A systematic literature review is a rigorous, replicable method for identifying, evaluating, and synthesizing all relevant research on a specific question. Unlike narrative reviews, systematic reviews follow explicit methodology to minimize bias and maximize transparency. This guide walks you through the process step by step.
What Is a Systematic Review?
A systematic review differs from a traditional narrative literature review in several key ways: Explicit methodology: Every step is documented and replicable. Comprehensive search: Aims to identify all relevant studies, not just convenient ones. Quality assessment: Critically evaluates study quality using standardized tools. Transparent reporting: Uses frameworks like PRISMA for clear documentation. Synthesis: May include meta-analysis (statistical combination of results).
The PRISMA Framework
PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) is the gold standard for reporting systematic reviews. The PRISMA 2020 statement includes a 27-item checklist and a flow diagram documenting your search and screening process.
Key PRISMA sections: Title, Abstract, Introduction (rationale, objectives), Methods (eligibility, search strategy, selection, data collection, quality assessment, synthesis), Results (search results, study characteristics, findings), Discussion (interpretation, limitations), and Other (registration, funding).
Step 1: Define Your Research Question
Use the PICO framework for clinical questions: Population, Intervention, Comparison, Outcome. Or SPIDER for qualitative reviews: Sample, Phenomenon of Interest, Design, Evaluation, Research type.
Your question should be specific enough to search systematically but broad enough to capture all relevant evidence.
Step 2: Develop Search Strategy
Identify databases: Use multiple databases relevant to your field (e.g., PubMed, PsycINFO, CINAHL, Scopus). Develop search terms: Include subject headings (MeSH terms) and free-text keywords. Use Boolean operators: Combine terms with AND, OR, NOT. Document everything: Keep detailed records of every search, including date, database, exact search string, and results count.
Step 3: Screen Studies
Title/abstract screening: Apply inclusion/exclusion criteria to identify potentially relevant studies. Full-text screening: Review full articles for final inclusion decisions. Use two reviewers: Have two people independently screen, then resolve disagreements. Track using PRISMA flow diagram: Document reasons for exclusion at each stage.
Step 4: Assess Quality
Use validated quality assessment tools: Cochrane Risk of Bias for randomized trials. Newcastle-Ottawa Scale for observational studies. CASP (Critical Appraisal Skills Programme) for various study types. GRADE for rating certainty of evidence.
Step 5: Extract Data
Create a standardized extraction form including: Study characteristics (authors, year, country). Methods (design, sample size, duration). Participants (demographics, inclusion criteria). Interventions/exposures. Outcomes measured. Key findings.
Step 6: Synthesize Findings
Narrative synthesis: Describe and compare findings across studies. Meta-analysis: Statistically combine results if studies are sufficiently homogeneous. Report heterogeneity: Use I² statistic to quantify variation between studies.
Common Pitfalls
1. Search strategy too narrow (missing relevant studies). 2. Inconsistent screening criteria. 3. Failing to assess study quality. 4. Inadequate documentation. 5. Not registering your protocol (on PROSPERO or similar).