Types of Systematic Review Protocols Explained for Researchers
Types of Systematic Review Protocols Explained for Researchers
The seven main types of systematic review protocols are: traditional systematic review, scoping review, rapid review, meta-analysis or meta-synthesis, umbrella review, integrative review, and living systematic review. Each type differs primarily in scope, synthesis depth, speed, and how strictly it follows a pre-specified protocol.
- Traditional systematic review: Exhaustive search, strict eligibility criteria, dual screening, risk-of-bias assessment. Registers on PROSPERO or OSF.
- Scoping review: Maps the breadth of a literature base rather than answering a focused clinical question. Needs a protocol but is often ineligible for PROSPERO; register on OSF instead.
- Rapid review: Streamlines or omits steps (single screener, limited databases) to meet policy deadlines. Protocol must document which shortcuts were taken and how bias is managed.
- Meta-analysis / meta-synthesis: Quantitative pooling (meta-analysis) or qualitative synthesis (meta-synthesis) of primary study results. Typically nested inside a traditional systematic review protocol.
- Umbrella review: Synthesizes existing systematic reviews and meta-analyses on a broad condition or problem. Protocol focuses on review-level eligibility and appraisal rather than primary studies.
- Integrative review: Combines diverse study designs to generate new frameworks or perspectives. Less restrictive eligibility; protocol must specify how heterogeneous evidence will be synthesized.
- Living systematic review: Continuously updated as new evidence emerges. Protocol must include an update schedule, amendment procedures, and version control.
Registration note: traditional systematic reviews, meta-analyses, and umbrella reviews typically register on PROSPERO. Scoping, integrative, and living reviews often use OSF Registries when PROSPERO eligibility is unclear.
Table of Contents
- What are the main types of systematic review protocols?
- What core elements does every protocol need?
- How do you choose the right protocol type for your question?
- Where should you register or publish your protocol?
- How long does a protocol-driven review actually take?
- How automation tools support protocol tasks
- Key Takeaways
- What most protocol guides get wrong
- Cut extraction time without cutting rigor
- Authoritative sources and templates for protocol development
What are the main types of systematic review protocols?
Understanding the differences between review protocol types is the first step toward choosing the right design for your research question. The type of review you plan determines what your protocol must specify, how long the work will take, and where you can register it.
Traditional systematic review
A traditional systematic review attempts to identify, appraise, and synthesize all empirical evidence meeting pre-specified eligibility criteria. The protocol must document the full search strategy across multiple databases, dual independent screening, data extraction procedures, risk-of-bias tools, and the planned synthesis approach. Cochrane methodology holds that the research question must dictate every methodological choice — the protocol locks those choices in before any searching begins.
When to use it: You have a focused, answerable question; you need findings that can inform clinical or policy decisions; you have a team and six months or more.
Registration: PROSPERO (health and social care outcomes) or OSF for other disciplines.
Scoping review
Scoping reviews map the breadth and type of literature on a topic, identify research gaps, and clarify concepts. They suit broad or exploratory questions where a focused PICO framework would be too restrictive. The protocol still needs eligibility criteria, a search strategy, and a charting plan, but the synthesis is descriptive rather than evaluative. PROSPERO often does not accept scoping reviews for registration; OSF Registries is the standard alternative.
When to use it: Your question is broad, the literature is heterogeneous, or you need to map what exists before designing a primary study.
Rapid review
Rapid reviews streamline or omit certain steps — fewer databases, single screener, restricted date ranges — to deliver findings within weeks rather than months. The protocol for a rapid review carries a specific obligation: it must state explicitly which shortcuts are taken and how residual bias is acknowledged. Funders and policymakers accept rapid reviews for time-sensitive decisions, but the trade-off in rigor must be transparent.
When to use it: A policy or clinical decision needs evidence within four to eight weeks and a full systematic review is not feasible.
Meta-analysis and meta-synthesis
Meta-analysis statistically pools effect sizes from primary studies; meta-synthesis interprets qualitative findings across studies. Both are typically embedded within a traditional systematic review protocol. The protocol must pre-specify the statistical model (fixed vs. random effects), heterogeneity thresholds, and subgroup analyses before data collection begins — otherwise the analysis risks becoming exploratory rather than confirmatory.
Umbrella review
An umbrella review synthesizes existing systematic reviews and meta-analyses rather than primary studies. It suits broad conditions where multiple competing interventions already have their own reviews. The protocol must define review-level eligibility criteria, specify how review quality will be appraised (e.g., AMSTAR-2), and explain how overlapping primary studies across reviews will be handled.
