Background To assess the relationship between sport and osteoarthritis (OA), and specifically to determine whether previous participation, in terms of level (elite or non-elite), type of sport, intensity or previous injury, was associated with OA. Methods This systematic review was developed using PRISMA guidelines. Databases were searched (to May 2016). Narrative review and meta-analysis (with
... [Show full abstract] risk ratio (RR) and 95% CIs) approaches were undertaken where appropriate. Study quality was assessed using GRADE. Results 46 studies were included. Narratively, 31 studies reported an increased risk of OA, with 19 demonstrating an increased risk in elite athletes. There was an increased risk after sports exposure (irrespective of type; RR 1.37; 95% CI 1.14 to 1.64; 21 studies). It remained uncertain whether there was a difference in risk of OA between elite and non-elite athletes (RR 1.37; 95% CI 0.84 to 2.22; 17 studies). The risk was higher in soccer (RR 1.42; 95% CI 1.14 to 1.77; 15 studies) but lower in runners (RR 0.86; 95% CI 0.53 to 1.41; 12 studies). 9 studies showed an association with the intensity of sport undertaken and OA. 5 studies demonstrated a higher prevalence of OA following meniscectomies and anterior cruciate ligament tears. Overall, the evidence was of GRADE 'very low' quality. Conclusions There was very low-quality evidence to support an increased relationship between sports participation and OA in elite participants. It is unclear whether there is a difference in risk between elite and non-elite participants with further prospective studies needed to evaluate this. Pooled findings suggested that significant injuries were associated with OA in soccer players. © 2016 BMJ Publishing Group Ltd & British Association of Sport and Exercise Medicine.