Handsearching for 'buried' randomized trials in Bahrain medical journals

Article type
Authors
Fedorowicz Z, Amin F, Eisinga A, Al-Sayyad J
Abstract
Background: The handsearching initiative of The Cochrane Collaboration has added many previously 'buried’ reports of randomized trials to the Cochrane Central Register of Controlled Trials (CENTRAL). Some were 'buried’ because of inconsistencies in indexing (indexing bias), which arise in the assignment of index terms and the inclusion - or not - of records for articles. Bahrain is known for buried archaeological treasures; the handsearching of the Bahrain Medical Bulletin (BMB) and the Journal of the Bahrain Medical Society (JBMS) has revealed a previously buried treasure trove of reports of trials.
Objectives: To identify reports of randomized trials by handsearching BMB and JBMS, both indexed in EMBASE but not in MEDLINE. To determine the value of the handsearch by assessing the precision and sensitivity of (i) the EMBASE index term RANDOMIZED CONTROLLED TRIAL and (ii) The Cochrane Collaboration’s systematic electronic search of EMBASE as means of identifying the reports we found by handsearching.

Methods: The BMB (1979 to December 2004) and JBMS (1989 to October 2004) were handsearched for reports of trials. These were classified as randomized controlled trials (RCTs) or controlled clinical trials (CCTs), according to Cochrane eligibility criteria. The reports identified were verified and submitted for future inclusion in CENTRAL. EMBASE and CENTRAL were searched to identify if the reports were already included in either.

Results: We checked 395 articles in the BMB, and found 12 RCTs and four CCTs. Of 97 articles in JBMS, we found 15 RCTs and 2 CCTs. Overall distribution by country of author was highest for Jordan (16), followed by Bahrain (5). Of the 33 reports of trials found, 23 were in EMBASE but only 6 of these had the index term RANDOMIZED CONTROLLED TRIAL. Of the 23 reports in EMBASE, 12 were in CENTRAL (Issue 1, 2005).

Conclusions: This handsearch of Bahrain journals should help to minimize the effects of publication bias by providing reports of trials not previously identified. Although these journals are indexed in EMBASE, the handsearch has ensured that reports of trials will not remain "buried" because of inconsistent indexing (indexing bias).