Durgical
EN
Open the app

Literature search

  • Concept

An orderly search for what is already known about a question, with controlled keywords, logical operators and several sources.

Updated

What it is

A literature search gathers what has been published, and what has not, on a question before a study is designed or a clinical doubt is answered. A good strategy combines controlled vocabulary (MeSH in English, DeCS in Spanish and Portuguese) with free text, joins the concepts with Boolean operators and searches several databases, grey literature included. It is documented step by step so that someone else can repeat it and get the same result.

Key facts

  • Structure: one block per concept of the question (P, I, C, O), each block with its synonyms joined by OR, and the blocks joined by AND.
  • Controlled vocabulary + free text: the controlled term finds well-indexed records; free text finds recent or poorly indexed ones.
  • Truncation (paediatr*) and phrases in quotation marks refine the search; filters (date, language) can leave out useful studies.

Worked example

  1. Question: severe dengue in children in Colombia. In an international database, the term dengue gives 12 400 records.
  2. Combined with AND Colombia: 1 150 records remain, only those that have both concepts.
  3. AND (child OR paediatrics OR adolescent) is added: 310. OR adds synonyms within a concept; AND requires every concept.
  4. A last-10-years filter leaves 182. In LILACS, with DeCS descriptors in Spanish, 95 appear.
  5. Duplicates between the two databases: 41. Unique records: 182 + 95 − 41 = 236.
  6. Title and abstract screening excludes 198, leaving 236 − 198 = 38 full texts. After reading them, 12 studies are included.
  7. Reading: without LILACS, 95 − 41 = 54 unique regional records would have been missed; the full strategy is reported so it can be repeated.
SourceWhat it holdsVocabulary
MEDLINEInternational biomedical journalsMeSH
LILACS (BVS)Latin American and Caribbean literature, including theses and reportsDeCS
Trial registries (WHO ICTRP)Registered trials, unpublished ones tooFree text
Grey literatureTheses, technical reports, conference abstractsVaries

Markers

  • MeSH
  • DeCS
  • AND
  • OR
  • NOT

In practice

  • A quick clinical question needs only a simple search of summarised sources; a systematic review needs an exhaustive, documented strategy.
  • A librarian or information specialist greatly improves the sensitivity of a search.
  • Searching in Spanish with DeCS helps to find Colombian and Latin American studies.

Common pitfalls

  • Using a single database and English-language articles only.
  • Joining synonyms with AND instead of OR, which loses almost all the records.
  • Not saving the strategy or the search date, so that nobody can repeat it.

Test yourself

Think of the answer, then tap the question to see it.

Which Boolean operator joins synonyms of the same concept?

OR, because it widens the search: it returns records with any of the terms.

What is DeCS and how does it differ from MeSH?

It is the controlled vocabulary of BIREME (BVS), derived from MeSH, with terms in Spanish, Portuguese, English and French.

One database gives 150 records, another 70, and 30 are duplicates. How many unique records are there?

150 + 70 − 30 = 190 unique records.

Why is grey literature searched?

Because studies with negative results are published less often; including theses, reports and registries reduces publication bias.

Add to my cards (4)

Connections

← All of epidemiology

Educational content on epidemiological and statistical methods; the examples use made-up numbers to teach the calculation. Not a clinical or legal guide: current Colombian rules are in the official sources.

Sources: Chapter 4: Searching for and selecting studies (Cochrane, 2025); Medical Subject Headings - Home Page (National Library of Medicine, 2026); DeCS – Descritores em Ciências da Saúde (BIREME/OPS/OMS, 2026).

Spotted a mistake?We are always improving for the community. Tell us what is wrong and our team will review it.Report a mistake

Representative model, for educational purposes only.