Should systematic reviews include searches for published errata?

Our objective was to perform a pilot study to estimate the proportion of published errata linked to randomized controlled trials (RCTs) that are worthwhile obtaining when doing a systematic review. medline was searched for records that had both 'randomized-controlled-trial' in the publication type field and 'erratum' in the comments field. One hundred records from four general medical journals were examined independently from two different perspectives. From the information specialist's perspective, 74% of the errata were considered worthwhile obtaining; these were mainly errors in tables or figures. Another 9% described less serious errors, but were worth obtaining if easily available. The other 17% were minor errors. From the perspective of the experienced reviewer/public health consultant, 5% of errata were classified as likely to affect a meta-analysis, and 10% as having significant errors that would affect the interpretation of the RCT, but no effect on a meta-analysis; 85% were not considered important enough to affect either. About 5% of errata to RCTs appeared to matter in terms of changing the final conclusions of a systematic review. However, the majority of errata were considered to be worthwhile obtaining, on the basis that having full and accurate data can reduce confusion and save reviewers time.