IDENTIFYING ECONOMIC EVALUATIONS IN MEDLINE AND EMBASE- HOW WELL DO PUBLISHED SEARCH FILTERS PERFORM?
Author(s)
Glanville J1, Kaunelis D2, Mensinkai S21York Health Economics Consortium Ltd, York, North Yorks, United Kingdom, 2Canadian Agency for Drugs and Technologies in Health, Ottawa, ON, Canada
Presentation Documents
OBJECTIVES: Health care decision makers assessing the cost-effectiveness of health care technologies seek evidence from economic evaluations. As well as searching economic evaluation databases such as NHS EED and HEED, researchers often search MEDLINE and EMBASE, using sets of search terms or search filters whose current performance is unclear. We tested the performance of search filters in identifying economic evaluations from MEDLINE and EMBASE. METHODS: A reference standard set of economic evaluations was identified from NHS EED published in 2000, 2003 and 2006. The MEDLINE and EMBASE records corresponding to those evaluations were retrieved. Search filters were identified from the InterTASC Information Specialists’ SubGroup website and from Canadian Agency for Drugs and Technologies in Health (CADTH) staff and reviewers. The sensitivity and precision of search filters in retrieving the reference standard records from MEDLINE and EMBASE were tested. RESULTS: A total of 2070 full economic evaluations were identified from NHS EED. Of these 1955 had matching records in Ovid MEDLINE and 1873 had matching records Ovid EMBASE. 13 MEDLINE and 8 EMBASE filters were identified. 3 filters achieved greater than 0.99 sensitivity in MEDLINE and four in EMBASE. Filters demonstrated low precision. CONCLUSIONS: This research provides new performance data on search filters to identify economic evaluations in MEDLINE and EMBASE and indicates which filters may assist dependent on searchers’ priorities. It remains difficult to identify economic evaluations reliably whilst achieving reasonable levels of precision.
Conference/Value in Health Info
2009-10, ISPOR Europe 2009, Paris, France
Value in Health, Vol. 12, No. 7 (October 2009)
Code
PMC27
Topic
Real World Data & Information Systems
Topic Subcategory
Health & Insurance Records Systems
Disease
Multiple Diseases