How to Write a Review Article

1. Understand the Types of Review Articles

2. Pre-Writing Steps

Choose Your Topic

• Pick a topic where there is enough published literature to synthesize
• Identify a clear research question or gap in knowledge
• Make sure your review adds something new - a summary, a resolution of conflicting findings, or a fresh perspective

Check the Target Journal

• Read the journal's aims, scope, and author guidelines
• Not all journals accept unsolicited review articles - some require an invitation from editors
• Check word limits, reference limits, and required sections

Define Your Scope

• Be specific: a review on "diabetes" is too broad; "GLP-1 agonists in type 2 diabetes and cardiovascular outcomes" is manageable
• Decide a time range for the literature (e.g., last 10-20 years, or all time)

3. Literature Search

Search Multiple Databases

• PubMed / MEDLINE (biomedical)
• Embase, Cochrane Library, Web of Science, Scopus
• Google Scholar for grey literature

Use Structured Search Terms

• Use MeSH (Medical Subject Headings) for PubMed searches
• Combine terms with Boolean operators: AND, OR, NOT
• Example: "heart failure"[MeSH] AND "exercise training"[MeSH]

For Systematic Reviews - Follow PRISMA

• Document your search strategy fully
• Create a PRISMA flow diagram showing records identified, screened, excluded, and included
• Register your protocol on PROSPERO before you start

Screen and Select Studies

• Screen titles/abstracts first, then full texts
• Apply pre-defined inclusion and exclusion criteria
• Use two independent reviewers to reduce bias
• Document reasons for exclusion

4. Structure of a Review Article (IMRaD Format)

Title

• Brief, informative, and keyword-rich
• State the type of review in the title if appropriate (e.g., "A Systematic Review and Meta-Analysis of...")

Abstract

• Usually structured: Background, Methods, Results, Conclusion
• 150-300 words typically
• Must stand alone and summarize the entire review

Introduction

• Define the topic and its importance
• Summarize existing knowledge briefly
• State the gap your review addresses
• State the objective of the review clearly at the end

Methods (critical for systematic reviews)

• Describe your search strategy (databases, date ranges, search terms)
• State inclusion and exclusion criteria
• Describe how studies were screened and selected
• Describe quality/risk of bias assessment tools used (e.g., Cochrane Risk of Bias, Newcastle-Ottawa Scale)

Results / Main Body

• Organize by themes, sub-topics, or chronology - not just a list of "study A found X, study B found Y"
• Critically compare and contrast findings
• Use tables and figures to summarize data
• Highlight areas of consensus and contradiction

Discussion

• Interpret and synthesize findings - what does the body of evidence actually show?
• Address contradictions in the literature
• Discuss limitations of existing evidence and of your review itself
• Suggest directions for future research

Conclusion

• Concise summary of the main message
• Clinical or practical implications
• Do not introduce new information here

References

• Be thorough and consistent - use a reference manager (Zotero, Mendeley, EndNote)
• Follow the journal's citation style (Vancouver, APA, AMA, etc.)

5. Writing Tips

• Synthesize, don't just summarize - the value of a review is your critical analysis, not a list of abstracts
• Use hedging language - "evidence suggests," "data indicate," rather than absolute claims
• Be balanced - present conflicting evidence fairly
• Avoid plagiarism - paraphrase and cite; do not copy text from sources
• Use subheadings - they improve readability and navigation
• Tables and figures - summary tables comparing studies are very useful for readers

6. Quality and Bias Assessment

• RCTs: Cochrane Risk of Bias tool (RoB 2)
• Observational studies: Newcastle-Ottawa Scale (NOS)
• Diagnostic studies: QUADAS-2
• Overall evidence body: GRADE framework (Grading of Recommendations Assessment, Development, and Evaluation)

7. Reporting Guidelines by Review Type

8. Common Mistakes to Avoid

• Searching only one database
• Not defining inclusion/exclusion criteria upfront
• Simply describing each paper in sequence without synthesis
• Ignoring negative or conflicting studies
• Failing to assess quality of included studies
• Not registering a systematic review protocol before starting

Quick Checklist Before Submission

• Clear research question defined
• Comprehensive literature search documented
• PRISMA flow diagram included (for systematic reviews)
• Studies critically appraised, not just described
• Conflict of interest statement included
• All tables/figures labeled and cited in text
• Abstract matches the body of the article
• Formatted per target journal guidelines