Objective: To assess the benefits and harms associated with biopsychosocial rehabilitation in patients with inflammatory arthritis and osteoarthritis (OA). Methods: We performed a systematic review and meta-analysis. Data were collected through electronic searches of Cochrane CENTRAL, MEDLINE, Embase, PsycInfo, and CINAHL databases up to March 2019. Trials examining the effect of biopsychosocial rehabilitation in adults with inflammatory arthritis and/or OA were considered eligible, excluding rehabilitation adjunct to surgery. The primary outcome for benefit was pain and total withdrawals for harm. Results: Of the 27 trials meeting the eligibility criteria, 22 trials (3,750 participants) reported sufficient data to be included in the quantitative synthesis. For patient-reported outcome measures, biopsychosocial rehabilitation was slightly superior to control for pain relief (standardized mean difference [SMD] −0.19 [95% confidence interval (95% CI) −0.31, –0.07]), had a small effect on patient global assessment score (SMD –0.13 [95% CI –0.26, –0.00]), with no apparent effect on health-related quality of life, fatigue, self-reported disability/physical function, mental well-being, and reduction in pain intensity ≥30%. Clinician-measured outcomes displayed a small effect on observed disability/physical function (SMD −0.34 [95% CI −0.57, −0.10]), a large effect on physician global assessment score (SMD −0.72 [95% CI −1.18, −0.26]), and no effect on inflammation. No difference in harms existed in terms of the number of withdrawals, adverse events, or serious adverse events. Conclusion: Biopsychosocial rehabilitation produces a significant but clinically small beneficial effect on patient-reported pain among patients with inflammatory arthritis and OA, with no difference in harm. Methodologic weaknesses were observed in the included trials, suggesting low-to-moderate confidence in the estimates of effect.