Planning your systematic review? Our PROSPERO Formatter tool structures your protocol fields for quick, error-free PROSPERO registration.
Key Takeaways
A systematic review protocol is a pre-specified plan that defines your research question, methods, and analysis strategy before you begin, preventing outcome-switching and selective reporting
The PRISMA-P (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols) checklist by Shamseer et al. (2015) provides a 17-item framework for complete protocol reporting
Key protocol sections include PICO/PCC framework, search strategy, eligibility criteria, data extraction plan, risk of bias assessment, and synthesis method
Want expert help planning your review?
We develop protocols, register on PROSPERO, build search strategies, and deliver the complete review. Start strong, finish published.
Planning Your Review? We Can Take It From Protocol to Publication.
Protocol development, PROSPERO registration, and a comprehensive search strategy. Our PhD experts plan and execute the entire review so you start on the right track.
Dr. Sarah Mitchell holds a PhD in Biostatistics from Johns Hopkins Bloomberg School of Public Health and has over 15 years of experience in systematic review methodology and meta-analysis. She has authored or co-authored 40+ peer-reviewed publications in journals including the Journal of Clinical Epidemiology, BMC Medical Research Methodology, and Research Synthesis Methods. A former Cochrane Review Group statistician and current editorial board member of Systematic Reviews, Dr. Mitchell has supervised 200+ evidence synthesis projects across clinical medicine, public health, and social sciences.
Need help building a focused research question before drafting your protocol? Try our free PICO Framework Builder to define your population, intervention, comparator, and outcome.
Research Gold helps researchers develop rigorous protocols, run comprehensive searches, and complete every stage of the systematic review process. Start your project today.
Planning Your Review? We Can Take It From Protocol to Publication.
Protocol development, PROSPERO registration, and a comprehensive search strategy. Our PhD experts plan and execute the entire review so you start on the right track.
Starting from $895. Final quote after we review your project.
Register your protocol on PROSPERO before you begin screening, registration creates a public, time-stamped record of your pre-specified methods
Protocol amendments are expected and acceptable, document every change with a date, rationale, and description of impact on the review
Research Gold's free PROSPERO Formatter and PICO Framework Builder tools help you draft and structure your protocol efficiently
A systematic review protocol template is a structured document that pre-specifies every methodological decision you will make before you begin your review. The protocol defines your research question, eligibility criteria, search strategy, screening process, data extraction plan, risk of bias assessment, and synthesis approach, all documented and registered before a single study is screened. It is the blueprint that separates a rigorous systematic review from an ad hoc literature summary.
At Research Gold, we always develop the protocol first, it is the blueprint that guides every subsequent decision in the review process. Without a registered protocol, reviewers and editors have no way to verify that your methods were pre-specified rather than shaped by the results you found. This guide walks you through every section of a systematic review protocol, maps each section to the PRISMA-P checklist, and shows you how to register on PROSPERO.
What Is a Systematic Review Protocol
The protocol serves the same function in evidence synthesis that a study protocol serves in clinical trials. Just as a clinical trial protocol prevents researchers from changing their primary outcome after seeing the data, a systematic review protocol prevents reviewers from adjusting eligibility criteria, analysis methods, or outcome definitions based on what the included studies happen to show. This transparency is fundamental to the credibility of the review.
A protocol is not a rough plan or a set of notes. It is a formal document that contains enough detail for another researcher to execute the review independently. When published or registered, it becomes a public commitment to a specific set of methods, and any departures from that commitment must be documented and justified.
The Cochrane Handbook for Systematic Reviews of Interventions (Chapter 3) states that writing the protocol is one of the most important steps in the review process. The Handbook emphasizes that the protocol should be developed iteratively with input from the review team, methodologists, and information specialists before it is finalized and registered.
Why You Need a Protocol
A pre-registered protocol prevents five concrete biases. Source: Cochrane Handbook v6.5; PRISMA-P 2015.
Writing a protocol takes time and effort, and researchers under pressure to publish sometimes question whether it is necessary. The evidence is clear: a protocol strengthens every aspect of your systematic review.
Prevents selective reporting. Without a pre-specified protocol, there is no way to verify that you did not change your eligibility criteria after seeing which studies were available, or add an outcome measure after discovering a significant result. Selective reporting undermines the entire purpose of systematic review methodology. A registered protocol creates a time-stamped record of your original intentions.
Reduces methodological bias. When decisions about inclusion criteria, search scope, and analysis methods are made in advance, they cannot be influenced by the data. This is the same principle that underpins pre-registration of clinical trials and has been widely adopted across the evidence synthesis community.
