Systematic Review vs Scoping Review: Differences, Protocols and Reporting

A systematic review answers a focused question, often about effectiveness, accuracy, prevalence or association, by finding, appraising and synthesizing all eligible studies. A scoping review maps the extent, range and nature of evidence on a broader topic, identifies key concepts and gaps, and clarifies how research has been conducted, usually without judging the quality of individual studies or pooling their results.

Put another way: a systematic review asks 'What does the evidence say about X?' while a scoping review asks 'What evidence exists about X, and what does it look like?' Scoping reviews are often done before a systematic review, to check whether enough studies exist and how questions should be framed, but they are also valuable studies in their own right, especially in emerging or complex fields.

The table summarizes the main differences. Specific methods vary by organization and field, so always check the guidance your department, funder or target journal follows, for instance the Cochrane Handbook for intervention reviews or the JBI Manual for Evidence Synthesis for scoping and other review types.

Start from your question, not from which review sounds more impressive. Examiners and editors look for fit between the aim, the question and the method. A scoping review is not a lesser systematic review, and a systematic review is not simply a bigger scoping review.

• Choose a systematic review if you want to know whether an intervention works, how accurate a test is, how common a condition is, or how strongly two factors are associated.
• Choose a systematic review if you intend to inform clinical guidance or policy decisions, where the quality of evidence matters.
• Choose a scoping review if the literature is large, diverse or poorly defined and you need to understand its shape first.
• Choose a scoping review if your aim is to clarify how a concept is defined or measured across studies.
• Choose a scoping review if you want to identify research gaps to justify a future study or systematic review.
• Consider other review types, such as rapid, umbrella, realist or qualitative evidence syntheses, if neither fits well.

Both review types should start with a protocol: a document written before the review begins that sets out the question, methods and planned analyses. A protocol reduces the risk of bias from decisions made after seeing the results, helps teams work consistently, and shows examiners and editors that the review was planned.

For systematic reviews, the PRISMA-P checklist describes what a protocol should contain. For scoping reviews, JBI provides a protocol template, and many journals publish scoping review protocols. In either case, write down what you will do if things do not go as planned, such as how you will resolve disagreements between reviewers or what you will do if too few studies are found for meta-analysis.

Deviations from a protocol are not a failure in themselves. They happen in most reviews. What matters is that you record each change, explain why it was made, and report it in the final paper.

• Background and rationale, including why this review is needed now.
• Review question in PICO, PCC or another structured format.
• Eligibility criteria: population, concept or intervention, context, study designs, languages, dates.
• Information sources: databases, registers, grey literature sources and citation searching.
• A draft search strategy for at least one database, ideally peer reviewed (for example using the PRESS checklist).
• Screening process: number of reviewers, pilot screening, software such as Covidence or Rayyan.
• Data extraction or charting form and the items you will collect.
• Critical appraisal tools (systematic reviews) or a statement of whether appraisal will be done (scoping reviews).
• Synthesis plan, including criteria for meta-analysis and planned subgroup or sensitivity analyses.

Registering a protocol creates a public, time-stamped record of your plans. It helps prevent duplication, discourages selective reporting, and is increasingly expected by journals.

PROSPERO, run by the Centre for Reviews and Dissemination at the University of York, is the best-known international register for systematic reviews with a health-related outcome. It has historically not accepted scoping reviews, and it expects registration before data extraction begins. Its eligibility rules and forms are updated from time to time, so read the current guidance on the PROSPERO site before you apply, and allow time for the record to be checked.

The Open Science Framework (OSF) Registries accept a wider range of study types, including scoping reviews and systematic reviews outside health. Researchers often upload the full protocol there, which gives a public record and a DOI. Other options include INPLASY and publishing the protocol in a journal that accepts protocols, such as some open-access review journals. Whatever you choose, cite the registration number in the final review.

PRISMA 2020 replaced the 2009 statement and was published in 2021 in several journals simultaneously, including the BMJ. It consists of a main checklist of 27 items, a separate checklist for abstracts, updated flow diagram templates for new and updated reviews, and an explanation and elaboration paper. The flow diagram tracks records from identification through screening to inclusion, and the 2020 version distinguishes database and register searches from other methods such as citation searching.

PRISMA-ScR, the PRISMA extension for scoping reviews, was published in 2018 in the Annals of Internal Medicine by Tricco and colleagues. It has 20 essential items and 2 optional items. It adapts PRISMA to the scoping purpose; for example, critical appraisal is optional and the synthesis item focuses on how results are charted and summarized. Scoping reviews typically still use a PRISMA-style flow diagram.

PRISMA also has other extensions, including PRISMA-P for protocols, PRISMA-S for reporting literature searches, and versions for diagnostic accuracy, network meta-analysis and individual participant data reviews. Check which apply to your design.

Systematic reviews of interventions in health usually follow the Cochrane Handbook for Systematic Reviews of Interventions, currently version 6.5 (2024), which covers everything from question formulation to risk of bias, meta-analysis and summary of findings tables with GRADE. Other disciplines have their own guidance, such as Campbell Collaboration standards for social policy and education.

Scoping review methodology developed from Arksey and O'Malley's 2005 framework, later refined by Levac and colleagues and by JBI. The JBI approach is now widely used and describes stages from defining the question to presenting results, with an optional consultation exercise with stakeholders. Citing the framework you followed, and following it consistently, is a simple way to strengthen your methods section.

Reviews almost always take longer than first planned. The table below gives rough ranges for a small team working part time; your timeline depends on the number of records, the team's experience, and the complexity of synthesis. A doctoral student working mostly alone should plan for the upper end or beyond.

Whichever review you choose, a few mistakes account for many rejected protocols and critical examiner comments.

• Calling a review 'systematic' when it lacks a protocol, dual screening or appraisal.
• Doing a scoping review but then drawing conclusions about effectiveness, which the method does not support.
• Searching only one or two databases, or omitting grey literature when it matters for the question.
• Not documenting search strings fully enough for others to reproduce them.
• Running a meta-analysis on studies too different in design, population or outcome to combine meaningfully.
• Registering after data extraction has started, which undermines the purpose of registration.
• Leaving the review so long that searches are outdated at submission; plan a search update.