Optimal Search Filters for Renal Information in EMBASE

Iansavichus, Arthur V.; Haynes, R. Brian; Shariff, Salimah Z.; Weir, Matthew; Wilczynski, Nancy L.; McKibbon, Ann; Rehman, Faisal; Garg, Amit X.

doi:10.1053/j.ajkd.2009.11.026

« Previous Next »American Journal of Kidney Diseases
Volume 56, Issue 1 , Pages 14-22, July 2010

Optimal Search Filters for Renal Information in EMBASE

Arthur V. Iansavichus, MLIS
- Division of Nephrology, University of Western Ontario, London, Canada
R. Brian Haynes, MD, PhD
- Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Canada
- Department of Medicine, McMaster University, Hamilton, Canada
Salimah Z. Shariff, BMath
- Division of Nephrology, University of Western Ontario, London, Canada
- Department of Epidemiology and Biostatistics, University of Western Ontario, London, Canada
Matthew Weir, MD
- Division of Nephrology, University of Western Ontario, London, Canada
- Department of Epidemiology and Biostatistics, University of Western Ontario, London, Canada
Nancy L. Wilczynski, PhD
- Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Canada
Ann McKibbon, MLS, PhD
- Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Canada
Faisal Rehman, MD
- Division of Nephrology, University of Western Ontario, London, Canada
Amit X. Garg, MD, PhD
- Division of Nephrology, University of Western Ontario, London, Canada
- Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Canada
- Department of Epidemiology and Biostatistics, University of Western Ontario, London, Canada
- Address correspondence to Amit X. Garg, MD, PhD, London Kidney Clinical Research Unit, Rm ELL-101, Westminster, London Health Sciences Centre, 800 Commissioners Rd E, London, Ontario N6A 4G5, Canada

Received 14 August 2009; accepted 27 November 2009. published online 15 March 2010.

Background

EMBASE is a popular database used to retrieve biomedical information. Our objective was to develop and test search filters to help clinicians and researchers efficiently retrieve articles with renal information in EMBASE.

Study Design

We used a diagnostic test assessment framework because filters operate similarly to screening tests.

Settings & Participants

We divided a sample of 5,302 articles from 39 journals into development and validation sets of articles.

Index Test

Information retrieval properties were assessed by treating each search filter as a “diagnostic test” or screening procedure for the detection of relevant articles. We tested the performance of 1,936,799 search filters made of unique renal terms and their combinations.

Reference Standard & Outcome

The reference standard was manual review of each article. We calculated the sensitivity and specificity of each filter to identify articles with renal information.

Results

The best renal filters consisted of multiple search terms, such as “renal replacement therapy,” “renal,” “kidney disease,” and “proteinuria,” and the truncated terms “kidney,” “dialy,” “neph,” “glomerul,” and “hemodial.” These filters achieved peak sensitivities of 98.7% (95% CI, 97.9-99.6) and specificities of 98.5% (95% CI, 98.0-99.0). The retrieval performance of these filters remained excellent in the validation set of independent articles.

Limitations

The retrieval performance of any search will vary depending on the quality of all search concepts used, not just renal terms.

Conclusions

We empirically developed and validated high-performance renal search filters for EMBASE. These filters can be programmed into the search engine or used on their own to improve the efficiency of searching.

Index Words: Kidney diseases, medical informatics, information retrieval, knowledge translation, EMBASE, nephrology