Systematic Review Protocol Template

A systematic review protocol is the plan for a review, written and finalized before screening or data extraction begins. It fixes the review question, eligibility criteria, search strategy, and analysis plan in advance so that later decisions can be checked against what was originally intended. This page gives a practical, section-by-section template for writing one, built on the structure of the PRISMA-P 2015 statement — the most widely used reporting standard specifically for systematic review and meta-analysis protocols.

This is a companion to, not a duplicate of, two other CASRAI pages: the PRISMA and Systematic Review Methodology guide covers the overall review process and the PRISMA 2020 standard for reporting a completed review, and the PROSPERO Registration term explains the registry submission process itself. This page focuses specifically on what goes inside the protocol document you write before either of those steps. (The PRISMA statement itself keeps evolving — see CASRAI’s coverage of the PRISMA 2026 revision for the most recent update to the completed-review reporting standard.)

Why write a protocol before starting the review

A prospective, written protocol serves three practical purposes:

It prevents outcome switching. Decisions made after seeing the results — narrowing eligibility criteria, dropping an outcome that didn’t show an effect, adding a post hoc subgroup analysis — are a recognized source of bias in systematic reviews. A dated protocol makes any deviation visible.
It supports registration. Registries such as PROSPERO require a structured protocol as the basis for the registration record, and reviewers/editors increasingly expect a registered protocol as a condition of considering the finished review.
It reduces duplicate effort. A registered protocol is visible to other teams working in the same area before the review is complete.

The PRISMA-P 2015 checklist as the basis for this template

PRISMA-P 2015 (Moher D, Shamseer L, Clarke M, et al. “Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement.” Systematic Reviews. 2015;4:1) is a 17-item checklist (26 including sub-items), organized into three sections: administrative information, introduction, and methods. It is the checklist PROSPERO and most methods-conscious journals expect a submitted protocol to satisfy. The template below follows that structure and expands each item into a fill-in-the-blank prompt.

Section-by-section protocol template

1. Administrative information

Title — State that the document is a protocol (not the review itself), e.g. “Protocol for a systematic review of [intervention] for [population/condition].” If this is an update of a previous review, say so explicitly.
Registration — Name the registry and record the registration number once assigned (e.g. a PROSPERO CRD-prefixed ID).
Authors and contributions — List all protocol authors with contact information, and identify the guarantor (the person accountable for the review as a whole).
Amendments — Reserve a section (even if initially blank) to log any change made to the protocol after registration, with the date and rationale for each change.
Support — State funding sources and the role, if any, the funder or sponsor had in designing the review.

2. Introduction

Rationale — Explain why this review is needed given what is already known: what existing reviews exist, what gap this one addresses, and why now.
Objectives — State the review question explicitly, framed with a structured question format. Health-related reviews typically use PICO (Population, Intervention, Comparator, Outcome) or the extended PICOTS criteria, which adds Timing and Setting.

3. Methods

Eligibility criteria — Define, in advance, the study and report characteristics (design, population, comparator, outcomes, publication status, language, date range) that determine whether a study is included.
Information sources — Name every database, registry, and grey-literature source you intend to search, and the planned date range of coverage.
Search strategy — Present a full draft search strategy for at least one database, including the search terms and Boolean logic, so it can be peer-reviewed and reproduced.
Study records — Describe (a) how records will be managed (the software/platform used for deduplication and screening), (b) the study selection process (how many reviewers screen independently, how disagreements are resolved), and (c) the data extraction process (piloted extraction form, number of extractors, verification method).
Data items — List every variable that will be extracted from each included study.
Outcomes and prioritization — Define primary and secondary outcomes, and state which are most important to interpreting the review’s findings.
Risk of bias in individual studies — State which tool will be used (e.g. the Cochrane RoB 2 tool for randomized trials, ROBINS-I for non-randomized studies) and how many reviewers will apply it.
Data synthesis — State the criteria that determine whether studies are combined quantitatively, the planned summary measures (e.g. risk ratio, mean difference), the model (fixed-effect vs. random-effects — see the Cochrane Handbook’s treatment of this choice), and how heterogeneity will be assessed. If quantitative synthesis is not planned or not always possible, describe the synthesis method that will be used instead.
Meta-bias(es) — Describe how publication bias and within-study selective outcome reporting will be assessed (e.g. funnel plots, comparison against trial registrations).
Confidence in cumulative evidence — State the approach for grading the strength of the body of evidence, most commonly GRADE.