Improves team coordination. For reviews with multiple authors, the protocol is the reference document that keeps everyone aligned. When a disagreement arises during screening, does this study meet the age criterion?, the protocol provides the answer. Without it, decisions become inconsistent and subjective.
Increases publication credibility. Many high-impact journals now require or strongly recommend protocol registration before they will consider a systematic review for publication. Editors and peer reviewers check whether your methods match your protocol. Deviations without justification raise red flags.
Avoids duplication. Registering your protocol on PROSPERO makes your planned review visible to the research community. Other teams considering a similar review can see that yours is underway, reducing wasted effort and unnecessary duplication.
Supports reproducibility. A detailed protocol allows other researchers to evaluate your methodological choices, replicate your search strategy, and assess whether your review can be updated in the future. Reproducibility is a cornerstone of the scientific method, and the protocol is where it begins for systematic reviews.
The PRISMA-P Checklist: 17 Items for Protocol Reporting
The PRISMA-P (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols) checklist was published by Shamseer et al. (2015) to standardize the content and reporting of systematic review protocols. It contains 17 items organized into three sections: administrative information, introduction, and methods. Every item should be addressed in your protocol.
#
PRISMA-P Item
Section
What to Include
1
Title identification
Admin
Identify the document as a protocol; include "protocol" in the title
2
Registration
Admin
Registration number and registry name (e.g., PROSPERO CRD42026XXXXXX)
3a
Authors and contact
Admin
Full names, affiliations, corresponding author contact details
3b
Contributions
Admin
Each author's planned contribution to the review
4
Amendments
Admin
Mechanism for documenting protocol amendments
5a
Sources of support
Admin
Shamseer et al. (2015) designed PRISMA-P so that each item maps directly to a section of the protocol document. Use this checklist as a writing guide: address each item in order, and your protocol will be complete. The EQUATOR Network hosts the full checklist with elaboration and explanation for each item.
Aligned with PRISMA-P 2015. Source: Shamseer et al., 2015, BMJ 349:g7647; Cochrane Handbook v6.5.
While the PRISMA-P checklist tells you what to report, the following sections describe how to write each component with sufficient methodological detail. Each section corresponds to one or more PRISMA-P items.
Research Question Using PICO or PCC
The research question is the foundation of your protocol. For intervention reviews, structure it using the PICO framework: Population, Intervention, Comparator, Outcome. For scoping reviews or qualitative evidence syntheses, use the PCC framework: Population, Concept, Context.
A well-constructed PICO question constrains every other decision in the protocol. The Population defines your eligibility criteria. The Intervention and Comparator define what you are comparing. The Outcome defines what you will measure and synthesize. If your PICO is vague, your protocol will be vague, and your review will suffer from inconsistent decision-making.
For example, a PICO question might read: "In adults aged 18 years and older with type 2 diabetes (P), does structured exercise training (I) compared with usual care (C) reduce HbA1c levels (O)?" Every word in that question carries methodological implications. "Adults aged 18 years and older" means you will exclude pediatric studies. "Structured exercise training" means you need to define what counts as structured. "Usual care" means you need to define what you accept as a comparator. "HbA1c levels" means you need studies that report this specific outcome measure.
Use our free try the pico framework builder to structure your research question before drafting the rest of the protocol. The tool helps you define each element precisely and generates a formatted PICO statement ready for your protocol document.
Pro Tip
Write with PRISMA-P, then register on PROSPERO
Draft your protocol using the PRISMA-P 17-item checklist as a writing guide. Once each item is addressed, transfer the relevant fields into the PROSPERO registration form. This ensures completeness before registration.
Pro Tip
Name your risk of bias tool explicitly
Do not write 'risk of bias will be assessed.' State the exact tool, RoB 2 for randomized trials, ROBINS-I for non-randomized studies, Newcastle-Ottawa Scale for observational studies. Reviewers and editors expect specificity.
Pro Tip
Pre-specify subgroup analyses
If you plan to analyze results by age group, severity, intervention dose, or geographic region, state those subgroup analyses in the protocol. Post-hoc subgroup analyses are viewed with skepticism and weaken your review's credibility.
Frequently Asked Questions
5
It is required by Cochrane for all Cochrane Reviews and strongly recommended by PRISMA 2020 for all systematic reviews. Many journals now require protocol registration before they will consider a systematic review manuscript for publication.
PRISMA-P (2015) guides protocol writing, what to include when you plan your review. PRISMA 2020 guides the reporting of the completed review. They are complementary: PRISMA-P is for the protocol stage, PRISMA 2020 is for the manuscript stage.
Yes. Protocol amendments are expected as reviews progress. The key requirement is transparency: document every change with the date, the original plan, the revised plan, and the rationale for the change.
