To use meta-analysis to determine the accuracy of anti-cyclic citrullinated peptide (CCP) antibody in diagnosis of patients with rheumatoid arthritis (RA) in a Chinese population, we searched MEDLINE and CNKI databases for studies published in English or Chinese between January 2000 and June 2010. Two investigators independently evaluated studies for inclusion, data extraction, and quality assessment. We used a random-effects model to combine estimates of sensitivity, specificity, positive likelihood ratio (LR+), negative likelihood ratio (LR-), and diagnostic odds ratio (DOR). One hundred and eighteen studies met our inclusion criteria. All studies were of high quality. The summary estimates for anti-CCP antibody in the diagnosis of RA in a Chinese population were as follows: sensitivity 0.65 (95% confidence interval (CI) 0.65-0.66), specificity 0.95 (95% CI 0.95-0.96), positive likelihood ratio (LR+) 15.84 (95% CI 13.55-18.54), negative likelihood ratio (LR-) 0.33 (95% CI 0.31-0.35), and diagnostic odds ratio (DOR) 51.60 (95% CI 43.64-61.01). With high specificity and moderate sensitivity, anti-CCP antibody tests play an important role in conforming the diagnosis of RA in a Chinese population.