Background: Identifying public health studies can be challenging. Interventions are often complex and involve a range of disciplines, with varied settings, communities and populations. There is a lack of uniform terminology and many relevant research designs and publication routes. The combination of these elements had led reviewers to adopt broad search strategies covering a huge range of information resources and yielding a very large number of search results. Earlier work (Booth 2012) has shown that public health reviews do not conform to the adage that Medline consistently delivers 80% or more of the included studies. So how much of a contribution does Medline make across the breath of public health topics and what is the best advice for a pragmatic but efficient search strategy?
Methods: For 15 Public Health reviews (nine Cochrane reviews and six reviews for the UK National Institute of Health & Care Excellence (NICE)), an exploration was carried out as to whether each included study was indexed in Medline. If not, further analysis was undertaken to see if the study was included in other databases searched, or may have required supplementary search methods such as reference list scanning, web site searching, citation tracking and expert contact.
Results: Early indications are that a search of Medline and the use of supplementary search techniques are crucial minimal requirements. Other generalist and topic specific databases are often also required but there are strong indications that a pragmatic approach to database choice can be made so long as other search strategies and methods are deployed appropriately.
This research is currently being extended to a wider range of public health systematic reviews to confirm or refute these findings and to explore if there are any differences across the topic spectrum. The results of this extended analysis and guidance for information specialists will be presented.
