SUMMARY
The medical literature is distorted by publication bias, duplicate publication and under-reporting of clinical trials. This can create problems in systematic reviews and meta-analysis which are increasingly used as the basis for evidence-based guidelines and therefore for treatment decisions. Various schemes for identifying trials and their associated publications are reviewed. These include trial registers and trial numbering systems. The inclusion of trial identifiers in all publications could greatly enhance the value of the medical literature and increase the robustness of systematic reviews and meta-analyses.