Where to register the finished protocol

Once drafted, most health-related systematic review protocols are registered with PROSPERO, the international prospective register of systematic reviews maintained by the Centre for Reviews and Dissemination at the University of York and funded by the UK’s National Institute for Health and Care Research (NIHR). PROSPERO registration is free, and — per PROSPERO’s own scope rules — is limited to reviews with a health-related outcome; it explicitly excludes scoping reviews and narrative/literature reviews. Reviews outside PROSPERO’s scope, or from other disciplines, commonly register on general-purpose platforms such as OSF Registries instead — see CASRAI’s guide to preregistering a study protocol for the general registration process across platforms. A protocol can also be published as a standalone protocol paper in a journal that accepts them, independent of or alongside registry submission.

Common mistakes to avoid

Vague eligibility criteria. “Studies of adults with diabetes” is not specific enough to apply consistently during screening — specify the diagnostic criteria, comparator, and outcome measures precisely.
An incomplete search strategy. A protocol that only names the databases, without the actual search string for at least one of them, does not meet PRISMA-P item 10 and cannot be properly peer-reviewed before the review starts.
No amendment log. Protocols do sometimes need to change legitimately (e.g. after a scoping search reveals the eligibility criteria need refinement). The problem is not amendment itself but failing to date and disclose it — build the log in from the start rather than adding it retroactively.
Registering after screening has started. Registries including PROSPERO expect registration before data extraction is complete, and ideally before screening begins; late registration undermines the point of a prospective plan.
Treating the protocol as identical to the eventual PRISMA 2020 report. The protocol states planned methods; the final report (governed by PRISMA 2020, a separate checklist) documents what was actually done, including any deviations from the plan.

Frequently asked questions

What is a systematic review protocol?

It is a document, finalized before screening or data extraction begins, that specifies a systematic review’s objectives, eligibility criteria, search strategy, and planned analysis methods. Its purpose is to fix the review’s methods in advance so later decisions can be checked against the original plan.

What’s the difference between a systematic review protocol and PRISMA?

PRISMA 2020 is a reporting checklist for the completed review. PRISMA-P 2015 is the separate, earlier-stage checklist specifically for the protocol document — this page’s template follows PRISMA-P. A review typically satisfies both checklists at different points in its lifecycle: PRISMA-P when the protocol is written and registered, PRISMA 2020 when the finished review is submitted for publication.

Do I need to register my protocol before starting a systematic review?

Registration is not universally mandated, but it is increasingly expected. Many health and medical journals ask whether a review was registered, and funders and evidence-synthesis groups (including Cochrane) treat prospective registration as standard practice. PROSPERO accepts registration up to the point data extraction is complete, but registering before screening begins is the practice that actually delivers the bias-reduction benefit.

What is PICO or PICOTS, and do I need it in the protocol?

PICO (Population, Intervention, Comparator, Outcome) is the standard framework PRISMA-P expects reviewers to use to state the review question explicitly (item 7, Objectives). PICOTS extends this with Timing and Setting. Using one of these frameworks makes the eligibility criteria section easier to write consistently — see the PICOTS Criteria entry for a fuller breakdown.

Can I deviate from the protocol once it’s registered?

Yes, but any deviation should be logged as a dated amendment with a stated rationale, and disclosed in the final review report. Undisclosed deviation — particularly changes that align conveniently with the results — is exactly the outcome-switching problem prospective registration is meant to make visible.

Related CASRAI pages

PRISMA and Systematic Review Methodology — the overall review process and the PRISMA 2020 reporting standard.
PROSPERO Registration — how to submit a protocol to PROSPERO.
PRISMA 2020 — the reporting checklist for the completed review.
PRISMA Flow Diagram — the study-selection flow diagram required by PRISMA 2020.
Cochrane Handbook for Systematic Reviews of Interventions — the methods handbook this template’s synthesis and risk-of-bias guidance draws on.
Protocol Paper — publishing the protocol itself as a standalone journal article.
Preregistration of a Study Protocol — the general registration process across platforms including OSF.
Literature Review vs. Systematic Review vs. Scoping Review — confirming a systematic review is the right review type before writing the protocol.
Systematic Review vs. Meta-Analysis — deciding whether the protocol should include a meta-analysis plan.