When to use it: You need a high-level synthesis of an already-reviewed field to compare competing interventions or map the review landscape.
Integrative review
Integrative reviews combine quantitative, qualitative, and theoretical literature to generate new frameworks. The protocol is less restrictive on study design but must be explicit about how heterogeneous evidence types will be synthesized and appraised. Registration on OSF is typical.
Living systematic review
A living systematic review updates continuously as new trials or studies are published. The protocol must specify the update trigger (e.g., a new eligible study), the version-control system, and the amendment reporting procedure. These reviews demand sustained team capacity and a clear governance structure.
Pro Tip: Before drafting your protocol, confirm whether your review type is eligible for PROSPERO. If it is not, register on OSF Registries immediately after finalizing your protocol draft — the timestamp establishes your a priori record.
What core elements does every protocol need?
A systematic review protocol is a formal a priori document that specifies the rationale, research question, and complete methods before any searching begins. The PRISMA-P 2015 statement organizes protocol content into 17 items across three categories: administrative information, introduction, and methods.
| PRISMA-P Category | Key Items |
|---|---|
| Administrative information | Title, registration number, protocol version, author contributions, funding |
| Introduction | Rationale, explicit research question (PICO or equivalent) |
| Methods | Eligibility criteria, search strategy, study selection, data extraction, risk-of-bias assessment, synthesis plan, meta-bias assessment, confidence in evidence |
PICO in practice
PICO (Population, Intervention, Comparison, Outcome) structures the research question and directly generates eligibility criteria. A well-formed PICO example:
- P: Adults aged 18+ with type 2 diabetes
- I: Structured aerobic exercise programs (≥8 weeks)
- C: Usual care or no exercise intervention
- O: HbA1c reduction at 3 and 6 months
Every eligibility criterion in the protocol should trace back to one of these four elements. For scoping reviews, the PCC framework (Population, Concept, Context) is the standard alternative.
Data extraction fields to pre-specify
Vague extraction fields are one of the most common protocol failures. Under-specified extraction fields force teams to revisit included studies mid-review, introducing inconsistency and rework. Pre-specify at minimum:
- Study design and setting
- Sample size and population characteristics
- Intervention details (dose, duration, delivery mode)
- Outcome measures and measurement tools
- Funding source
- Risk-of-bias rating per domain
Synthesis and risk-of-bias planning
The protocol must name the risk-of-bias tool you will use (e.g., Cochrane RoB 2 for randomized trials, ROBINS-I for non-randomized studies) and state whether synthesis will be quantitative (meta-analysis) or narrative. Deciding this after data collection invites selective reporting. The seven-step SR process — define scope, search, select, extract, assess risk of bias, synthesize, assess certainty — maps directly onto the sections your protocol must cover.
Pro Tip: Run a pilot extraction on five to ten papers before finalizing your extraction form. Inconsistencies caught at that stage cost hours, not weeks.
How do you choose the right protocol type for your question?
Start with four questions about your research situation:
- How narrow is your research question? (Focused PICO → systematic review; broad concept → scoping review)
- How much time do you have? (Weeks → rapid review; months → traditional; ongoing → living)
- Do prior systematic reviews already exist on this topic? (Yes → umbrella review)
- Will synthesis be quantitative? (Yes → plan for meta-analysis within the protocol)
| Dimension | Traditional SR | Scoping Review | Rapid Review | Umbrella Review | Living SR |
|---|---|---|---|---|---|
| When to use | Focused clinical/policy question | Broad or exploratory question | Urgent policy need | Field already has multiple SRs | Rapidly evolving evidence base |
| Scope | Narrow, PICO-defined | Broad, concept-driven | Narrow, time-constrained | Review-level, broad condition | Narrow, continuously updated |
| Methods required | Full search, dual screen, RoB, synthesis | Broad search, charting, no RoB required | Streamlined search, documented shortcuts | Review appraisal (AMSTAR-2), overlap analysis | Full SR methods + update triggers |
| Timeline | 12 months | 6–12 months | 4–8 weeks | 6–18 months | Ongoing |
| Resource intensity | High (team of 3) | Moderate (team of 2–3) | Low–moderate (1–2 reviewers) | High (methodologist required) | Very high (sustained team) |
| Registration eligibility | PROSPERO, OSF | OSF (PROSPERO often ineligible) | PROSPERO or OSF | PROSPERO | PROSPERO or OSF |
Three scenarios
Scenario 1: A PhD student asks whether mindfulness-based interventions reduce anxiety in college students. The question is focused, a PICO fits cleanly, and no recent umbrella review exists. A traditional systematic review protocol is the right choice. Register on PROSPERO.
