PICO, PICOS and SPIDER: a comparison study of specificity and sensitivity in three search tools for qualitative systematic reviews.

TitlePICO, PICOS and SPIDER: a comparison study of specificity and sensitivity in three search tools for qualitative systematic reviews.
Publication TypeJournal Article
Year of Publication2014
AuthorsMethley AM, Campbell S, Chew-Graham C, McNally R, Cheraghi-Sohi S
JournalBMC health services research
Volume14
Issue1
Pagination579
Date Published2014 Nov 21
ISSN1472-6963
AbstractBackgroundQualitative systematic reviews are increasing in popularity in evidence based health care. Difficulties have been reported in conducting literature searches of qualitative research using the PICO search tool. An alternative search tool, entitled SPIDER, was recently developed for more effective searching of qualitative research, but remained untested beyond its development team.MethodsIn this article we tested the `SPIDER¿ search tool in a systematic narrative review of qualitative literature investigating the health care experiences of people with Multiple Sclerosis. Identical search terms were combined into the PICO or SPIDER search tool and compared across Ovid MEDLINE, Ovid EMBASE and EBSCO CINAHL Plus databases. In addition, we added to this method by comparing initial SPIDER and PICO tools to a modified version of PICO with added qualitative search terms (PICOS).ResultsResults showed a greater number of hits from the PICO searches, in comparison to the SPIDER searches, with greater sensitivity. SPIDER searches showed greatest specificity for every database. The modified PICO demonstrated equal or higher sensitivity than SPIDER searches, and equal or lower specificity than SPIDER searches. The modified PICO demonstrated lower sensitivity and greater specificity than PICO searches.ConclusionsThe recommendations for practice are therefore to use the PICO tool for a fully comprehensive search but the PICOS tool where time and resources are limited. Based on these limited findings the SPIDER tool would not be recommended due to the risk of not identifying relevant papers, but has potential due to its greater specificity.
DOI10.1186/s12913-014-0579-0
Alternate JournalBMC Health Serv Res