Purpose: This meta-analysis study aimed to investigate the effects of cognitive behavioral therapy on anxiety, depression and self-efficacy of patients with arthritis. Methods: The review searched from PubMed, Ovid-MEDLINE, EMBASE, CINAHL, Cochrane Library and Korean databases. Key words for searching included ‘cognitive behavioral therapy’, ‘rheumatoid arthritis’, ‘osteoarthritis’ and the articles published from 2000 to June 2018 were selected. Cochrane s Risk of Bias for randomized studies was used to assess methodological quality. Data were analyzed by the R version 3.5.1 program. Results: Thirteen studies with a total of 1,135 participants met the inclusion criteria. There were significant differences effect of cognitive behavioral therapy (effect size, ES=-0.30), anxiety (ES=-0.32), depression (ES=-0.38), self-efficacy (ES=0.46). As a result of the moderator analysis, the longer the duration, the greater the effect of cognitive behavior therapy. Conclusion: The results of the study showed that cognitive behavioral therapy is effective in reducing anxiety, depression, and increasing self-efficacy in patients with arthritis. As a result, it can be expected to be used as basic data for applying cognitive behavioral therapy by patients with arthritis.