Purpose: To develop and validate search filters for MEDLINE via PubMed according to two categories of the NLINKS-EBN matrix.
Methods: The search results of the search filters were compared to a gold standard.
Findings: The usage of nursing classification terms for the literature search in evidence-based nursing (EBN) is still limited because taxonomies are neither widely used in nursing literature nor applied for indexing by MEDLINE. The proposed filters achieved a sensitivity of 96% and a specificity of 94% for "secondary data" and a sensitivity of 87% and a specificity of 73% for "diagnostic tests."
Conclusions: The usage of database-specific search filters are a reliable and valid method to search for nursing classification terms in medical databases.