Skip to main content

Languages

Beyond titles and abstracts; systematic full text search for identifying publications addressing shared decision making

Date and Location

Session: 

P2.052

Date

Saturday 21 September 2013 - 10:30 - 12:00

Location

Presenting author and contact person

Presenting author

Xavier Blanc

Contact person

Xavier Blanc
Abstract text
Background: Full text search permits to identify articles whose keywords appear not only in the title and abstract, but also in the discussion. Those articles, such as editorials and debates, have an impact on readers and serve the emergence of new concepts, such as shared decision making (SDM). By retrieving them, full text search may be a great tool for medical research. Objectives: To assess the validity and feasibility of full text search in major medical journals for identifying publications addressing SDM and to compare the results with a traditional PubMed search. Methods: Databases on websites of 15 high impact medical journals were searched with full text function, defined as website full text search. Publications of any type from 1996 to 2011 containing the phrase “shared decision making” or five other variants were included. The search performance was compared with a PubMed search using similar strategy. As a validation dataset, a locally stored full text corpus was made with the authorization of 6/15 journals to collect their published materials and was searched with an automated script. This was defined as downloaded full text search. Results: The website full text search identified 1285 SDM publications in 15 journals. This was ten times more than through a PubMed search of abstracts only (119 publications). When limited to the 6 collaborating journals, 614 publications were found, while the downloaded full text search retrieved 613 publications. The matching rate was 89.4% between both full text searches. Conclusions: Full text search on medical journals websites was valid and feasible. It permitted a thorough retrieval of SDM publications. However, an easy access to a larger corpus of full text publications is needed to make the most of full text search, and more generally of text mining. This could be fostered by a greater collaboration with publishers.