CI 1.02-1.26). The results suggest that the IRF5 rs2004640 polymorphism is associated with rheumatoid arthritis especially when the dominant genetic model is applied.