Detailed enough for another researcher to reproduce it. Include the databases you will search, the date range, language restrictions, and the full search string for at least one database (typically MEDLINE). The complete strategy for all databases can be appended as a supplement.
PROSPERO is the most widely used free registry for systematic review protocols. You can also publish protocols as peer-reviewed articles in journals such as Systematic Reviews, BMJ Open, or JMIR Research Protocols.
Data extraction process: forms, piloting, variables to extract
12
Outcomes and prioritization
Methods
Primary and secondary outcomes, how they will be measured
13
Risk of bias in individual studies
Methods
Tool(s) to be used (RoB 2, ROBINS-I, NOS), domains assessed
14
Data synthesis
Methods
Narrative synthesis plan, meta-analysis model if applicable, heterogeneity assessment
15a
Meta-bias(es)
Methods
Plans to address publication bias and selective outcome reporting
15b
Confidence in cumulative evidence
Methods
GRADE or equivalent certainty assessment approach
16
Dissemination plans
Methods
How results will be published and shared
17
Ethics and dissemination
Methods
Ethical approval (usually not required) and dissemination strategy
Eligibility Criteria
Eligibility criteria translate your PICO or PCC framework into explicit inclusion and exclusion rules. Every study you encounter during screening will be evaluated against these criteria, so they must be specific, measurable, and unambiguous.
State your criteria across the standard PICOS dimensions plus study design:
Dimension
Inclusion Criteria
Exclusion Criteria
Population
Adults aged 18+ with confirmed type 2 diabetes
Type 1 diabetes, gestational diabetes, prediabetes
Intervention
Structured exercise programs (minimum 4 weeks duration, at least 2 sessions per week)
Active comparator interventions (e.g., dietary programs)
Outcome
HbA1c measured at any time point post-intervention
Studies not reporting HbA1c
Study design
Randomized controlled trials
Non-randomized studies, case reports, reviews
Other
Published in English, peer-reviewed
Conference abstracts without full-text, animal studies
Each criterion should be justified. Why exclude non-English studies? Because resources for translation are unavailable and the expected impact on findings is minimal. Why require a 4-week minimum? Because shorter interventions are unlikely to produce measurable changes in HbA1c. Peer reviewers will ask these questions, answer them in the protocol.
The search strategy is arguably the most technically demanding section of the protocol. It must be comprehensive enough to identify all relevant studies, reproducible enough for another researcher to execute independently, and documented with sufficient detail for peer review.
Your protocol should specify the databases you will search (e.g., MEDLINE via PubMed, Embase via Ovid, CENTRAL, PsycINFO, CINAHL), any supplementary sources (trial registries, grey literature databases, reference lists of included studies, citation searching), the date range for the search, and any language or publication type restrictions.
Include the full search string for at least one database. For MEDLINE, this typically combines MeSH terms and free-text terms using Boolean operators (AND, OR, NOT). The search strategy is usually developed in collaboration with an information specialist or librarian, their expertise ensures that your search is both sensitive (capturing all relevant studies) and specific (excluding irrelevant records).
A well-written search strategy section looks something like this: "We will search MEDLINE (via PubMed), Embase (via Ovid), CENTRAL (via Cochrane Library), CINAHL (via EBSCOhost), and PsycINFO (via Ovid) from inception to the date of the search. The search strategy was developed in consultation with a research librarian and combines controlled vocabulary terms with free-text synonyms. The full MEDLINE search strategy is provided in Appendix 1. We will also search ClinicalTrials.gov and the WHO International Clinical Trials Registry Platform for ongoing or completed trials. Reference lists of included studies and relevant systematic reviews will be screened for additional eligible studies."
The screening section describes how you will select studies from the search results. State the screening software you will use (e.g., Covidence, Rayyan, EPPI-Reviewer), the number of reviewers who will independently screen titles and abstracts, the process for resolving disagreements, and whether you will pilot the screening criteria on a sample of records before full screening begins.
Best practice requires two independent reviewers to screen all records at both the title/abstract stage and the full-text stage. Disagreements should be resolved by discussion, with a third reviewer available to arbitrate if consensus cannot be reached. Pilot screening on a random sample of 50-100 records helps calibrate the team and identify ambiguities in the eligibility criteria before committing to full screening.
State the anticipated flow of records. While you will not know exact numbers at the protocol stage, you can describe the process: "Records identified from database searches will be imported into reference management software and deduplicated. Unique records will be screened independently by two reviewers at the title and abstract level. Records selected by either reviewer will proceed to full-text screening. Full-text articles will be assessed independently by two reviewers against the eligibility criteria. Reasons for exclusion at the full-text stage will be recorded and reported in a PRISMA flow diagram."
