A scoping review is conducted in nine sequential steps set out by the Joanna Briggs Institute (JBI), building on Arksey and O’Malley’s original 2005 five-stage framework: define the question with a PCC (Population, Concept, Context) framework, set eligibility criteria, plan and run a systematic multi-database search, screen and select evidence, chart (extract) the data into a standardized table, and analyze, present, and summarize the results against PRISMA-ScR’s 20-item reporting checklist. Unlike a systematic review, no risk-of-bias appraisal is required, and the protocol cannot be registered with PROSPERO. This guide walks through each step with a worked charting table.
For the definitional distinction between a scoping review, a systematic review, and a narrative literature review, see CASRAI’s side-by-side comparison and the scoping review dictionary entry — this page assumes you have already decided a scoping review is the right tool and focuses on how to actually run one.
Is a scoping review the right choice for your question?
Before starting, confirm the fit. A scoping review is the appropriate choice when the goal is to map the extent, range, and nature of evidence on a broad or emerging topic, clarify how key concepts are defined and used across a literature, identify gaps for future primary research, or establish whether a narrow-enough question and enough primary studies exist to justify a subsequent systematic review.
When a scoping review is the wrong choice
Choose a systematic review instead — not a scoping review — when the goal is any of the following:
- Answering an effectiveness question. If the review needs to establish whether an intervention works, by how much, or compared with what, that requires pooled effect estimates and formal risk-of-bias appraisal, which a scoping review deliberately does not produce.
- Informing a clinical guideline, policy decision, or funding recommendation. Bodies that develop practice guidelines (e.g., Cochrane-affiliated groups, guideline panels using GRADE) require the quality appraisal a scoping review omits by design.
- Evaluating diagnostic accuracy, prognosis, or patient/participant experience against a defined comparator. These need a focused, PICO-shaped question and critical appraisal, not a descriptive map.
- The question is already narrow and answerable, and enough primary studies exist to synthesize. Running a scoping review on a question that is really a systematic-review question wastes the extra rigor a systematic review would have delivered, and reviewers will ask why appraisal was skipped.
If the literature is broad, heterogeneous, or too immature for formal synthesis, and the goal is genuinely descriptive, proceed with the steps below.
The JBI nine-step framework
The JBI framework (Peters MDJ, Godfrey C, McInerney P, et al., “Chapter 11: Scoping Reviews,” in Aromataris E, Munn Z, eds., JBI Manual for Evidence Synthesis, 2020) refines and formalizes Arksey and O’Malley’s original 2005 stages — identifying the research question; identifying relevant studies; study selection; charting the data; and collating, summarizing, and reporting results, with an optional sixth consultation stage — into nine explicit steps:
- Identify the research question and objectives using the PCC framework
- Develop and document eligibility (inclusion/exclusion) criteria
- Plan and document the search strategy
- Conduct the search across databases and grey literature
- Select evidence through title/abstract and full-text screening
- Extract (“chart”) the data using a standardized, piloted form
- Analyze the extracted evidence descriptively
- Present results in tables, charts, or diagrams mapped to PCC
- Summarize the evidence and discuss its implications, including gaps
Step 1 — Define the question with PCC, not PICO
Systematic reviews frame questions with PICO (Population, Intervention, Comparison, Outcome), which presumes a comparator and an outcome to measure. Scoping reviews use PCC — Population, Concept, Context — because the question is exploratory rather than comparative. A workable PCC question names who or what is being studied (Population), what is being examined about them (Concept — an intervention, phenomenon, gap, or practice), and the setting, geography, or discipline the review is bounded to (Context).
| PCC element | What it captures | Worked example |
|---|---|---|
| Population | Who or what the evidence is about | Early-career academic researchers |
| Concept | The core idea, intervention, or phenomenon of interest | Use of AI writing-assistance tools in manuscript preparation |
| Context | Setting, discipline, geography, or timeframe boundary | Peer-reviewed STEM journals, 2022 onward |
Step 2 — Set eligibility criteria
Translate the PCC elements into explicit inclusion and exclusion criteria before screening begins: source types (empirical studies, grey literature, conference proceedings), publication date range, language, and any Population/Concept/Context boundaries decided in Step 1. Document these criteria in the protocol — changing them mid-review without recording the change and rationale undermines the review’s reproducibility, the same concern that drives systematic-review protocol discipline.
Step 3–4 — Plan and run the search
Scoping reviews use a systematic, reasonably comprehensive search strategy, typically developed with a research librarian, run across at least two or three subject-relevant databases plus grey literature sources (preprint servers, dissertation repositories, conference abstracts, relevant organizational websites). JBI guidance describes the search as unfolding in stages: an initial limited search of one or two databases to identify text words and index terms, a second, more exhaustive search across all chosen databases and sources using the refined terms, and a hand-search of the reference lists of included sources. The full search strategy for at least one database, including every search string used, must be reported (a PRISMA-ScR requirement) so the search is reproducible.
Step 5 — Select evidence
Screening proceeds in two stages: title/abstract screening against the eligibility criteria, followed by full-text screening of everything that survives the first pass. JBI and PRISMA-ScR both recommend at least two independent reviewers for screening, with a documented process for resolving disagreement (discussion, or a third reviewer as tiebreaker). Pilot-test the screening criteria on a small sample of records first — a common source of rework is discovering mid-screening that the criteria are ambiguous and need to be re-applied from the start.
Step 6–8 — Chart the data (the centerpiece step)
“Charting” is the scoping-review-specific term for data extraction: pulling standardized information out of each included source into a shared table, then analyzing and presenting it descriptively. This is the step most reviewers underestimate — a charting form built after screening starts, instead of piloted before it, is one of the most common causes of having to re-extract data from already-screened sources.
