MeSH introduction (By Terry Harrison)Main points
  • Many databases, such as Medline, use a thesaurus to enable more effective searching. A thesaurus is a controlled vocabulary that is used to index information from journals. Instead of allowing numerous variations in words and phrases it seeks to 'control' the vocabulary by grouping related concepts under a single preferred term. A thesaurus will typically contain keywords, definitions of those keywords and cross-references between keywords.
  • There are several reasons for using the thesaurus of a database. As a thesaurus is a controlled vocabulary, all indexers use the same standard terms to describe a subject area, regardless of the term the author has chosen to use. So, for example, all items relating to heart attacks are indexed as "myocardial infarction", whether or not the author has actually used the term "myocardial infarction". Each of the MeSH (Medical Subject Headings) terms or keywords represents a single concept appearing in the medical literature. As important new concepts appear, a new MeSH keyword is created. When a new reference is added to Medline, indexers choose and add the appropriate MeSH keywords (usually 10 to 20) to represent the contents of the article. If you look at a Medline reference, you will notice the MeSH keywords towards the end.
  • To complicate matters, different databases have different thesauri (i.e., an equivalent of MeSH). CINAHL, EMBASE and PsycInfo each have their own thesauri - consequently, you cannot merely search MeSH within Medline and expect to apply the same MeSH terms for other databases!
  • Where a database does not have a controlled vocabulary (thesaurus) you are free, of course, to search using any terms (MeSH or non-MeSH).
  • Note, too, that when using a controlled vocabulary facility you often have the option to 'explode' the search so as to retrieves all documents relating to distal subject terms.
  • In addition, you sometimes have the option to Focus your search - i.e., choosing citations that have the subject as the major component (see below). There are also subheadings (different to each MeSH term) to choose from, if needed.
  • If you select both Explode and Major Concept, you retrieve all references indexed to your term (and its distal terms) as well as all articles for which the subject heading is a major point of the article.
For more information on MeSH, see: and to access MeSH directly go to MeSH treeThe keywords that make up MeSH are arranged into hierarchical structures called trees, starting with broad terms which branch off into increasingly narrower or more specific terms. Indexers are instructed to always use the most specific term when assigning terms to an article. Thus, an article covering obesity in diabetics will be indexed using the term OBESITY IN DIABETES and not using the broader term DIABETES MELLITUS. An article indexed as DIABETES MELLITUS will usually discuss diabetes in general without going into very much detail on specific aspects. The tree structure is helpful when searching as it helps you to identify broader and narrower terms.What if you have no MeSH keyword?Although keyword searching using MeSH terms is a more precise way of searching, this won't always work for you. There may be times when there is no MeSH keyword for the subject you are searching; for example, until 1997, there was no MeSH keyword for "evidence based medicine". In the absence of MeSH keywords, you need to search free text. Free text – or ‘words anywhere’ - searching is a method of searching by using words and phrases from the title, abstract and keywords of references. Free text searching is also useful for broadening your search. Also, some databases do not map against a controlled thesaurus (MeSH or otherwise) and so with these you have no choice but to apply keyword searching. Finally, you can always opt for a combination of MeSH (or subject) searching with free text searching if your search term is obscure and you want to get as many results as possible.More Search Tips
  • You will generally find that long search strings can be non-productive. If one element is wrong, the whole search becomes invalid. By taking a step-by-step approach, you are able to see what a particular database has got and what it hasn't, and so refine your search strategy accordingly.
  • If a database has an Advanced search facility, use it (provides more options)
  • Choose subheadings only as needed (and if available). If results are too many, consider applying selective subheadings.
  Terence Harrison