Should You Include Broader Terms as Well as Specific Terms?

Should your search use the umbrella term, the specific terms, or both?

Suppose your research concerns a specific concept, population, intervention, or phenomenon. You know its precise name, but it also sits beneath a broader category. Should that broader category go into the search too?

The reverse problem is equally common. Perhaps your research concerns a broad category containing many specific forms. Is searching the umbrella term enough, or do you need to list the narrower terms individually?

There is no universal rule that broader plus specific is always better. The right choice depends on how the literature is written, how the database indexes concepts, how much sensitivity the search requires, and what your eligibility criteria actually include.

Include broader terms when they improve retrieval without changing the concept

Broader and narrower terms represent different levels of a concept hierarchy

A broader term is not simply another synonym

Synonyms are alternative expressions for substantially the same concept. Broader and narrower terms have a hierarchical relationship instead.

In a thesaurus, a broader term represents a more general concept, while a narrower term represents a more specific concept that falls beneath it. The USGS Thesaurus, for example, distinguishes narrower terms as more specific concepts that are types, parts, or instances of a broader concept and treats related terms as a separate relationship.

This distinction matters because adding a synonym generally provides another linguistic route to the same concept. Adding a broader term can change what qualifies for retrieval.

Broader terms can protect against underspecific descriptions

A relevant paper does not always name the most specific concept in its title or abstract. Authors may describe an intervention, population, technology, or phenomenon using an umbrella category even though the study itself concerns the specific form you care about.

In that situation, searching only the precise term can miss the paper. A broader free-text expression may provide an additional retrieval route.

This is one reason broader terminology can improve sensitivity, often called recall: the proportion of all relevant records that the search successfully retrieves. Empirical work on indexing has likewise found circumstances in which broader terms improve retrieval of relevant documents, although their behavior can differ among databases and indexing systems.

The cost of a broader term is usually more noise

Imagine that your eligible intervention is one specific member of a much larger intervention family. Searching the family name may retrieve your target studies, but it may also retrieve studies of every other member of that family.

The search becomes more sensitive but potentially less precise. Precision refers to the proportion of retrieved records that are relevant. In practical terms, lower precision means more screening.

Specific terms matter when authors name the subtype rather than the category

The opposite problem also occurs. A relevant article may mention only a specific subtype, product, technique, diagnosis, technology, species, or intervention without using the broader category name you chose for the search.

If the database is searching free text, the fact that you know the subtype belongs to the broader category does not guarantee that the database knows it. Unless automatic term mapping or another search feature intervenes, searching one string does not conceptually infer every member of the category.

Specific terms may therefore be needed alongside the broader term when those specific names plausibly appear without the umbrella expression.

Controlled vocabulary changes the calculation

Many database thesauri organize concepts hierarchically. A subject heading may have narrower headings beneath it, and some databases allow the researcher to "explode" the broader heading so that narrower concepts are included automatically.

Guidance on systematic search development explains that subject-heading trees arrange broader concepts above narrower ones and that exploding a heading retrieves the selected heading together with its narrower descendants.

PubMed's MeSH system provides a familiar example. MeSH terms are hierarchically arranged, and PubMed generally includes more specific terms beneath a MeSH heading through automatic explosion unless that behavior is deliberately restricted.

This is a good example of why understanding controlled vocabularies matters before translating a conceptual search plan into database syntax.

Free-text broader terms and broader subject headings are not the same thing

If you type a broad word into a title-and-abstract field, the database searches for that expression according to its free-text rules. If you select a broader controlled-vocabulary heading and explode it, the database may retrieve records indexed with multiple narrower concepts even when the broader phrase itself does not appear in the title or abstract.

These mechanisms should not be conflated.

Do not confuse broader terms with merely related concepts

A broader term contains the target concept hierarchically. A related concept has an associative relationship but is not necessarily above or below it.

That difference matters. Adding an umbrella category because your target genuinely belongs beneath it is conceptually different from adding a neighboring construct because relevant papers sometimes discuss both.

If you are considering the latter, the more appropriate question is whether to include related concepts that are not exact synonyms.

Your eligibility criteria should determine how broad the search may reasonably become

Suppose your review includes every subtype within a broad category. A broader term may represent the eligibility criteria very naturally, supplemented by specific subtype names to protect against free-text variation.

If your review includes only one subtype, however, the umbrella term represents a larger set than the eligible concept. It may still be useful for sensitivity, but its retrieval cost could be substantial.

This is not inherently wrong. Searches for systematic reviews are often deliberately broader than the final eligibility criteria because screening can remove false positives. But the broadening should be purposeful rather than accidental.

Test what the broader term actually contributes

You do not have to settle the issue entirely in the abstract. During search development, compare retrieval with and without the broader expression.

Inspect records found uniquely because of the broader term. Are any relevant? Are they records your specific terms genuinely miss? If the broader expression adds several valuable studies, its inclusion has an empirical justification. If it adds thousands of records without meaningful relevant retrieval, reconsider its value or whether it can be searched more precisely.

The same principle applies to specific terms. A very long list of subtypes is not automatically superior if most contribute nothing and the database's indexing already captures them.

Testing an umbrella term against specific terminology

The decision is therefore not based on a rule that broad terms are good or bad. It is based on what the term contributes to retrieval in the context of your particular question.

Common mistakes with broader and specific terms

Choose the level of specificity deliberately

For each concept block, ask how authors are likely to describe eligible studies. Could they use only the umbrella category? Could they name only a subtype? Does the database's controlled vocabulary already connect those levels?

The best search often uses more than one level of specificity, but each level should have a retrieval purpose.

Before combining broader and specific search terms

Questions about broader and specific search terms

Use broader terms for a reason, not as insurance

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.