Build and pilot the charting form before full extraction begins, testing it on a handful of included sources with more than one reviewer to confirm the categories are unambiguous and complete. A typical charting table records, at minimum:
| Field | What it captures |
|---|---|
| Author(s), year, country | Basic bibliographic and provenance data |
| Study design / source type | Empirical study, conceptual paper, guideline, grey literature report, etc. |
| Population / sample | Who or what was studied, and how many |
| Concept addressed | Which aspect of the concept of interest the source examines |
| Context | Setting, discipline, or geography the source is situated in |
| Key findings relevant to the review question | Descriptive summary — not an effect size or pooled result |
| Gaps or limitations noted | What the source itself flags as unresolved or out of scope |
Analysis at this stage stays descriptive — frequency counts (how many sources address each sub-concept, by year or by context), basic categorization, and narrative description of patterns. This is a deliberate methodological boundary: thematic synthesis or pooled quantitative analysis belongs to a systematic review or meta-analysis, not a scoping review. Present the charted data as tables, and where useful, a supplementary visual map (e.g., a bubble chart of source count by concept and context) aligned to the PCC elements from Step 1.
Step 9 — Summarize and discuss
Close by relating the charted evidence back to the original PCC question: what does the map show about the extent and nature of the evidence, what concepts or contexts are under-represented, and what does that imply for practice or for a future, more narrowly scoped systematic review. This is also where the JBI framework’s optional consultation exercise fits, if used — a structured check-in with relevant stakeholders (researchers, librarians, content experts, or, in health-related topics, patients, clinicians, or policymakers) at protocol, execution, or dissemination stage, intended to sense-check relevance and surface evidence the database search may have missed. It is optional, not a required stage, and should be reported as a distinct step if used.
Registering the protocol
Write and, ideally, publicly register the protocol before screening begins — the same reproducibility rationale that applies to systematic reviews. The one detail that consistently trips people up: PROSPERO does not accept scoping-review protocol registrations — it is scoped to systematic reviews (plus rapid and umbrella reviews) with a health-related outcome. Register a scoping-review protocol instead through the Open Science Framework (OSF), INPLASY, or by publishing the protocol itself as a standalone, peer-reviewed protocol paper before the review is conducted.
Reporting with PRISMA-ScR
Report the completed review against PRISMA-ScR — the PRISMA Extension for Scoping Reviews (Tricco AC, Lillie E, Zarin W, et al., “PRISMA Extension for Scoping Reviews (PRISMA-ScR): Checklist and Explanation,” Annals of Internal Medicine, 2018;169(7):467–473, DOI: 10.7326/M18-0850). It sets out 20 essential items plus 2 optional items (critical appraisal of individual sources, and the implications of that appraisal), uses PCC rather than PICO framing throughout, and — unlike PRISMA 2020 for systematic reviews — does not require a risk-of-bias assessment, consistent with a scoping review’s descriptive purpose. Most journals that publish scoping reviews now expect a completed PRISMA-ScR checklist and flow diagram submitted alongside the manuscript. See the CASRAI PRISMA 2020 dictionary entry and PRISMA and systematic review methodology guide for how the base PRISMA statement and its extensions relate to each other.
Common mistakes
- Treating it as a shortcut systematic review. A scoping review is not a systematic review done faster — it answers a genuinely different question and should not substitute for one when an effectiveness or appraisal question is really what’s being asked.
- Skipping the charting-form pilot. Building the extraction categories after full-text screening has already started is the single most common source of costly re-work.
- Attempting a quantitative synthesis. Pooled effect estimates and meta-analytic synthesis are out of scope for a scoping review by design; if the charted data turns out to support that kind of synthesis, that is itself a finding — it suggests a systematic review or meta-analysis may now be feasible.
- Trying to register with PROSPERO. It will be declined; use OSF, INPLASY, or a published protocol paper instead.
- Reporting against PRISMA 2020 instead of PRISMA-ScR. The two checklists diverge specifically because the review types diverge; using the wrong one is a common cause of reviewer or editor pushback.
Frequently asked questions
How many reviewers does a scoping review need?
At minimum two independent reviewers for title/abstract and full-text screening, per JBI and PRISMA-ScR guidance, with a documented process (discussion, or a third reviewer) for resolving disagreements. A single-reviewer scoping review is possible in principle but weakens the review’s credibility and is generally discouraged for anything intended for peer-reviewed publication.
Does a scoping review need a risk-of-bias or quality appraisal?
No — this is one of the defining differences from a systematic review. PRISMA-ScR includes critical appraisal of individual sources as one of its two optional items, not a required one. If quality appraisal is genuinely needed to answer the review’s question, that is usually a signal the review is better framed as a systematic review.
Can a scoping review include grey literature?
Yes, and it typically should. Grey literature (theses, conference proceedings, government and organizational reports, preprints) often captures emerging or non-journal evidence that a scoping review’s broad, exploratory purpose is specifically well suited to surface — more so than a narrowly scoped systematic review, which may restrict itself to peer-reviewed sources depending on its protocol.
Where do I register a scoping review protocol if not PROSPERO?
The Open Science Framework (OSF) and INPLASY both accept scoping-review protocol registrations. A protocol can also be published as a standalone, peer-reviewed protocol paper ahead of the review itself.
How is a scoping review different from a rapid review?
They answer different needs: a scoping review trades off depth of appraisal for breadth of mapping on a broad question, while a rapid review keeps a systematic review’s narrow, appraisal-driven structure but compresses the timeline and methods (e.g., single-reviewer screening, narrower search) to meet a time-sensitive decision. See the scholarly publishing hub for CASRAI’s broader coverage of review and reporting standards.
