Should systematic reviews include searches for published errata

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