Searching KoreaMed: increasing the discoverability of trials conducted and published in Korea




Poster session 4


Tuesday 25 October 2016 - 15:30 to 16:00


All authors in correct order:

Kim M1, McDonald S2
1 Monash University, Australia
2 Cochrane Australia, Australia
Presenting author and contact person

Presenting author:

Steve McDonald

Contact person:

Abstract text
Background: KoreaMed provides access to over 235,000 articles published in around 230 Korean journals. It is estimated that fewer than 10% of these journals are indexed in MEDLINE. Although KoreaMed is an open access database, it is only recently that digital object identifiers (DOIs) and links to full-text content have been added to KoreaMed records.

Objectives: To search KoreaMed for reports of randomised trials (RCTs) published in Korean journals and to make these available through CENTRAL in the Cochrane Library (in partnership with the publishers of KoreaMed).

Methods: Using a sensitive search strategy comprising free-text terms in English and Korean, plus publication type terms, we retrieved citations from the early 1970s onwards. One author (MK) screened citations and referred queries to the second author (SM). Records identified as reporting a randomised (or possible randomised) trial were then independently checked. We also looked at the journals that published RCTs and the accuracy of the indexing of trials in KoreaMed.

Results: Our search strategy retrieved 7645 citations, of which 3319 were identified as being reports of RCTs. Relatively few trials (n = 46) were published before 1990, but the number increased rapidly thereafter: 850 trials were identified in the 1990s and over 1750 in the 2000s. The Korean Journal of Anesthesiology published 41% (n = 1374) of all RCTs identified. A fifth of records (n = 671) were correctly indexed with RCT as a Publication Type. We identified over 450 citations that had the RCT tag applied erroneously. The KoreaMed records were added to the Cochrane Library in 2015. Following the inclusion of DOIs in a subset of journals in KoreaMed, we have recently added DOIs to about a third of the trials we identified, thus helping to improve the discoverability of these trials.

Conclusions: KoreaMed indexes a significant proportion of the Korean medical literature. By systematically searching the database for trials and including these in the Cochrane Library we are improving access to Korean trials and increasing the likelihood that systematic reviews will consider Korean trials for inclusion.