Full publication of trials initially reported as abstracts in the Australian and New Zealand Journal of Medicine 1980-2000.

Abstract Failure to identify all relevant reports of controlled trials is a potential source of bias in systematic reviews of health-care interventions. The present study aims to identify how many reports of trials -initially published as conference abstracts in the Australian and New Zealand Journal of Medicine were subsequently published in full. We identified trial reports by handsearching conference abstracts published in the Journal from 1980 to 2000. We then searched the Cochrane Controlled Trials Register and PubMed to determine how many of these had been subsequently published in full. A total of 962 reports of controlled trials was identified from the conference proceedings of 17 medical societies. Of these, 589 (61%) reports of trials were subsequently published in full, and on average within 1-2 years. Handsearching conference abstracts identified a large number of reports of controlled trials, over one-third of which were unpublished and therefore not easily accessible.