Data Extraction
The data extraction section specifies what data you will collect from each included study and how. Define the variables to be extracted, typically study characteristics (author, year, country, setting, design), participant characteristics (sample size, age, sex, disease severity), intervention details (type, duration, frequency, intensity), comparator details, and outcome data (means, standard deviations, event counts, follow-up duration).
Describe the extraction process: will you use a standardized form? Will it be piloted? Will two reviewers extract data independently, or will one extract and a second verify? State how you will handle missing data, will you contact study authors? Use imputation methods? Exclude studies with insufficient data?
The protocol should include a draft data extraction form or at minimum list every variable that will be extracted. This level of detail prevents post-hoc additions of variables that were not part of the original plan.
Risk of Bias Assessment
Name the specific risk of bias tool you will use for each study design included in your review. Do not write "risk of bias will be assessed", this is too vague. State the exact tool and version.
Study Design
Risk of Bias Tool
Reference
Randomized controlled trials
RoB 2 (Revised Cochrane Risk of Bias Tool)
Sterne et al. (2019)
Non-randomized studies of interventions
ROBINS-I
Sterne et al. (2016)
Observational (cohort, case-control)
Newcastle-Ottawa Scale (NOS)
Wells et al.
Diagnostic accuracy studies
QUADAS-2
Whiting et al. (2011)
Qualitative studies
CASP Qualitative Checklist
Critical Appraisal Skills Programme
Describe the assessment process: two independent reviewers, with disagreements resolved by discussion or a third reviewer. State how risk of bias results will be incorporated into your synthesis, will you conduct sensitivity analyses excluding high risk-of-bias studies? Will you use GRADE to downrate certainty of evidence based on risk of bias?
Synthesis Method
The synthesis section describes how you will combine and present the findings of included studies. This is where you distinguish between narrative (qualitative) synthesis and statistical (quantitative) meta-analysis, and where you pre-specify the details of each approach.
For narrative synthesis, describe the framework you will follow (e.g., Synthesis Without Meta-analysis, SWiM guidelines by Campbell et al., 2020). State how you will group and compare studies, what visual displays you will use (harvest plots, effect direction plots), and how you will assess the consistency of findings across studies.
For meta-analysis, specify the statistical model (random-effects or fixed-effect, with justification), the effect measure (risk ratio, odds ratio, mean difference, standardized mean difference), the software you will use (R metafor, RevMan, Stata), and how you will assess and explore heterogeneity (I-squared statistic, prediction intervals, subgroup analyses, meta-regression).
Pre-specify the conditions under which you will or will not conduct a meta-analysis. A common statement is: "We will conduct meta-analysis when at least two studies report the same outcome using comparable measures and the clinical and methodological heterogeneity across studies is judged to be sufficiently low. When meta-analysis is not appropriate, we will present results using narrative synthesis following the SWiM guidelines."
State your planned subgroup analyses and sensitivity analyses. Subgroup analyses might compare results by age group, intervention duration, or risk of bias category. Sensitivity analyses might exclude studies with high risk of bias, studies with imputed data, or studies published only as conference abstracts.
Registering Your Protocol on PROSPERO
PROSPERO (International Prospective Register of Systematic Reviews) is a free, publicly accessible database maintained by the Centre for Reviews and Dissemination at the University of York. Registering your protocol on PROSPERO creates a permanent, time-stamped record of your pre-specified methods. PROSPERO is the registry for systematic review protocols and is recognized by all major journals and reporting guidelines.
When to register. Register your protocol before you begin screening. PROSPERO will not accept registrations for reviews that have already completed data extraction. The ideal time to register is after you have finalized your protocol and piloted your search strategy, but before you run the final search and begin screening records.
What to include. The PROSPERO registration form asks for your review title, team members, research question, eligibility criteria, information sources, search strategy summary, study selection process, data extraction plan, risk of bias tool, synthesis plan, and funding source. Most of these fields map directly to sections in your protocol document.
Registration process. Create an account at the PROSPERO website (crd.york.ac.uk/prospero), complete the online form, and submit. A PROSPERO editor will review your submission and may request clarifications. Once accepted, your review receives a unique registration number (e.g., CRD42026XXXXXX) that you include in your manuscript when submitting the completed review for publication.
Our free PROSPERO Formatter tool helps you organize and format your protocol information for the PROSPERO registration form. It maps your protocol sections to the PROSPERO fields so you do not miss any required information. For a detailed registration walkthrough, see our guide on PROSPERO registration step by step.
