Testing search strategies for systematic reviews in the Medline literature database through PubMed

Abstract

BackgroundA high-quality electronic search is essential in ensuring accuracy and completeness in retrieved records for the conducting of a systematic review.ObjectiveWe analysed the available sample of search strategies to identify the best method for searching in Medline through PubMed, considering the use or not of parenthesis, double quotation marks, truncation and use of a simple search or search history.MethodsIn our cross-sectional study of search strategies, we selected and analysed the available searches performed during evidence-based medicine classes and in systematic reviews conducted in the Botucatu Medical School, UNESP, Brazil.ResultsWe analysed 120 search strategies. With regard to the use of phrase searches with parenthesis, there was no difference between the results with and without parenthesis and simple searches or search history tools in 100% of the sample analysed (P = 1.0). The number of results retrieved by the searches analysed was smaller using double quotations marks and using truncation compared with the standard strategy (P = 0.04 and P = 0.08, respectively).ConclusionsThere is no need to use phrase-searching parenthesis to retrieve studies; however, we recommend the use of double quotation marks when an investigator attempts to retrieve articles in which a term appears to be exactly the same as what was proposed in the search form. Furthermore, we do not recommend the use of truncation in search strategies in the Medline via PubMed. Although the results of simple searches or search history tools were the same, we recommend using the latter.