Objective: To investigate the expression level and application value of anti-carbamylated protein (CarP) antibody in rheumatoid arthritis (RA). Methods: Demographic data and laboratory test results of RA patients, non-RA patients and healthy controls in the physical examination center were reviewed from December 2018 to June 2019 in the Rheumatology and Immunology Department of the People' s Hospital of Xinjiang Uygur Autonomous Region. The serum concentrations of anti-CarP antibodies in all the subjects were measured by ELISA and statistically analyzed. Results: A total of 259 subjects were included in this study, including 158 in the RA group (45 serum-negative RA patients), 59 in the non-RA group and 42 in the healthy control group. The concentration of anti-CarP antibody in RA group [8.31 (5.22, 15.26) U/mL] was higher than that in non-RA group [4.50 (3.35, 5.89) U/mL] and healthy control group [3.46 (2.76, 4.92) U/mL]. The concentration of anti-CarP antibody in non-RA group was not significantly different from that in healthy control group (P=0.10). Receiver operating characteristic (ROC) curve analysis showed that the sensitivity of anti-CarP antibody in the diagnosis of RA was 58.2%, and the specificity was 93.1%. The sensitivity of the combined detection of anti-CarP antibody, anti-cyclic peptide containing citrulline (CCP) antibody and rheumatoid factor (RF) was 82.3%, and the specificity was 96.5%. The positive rate of anti-CarP antibody in serum-negative RA patients was 44.4% (20/45). Univariate Logisitic regression analysis showed that age, C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), RF, glucose-6-phosphate isomerase (GPI), anti-CCP antibody and anti-CarP antibody were risk factors for RA. Multivariate Logisitic regression analysis showed that anti-CCP antibody and anti-CarP antibody were independent risk factors for RA. Spearman correlation analysis showed that there was no significant correlation between anti-CarP antibody and swollen joint count (SJC), tenderness joints count (TJC), ESR, disease activity score for 28 joints (DAS28), clinical disease activity index (CDAI), simplified disease activity index (SDAI). The concentration of anti-CarP antibody in RA with bone erosion (n=88) was higher than that in RA without bone erosion (n=70), and there was significant difference between the two groups (P < 0.05). Conclusions: Anti-CarP antibody is an effective serological marker for the diagnosis of RA. The combined detection of RF, anti-CCP antibody and anti-CarP antibody can improve its diagnostic value, and anti-CarP antibody may be an effective assistant diagnostic tool for serum negative RA. The high serum concentration of anti-CarP antibody in patients with RA may indicate an increased risk of bone erosion and should be treated early, but further cohort studies are needed for follow-up observation.