We have all been there. You type a perfectly reasonable
search into one of the medical databases, hit enter, and find yourself
overwhelmed by thousands of irrelevant database results — or frustrated that a
search returns almost nothing. Something is going wrong — but what?
The answer is that you're speaking a different language from
the database. Medical databases aren't like Google. They aren't built to
interpret natural language and guess what you mean. They're built around a
precise, controlled vocabulary — and unless you know how to use it, you can
leave the best evidence buried.
This post will walk you through subject headings:
what they are, why they exist, how to find the right ones, and how to use them
to make your searches more comprehensive.
What Are Subject Headings?
Every article indexed in a medical database is read by a
trained indexer — a human expert who assigns standardised labels to describe
what that article is about. These labels are called subject headings,
and they come from a fixed, carefully maintained list of approved terms.
Different databases use different systems:
Think of it like a library catalogue. If every librarian
labelled books differently — one calling it "heart attack," another
"myocardial infarction," another "MI" — finding everything
on the topic would be more difficult. Subject headings solve this by insisting
that every article about a heart attack gets the same tag, no matter what term
the authors use.
All those terms — heart attack, MI, coronary thrombosis, cardiac infarction —
feed into a single subject heading. Search the heading, and you capture them
all.
Subject Headings vs. Keywords: What's the Difference?
A keyword search asks: "does this word appear
somewhere in the article?"
A subject heading search asks: "was this article tagged as being about
this concept?" These are very different questions.
Keywords (sometimes called free-text or text
words) search for the word you type — in the title, abstract, or sometimes the
full text. They are flexible and can pick up new terminology, but if you miss a
synonym, you miss the evidence.
Subject headings search the controlled
vocabulary tags assigned by the indexer. They're consistent, precise, and
enormously powerful — but they require you to know the right heading to use.
***TOP TIP*** For a thorough search — especially a
systematic review or comprehensive literature review — always use both subject
headings and keywords together. Subject headings give you comprehensiveness;
keywords catch what the subject headings miss. Combining both in a single
search is the gold standard approach recommended and practiced by Information Specialists.
How to Find the Right Subject Heading
The good news: you don't have to memorise every single
subject heading. Every database has a built-in thesaurus tool to help you find
the right heading for your concept.
Step 1: Use the thesaurus or index tool
In Ovid MEDLINE or Ovid Embase, there's a dedicated Map Term
to Subject Heading feature, or you can use the Term Finder tool.
In CINAHL via EBSCOhost, use the CINAHL Headings browser. In PubMed, go to the MeSH Database (found under "Explore" in the menu). Type your concept in plain language and let the system suggest headings.
Step 2: Read the scope note
Every subject heading comes with a scope note — a brief
definition explaining exactly what it covers and, crucially, what it does not cover.
Always read this. It tells you whether a heading matches your concept or
whether you need a different one (or several). It is also a good place to find
alternative keywords
Step 3: Check the entry terms
Entry terms (sometimes called "see also" terms) are all the synonyms
and variant terms that map onto this heading. If you can see your keyword in
the entry terms list, you know you've found the right heading.
Step 4: Look at the tree structure
Subject headings exist within a hierarchy — a branching tree
from the very broad down to the very specific. Viewing the tree shows you what
sits above your heading (broader concepts) and below it (more specific ones).
This is invaluable for deciding how wide or narrow you want your search to be.
Step 5: Test with a known article
If you have a highly relevant article already, look at its
subject headings in the database record. This is an excellent way to verify you
have the right heading and often reveals additional headings you hadn't
considered.
Broadening and Narrowing Your Search
One of the most powerful features of subject headings is the
ability to control the scope of your search with surgical precision. You can
deliberately cast a wider or narrower net depending on what your research
question demands.
Exploding a heading — to broaden
Most databases allow you to "explode" a subject
heading, which means your search automatically includes that heading and
all the narrower terms beneath it in the hierarchy. This is enormously
useful when you want to be comprehensive.
By exploding "Antidepressant Agent," a single
subject heading retrieves articles about all antidepressants — including
specific drugs — without you needing to list them individually. This is far
more reliable than trying to think of every synonym yourself.
Focusing a heading — to narrow further
In some databases, particularly those using the Ovid interface, you can "focus" a subject heading (often marked with an asterisk, e.g. *Hypertension). This restricts results to articles where that heading is considered a major topic — meaning the article is primarily about that concept, rather than merely mentioning it in passing. This is useful when precision matters more than comprehensiveness.
Subheadings (qualifiers) — to narrow
Subject headings can be refined using subheadings (also
called qualifiers) — these are standard modifiers that specify a particular
aspect of the topic. When you select a subject heading, the database will
typically offer you a menu of relevant subheadings to apply.
Common subheadings include:
- /drug
therapy - treatment with medications, e.g. Hypertension/drug therapy retrieves
only articles about treating hypertension with drugs
- /diagnosis
- diagnostic aspects, e.g. Depression/diagnosis focuses on
identifying and diagnosing depression
- /surgery
- surgical treatment, e.g. Breast Neoplasms/surgery limits to
surgical management of breast cancer
- /prevention
& control – preventive measures
- /epidemiology
- incidence, prevalence, distribution
- /adverse
effects - unwanted effects of an intervention
Applying a subheading dramatically reduces noise while retaining highly relevant results.
The tree structure works in both directions. If your search
is returning too little, move up the tree to a broader heading
and explode it. If you're drowning in results, move down the
tree to a more specific heading, or add a subheading to constrain the aspect
you care about.
Differences Between Databases
It's tempting, once you've learned MeSH, to assume the same
headings will work in Embase or CINAHL. They won't — not exactly. Each database
has its own vocabulary, and the headings, though often similar in concept,
differ in terminology, hierarchy, and scope.
Embase's Emtree tends to have more granular terms for drugs and pharmacological interventions — useful if your search involves specific medications. CINAHL Headings include terms specific to nursing practice, patient care, and allied health that don't always exist in MeSH. If you are using multiple databases, you will need to translate your strategy into each database's own vocabulary. This can take some time but is a mark of a rigorous, comprehensive search.
Where to get help
if you need help contact the Clinical Librarians for help – we know the quirks of each thesaurus and can translate strategies reliably between databases.
Email: mtw-tr.clinical.librarians@nhs.net
