To write a review article a journal will publish, choose the review type that fits your question (narrative, scoping, systematic, meta-analysis or bibliometric), confirm nobody has reviewed the same question recently, check that the journal accepts unsolicited reviews, and write a paper with a stated aim, a described method, a body organised by theme and a research agenda. A review that only summarises papers will not get past most editors.
For many PhD scholars a review is the first paper, because the reading is already done for the thesis. That instinct is sound, but a thesis chapter and a review paper do different jobs, and the commonest problem in a first review is a chapter that was never rewritten. Last reviewed September 2026.
Which type of review article should you write?
Decide this first. It sets how long the work takes and which journals will look at it. Our guide to types of research articles defines each kind; the table is about what each one demands of you.
| Review type | What it needs | What journals expect | Reporting standard |
|---|---|---|---|
| Narrative | Deep knowledge of a field and a clear argument about it | A stated aim, a described search, a point of view; often invited from experts | No PRISMA; SANRA as a self-check |
| Scoping | A broad question, a protocol, a documented search, charting of studies | A map of the evidence and its gaps, not a verdict on what works | PRISMA-ScR |
| Systematic | A focused question, registered protocol, two independent screeners, risk-of-bias assessment | Reproducible methods, a flow diagram, conclusions limited by study quality | PRISMA 2020 |
| Meta-analysis | A systematic review plus enough similar studies with comparable effect sizes | Correct pooling, heterogeneity and sensitivity analyses | PRISMA 2020 |
| Bibliometric | A Scopus or Web of Science export and a mapping tool | Interpretation of the maps, not just the maps | No single agreed standard; report data and settings in full |
For most second- or third-year scholars I’d suggest a scoping review, or a well-built narrative review in a field you now know better than most. A full systematic review is excellent if a colleague will screen independently; the PRISMA method is set out step by step in our systematic review guide, so I won’t repeat it.
Scoping reviews grew from the framework of Arksey and O’Malley (International Journal of Social Research Methodology, 2005). They answer “what research exists on this, and where are the holes?” rather than “does this work?”. After two years of reading, that is often exactly the question you can answer.
A word on bibliometric reviews
These have become very common in Indian management and engineering departments. You export records from Scopus or Web of Science and map citation or keyword networks with VOSviewer (from CWTS, Leiden University) or Biblioshiny, the point-and-click interface of the free R package bibliometrix (Aria and Cuccurullo, Journal of Informetrics, 2017). The software is the easy part. Ten network maps with a paragraph describing their colours tell an editor nothing. What gets published is the reading of the maps: why these clusters formed, which strand is fading, what the field ignores.
How do you check whether your topic has already been reviewed?
An afternoon’s work, and it can save you months.
- Search Scopus and Web of Science for your topic terms, filter the document type to Review, and look at the last five years.
- Search PROSPERO for registered health reviews and OSF Registries for protocols in any field. A registered protocol means someone is working on it now.
- Read the two or three closest reviews properly, noting their search dates, scope and conclusions.
Finding a review changes your plan rather than ending it. If the last one searched up to 2021 and plenty has appeared since, an update is defensible. If it covered only trials and you want observational evidence, or it was global and Indian evidence behaves differently, that is a new question. Writing the same review again is not an option; the reviewers have usually read the first one (sometimes they wrote it).
Can you turn your thesis literature review into a review paper?
Yes, but you are rewriting, not trimming. Your chapter was written for a supervisor, a DC and two examiners, and it argues toward your gap. The paper is written for an editor and strangers in your field. Five things change:
- Argument. The paper needs its own aim, one that matters to someone who will never read your thesis.
- Framework. Chapter sections follow your objectives. Paper sections follow the logic of the field: a framework, mechanisms, or the questions the literature has tried to answer.
- Contribution. Say what readers get: a classification, a model, a reconciliation of conflicting findings, a map of gaps.
- Recency. Rerun the searches. A first-year chapter misses the last two years, and reviewers notice.
- The “so what”. End with a research agenda and, where it fits, implications for practice.
How the chapter itself is built is in our guide to writing a literature review, and summary versus synthesis at paragraph level is shown in literature review examples and format. In a paper the synthesis rule applies more strictly, not less.
What is the outline of a review paper?
Most reviews in the sciences and social sciences follow this shape. Check two recent reviews in your target journal and match their headings.
Title
Topic plus review type, for example “…: a scoping review”.
Abstract
Aim, how the literature was found, what was found, main gaps.
Introduction
Why a review is needed now, what earlier reviews covered, and the aim in one sentence.
Methods
Databases, dates, search terms, inclusion criteria. Short for a narrative review, but present.
Body
Sections by theme, framework or question, never by author.
Synthesis table
The key studies or the evidence for each theme on one page.
Research agenda
Specific questions that follow from the gaps, with the design that would answer each.
Conclusion
What the evidence shows and how strong it is. No new material.
Scholars leave the methods section out of narrative reviews, and reviewers ask for it first. Four or five sentences on where, when and how you searched will do.
