Systematic reviews are among the most cited and most respected types of academic publications. They synthesize all available evidence on a research question using a transparent, reproducible methodology. In 2026, journals increasingly require adherence to the PRISMA 2020 guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analyses).
This guide walks you through every step — from formulating your research question to publishing your review in a top journal.
What Is a Systematic Review?
A systematic review differs from a traditional literature review in several critical ways:
| Feature | Traditional Literature Review | Systematic Review |
|---|---|---|
| Search Strategy | Informal, may miss studies | Comprehensive, documented search across multiple databases |
| Study Selection | Author's choice | Pre-defined inclusion/exclusion criteria |
| Bias Assessment | Not required | Mandatory quality appraisal of each study |
| Reproducibility | Low | High (protocol + detailed methodology) |
| Evidence Level | Low | Highest level of evidence |
Step 1: Formulate Your Research Question
Use the PICO framework for clinical/healthcare topics or PEO/SPIDER for qualitative research:
- P — Population (Who are you studying?)
- I — Intervention (What treatment/approach?)
- C — Comparison (What is the alternative?)
- O — Outcome (What do you measure?)
Example: "In patients with Type 2 diabetes (P), does AI-assisted diagnostic imaging (I) compared to traditional radiology (C) improve early detection rates (O)?"
Step 2: Register Your Protocol
Before starting your search, register your protocol on PROSPERO (for health-related reviews) or OSF (for any field). This demonstrates transparency and prevents duplication. Many Q1 journals now require protocol registration.
Step 3: Develop a Comprehensive Search Strategy
Search at minimum 3-4 databases:
- PubMed/MEDLINE — Biomedical and health sciences
- Scopus — Broad multidisciplinary coverage
- Web of Science — Citation tracking and impact analysis
- IEEE Xplore — Engineering and computer science
- Cochrane Library — Clinical trials and healthcare
- Google Scholar — Supplementary search for grey literature
Use Boolean operators (AND, OR, NOT) and MeSH terms or controlled vocabulary. Document every search string, database, date, and number of results.
Step 4: Study Screening and Selection
The screening process follows these stages:
- Remove duplicates across databases
- Title and abstract screening — Apply inclusion/exclusion criteria
- Full-text screening — Read remaining papers in full and apply criteria
- Document reasons for exclusion at the full-text stage
Use tools like Rayyan, Covidence, or ASReview for efficient screening. Two independent reviewers should screen with a third resolving disagreements.
Step 5: Data Extraction
Create a standardized extraction form capturing:
- Study characteristics (author, year, country, design)
- Population details (sample size, demographics)
- Intervention/exposure details
- Outcomes and results (effect sizes, confidence intervals, p-values)
- Quality assessment scores
Step 6: Quality Assessment
Assess the risk of bias in each included study using validated tools:
- Cochrane Risk of Bias (RoB 2) — For randomized controlled trials
- ROBINS-I — For non-randomized studies
- Newcastle-Ottawa Scale (NOS) — For observational studies
- JBI Critical Appraisal Checklist — For qualitative studies
- QUADAS-2 — For diagnostic accuracy studies
Step 7: Data Synthesis and Meta-Analysis
If studies are sufficiently homogeneous, conduct a meta-analysis:
- Calculate pooled effect sizes (odds ratios, mean differences, hazard ratios)
- Create forest plots to visualize results
- Assess heterogeneity using I² statistic and Q test
- Perform subgroup analyses and sensitivity analyses
- Check for publication bias using funnel plots and Egger's test
If meta-analysis isn't appropriate, use narrative synthesis with structured tables.
Step 8: Report Using PRISMA 2020
The PRISMA 2020 checklist has 27 items across all sections. Key requirements include:
- PRISMA flow diagram — Documenting the selection process with exact numbers
- Search strategy — Full search strings for at least one database
- Risk of bias summary — Visual representation across all studies
- Certainty of evidence — GRADE assessment for each outcome
Need Systematic Review Support?
Systematic reviews are complex, multi-month projects. At DeepDivers, we offer end-to-end systematic review support — from protocol registration and database searches to data extraction, meta-analysis, and manuscript writing.
