OBJECTIVES: Embase is one of the most commonly searched bibliographic databases when undertaking systematic reviews of healthcare interventions. Using a combination of text searching and index searching is accepted best practice when designing Embase search strategies. As the number of Emtree index terms assigned to each Embase record has increased over time, there has been a corresponding increase in overall search result numbers and, therefore, the size of the workload for reviewers. The objective of this research was to investigate whether focusing Emtree indexing terms using the ‘restrict to focus’ function, so that only those articles where the index term is key to that article are retrieved, could reduce the number of records for screening without loss of included studies.

METHODS: Embase searches conducted in three selected clinical reviews undertaken by the review team were retrospectively compared with searches in which the Emtree terms were focused. The records retrieved by the focused Emtree searches were checked to see what proportion of included studies identified by the original unfocused search strategy were retrieved. The subjects of reviews investigated were treatments for invasive fungal infections, type 2 diabetes mellitus, and acute lymphoblastic leukaemia.

RESULTS: The data collected was analysed to identify total results with and without focusing Emtree terms, and to calculate the yield of included records. The focused searches retrieved 100% of the included papers in each review but results retrieved to screen were reduced by up to 30%.

CONCLUSIONS: Focusing Emtree terms can potentially reduce screening burden in clinical systematic reviews without lowering retrieval of relevant records. Reducing the number of results in Embase searches without a loss of sensitivity could improve efficiency by reducing time spent and costs when undertaking systematic reviews.