This systematic review aimed to synthesise the content, structure, and delivery characteristics of effective yoga interventions for managing osteoarthritis symptoms, including joint pain and joint function. JBI guidelines were followed. 17 databases were searched for randomised controlled trials (RCTs) assessing yoga’s effectiveness on osteoarthritis symptoms. Meta-analyses and a narrative synthesis were conducted to address the objective. The systematic review and meta-analysis included 18 and 16 articles (representing 16 and 14 RCTs), respectively. Overall, the included studies had low methodological quality scores. 10 of 14 yoga interventions effectively reduced pain (standardised mean difference (SMD) − 0.70; 95% confidence interval (CI) − 1.08, − 0.32) and/or improved function (− 0.40; − 0.75, − 0.04). Notably, 8 effective interventions had centre-based (supervised, group) sessions, and 6 included additional home-based (unsupervised, individual) sessions. Effective interventions included 34 yogic poses (12 sitting, 10 standing, 8 supine, 4 prone), 8 breathing practices, and 3 meditation and relaxation practices. 8 interventions included yogic poses, and 7 also incorporated breathing practices and/or meditation and relaxation practices. 4 interventions included yogic pose modifications for osteoarthritis. The median duration of centre-based sessions was 8 weeks and each session was around 53 min, mostly delivered once a week. The median duration of home-based sessions was 10 weeks and each session was 30 min, usually instructed to practice 4 times a week. Given previous studies’ limitations, a high-quality long-term RCT should be conducted using synthesised findings of previous effective yoga interventions.