A systematic review protocol is a detailed, pre-specified plan that defines the objectives, methods, and analytical approach for a systematic review before the research begins. PRISMA-P (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols) is the internationally recognized checklist for writing and reporting systematic review protocols. It contains 17 items organized into three sections (administrative information, introduction, and methods) that ensure transparency, reduce bias from post-hoc methodological changes, and allow peer evaluation of the planned methodology.
Writing a protocol before conducting your systematic review is not optional good practice; it is a methodological requirement expected by journals, funding bodies, and the Cochrane Collaboration. Registration on PROSPERO or another public registry provides a timestamp that demonstrates your methods were established before data collection began.
What Is PRISMA-P and Why Does It Matter?
PRISMA-P was published in 2015 as a companion to the PRISMA statement for completed reviews. While the complete PRISMA 2020 checklist governs reporting of finished systematic reviews, PRISMA-P governs reporting of planned systematic reviews. The two checklists are complementary, and a well-written PRISMA-P protocol makes it substantially easier to produce a PRISMA-compliant final report.
A registered protocol serves several critical functions:
- Reduces selective reporting bias: Pre-specifying outcomes and analyses prevents cherry-picking results
- Prevents unnecessary duplication: Other research teams can check registries before starting similar reviews
- Enables peer feedback: Published protocols can receive methodological critique before the review is conducted
- Increases transparency: Readers can compare the final review against the original protocol to identify deviations
- Strengthens journal submissions: Many journals now require protocol registration numbers in systematic review manuscripts
The 17 PRISMA-P Checklist Items
Section 1: Administrative Information
Item 1: Title: Identify the report as a protocol for a systematic review. Include "protocol" and "systematic review" in the title. If a meta-analysis is planned, state this explicitly. Example: "Protocol for a Systematic Review and Meta-Analysis of Cognitive Behavioral Therapy for Insomnia in Older Adults."
Item 2: Registration: Provide the registration number and registry name (e.g., PROSPERO CRD42026123456). If not yet registered, state your intention to register and the target registry.
Item 3: Authors: List all authors with affiliations, including the guarantor (the person responsible for the review).
Item 4: Amendments: If the protocol has been amended since initial registration, describe each amendment with the date, rationale, and description of the change.
Item 5: Support: Describe sources of financial and non-financial support for the review, and the role of funders in the review process.
Section 2: Introduction
Item 6: Rationale: Describe the rationale for the review in the context of existing knowledge. What gap does this review address? Why is an updated or new synthesis needed? Cite relevant existing reviews and explain their limitations.
Item 7: Objectives: State the specific objectives and research question(s) using the PICO framework or another appropriate structure. The objectives must be precise enough to be answerable through systematic evidence synthesis.
Section 3: Methods
Item 8: Eligibility criteria: Specify inclusion and exclusion criteria for studies, covering:
- Population characteristics and clinical context
- Intervention or exposure details
- Comparator conditions
- Outcome measures (primary and secondary)
- Study designs eligible for inclusion
- Time frame, language, and publication status restrictions
Item 9: Information sources: List all databases, registers, and other sources you plan to search. Include the date range and any restrictions. For a comprehensive review, this typically includes at least PubMed/MEDLINE, Embase, and a domain-specific database.
Item 10: Search strategy: Present the complete search strategy for at least one database. Include all search terms, Boolean operators, MeSH headings, and filters. The search should be developed with input from an information specialist or research librarian.
Item 11: Study records: Describe:
- Data management: How records will be stored and deduplicated
- Selection process: How many reviewers will screen, at what stages, and how disagreements will be resolved
- Data collection: What data extraction forms will be used and how they will be piloted
Item 12: Data items: List all variables for which data will be extracted, including study characteristics, participant demographics, intervention details, outcome measures, and potential effect modifiers.
Item 13: Outcomes and prioritization: List primary and secondary outcomes with their definitions, measurement instruments, and time points. Explain any outcome prioritization approach.
Item 14: Risk of bias: Describe the tool(s) you will use to assess risk of bias or methodological quality (e.g., Cochrane RoB 2 for RCTs, Newcastle-Ottawa Scale for observational studies). State how many reviewers will assess each study. For a full comparison of tools, see our guide on quality assessment in systematic reviews.
Item 15: Data synthesis: Describe your planned synthesis approach:
- Will you conduct a meta-analysis? Under what conditions?
- What statistical model will you use (fixed-effect vs random-effects)?
- How will you assess and handle heterogeneity?
- If meta-analysis is not appropriate, how will narrative synthesis be structured?
- Are subgroup analyses or meta-regression planned?
Item 16: Meta-bias: Describe any planned assessment for biases across studies, including publication bias (funnel plots, Egger's test) and selective outcome reporting.
Item 17: Confidence in cumulative evidence: Describe the approach for assessing certainty of evidence across outcomes (e.g., GRADE framework).
Step-by-Step Protocol Writing Process
Step 1: Formulate Your Research Question
Use PICO (or PICOS, PEO, SPIDER, depending on your review type) to structure a focused, answerable question. A well-formulated question directly determines your eligibility criteria, search strategy, and planned analyses.
