Abstract

This article discusses techniques that are appropriate when developing search strategies for systematic reviews of medical tests. This includes general advice for searching for systematic reviews and issues specific to systematic reviews of medical tests. Diagnostic search filters are currently not sufficiently developed for use when searching for systematic reviews. Instead, authors should construct a highly sensitive search strategy that uses both controlled vocabulary and text words. A comprehensive search should include multiple databases and sources of grey literature. A list of subject-specific databases is included in this article.
Hide

Abstract

Key Points • A librarian or other expert searcher should be involved in the development of the search • Sources of grey literature including regulatory data, clinical trial registries and conference abstracts should be searched in addition to bibliographic databases. • Requests should be made to industry to request additional sources of unpublished data. • For the main published literature search, more than one bibliographic database needs to be searched. • Searches should be carefully documented and fully reported. Introduction While, this article both describes and advises on the process of literature searching in support of comparative effectiveness reviews (CERs) for the Effective Health Care Program, it does not address searching for previously published systematic reviews, which is discussed in other articles in this series. Searches to support systematic reviews often require judgment calls about where to search, how to balance recall and precision, and when the point of diminishing returns has been reached. Searchers with more experience with complex search strategies are better equipped to make these decisions. A number of reviews of the quality of systematic reviews suggest that those reviews that employed a librarian or other professional searcher had better reporting of and more complex search strategies. Table 1 describes the various search activities discussed in this paper and identifies who is responsible for performing each of these tasks. As is evident from the table, the EPC conducting the review is responsible for most of these activities. Because the EPC is involved in the development of the Key Questions, is familiar with the literature, and consults with experts regarding studies relevant to the topic, the EPC is in the best position to develop the required search strategies. Because grey literature can provide primary documents to verify published results, EPCs should routinely search regulatory data, clinical trial registries, and conference papers and abstracts for all CERs. This has been a centralized activity conducted by the Scientific Resource Center (SRC), but is now an activity conducted by the EPCs. However, one aspect of the search strategy benefits from centralization. Centralizing the request to drug and device manufacturers for data on their products—what we call the Scientific Information Packet (SIP)—ensures that all requests to industry are conducted in the same manner; this also minimizes or eliminates contact between manufacturers and the EPC involved in writing the report.
Hide

Abstract

This chapter addresses the identification, screening, data collection, and appraisal of the individual studies that make up a systematic review's (SR's) body of evidence. The committee recommends six related standards. The search should be comprehensive and include both published and unpublished research. The potential for bias to enter the selection process is significant and well documented. Without appropriate measures to counter the biased reporting of primary evidence from clinical trials and observational studies, SRs will reflect and possibly exacerbate existing distortions in the biomedical literature. The review team should document the search process and keep track of the decisions that are made for each article. Quality assurance and control are critical during data collection and extraction because of the substantial potential for errors. At least two review team members, working independently, should screen and select studies and extract quantitative and other critical data from included studies. Each eligible study should be systematically appraised for risk of bias; relevance to the study's populations, interventions, and outcomes measures; and fidelity of the implementation of the interventions.
Hide