Email updates

Keep up to date with the latest news and content from BMC Medical Informatics and Decision Making and BioMed Central.

Open Access Highly Accessed Study protocol

An overview of the design and methods for retrieving high-quality studies for clinical care

Nancy L Wilczynski1, Douglas Morgan1, R Brian Haynes123* and the Hedges Team1

Author affiliations

1 Health Information Research Unit, Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada

2 Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada

3 Michael G. DeGroote School of Medicine, McMaster University, Hamilton, Ontario, Canada

For all author emails, please log on.

Citation and License

BMC Medical Informatics and Decision Making 2005, 5:20  doi:10.1186/1472-6947-5-20

Published: 21 June 2005

Abstract

Background

With the information explosion, the retrieval of the best clinical evidence from large, general purpose, bibliographic databases such as MEDLINE can be difficult. Both researchers conducting systematic reviews and clinicians faced with a patient care question are confronted with the daunting task of searching for the best medical literature in electronic databases. Many have advocated the use of search filters or "hedges" to assist with the searching process. The purpose of this report is to describe the design and methods of a study that set out to develop optimal search strategies for retrieving sound clinical studies of health disorders in large electronics databases.

Objective

To describe the design and methods of a study that set out to develop optimal search strategies for retrieving sound clinical studies of health disorders in large electronic databases.

Design

An analytic survey comparing hand searches of 170 journals in the year 2000 with retrievals from MEDLINE, EMBASE, CINAHL, and PsycINFO for candidate search terms and combinations. The sensitivity, specificity, precision, and accuracy of unique search terms and combinations of search terms were calculated.

Conclusion

A study design modeled after a diagnostic testing procedure with a gold standard (the hand search of the literature) and a test (the search terms) is an effective way of developing, testing, and validating search strategies for use in large electronic databases.