Test your question: Can it be answered by synthesizing existing primary studies? Is it narrow enough to be feasible but broad enough to find sufficient evidence?
Step 2: Conduct a Preliminary Scoping Search
Before writing the full protocol, run a preliminary search to assess the volume and type of available literature. This helps you:
- Estimate feasibility (are there enough studies?)
- Refine your research question
- Identify appropriate study designs to include
- Determine whether meta-analysis is likely to be feasible
- Identify relevant search terms and MeSH headings
Step 3: Develop the Search Strategy
Work with a research librarian to develop a comprehensive search strategy. The search should be sensitive enough to capture all relevant studies without being so broad that screening becomes unmanageable. Document the complete strategy for at least one database in your protocol.
Step 4: Define Eligibility Criteria
Write explicit, operationalized inclusion and exclusion criteria. Each criterion should be clear enough that two independent reviewers can apply it consistently. Pilot your criteria on a sample of abstracts to test inter-rater agreement.
Step 5: Plan Your Screening Process
Define how screening will proceed through title/abstract review and full-text assessment. Specify the number of reviewers, the conflict resolution process, and any screening tools you will use. For detailed guidance, read our article on screening best practices for systematic reviews.
Step 6: Design Data Extraction Forms
Create standardized forms that capture all variables listed in your protocol. Pilot the forms on 2-3 studies before full extraction. Decide whether extraction will be performed by one reviewer with verification or two independent reviewers.
Step 7: Plan the Analysis
Specify your synthesis approach in detail. If meta-analysis is planned, state the statistical model, the software you will use, and the conditions under which you would not pool studies. Describe planned subgroup analyses and sensitivity analyses.
Step 8: Write and Register
Draft the complete protocol following all 17 PRISMA-P items. Submit to PROSPERO (for health-related reviews) or another appropriate registry. Consider publishing the protocol in a peer-reviewed journal for additional scrutiny.
Registering Your Protocol on PROSPERO
PROSPERO is the international prospective register of systematic reviews maintained by the Centre for Reviews and Dissemination at the University of York. Registration is free and involves completing an online form that maps closely to PRISMA-P items.
Key PROSPERO requirements:
- The review must be about health or health-related topics
- The review must have a health-related outcome
- Data extraction must not have started at the time of registration
- The protocol is reviewed by PROSPERO staff before publication
For non-health systematic reviews, alternatives include OSF Registries, Research Registry, or the Campbell Collaboration's registry for social science reviews.
Common Protocol Mistakes to Avoid
-
Vague eligibility criteria: "Relevant studies will be included" is not a criterion. Each criterion must be specific and operationalized.
-
Incomplete search strategy: Listing databases without providing the actual search terms. At least one complete strategy must be documented.
-
No plan for disagreements: Failing to specify how conflicts between reviewers will be resolved at each stage.
-
Missing outcome definitions: Listing outcomes without defining how they will be measured or at what time points.
-
No contingency for heterogeneity: Stating that meta-analysis will be performed without specifying conditions under which it would not be appropriate.
-
Registering after starting: Submitting to PROSPERO after screening has already begun undermines the purpose of protocol registration.
From Protocol to Final Report
Your protocol directly informs your final PRISMA 2020 report. The eligibility criteria, search strategy, and synthesis methods documented in the protocol should be reproduced in the methods section of your completed review. Any deviations from the protocol must be documented and justified.
When you reach the reporting stage, create your PRISMA flow diagram using our free PRISMA 2020 diagram builder to document the study selection process that your protocol predefined.
For the complete process of conducting the systematic review after your protocol is finalized, see our step-by-step guide to conducting a systematic review.
Frequently Asked Questions
Is protocol registration mandatory for systematic reviews?
It is not universally mandatory, but it is strongly recommended and increasingly required. Many journals require a protocol registration number for submission, PROSPERO registration is expected for Cochrane reviews, and funding bodies typically require it. Unregistered reviews are viewed with greater skepticism regarding methodological transparency.
Can I change my protocol after registration?
Yes, but all amendments must be documented with the date, description, and rationale. PROSPERO allows amendments and tracks version history. Minor clarifications are expected; major changes to primary outcomes or eligibility criteria after data collection has begun require strong justification.
What is the difference between PRISMA-P and PRISMA 2020?
PRISMA-P guides the reporting of planned systematic reviews (protocols), while PRISMA 2020 guides the reporting of completed systematic reviews. PRISMA-P has 17 items focused on planned methods; the all 27 PRISMA checklist items cover results and discussion in addition to methods.
How long should a systematic review protocol be?
A typical protocol is 3,000-5,000 words. It should be detailed enough that another research team could replicate your planned methods. Published protocols in journals like Systematic Reviews or BMJ Open typically range from 2,500-4,000 words excluding references.
Should I publish my protocol in a journal?
Publishing your protocol in a peer-reviewed journal provides additional methodological scrutiny and a citable reference. Journals that accept systematic review protocols include Systematic Reviews, BMJ Open, JMIR Research Protocols, and Medicine. However, PROSPERO registration alone is sufficient for most purposes.