Scenario 2: A health policy team needs evidence on telehealth adoption barriers within six weeks for a congressional briefing. A rapid review protocol is appropriate. The protocol must document which databases were searched, that only one screener was used, and how that limitation affects confidence in findings.
Scenario 3: A research team wants to map what is known about digital literacy interventions across K–12 settings before designing a primary study. The question is exploratory, the literature is heterogeneous, and no synthesis is planned. A scoping review protocol fits. Register on OSF.
Where should you register or publish your protocol?
Registering a protocol before searching begins is what separates a protocol-driven systematic review from a narrative literature review. Registration creates a public, timestamped record that protects against selective outcome reporting.
Registries:
- PROSPERO: The primary international registry for health and social care systematic reviews. Accepts traditional systematic reviews, meta-analyses, and some umbrella reviews. Scoping reviews and non-health reviews are frequently outside its scope; teams register those on OSF instead.
- OSF Registries (Open Science Framework): Accepts all review types, any discipline. Free, immediate registration. Generates a DOI-linked record. The standard choice when PROSPERO is ineligible.
Journals that publish protocols:
- BMJ Open: Publishes standalone systematic review protocols across health disciplines. Requires PRISMA-P adherence and a completed checklist.
- Systematic Reviews (BioMed Central): Dedicated to publishing protocols and completed reviews. Peer-reviewed; a published protocol strengthens the credibility of the eventual review.
Registration steps:
- Complete all PRISMA-P items before submitting — reviewers and editors use the 17-item checklist to assess completeness.
- Submit to PROSPERO or OSF before running your first database search.
- Record the registration number in your protocol document and in every subsequent manuscript.
- If you amend the protocol after registration, log the amendment with a date and justification in the registry record. Report all amendments transparently in the final review.
How long does a protocol-driven review actually take?
Realistic timelines prevent the most common planning mistake: underestimating how long screening and extraction take relative to searching.
| Review Phase | Traditional SR | Scoping Review | Rapid Review |
|---|---|---|---|
| Protocol development | 4–8 weeks | 2–4 weeks | 1–2 weeks |
| Database searching | 1–2 weeks | 1–2 weeks | 3 days |
| Screening (title/abstract + full text) | 8 weeks | 4–8 weeks | 1–2 weeks |
| Data extraction | 8–12 weeks | 4–6 weeks | 1–2 weeks |
| Synthesis and write-up | 8–12 weeks | 4–8 weeks | 1–2 weeks |
| Total (approximate) | 12 months | 6–12 months | 4–8 weeks |
Core team roles
A traditional systematic review needs at minimum: a methodologist (designs the protocol and synthesis plan), a librarian (builds and runs the search strategy), two independent screeners (title/abstract and full-text), and a statistician if meta-analysis is planned. Scoping reviews can function with two to three people. Rapid reviews sometimes use a single screener, but the protocol must document that decision.
Living systematic reviews require a sustained governance structure: a standing team, a defined update trigger, and a version-control system for the protocol itself. The literature review workflow for a living review is effectively never-ending, which makes automation particularly valuable for maintaining consistency across update cycles.
Resource planning tip: budget time for a pilot screening round before full screening begins. Teams that skip the pilot routinely discover eligibility disagreements at the 30% mark, forcing a restart of the screening process.
How automation tools support protocol tasks
Every phase of a protocol-driven review involves repetitive, error-prone tasks. AI workflow automation can reduce manual effort at several of these stages without compromising the pre-specified methods the protocol requires.
| Protocol Task | What Automation Can Do |
|---|---|
| Reference import and deduplication | Import CSV/RIS files from Scopus or Web of Science; flag duplicates automatically |
| Title/abstract screening | AI-assisted relevance scoring against pre-specified eligibility criteria |
| Data extraction | AI reads abstracts and fills structured extraction fields |
| Label normalization | Standardizes inconsistent terminology across extracted records |
| PRISMA flow generation | Tracks inclusion/exclusion counts at each stage |
| Table and chart export | Exports enriched CSV or PNG visualizations for publication |
Automation does not replace methodological judgment. Every automated output should be spot-checked against source documents, and the protocol should specify how automated steps will be audited. That said, under-specified extraction fields and inconsistent labeling are among the most common protocol failures — exactly the problems that structured extraction tools address.
