We aimed to evaluate the influence of supervision, athlete age and sex and programme duration and adherence on exercise-based injury prevention programme effectiveness in sport. Databases were searched for randomized controlled trials evaluating exercise-based injury prevention programme effectiveness compared to "train-as-normal". A random effects meta-analysis for overall effect and pooled effects by sex and supervision and meta-regression for age, intervention duration and adherence were performed. Programmes were effective overall (risk ratio (RR) 0.71) and equally beneficial for female-only (0.73) and male-only (0.65) cohorts. Supervised programmes were effective (0.67), unlike unsupervised programmes (1.04). No significant association was identified between programme effectiveness and age or intervention duration. The inverse association between injury rate and adherence was significant (β=-0.014, p = 0.004). Supervised programmes reduce injury by 33%, but there is no evidence for the effectiveness of non-supervised programmes. Females and males benefit equally, and age (to early middle age) does not affect programme effectiveness.
Keywords: Sport injury prevention; age; sex; supervision.