Background <p>Preoperative educational interventions are often implicitly regarded as a useful and cost-effective means of improving arthroplasty care. However, previous research in this area is limited – both in terms of age and clinical heterogeneity. This review adds to the existing literature by investigating the relationship between educational interventions and clinically relevant outcome measures, whilst also incorporating a larger sample size, and more stringently defined eligibility criteria. The primary outcome was preoperative anxiety. Secondary outcomes were patient knowledge score, postoperative pain, time-to-discharge, quality-of-life, and surgical complication rate.</p> Methods <p>A systematic review and meta-analysis of 19 randomised controlled trials. Inclusion criteria specified an adult population receiving elective hip or knee arthroplasty, an intervention group participating in a structured educational intervention, and a control group receiving no education beyond routine preoperative consenting. Outcome data were independently extracted by the review team, before being pooled for statistical analysis using comparison of mean differences and a random effects model.</p> Results <p>Results showed small, but statistically significant, improvements for the outcomes of patient anxiety (SMD -0.37, 95% CI -0.63 to -0.11), knowledge (SMD +0.37, 95% CI +0.05 to +0.69), and pain (SMD -0.31, 95% CI -0.47 to -0.15). However, the results for other outcome measures were not statistically significant. </p> Conclusions <p>Findings show a trend in favour of education for reducing anxiety and postoperative pain, though effect sizes were small, and the amount of practical benefit remains questionable due to inter-trial clinical heterogeneity. It was also observed that multi-format interventions may be beneficial for patients with greater anxiety, learning needs, or pain. Further research is necessary due to a lack of trials with reproducibly described interventions. </p>