Papersynapse is built around this workflow. Researchers import references from Scopus or Web of Science, define custom extraction fields that match their protocol, and let the AI read abstracts and populate structured tables. The platform handles label normalization across records, generates exportable tables and charts, and supports team collaboration so multiple reviewers work from the same dataset. For teams running literature review automation on large corpora, Papersynapse processes up to 200 papers in under two minutes — a meaningful difference when a traditional systematic review yields hundreds of included studies.
Key Takeaways
Every systematic review protocol should pre-specify its research question, eligibility criteria, search strategy, extraction fields, risk-of-bias tools, and synthesis plan before any searching begins.
| Point | Details |
|---|---|
| Match type to question | Choose your protocol type based on question narrowness, available time, and whether prior reviews exist. |
| PRISMA-P is the standard | The 17-item PRISMA-P checklist covers every section a complete protocol requires, from rationale to synthesis plan. |
| Register before you search | Submit to PROSPERO (health reviews) or OSF (all types) before running your first database query to establish your a priori record. |
| Document every amendment | Log protocol changes with a date and justification; report all deviations transparently in the final review. |
| Papersynapse supports extraction | Papersynapse automates abstract screening, data extraction, and label normalization, reducing manual effort across protocol-defined tasks. |
What most protocol guides get wrong
The conventional advice is to spend more time on the protocol so the review goes smoothly. That is true, but it misses the more specific problem: most protocol failures are not about length or effort — they are about the wrong things being specified in detail.
Teams write elaborate rationale sections and then leave the extraction form as a vague list of “relevant outcomes.” They specify three databases in the protocol and then search twelve, or vice versa, without documenting why. They pre-specify a random-effects model and then switch to fixed-effects when heterogeneity is low, with no amendment record. These are not effort failures. They are precision failures in the parts of the protocol that actually constrain bias.
The amendment problem is underappreciated. Protocols are not immutable — justified changes are expected and acceptable. What is not acceptable is undocumented change. A protocol amendment that is logged, dated, and justified in the registry record is a sign of methodological integrity. An undocumented deviation discovered during peer review is a credibility problem. The distinction is entirely about transparency, not about whether the change was reasonable.
There is also a tendency to treat registration as a formality rather than a commitment. Researchers sometimes register a protocol, then substantially revise the eligibility criteria or outcomes after seeing the search results, without updating the registry. That practice defeats the purpose of pre-registration. The reproducible review methodology that makes systematic reviews trustworthy depends on the protocol record matching what was actually done.
One practical fix: treat your extraction form as a pilot document. Run it on ten papers before finalizing. Every field that generates disagreement between two extractors is a field that needs a clearer operational definition in the protocol. Catching that before full extraction saves weeks.
Cut extraction time without cutting rigor
Researchers who have spent weeks manually categorizing hundreds of abstracts know the real bottleneck in a systematic review is not the search. It is the extraction. Papersynapse addresses that directly: import your references from Scopus or Web of Science, define extraction fields that match your protocol, and the AI fills structured tables from abstract text.
The platform keeps your workflow PRISMA-compliant by tracking inclusion and exclusion counts at each stage, normalizing labels across records, and exporting enriched tables and charts ready for publication. Teams collaborate in real time on the same dataset, which eliminates the version-control problems that plague multi-author reviews. For researchers who need to move from a registered protocol to a completed extraction without losing consistency, Papersynapse is worth a close look. Start with the free tier and see how far automated extraction gets you before committing to a paid plan.
Authoritative sources and templates for protocol development
These are the primary resources to consult when drafting, registering, or evaluating a systematic review protocol.
- PRISMA-P 2015 statement: The definitive 17-item reporting guideline for systematic review protocols. Use it as your drafting checklist and attach the completed checklist to any journal submission.
- PROSPERO: Register health and social care systematic reviews at the NIHR PROSPERO registry before searching begins. The registration form maps closely to PRISMA-P items.
- OSF Registries (Open Science Framework): Use for scoping reviews, non-health reviews, or any review type ineligible for PROSPERO. Free, immediate, and generates a citable DOI.
- Cochrane Handbook for Systematic Reviews of Interventions: The methodological reference for Cochrane protocol development. Chapters on eligibility criteria, search strategy, and synthesis are directly applicable to non-Cochrane reviews.
- BMJ Open and Systematic Reviews (BioMed Central): Both journals publish standalone protocols. Submitting a protocol for peer review adds an external quality check before the full review begins.
- University library protocol templates: Many research libraries (University of Washington, OHSU, Emory) maintain protocol templates and review-type guides tailored to their institutional databases and citation managers. Check your institution’s library guide before building your search strategy from scratch.
- PhD protocol guide: Papersynapse’s step-by-step protocol template for PhD researchers, covering question framing through registration.