Protocol Amendments
No protocol survives contact with the evidence base entirely unchanged. Protocol amendments are a normal part of the systematic review process, the important thing is not to avoid changes, but to document them transparently and justify them clearly.
Common reasons for protocol amendments include discovering that the pre-specified search strategy misses a relevant database, realizing that an eligibility criterion is ambiguous and needs clarification, finding that no studies report the planned primary outcome (necessitating a change in outcome hierarchy), or deciding that a pre-specified subgroup analysis is not feasible due to insufficient data.
For each amendment, document the following: the date of the change, the original protocol statement, the revised statement, and the rationale for the change. If the review is registered on PROSPERO, update the registration record to reflect the amendment. If the protocol is published as a journal article, note the amendment in the final review manuscript.
The Cochrane Handbook (Chapter 3) distinguishes between amendments made before encountering the evidence (generally unproblematic) and amendments made after seeing the data (potentially introducing bias). Amendments made after data extraction has begun require particular scrutiny and should be clearly flagged as post-hoc changes.
Transparency about amendments actually strengthens your review's credibility. Readers understand that methods sometimes need adjustment, they are more concerned about undisclosed changes than about documented ones. A review that acknowledges its protocol deviations is more trustworthy than one that claims perfect adherence to an implausible degree.
Common Mistakes in Protocol Writing
Even experienced researchers make avoidable errors when writing systematic review protocols. Awareness of these pitfalls helps you produce a stronger, more defensible protocol from the start.
Vague eligibility criteria. Stating "adults with diabetes" without specifying type 1 versus type 2, age range, or diagnostic criteria creates ambiguity that leads to inconsistent screening decisions. Every criterion should be specific enough that two independent reviewers would make the same inclusion/exclusion decision for any given study.
Missing the search strategy. Some protocols describe which databases will be searched but omit the actual search string. The PRISMA-P checklist requires a complete search strategy for at least one database. Without it, your search cannot be evaluated for sensitivity and specificity, and it cannot be reproduced.
Unnamed risk of bias tools. Writing "quality assessment will be performed" tells the reader nothing. Name the tool (RoB 2, NOS, ROBINS-I), specify who will use it (two independent reviewers), and describe how the results will inform your synthesis (sensitivity analysis, GRADE assessment).
No synthesis plan. Stating that "results will be synthesized" without specifying narrative versus meta-analysis, the statistical model, the effect measure, or heterogeneity assessment is insufficient. The synthesis plan is one of the most important sections of the protocol because it is the section most susceptible to post-hoc manipulation.
Ignoring PRISMA-P. Some researchers write protocols based on their own idea of what a protocol should contain, missing critical items like amendment procedures, dissemination plans, or confidence in cumulative evidence (GRADE). Using the PRISMA-P checklist as a writing guide ensures completeness.
Registering too late. PROSPERO requires registration before data extraction begins. Researchers who delay registration until the review is nearly complete miss the entire point of prospective registration, and may have their registration rejected. Register early, ideally before screening begins.
The difference between a publishable systematic review and a desk-rejected one often traces back to the protocol. A rigorous, PRISMA-P-compliant protocol registered on PROSPERO signals to editors and reviewers that your review was planned with the same care and transparency expected of any prospective research study.
For a comprehensive overview of the PRISMA 2020 reporting guidelines that govern how you report the completed review, see our guide on PRISMA 2020 guidelines explained. And for the detailed steps of conducting the review itself, from search execution through to manuscript preparation, our our guide to step-by-step systematic review guide picks up where the protocol leaves off. Researchers working through the PROSPERO registration process specifically should consult our dedicated learn about prospero registration guide, and those still refining their research question will benefit from our learn about pico framework guide.
Diagnostic test accuracy meta-analysis pools paired sensitivity and specificity estimates across studies using bivariate or HSROC models. This guide covers when bivariate methods are required, the role of the summary ROC curve, threshold effects, QUADAS-2 risk of bias, and software options including R mada, R metafor, and Stata midas.
Individual participant data meta-analysis pools raw participant-level data across studies instead of aggregate effect estimates. This guide covers when an IPD meta-analysis is justified, one-stage versus two-stage models, data acquisition and harmonization, software (Stata ipdmetan, R metafor and ipdma), reporting under PRISMA-IPD, and common pitfalls in subgroup and interaction testing.
Research Gold's medical translation service delivers publication-grade translations of manuscripts, clinical trial protocols, regulatory submissions, and patient-reported outcome instruments. Translators are PhD-credentialed and ATA-certified, working through a two-translator translation, edit, proofread workflow. English-Arabic is our deepest pair, with strong coverage of Spanish, Mandarin, French, German, Portuguese, Japanese, Korean, and Turkish. From $0.12 per source word.
