Rheumatoid arthritis (RA) and depression are conditions which commonly co-exist. Recent longitudinal studies now suggest a bidirectional association between these disorders, with inconsistent results. We conducted a systematic review and meta-analysis to examine this relationship. Three electronic databases (PubMed, Embase and PsycINFO) were searched from inception to September 4, 2021 for cohort studies evaluating either the risk of depression in RA patients or the risk of RA in patients with depression, as well as the secondary outcome of all-cause mortality risk in RA patients with depression. A random effects model was used to summarize the included studies. Eleven cohort studies were included, comprising a total of 39,130 RA patients, 550,782 patients with depression and 7,802,230 controls. RA patients had a 47% greater risk of incident depression compared to controls, while patients with depression had a 34% greater risk of developing RA. Subgroup analysis by age was only significant in the ≥60 years old age group. RA patients with depression had an 80% increased risk of all-cause mortality compared to those without depression. The results may have been confounded by factors such as differing methods of depression ascertainment across studies and overlap in presentation between the two conditions. There exists a bidirectional association between RA and depression especially in the elderly which increases mortality risk. This invites the need for clinicians to screen and be vigilant for the presence of these conditions. •Our meta-analysis establishes a bidirectional association between rheumatoid arthritis and depression.•Rheumatoid arthritis patients with depression had an 80% increased risk of mortality.•Clinicians should screen patients with RA for depression and vice versa.•Anti-cytokine therapies represents a potential alternative for the treatment of depression.