A systematic literature review (SLR) answers a focused question by finding, judging and combining all the relevant studies, using methods that someone else could repeat. Done well, it is one of the most useful and most cited kinds of paper, and it makes an excellent foundation for a thesis.
Systematic, narrative or scoping?
- A narrative review summarises the literature the author chooses to discuss. It can be insightful, but readers cannot tell what was left out
- A systematic review follows a pre-set plan: a defined question, a documented search, clear inclusion rules and a transparent selection process. That transparency is what reviewers look for
- A scoping review maps what research exists on a broad topic, without judging the quality of each study in depth. It is a good choice when a field is new or scattered
Choose the type that fits your question, and name it correctly in your title. Calling a narrative review "systematic" is one of the quickest ways to a rejection.
The main steps
- Frame the question. Frameworks help you make it precise. PICO (Population, Intervention, Comparison, Outcome) suits health and intervention questions; SPIDER or PEO (Population, Exposure, Outcome) suit qualitative and social-science questions.
- Write a protocol. Set out your question, search strategy, criteria and methods before you start. Health-related reviews are often registered on PROSPERO; reviews in other fields can be registered on the Open Science Framework (OSF). Registration shows that you did not change your plan after seeing the results.
- Search several databases. For example Scopus, Web of Science, PubMed and Embase, plus subject databases for your field. Record the exact search string for each database and the date you ran it.
- Screen the results. Remove duplicates, then screen titles and abstracts, then full texts, ideally with two reviewers working independently. Record the reason for every full-text exclusion.
- Extract the data. Use a standard form for each included study, and test the form on a few studies first.
- Assess quality or risk of bias. Use a tool suited to the study designs, such as RoB 2 for randomised trials, ROBINS-I for non-randomised studies, or the CASP and JBI checklists for qualitative and other designs.
- Synthesise. Combine the findings narratively or, where studies are similar enough, statistically in a meta-analysis. Organise the synthesis by theme or outcome, not study by study.
- Report. Follow PRISMA 2020.
What PRISMA 2020 is
PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) is a reporting guideline. It does not tell you how to do the review; it tells you what to report so that readers can judge it. The 2020 update includes:
- a 27-item checklist covering the title, abstract, introduction, methods, results and discussion
- a flow diagram showing how many records were found, screened, excluded (with reasons) and finally included
- a separate checklist for the abstract
Many journals ask you to submit the completed checklist with your manuscript, with page numbers for each item. There are also extensions for particular kinds of review, including PRISMA-P for protocols, PRISMA-ScR for scoping reviews and PRISMA-S for reporting searches.
Useful tools
- Reference managers such as Zotero, Mendeley or EndNote, to collect records and remove duplicates
- Screening tools such as Rayyan or Covidence, which let two reviewers screen independently and resolve conflicts
- The PRISMA flow diagram template, available free from the PRISMA website
- GRADE, for rating the certainty of evidence in health reviews
Common mistakes
- A question that is too broad to answer in one review
- Searching only one database, or only Google Scholar
- Not recording the search strings and dates, so the search cannot be repeated
- Inclusion criteria changed after seeing the results
- No quality or risk-of-bias assessment
- Describing studies one by one instead of synthesising them
- A flow diagram whose numbers do not add up
How long does it take?
Usually several weeks to a few months, depending on the number of records and included studies, and whether you are working alone or with a second reviewer. Searches older than about a year may need updating before submission. Plan for this, especially if the review is a chapter of your thesis.
Is a systematic review right for you?
It is a strong choice if your field has enough primary studies to combine, and if your question would genuinely benefit from a careful summary of the evidence. If only a handful of studies exist, a scoping review or a narrative review may be more honest and more useful.
Need help? Our systematic literature review service supports every step, from the protocol and search strategy to a PRISMA-compliant manuscript, with you as the author throughout. For the statistics in a meta-analysis, see our data analysis support.