An illustrative outline
A made-up scoping review in environmental engineering. The topic is invented and nothing here describes real findings.
Title: Agricultural-residue adsorbents for removing fluoride from groundwater: a scoping review.
- Introduction. Why fluoride in drinking water matters in parts of India, what earlier reviews covered, and the aim: to map which residues have been tested, under what conditions, and what is missing before field use.
- Methods. Scopus and Web of Science, 2010 to the search date; string in a supplementary file; criteria; reported to PRISMA-ScR.
- Adsorbents by source material. Rice husk, coconut shell, bagasse and so on.
- Test conditions. Synthetic solutions against real groundwater; batch against column studies.
- Regeneration, cost and disposal. The theme studies tend to skip, which is why it gets a section.
- Synthesis table. One row per residue: studies found, water type, test scale, regeneration reported.
- Research agenda. Field trials with real groundwater; costing that includes spent-adsorbent disposal.
- Conclusion.
Sections 3 to 5 would never appear in a chapter shaped around one scholar’s objectives. They follow what a water engineer reading the journal wants to know.
What do editors look for, and why do review papers get desk-rejected?
The first reason has nothing to do with quality.
Springer Nature’s guidance for authors says the same: some journals do not consider unsolicited reviews, so check the website, and if it is silent, send the editor a pre-submission enquiry before you write. That is a short email with a working title, the aim, why the review is needed now and how you will search; our journal cover letter guide helps with the tone.
Beyond that, reviews are declined for the same reasons as any paper (see why papers get desk-rejected) plus their own: a recent review already exists, the paper summarises without concluding, there is no method so the selection looks cherry-picked, or the references stop three years ago.
Reporting standards reviewers will check
Systematic reviews and meta-analyses: PRISMA 2020, with its 27-item checklist. Scoping reviews: PRISMA-ScR (Tricco and colleagues, 2018), 20 essential and 2 optional items. Narrative reviews have no reporting guideline of that kind, but SANRA (Baethge, Goldbeck-Wood and Mertens, Research Integrity and Peer Review, 2019) gives editors a six-item scale: importance of the review, aims, description of the literature search, referencing, scientific reasoning, and presentation of relevant endpoint data. Each is scored 0, 1 or 2. Score your own draft before an editor does.
- The journal accepts unsolicited reviews, or an editor has replied to your pre-submission enquiry.
- No review on the same question has appeared recently, or yours says clearly what it adds.
- There is a methods section, even though the review is narrative.
- Most paragraphs cite several studies and make a point about them.
- Studies from the last year or two are included.
- The research agenda names specific questions, not “more research is needed”.
- You have scored the draft against SANRA, PRISMA-ScR or PRISMA 2020 and fixed the weak items.
For picking the journal, how to choose a journal applies, with one extra check: how many reviews has it published in the last year?
How long does a review article take to write?
Longer than scholars expect, even with the reading done. These are rough ranges from experience, not from a study:
- Narrative review from a finished thesis chapter: two to four months to a submittable draft.
- Scoping review: four to eight months, mostly searching, screening and charting.
- Systematic review with meta-analysis: often a year or more.
Peer review comes on top; how long journal publication takes gives realistic ranges. If your university wants a published paper before thesis submission, start the review in your second year.
One last thing. A viva examiner can ask about any study in your review, so cite only papers you have read, and draw only conclusions you can defend in your own words.
Sources
- Baethge C, Goldbeck-Wood S, Mertens S. SANRA: a scale for the quality assessment of narrative review articles. Research Integrity and Peer Review 2019;4:5
- PRISMA-ScR: Tricco et al., Annals of Internal Medicine 2018
- Page MJ et al. The PRISMA 2020 statement. BMJ 2021;372:n71
- Arksey H, O’Malley L. Scoping studies: towards a methodological framework. IJSRM 2005;8(1):19–32
- Springer Nature: references and article types (unsolicited reviews, pre-submission enquiries)
- Alcohol Research: Current Reviews: information for authors
- VOSviewer (CWTS, Leiden University)
- bibliometrix and Biblioshiny
FAQ
Questions scholars ask
Can a PhD scholar write a review article without being invited?
Yes, at journals that accept unsolicited reviews, and many do. Read the author guidelines first; if they are silent, email the editor a short proposal before writing.
Does a review article count toward the PhD publication requirement?
Your university decides, not UGC. Some ordinances count any peer-reviewed paper; others expect original research. Ask the research cell in writing. See types of research articles.
Is a narrative review easier to publish than a systematic review?
Easier to write, often harder to publish for an early-career author, because many journals prefer narrative reviews from invited experts. A scoping or systematic review with a transparent method is easier for an editor to trust.
Do I need a methods section in a narrative review?
Include one. A few sentences on databases, dates, search terms and selection criteria let reviewers judge whether your selection is fair, and SANRA scores the search description as one of its six items.
Can I publish a review from my thesis chapter and keep the chapter in my thesis?
Usually yes, if you disclose it. Cite the published review in the thesis, tell your supervisor and follow your university’s rules on prior publication. The submission process itself is covered in how to publish a research paper.
