This review aimed to quantify the benefit of patient-based educational interventions in the management of cancer pain. We undertook a systematic review and meta-analysis of experimentally randomised and non-randomised controlled clinical trials identified from six databases from inception to November 2007.Two reviewers independently selected trials comparing intervention (formal instruction on cancer pain and analgesia on an individual basis using any medium) to usual care or other control in adults with cancer pain. Methodological quality was assessed, and data extraction undertaken by one reviewer with a second reviewer checking for accuracy. We used random effects model to combine the effect estimates from studies. Main outcome measures were effects on knowledge and attitudes towards cancer pain and analgesia, and pain intensity. Twenty-one trials (19 randomised) totalling 3501 patients met inclusion criteria, and 15 were included in the meta-analysis. Compared to usual care or control, educational interventions improved knowledge and attitudes by half a point on 0-5 rating scale (weighted mean difference 0.52, 95% confidence interval 0.04-1.0), reduced average pain intensity by over one point on 0-10 rating scale (WMD -1.1, -1.8 to -0.41) and reduced worst pain intensity by just under one point (WMD -0.78, -1.21 to -0.35). We found equivocal evidence for the effect of education on self-efficacy, but no significant benefit on medication adherence or on reducing interference with daily activities. Patient-based educational interventions can result in modest but significant benefits in the management of cancer pain, and are probably underused alongside more traditional analgesic approaches.