Searching

Filters (also called hedges) are used to narrow a search and/or focus in on a specific publication type, level of evidence, and/or subject area. Filters are constructed using specific database syntax (rules and commands). Strategies are developed with rules and commands used to search a variety of databases and interfaces including PubMed, Ovid, Embase, and CINAHL. Some filters are adjusted to be more sensitive (accurate identification) or specific (accurate exclusion).

Several research and healthcare groups produce pre-tested strategies that identify high quality evidence or focus in on a particular health care field or area. Using pre-existing filters or strategies can assist in managing the vast amounts of literature indexed in the major medical databases, and provide you with a smaller number of articles to start with when searching.

When using pre-existing filters be mindful of:

• Authority of filter or strategy creator  
• Specificity versus sensitivity of results
• Syntax of filter and what database it was developed for.

While it is important to be aware of the authority and accuracy of a specific filter, using a filter that has not been validated or tested is not incorrect. It is important to recognize and identify the impact that using a particular filter may have on your search results. If a particular filter works well in a given context and does not appear to be producing a high number of irrelevant results or excluding relevant items, by all means use that filter.

Several trusted groups have developed filters to narrow search results by methodology, publication type or question category (therapy, diagnosis, etiology, prognosis, clinical prediction guides).

Australia's CareSearch Palliative Care Knowledge Network has developed several filters to focus search results on subjects relevant to palliative care. All filters in PubMed syntax.

Several trusted groups have developed filters to focus search results on particular conditions.