PROSPERO Registration: Step-by-Step Guide for 2026

PROSPERO registration is the process of submitting your systematic review protocol, including your research question PICO, search strategy, eligibility criteria, and analysis plan, to the international prospective register of systematic reviews maintained by the Centre for Reviews and Dissemination (CRD) at the University of York. Registration is free, approval typically takes 2 to 5 business days, and your record receives a unique PROSPERO registration number (format CRD42XXXXXXX) that journals and peer reviewers use to verify your review was planned before results were known.

We have completed hundreds of PROSPERO registrations, the most common issue we see is researchers registering after screening has already begun, which triggers the retrospective flag and weakens the credibility of an otherwise sound protocol. This guide walks through every field with weak versus strong examples so you get it right the first time.

How PROSPERO Compares to Other Registries

The purpose of protocol registration is straightforward: by recording your research question, methods, and analysis plan before you see the data, you demonstrate that your conclusions were not shaped by results. This transparency is the foundation of credible evidence synthesis. Every evidence synthesis protocol, whether for a systematic review, scoping review, or rapid review, benefits from prospective registration.

PROSPERO currently holds over 500,000 registered protocols from researchers worldwide. Each registration receives a unique identifier (CRD42XXXXXXX) that you cite in your manuscript abstract, methods section, and PRISMA flow diagram. Journals, peer reviewers, and guideline developers use this identifier to verify that your review was planned prospectively.

For a complete overview of the protocol development process that feeds into registration, see our systematic review protocol guide.

Why You Should Register on PROSPERO

PROSPERO registration prevents outcome-switching, demonstrates methodological rigor to journals, and reduces duplication of effort across the research community. A systematic review requires protocol registration to establish credibility before results are known.

Prevents outcome-switching. When you register your primary outcomes, analysis methods, and eligibility criteria before screening, you create a public record that reviewers can check. If your published review deviates from the registered protocol without documented amendments, peer reviewers will flag it, and rightly so. Prospective registration eliminates the suspicion that you adjusted your methods after seeing which results looked favorable.

Demonstrates rigor to journals. Many high-impact journals, including BMJ, The Lancet, JAMA, and all Cochrane publications, either require or strongly recommend PROSPERO registration. Submitting a manuscript with a PROSPERO ID signals to editors that your review followed established methodological standards. Even journals that do not mandate registration view it favorably during peer review.

Reduces duplication. Before starting a new systematic review, researchers are expected to search PROSPERO CRD for ongoing reviews on the same topic. The ability to register systematic review protocols publicly is one of PROSPERO's most important contributions to research transparency. Your registration helps the broader community avoid duplicating effort, and it helps you identify potential collaborators working on similar questions.

Strengthens your peer review defense. When reviewers question your methodology, pointing to a prospectively registered protocol is one of the strongest responses available. The protocol was locked in before you had results, that is difficult to argue with.

For guidance on structuring the research question that anchors your registration, see our our guide to pico framework guide.

When to Register on PROSPERO

You should register your systematic review on PROSPERO after finalizing your protocol but before beginning title and abstract screening. This is the optimal window, your methods are set, but you have not yet been exposed to study results that could bias your approach.

Before screening begins (prospective). The gold standard. CRD explicitly designed PROSPERO for prospective registration, recording your plan before you encounter the evidence. Registrations submitted before screening carry no flags or caveats.

After screening has started (retrospective). If you register after screening has begun, CRD will flag your registration as retrospective. This flag appears publicly on your PROSPERO record and in any exports. While retrospective registration is still better than no registration at all, it weakens your methodological credibility because it cannot rule out the possibility that results influenced your protocol decisions. PROSPERO registration should occur before study screening begins, retrospective registrations are flagged by CRD, indicating that the protocol was finalized after seeing study results (CRD, University of York).

Cochrane reviews. If you are conducting a Cochrane systematic review, your protocol is registered through the Cochrane Database of Systematic Reviews (CDSR) rather than PROSPERO. Cochrane has its own editorial and registration process. Do not register a Cochrane review on PROSPERO.

Data extraction or later. If you have already completed data extraction or analysis before considering registration, PROSPERO will still accept your submission, but the retrospective flag will be prominent. At this stage, registration serves as a record rather than a prospective commitment.

