Systematic Review: Steps, PRISMA 2020 and Search Strategy

Reviewed by Prof. Dr. Alexander Holtermann · Updated 16 October 2026 · 5 min read

A systematic review answers a focused question by searching, selecting and appraising all relevant studies with an explicit, reproducible method (PRISMA 2020).

Key takeaways

A systematic review uses a predefined, transparent and reproducible method to identify, select, appraise and synthesize all relevant studies on a focused question.
PRISMA 2020 (Page et al., 2021) is the reporting guideline: a 27-item checklist and a flow diagram showing how many records were identified, screened, excluded and included.
Core steps: question (e.g. PICO), protocol, search strategy in several databases, screening by two reviewers, data extraction, risk-of-bias appraisal, synthesis.
A systematic literature review in a dissertation is often a scaled-down version; be transparent about limits such as a single screener.
A systematic review is not the same as a meta-analysis: meta-analysis is the statistical pooling of results, which may or may not be part of a review.

Systematic review vs other literature reviews

A narrative literature review summarizes the literature the author considers relevant; selection is not documented in a reproducible way. A systematic review follows explicit rules for every step, so that another researcher could repeat the search and arrive at the same set of studies.

Type	Aim	Method	Typical use
Narrative review	Overview, background	Selective, not reproducible	Literature review chapter
Systematic review	Answer a focused question comprehensively	Protocol, systematic search, screening, appraisal	Standalone study, health and social sciences
Meta-analysis	Pooled effect estimate	Statistical synthesis of quantitative studies	Part of a systematic review
Scoping review	Map the extent and nature of evidence	Systematic search, usually no quality appraisal	Emerging or broad topics
Rapid review	Timely evidence for decisions	Streamlined systematic methods	Policy and practice

Step 1: A focused question and a protocol

A systematic review needs a narrow, answerable question. In health research, PICO is the standard framework; for qualitative evidence, SPIDER or PICo are alternatives.

Write a protocol before searching: question, eligibility criteria, databases, search strings, screening procedure, appraisal tool, synthesis method.
For published reviews, protocols are often registered in public registries such as PROSPERO (for health-related reviews) or on open science platforms.
For a dissertation, ask your supervisor whether a protocol in the appendix is sufficient.

Element	Meaning	Example
P – Population	Who?	Undergraduate students
I – Intervention	What is done?	Mindfulness-based programs
C – Comparison	Compared with?	No intervention or waitlist
O – Outcome	What is measured?	Self-reported stress

Step 2: The search strategy

Search several bibliographic databases relevant to your discipline (for example databases for medicine, psychology, education or social sciences). Build search strings from keywords and synonyms for each PICO element, combine synonyms with OR and elements with AND, and use truncation and controlled vocabulary (subject headings) where the database offers it.

Document each search: database, date, exact string, filters, number of hits.
Add supplementary searches: reference lists of included studies (backward citation searching), citing articles (forward), grey literature, trial registries.
Consult a subject librarian. Most universities offer search support for systematic reviews.
Export results to a reference manager and remove duplicates before screening.

Step 3: Screening and selection

Screening happens in two stages: first titles and abstracts, then full texts of potentially relevant studies. Apply your predefined inclusion and exclusion criteria (population, design, outcomes, language, publication period). Ideally two reviewers screen independently and resolve disagreements by discussion; report agreement. Record reasons for excluding full texts.

The PRISMA 2020 flow diagram documents the process: records identified (by source), duplicates removed, records screened, records excluded, reports sought and assessed, reports excluded with reasons, and studies included.

PRISMA stage	Example numbers
Records identified from databases	1,248
Duplicates removed	312
Records screened (title/abstract)	936
Records excluded	871
Full-text reports assessed	65
Reports excluded (with reasons)	47
Studies included in review	18

Step 4: Data extraction and risk of bias

Use a standardized extraction form: authors, year, country, design, sample, intervention, outcomes, measures, key results. Pilot the form on two or three studies.

Assess risk of bias or methodological quality with an established tool suited to the study designs, for example tools for randomized trials, for non-randomized studies or checklists for qualitative studies. Do not invent your own scoring system. Report the appraisal per study, and consider how quality affects your conclusions.

Step 5: Synthesis and reporting

Choose a synthesis method that fits your data:

Meta-analysis: if studies are sufficiently similar in design, intervention and outcome, pool effect sizes statistically and assess heterogeneity.
Narrative synthesis: structured textual and tabular synthesis when pooling is not appropriate. Group studies, describe patterns and explain differences.
Qualitative evidence synthesis: thematic synthesis or meta-ethnography for qualitative studies.
Report following the PRISMA 2020 checklist (27 items, covering title, abstract, introduction, methods, results, discussion and other information such as registration and funding).

Systematic reviews as a dissertation

Many programs, especially in health, psychology and education, accept a systematic literature review as a dissertation. Typical simplifications: one main reviewer with a second reviewer checking a sample (for example 10–20% of records), a limited number of databases and a narrative synthesis. These are acceptable if you name them as limitations. Plan the time: searching and screening alone often take four to six weeks.

Requirements vary greatly between institutions, from the number of databases to whether a protocol must be approved. Your university’s own guidelines take precedence. Upload them to Professor Digital and the plan is tailored to them. The research question tool helps you sharpen a PICO question.

Frequently asked questions

What is the difference between a systematic review and a literature review?

A traditional literature review selects sources by the author’s judgment. A systematic review follows a predefined, documented method for searching, screening and appraising studies, so it can be reproduced.

What is PRISMA 2020?

The Preferred Reporting Items for Systematic reviews and Meta-Analyses, updated in 2020 (Page et al., 2021). It consists of a 27-item checklist and a flow diagram for reporting systematic reviews.

Can I do a systematic review for my dissertation?

Often yes, especially in health, psychology and education. Check your program’s rules and be transparent about simplifications such as a single screener.

How many databases should I search?

At least two to three databases relevant to your field, plus supplementary methods like citation searching. Your supervisor or subject librarian can advise on the right ones.

Is a meta-analysis the same as a systematic review?

No. A meta-analysis is a statistical method for pooling results. It is often part of a systematic review, but a review can also use a narrative synthesis.

Sources and further reading

Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., et al. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ, 372, n71.
Clark, T., Foster, L., Sloan, L., & Bryman, A. (2021). Bryman’s Social Research Methods (6th ed.). Oxford University Press.

Your university’s examination regulations and style guide always take precedence.