Skip to Main Content

The Library: Psychological Therapies

What is a systematic review?

Systematic reviews aim to collect and summarise the evidence on a particular topic in order to answer a specific research question. The evidence is collected using a comprehensive literature search. A systematic review is carried out in a methodical way, using a search strategy in multiple databases. The literature search should be documented in detail so that anyone reading your systematic review could replicate your results exactly.

This page gives general guidance on how to carry out each stage of the literature search for your systematic review, but please check your module handbook or assignment guidance in case your lecturer has given you more specific instructions.

Before reading this guide, make sure that you have actually been asked to do a systematic review, rather than a literature review which takes a systematic approach. Speak to your lecturer if you are not sure.

For help with any stage of the literature search for your systematic review, please contact your Academic Librarian.

Literature searching for your systematic review

Define your topic

Every systematic review starts with a question. The goal of a systematic review is to find as much evidence as possible in order to answer the question.

Your question needs to be:

  • Specific – what exactly do you want to find out? If your question is too vague, it will be difficult to answer.
  • On a topic where there is enough existing research. If your topic is too narrow or too recent, there may not be enough evidence available for a systematic review.
  • Something you can realistically answer in this project. If your topic is too broad, it is hard to know when to stop searching and you may get an unmanageable number of results.
  • Focused on research published in a specific time period – e.g. within the last 5-10 years. Some topics may go back further if appropriate.

To help you choose a topic, start by doing some simple searches to get an idea of how much research is available and whether you need to narrow down or expand your question.

Models such as PICO are often used to help map out your research topic. PICO can be a useful framework, but don't worry too much if your question does not exactly fit the PICO model.

Develop a search strategy

Once you have decided on a topic for your systematic review, you will need to develop a search strategy. This involves:

  • Breaking down your topic into the keywords and phrases that you will search for
  • Identifying any relevant synonyms (words that mean the same thing) for your keywords. For example, if one of your keywords is nutrition, you may want to include similar words such as dietary or food intake.
  • Thinking about advanced search techniques you could use, including phrase searching, truncation, or Boolean searching.

For more information on developing a search strategy, see our guide to finding information.

Below you will find an example search log document to show how you could plan and keep track of your search. The PDF version has been partially filled in to show what the completed search log would look like. The Word version is a blank copy which you can download, edit and customise.

Where to search

Systematic reviews are usually carried out in academic databases. You can find a list of key databases on your subject page, or a list of all library databases in our Databases A-Z.

You can test out your search strategy in different databases to see how effective the search is and whether the database contains relevant information for your topic. Bear in mind that each database works slightly differently, so you may need to modify your search for each database.

You should search multiple databases (usually at least 3) for your systematic review in order to collect a wider variety of information.

Citation searching is often used in systematic reviews. Citation searching means following up references or citations. You can do this by looking at the reference lists of articles you have found, by using a citation based database such as Scopus, or in Google Scholar

ACCORDION: Converted tabs - this tab opens by default
ACCORDION OPTIONS: enable collapse / start fully collapsed

Managing your results

Doing a systematic review means dealing with a large number of search results. It’s important to have a system for organising your results. This means that all of your references should be stored in one place, along with information on which results came from which database. You will need this information for documenting your search (more on this below) and for referencing when you write up your systematic review.

You may wish to use a reference manager such as Zotero or EndNote. The advantage of using a reference manager is that they usually have a plug-in for MS Word which allows you to insert saved references and format them in the correct referencing style for your subject.

Documenting your search

The literature search for your systematic review needs to be documented in detail, so that anyone could replicate your exact search using the details you have provided. Remember to keep track of:

  • Which databases you used
  • The exact search strategy you used in each database
  • Any filters or limits you put on the search e.g. publication dates
  • The number of search results you found in each database
  • Any additional search results you found without using a database, e.g. grey literature or articles you found by citation searching.

One way to document your search process (as well as the rest of your systematic review) is by using the PRISMA flowchart. You can use this template flowchart to record exactly how you carried out your systematic review, and include the flowchart as an appendix.

 

Narrowing down your results

In a systematic review you will need to be very selective about which sources are included in your final results set. Narrowing down your results is an important part of the process. This is often done using inclusion criteria: a list of requirements your results must meet in order to be included in the final review.

Some examples of typical inclusion criteria are:

  • Research must focus on a specific population (e.g. women, footballers, diabetics)
  • Research must have been carried out in a specific country or area
  • Research must have been published within a specific time frame

Sometimes exclusion criteria are also used to rule out results, e.g. if the research was not carried out in the specified time frame it would be excluded from the results.

There are three main ways to apply your inclusion/exclusion criteria:

  1. Using filters and limits in the database search (e.g. date limits)
  2. Using keywords within the search (e.g. including keywords such as diabetic if you want to focus on people with diabetes)
  3. Sorting through your results and filtering out items which do not meet your criteria. This can be a lengthy process. Remember to document how many results from each database you have included or excluded (e.g. using the PRISMA flowchart).

Critical appraisal

Once you have a set of results which meet your inclusion criteria, the next step is critical appraisal. This means carefully examining the quality of the studies you have chosen and identifying any potential issues with the study design or reporting of results.

There are many critical appraisal tools available to help you do this, such as CASP checklists. A completed CASP checklist for each article could be included in your appendices.

Only the articles which meet your critical appraisal standards should be included in the final review. Again, remember to document how many articles were ruled out due to critical appraisal.

Synthesising the evidence

By this stage you should have a final set of articles which meet your inclusion criteria and are of a high standard according to your critical appraisal. The next step is to read each article carefully and consider the findings and what they mean for your research question. When writing up the findings of your systematic review, think about:

  • How much evidence is available and is it sufficient to answer your research question? 
  • Are there any gaps in the literature where further research might be needed?
  • What are the strengths and weaknesses of the studies you have included?
  • Overall, what conclusions can you draw from the evidence you have found in order to answer your research question?

Check your module handbook or assignment brief for any specific guidance from your lecturer on how to write up your systematic review. You could also read some existing systematic reviews to get an idea of the typical structure and layout. An example of a published systematic review is linked below.