Want your PROSPERO registration handled professionally? Our team writes and submits complete protocol registrations, including search strategy documentation and methodology sections that satisfy reviewers. arrange a free research project discussion, or see our systematic review registration support.

Step-by-Step PROSPERO Registration Walkthrough

The PROSPERO registration form contains 22 required fields that map to the PRISMA-P protocol checklist (Shamseer et al., 2015), covering the complete methodological plan for the systematic review. Below is a field-by-field walkthrough of the 10 major steps.

Step 1, Create a PROSPERO Account

Navigate to crd.york.ac.uk/prospero and click "Register." You will need an institutional email address, your ORCID (recommended but not required), and your institutional affiliation. Account creation takes approximately five minutes. Once your account is confirmed via email, you can log in and begin a new registration.

If you already have a PROSPERO account from a previous registration, use the same credentials. Multiple registrations can be managed from a single account.

Step 2, Start a New Registration

After logging in, click "New Registration" from your dashboard. PROSPERO will ask whether your review has already started screening. Answer honestly, selecting "Yes" when screening is underway triggers the retrospective flag. Selecting "No" when screening has not begun ensures your registration is marked as prospective.

You will also confirm that your review is health-related. If it is not, PROSPERO will redirect you to alternative registries such as OSF.

Step 3, Review Title and Research Question (PICO/PCC)

Enter your review title exactly as it will appear in your manuscript. The title should be descriptive and include key terms that help other researchers find your registration when checking for duplication.

For your research question, use the PICO framework (Population, Intervention, Comparator, Outcome) for intervention reviews or the PCC framework (Population, Concept, Context) for scoping reviews. PROSPERO provides separate fields for each PICO element. Be specific: "Adults aged 18+ with type 2 diabetes" is stronger than "Patients with diabetes."

The PICO framework structures research questions for PROSPERO registration, each element maps to a dedicated registration field. For detailed guidance on constructing your PICO, see our our guide to pico framework guide.

Step 4, Anticipated Start and Completion Dates

Enter the date you plan to begin screening (or have already begun, if registering retrospectively) and your anticipated completion date. These dates help CRD track the progress of registered reviews and allow other researchers to gauge when your results might become available.

Be realistic. Systematic reviews typically take 6–18 months from protocol to submission. If your timeline changes, you can update these dates through the amendment process.

Step 5, Search Strategy Details

Describe your search strategy in sufficient detail that another researcher could replicate it. This includes:

• Databases: List every database you plan to search (e.g., MEDLINE via PubMed, Embase, CINAHL, Cochrane CENTRAL, PsycINFO, Scopus). List ALL databases, omitting one that you later search raises questions about post-hoc additions.
• Search terms: Provide your key search terms, Boolean operators, and any controlled vocabulary (MeSH terms, Emtree terms). You do not need to paste full search strings, but enough detail to demonstrate rigor.
• Date restrictions: State any date limits on your search and justify them.
• Language restrictions: Specify whether you will include non-English studies and how you will handle translation.
• Grey literature: Describe any plans to search grey literature sources (conference abstracts, theses, government reports, clinical trial registries).
• Pilot search date: Include the date you ran your pilot search. This demonstrates that your strategy has been tested and refined.

For help building an effective search strategy protocol, see our Boolean search strategy guide.

Step 6, Eligibility Criteria (Inclusion/Exclusion)

State your eligibility criteria clearly and completely. PROSPERO provides fields for both inclusion and exclusion criteria. Structure these around your PICO elements:

Be explicit about study design restrictions. If you are including only randomized controlled trials, state it. If you are including observational studies, specify which types (cohort, case-control, cross-sectional).

Step 7, Data Extraction Plan

Describe what data you will extract from included studies and how. PROSPERO expects you to specify:

• Data items: List the variables you will extract (study characteristics, participant demographics, intervention details, outcome measures, effect sizes, confidence intervals).
• Extraction method: State whether you will use a standardized form, piloted on a subset of studies, with independent extraction by two reviewers.
• Disagreement resolution: Describe how discrepancies between extractors will be resolved (discussion, third reviewer, consensus).

The data extraction plan is one of the fields reviewers examine most closely. A vague description ("Data will be extracted") is insufficient. Specify the tool or form you will use and the process for ensuring accuracy.

