Skip to Main Content
It looks like you're using Internet Explorer 11 or older. This website works best with modern browsers such as the latest versions of Chrome, Firefox, Safari, and Edge. If you continue with this browser, you may see unexpected results.

Systematic Reviews: Systematic Search Techniques

An introductory guide to Systematic Reviews

Planning Your Search

A search plan would comprise of the following steps:

  • Step 1: The PICO Framework - Identify the concepts of the research question and how those concepts will be combined.
  • Step 2: Searching in Bibliographic Databases: A list of several databases, of which the search strategy would be carefully translated for and then conducted. Cochrane recommends using at least three. In Medicine, these are typically PubMed/MEDLINE, Embase and Cochrane Library. Depending on the research topic, a multidisciplinary database such as Web of Science, or subject-specific databases such as CINAHL, PsycInfo or ERIC can also be included. In developing your search strategy, consider the following:
    • Controlled Vocabulary
    • Keyword & Phrase Searching
    • Truncation, Wildcards & Proximity Operators
    • Boolean Operators
    • Search Limits and Filters

For detailed information on these topics, you can refer to the boxes on the right of this page for more information.

  • Step 3: Grey Literature: A grey literature search should also be included. These could be clinical trials, dissertations and conference proceedings, which require other types of resources. This literature search can also be supplemented with hand searching of journals, suggestions from subject experts or colleagues or a citation analysis - an analysis of which articles have cited an older, still relevant study.

Sensitivity, Specificity and Precision

The aim of a systematic search is to identify all relevant studies on any given topic. Therefore, the search strategy of a systematic review should be rigorously developed such that it is highly sensitive in order to find all these potential relevant articles. According to the Cochrane Handbook, sensitivity is defined as, "the number of relevant reports identified, divided by the total number of relevant reports in existence." A search with high sensitivity should therefore retrieve all relevant studies on a topic. You may likely end up with a large amount of references, out of which a high percentage will be irrelevant.

Preliminary searching

Before you create a search strategy, it is a good idea to see what has already been done i.e. if there are published reviews or on-going reviews on the topic, or if sufficient studies have been conducted on the topic. The methods sections of reviews and any appended search strategies can be useful.

Controlled Vocabulary

Controlled vocabulary is a list of terms under which items in a database are indexed. In many databases, articles are for better retrieval indexed with subject headings or controlled terms from a thesaurus. In PubMed, the controlled vocabulary is MeSH: Medical Subject Headings. Examples of controlled vocabulary include MeSH in Medline, Emtree in Embase, Subject Headings in CINAHL and ERIC, and Subject Terms in PsycINFO.

You need to find subject headings to build a good search strategy. Search for your key concepts in the controlled vocabulary and check what subject headings your key articles are indexed with to find relevant terms. Explore the hierarchies and related terms. Please note that there might be several subject headings for closely related concepts, and that these headings might be part of different hierarchies.

Check if some of the terms are new as subject headings. If so, you might also consider including the previous subject heading used for indexing the concept.

Avoid limits for subject headings, such as subheadings or main concept, for example MeSH Major Topic in Pubmed.

The default setting in most databases is the exploding of subject headings, i.e. narrower terms of the subject headings are included. In general, this functionality is helpful, but you might in some cases consider if a subject heading should be searched without being exploded, i.e. not including narrower terms.

Karolinska Institutet University Library (2022). Systematic reviews. :
Jewell, S. (2017). Assembling the pieces of a systematic review - a guide for librarians. Rowman & Littlefield. and Jewell,

Keyword & Phrase Searching

Keywords searching should be used in conjunction with controlled vocabulary searching for greater sensitivity, This optimises the specificity potential of controlled vocabulary and improves sensitivity by retrieving articles where variant terms were used. It ensures that articles not tagged with subject headings, or indexed with other subject headings are also retrieved.

Quotation marks are useful for keeping words in phrases together also known as phrase searching. Use quotation marks (" ") around the phrase that you need to find in order to return records containing the exact phrase enclosed.

For systematic searching, care must be taken to ensure that a more specific search does not miss out relevant literature e.g. searching "[typeA] behavior" will miss out articles using "[typeA] behaviors" as well as "[typeA] behaviour".

Phrase searching may work differently for particular databases. Always check the database help file for how its phrase searching works.

Truncation, Wildcards and Proximity Operators

Truncation retrieves records with spelling variants containing a common root word. The symbol for truncation is usually an asterisk (*).

Apply truncation with care so as to retrieve optimal relevant results, e.g. app* can retrieve apps, application, apple, apply etc. 

Wildcard symbols $ or # or ? replace single character in a wildcard search, e.g. wom?n can retrieve woman, women, etc.

Proximity operator searches allow you to find terms that are within a set number of words of each other, e.g. oxygen NEAR/3 treatment in Web of Science.

 Always check the specific database help file for how its truncation, wildcards and proximity operators work.

Boolean Operators

Boolean operators allow setting of relationships between keywords or concepts. The most commonly used Boolean commands are ANDOR, and NOT. Parentheses can be used to define the logical relationship further. Using boolean operators can make your searches more precise and efficient.

In general use the 'OR' operator to join synonyms or related concepts i.e. finds any of the terms, and use the 'AND' operator to combine different concept blocks i.e. find all of the terms. Using the NOT Operator removes concepts so consider if necessary or use with care to avoid removing relevant results.

Always check the specific database help file for examples of their boolean operator search syntax.

Search Limits and Filters

Search filters may be customised (validated search filters) or built-in to particular databases as features. Exercise caution when using filters to avoid removing relevant results.

Further reading:

ISSG Search Filters Resource

NUS Libraries Libguide Search Filters