Holland abstract

ISSG structured abstract for:

Holland JL, Wilczynski NL, Haynes RB. Optimal search strategies for identifying sound clinical prediction studies in EMBASE. BMC Medical Informatics and Decision Making 2005;5:11.

Structured abstract prepared by: Anne Fry-Smith

Objective:

The filters are designed to identify methodologically sound studies of clinical prediction guides (CPGs) in EMBASE .

Methods:

The authors identified a gold standard of 163 records of which 69 were methodologically sound, by hand searching 55 journals in 2000. The search terms for the filters were collected from relevant records and from consulting widely with experts in the field. The search strategies were developed using the entire gold standard database.

Results:

Several filters are offered. The most sensitive scored 97.1% and the most precise scored 10.8%. The most specific scored 98.8% and the best compromise between sensitivity and specificity scored 91.3% sensitivity and 90.2% specificity (2.3% precision).

Discussion:

Similarities with Wong et al (2003) for CPGs for MEDLINE were discussed with regard to low precision and the best identified search strategies. Precision might be improved by the application of AND and AND/NOT combinations or multivariable statistical techniques, but this remains to be determined. Additional criteria could have been incorporated into the derivation of search strategies but further research would be needed to determine the performance of these search strategies.

ISSG commentary:

No commentary provided.