Step 8, Risk of Bias Approach

Name the specific risk of bias tool you will use. Do not write "An appropriate tool will be used", name it explicitly:

Specify how many reviewers will independently assess risk of bias, how disagreements will be resolved, and how results will be presented (summary table, traffic light plot, or both). The risk of bias approach is a critical field, vague answers here are the single most common reason CRD requests revisions to a registration.

Step 9, Synthesis Method (Narrative or Meta-Analysis)

Describe your planned synthesis method. If you plan a meta-analysis, specify:

• Effect measure: Risk ratio, odds ratio, mean difference, standardized mean difference, or hazard ratio.
• Model: Fixed-effect or random-effects (and justify your choice).
• Heterogeneity assessment: I-squared statistic, Cochran Q test, prediction intervals.
• Subgroup analyses: Pre-specify any planned subgroup analyses and the rationale for each.
• Sensitivity analyses: Describe what sensitivity analyses you will conduct (e.g., excluding high risk-of-bias studies, excluding outliers).
• Publication bias: State whether you will assess publication bias (funnel plot, Egger test) and the minimum number of studies required.

If a meta-analysis is not feasible (due to high heterogeneity or insufficient studies), describe your narrative synthesis approach. The Synthesis Without Meta-analysis (SWiM) reporting guideline (Campbell et al., 2020) provides a useful framework. For a detailed walkthrough of meta-analysis methodology, see our complete guide to meta-analysis.

Step 10, Review and Submit

Before submitting, review every field for completeness and consistency. Common issues that delay CRD approval include:

• Inconsistencies between the PICO fields and the eligibility criteria
• Vague risk of bias descriptions that do not name a specific tool
• Missing databases in the search strategy field
• Eligibility criteria that do not specify study design restrictions

Once you are satisfied, click "Submit for Assessment." CRD staff will review your registration and either approve it (typically within 2–5 business days) or return it with requests for clarification. You will receive an email notification when your registration is approved, along with your unique PROSPERO registration number (CRD42XXXXXXX).

Weak vs Strong PROSPERO Entries: Field-by-Field Examples

The single most common reason the CRD returns a registration for revision is vagueness, not error. The fields below are where reviewers most often ask for more detail. For each, here is the kind of entry that triggers a revision request next to one that is accepted as submitted.

Risk of bias assessment. This is the field CRD flags most often.

• Weak: "Risk of bias will be assessed using a standard tool."
• Strong: "Risk of bias in randomized trials will be assessed with the Cochrane RoB 2 tool across its five domains; non-randomized studies will be assessed with ROBINS-I. Two reviewers will assess each study independently, with disagreements resolved by a third reviewer. Overall risk-of-bias judgements will inform sensitivity analyses excluding high-risk studies."

The strong version names the tool, ties it to the design, and states the process and how the result will be used. The weak version names nothing the CRD can hold you to.

Synthesis / data synthesis method.

• Weak: "Data will be combined and analyzed."
• Strong: "If studies are sufficiently homogeneous clinically and statistically (I-squared below 75%), we will pool effects in a random-effects meta-analysis (REML estimator) reporting the pooled estimate, 95% confidence interval, and 95% prediction interval. Where pooling is inappropriate, we will conduct a structured narrative synthesis following SWiM guidance."

Pre-committing to a decision rule (the I-squared threshold and the fallback) is exactly what prevents an accusation of outcome-driven analysis later.

Search strategy.

• Weak: "We will search the major databases."
• Strong: "We will search MEDLINE (Ovid), Embase, CENTRAL, and CINAHL from inception to the search date, with no language restriction, supplemented by reference-list checking of included studies and a grey-literature search of ClinicalTrials.gov. The full MEDLINE strategy is provided in a supplementary file."

Naming the databases, the platform, the date limits, and the supplementary strategy file signals a search a reviewer could reproduce.

Eligibility criteria.

• Weak: "Adults with the condition who received the intervention."
• Strong: "Population: adults aged 18 or older with a clinical diagnosis of type 2 diabetes. Intervention: any structured exercise program of at least 12 weeks. Comparator: usual care or no exercise. Outcomes: change in HbA1c (primary); body mass index and adverse events (secondary). Designs: randomized controlled trials only."

A PICO that a second reviewer could apply to a borderline abstract without asking you a question is the standard to aim for.

