Adapting the National Institute for Health and Care Excellence (NICE) Organization for Economic Co-operation and Development (OECD) Countries’ Geographic Search Filters (MEDLINE and Embase, Ovid): Developing Low-and Middle-income Countries’ (LMIC) Filters
Objectives: The National Institute for Health and Care Excellence (NICE) Organisation for Economic Co-operation and Development (OECD) countries’ filters retrieve evidence for OECD country-focused research topics from MEDLINE and Embase (Ovid). The innovative filters work by excluding database records with only non-OECD country subject headings using the NOT Boolean operator. The filters are also applied to literature search strategies with the NOT operator.
This presentation will discuss how the NICE OECD countries' filters can be adapted for other country groups, using the creation of novel filters for low-and middle-income countries (LMICs) as an example. They were developed to improve searching effectiveness for LMIC-focused research topics. Validated LMIC filters were not previously available.
Methods: The LMIC filters were developed by rearranging and amending the OECD countries filters’ subject headings to find evidence about LMICs by excluding database records with only headings for World Bank high-income countries.
LMIC references from 25 reviews generated Gold Standard (GS) sets for MEDLINE (n = 421) and Embase (n = 415). The filters were validated by calculating their recall against these sets.
Results: The MEDLINE LMIC filter achieved 99.8% recall against the MEDLINE GS set and the Embase LMIC filter achieved 98.5% recall against the Embase GS set.
Conclusions: The novel LMIC geographic search filters find evidence about the countries effectively. The filters also demonstrate that the NICE OECD countries’ filters can be adapted to find evidence about other country groups successfully.
Validated geographic search filters for the following regions have been previously developed and published: Africa, Spain, the United Kingdom, and OECD countries. A validated filter for the United States is currently in development. Examples of how to adapt and validate the NICE OECD countries’ filters for additional country groups (such as continents, or multiple nations with a geographic, political, or economic relationship) will be provided in the presentation. It is hoped that this presentation will lead to the development of more validated geographic filters for context-sensitive research topics.