Anticipated outcomes (primary vs secondary). State the single primary outcome explicitly and list secondary outcomes separately. Registering five "primary" outcomes is itself a flag, because it leaves room to promote whichever outcome turns out significant. Pre-specifying one primary outcome is the protection against that suspicion.

Use these as templates rather than copy-paste text, then continue to the submission-tightening tips below.

Tips for a Strong PROSPERO Registration

A strong registration goes beyond simply filling in the 22 required fields. The following practices, drawn from our experience with hundreds of registrations, will strengthen your submission and reduce the likelihood of CRD requesting revisions.

Use your PRISMA-P checklist to prepare answers. The 17-item PRISMA-P checklist (Shamseer et al., 2015) maps directly to PROSPERO's registration fields. Complete PRISMA-P first, and you will have all the information organized before you open the registration form. For a complete walkthrough of the PRISMA-P protocol checklist, see our PRISMA-P guide.

Be specific about your risk of bias tool. Name it: RoB 2, ROBINS-I, Newcastle-Ottawa Scale, QUADAS-2, or CASP. Do not write "An appropriate tool will be selected." CRD reviewers will return your registration if the risk of bias approach is vague.

List ALL databases you plan to search. If you search five databases, list all five. If you add a database after registration, document it as an amendment. Omitting databases from your registration and later including them in your published review raises questions about selective searching.

Include your pilot search date. Adding the date you ran your pilot search demonstrates that your search strategy has been tested and refined. This small detail signals methodological maturity to CRD reviewers and, later, to journal peer reviewers.

Pre-specify your subgroup and sensitivity analyses. Post-hoc subgroup analyses are one of the most common sources of outcome-switching. By registering them in PROSPERO, you establish that these analyses were planned before results were available. If you want a pre-specified statistical analysis plan written before you register, a biostatistician can define your primary model, subgroups, and sensitivity analyses so CRD reviewers see a fully planned synthesis.

What Happens After Submission

After you click "Submit for Assessment," your registration enters the CRD review queue. Understanding this process helps you plan your timeline and avoid unnecessary delays.

CRD review (2–5 business days). A CRD reviewer examines your registration for completeness and clarity. They are not evaluating the quality of your research question or methodology, they are checking that you have provided sufficient detail in every required field. If your submission is complete, you receive approval and your PROSPERO registration number via email.

Revision requests. If CRD identifies gaps, they return the registration with specific feedback. Common requests include: name your risk of bias tool, clarify your eligibility criteria, or expand your search strategy description. Address these promptly to avoid delays. Most revision requests can be resolved within 24 hours.

Amendments and updates. After your registration is approved, you can submit amendments at any time. Document what changed, why, and when. PROSPERO maintains a version history, so all amendments are transparent. Transparent protocol amendments are expected, they demonstrate methodological integrity rather than weakness.

Linking to your published review. Once your review is published, update your PROSPERO record with the publication DOI. This creates a direct link between your protocol and your published findings, allowing readers to verify that you followed your prospectively registered plan.

How PROSPERO IDs appear in manuscripts. Your PROSPERO registration number appears in three locations: the manuscript abstract (typically as a parenthetical note), the Methods section (when describing protocol registration), and the PRISMA 2020 flow diagram (in the identification box). A systematic review follows PRISMA 2020 reporting guidelines, and the registration number is a required element of compliant reporting. For a complete breakdown of PRISMA reporting, see understanding our complete systematic review guide.

Whether you are registering your first systematic review protocol or your twentieth, the process is the same: finalize your protocol, complete the 22-field registration form with specific and transparent detail, and submit before screening begins. PROSPERO registration is one of the simplest steps in the systematic review process, yet it is also one of the most impactful, a single registration number carries your methodological credibility from protocol through publication.

For researchers who want professional support with protocol development and registration, Research Gold handles the entire process, from structuring your PICO to submitting a complete PROSPERO registration. Use our try our prospero registration formatter to organize your protocol fields, or explore professional systematic review writing assistance for end-to-end support.

PROSPERO registration requirements can vary. Understand whether Cochrane and non-Cochrane reviews handle registration differently.

PROSPERO now accepts registrations for living systematic reviews, which are continuously updated as new evidence emerges.

Registration timing matters. Understanding how long the full systematic review process takes helps you set realistic milestones in your PROSPERO registration.