PROSPERO Protocol Helper
A PROSPERO protocol helper turns a draft PICO and a small set of study-design choices into paste-ready text for all 22 required PROSPERO registration fields. The helper proposes named defaults for risk-of-bias tools and synthesis methods, enforces PROSPERO’s per-field word limits, and runs entirely in the browser.
Every input stays in your browser. The privacy badge shows zero network requests after page load, so unpublished PICOs and protocol drafts never leave your device.
PROSPERO accepts roughly 85 percent of submissions, but the median time to acceptance hides a long tail of “amendment requested” responses. Most of those amendments come from five fields: search strategy, eligibility criteria, risk of bias, data extraction, and synthesis plan. Each one is fixable upfront with named methodology and the right level of detail. A PROSPERO protocol template that just shows blank fields does not help. A helper that drafts every field from a PICO and named methodology choices does.
This page is that helper. Enter your PICO, pick study designs and synthesis approach, then generate paste-ready text for all 22 PROSPERO required fields. Word counts on every field show how each draft compares against PROSPERO’s published limits. Copy field-by-field into the CRD York registration form, or download the whole draft as plain text and revise it locally first.
What this helper does
PROSPERO has 22 required fields. You fill six (a PICO, study designs, search dates). The helper drafts text for the rest, with named methodology defaults that match what CRD York reviewers actually expect. Word counts on every field show how close you are to PROSPERO’s published limits, so the draft fits before you paste.
The helper does not submit your protocol on your behalf. CRD York reviews submissions through their registration form, and the registration ID comes from them. The job here is to compress the drafting time from an afternoon to twenty minutes, and to make sure the five fields that most often trigger amendment requests come out tight on the first attempt.
Three rules govern every generator:
- Use a named methodology where one exists (ROB 2 for randomised studies, ROBINS-I for observational, AMSTAR-2 for reviews of reviews) rather than “an appropriate tool”.
- Respect per-field word limits. Each generated field carries a count badge that turns amber at 80 percent of the limit and red at 100 percent.
- Stay in the researcher’s voice. Outputs read as a draft a graduate student wrote, not as filled-in template fields.
The 22 required PROSPERO fields, mapped to PRISMA-P
PROSPERO requires information across five sections. PRISMA-P 2015 (Moher et al.) sets the reporting standard each field maps to.
| PROSPERO section | Fields | PRISMA-P items |
|---|---|---|
| Administrative | 9 fields: title, named contact, team, funding, conflicts, country, language, stage, anticipated dates | 1a–1e, 2, 3 |
| Question and eligibility | 6 fields: review question, condition/domain, population, intervention, comparator, study types | 8, 9 |
| Search | 2 fields: searches, search strategy | 10, 11a, 11b, 11c |
| Extraction and analysis | 5 fields: primary outcome, secondary outcomes, data extraction, risk of bias, data synthesis | 12, 13a–13d, 14, 15a–15c |
Five fields drive almost every amendment request CRD York returns: the search strategy, the eligibility criteria, the risk-of-bias plan, the data-extraction plan, and the synthesis plan. The helper drafts all five with named methods by default, which is the single biggest unlock for first-attempt acceptance.
The full registration form lives at crd.york.ac.uk/prospero/documents/PROSPERO%20registration%20form.pdf. Read the field labels in there once. After that, the helper’s output panel uses the same labels.
Strong vs weak text in the five fields that drive amendment requests
| Field | Weak (likely amendment request) | Strong (likely first-pass accept) |
|---|---|---|
| Search strategy | “We will search PubMed and Embase using relevant terms” | “MEDLINE (via PubMed) and Embase (Ovid) using controlled vocabulary (MeSH, Emtree) and free-text terms for the population and intervention. The strategy will be peer-reviewed using PRESS before the final search.” |
| Eligibility criteria | “Studies of adults” | “Adults aged 18 or older diagnosed with type 2 diabetes. Studies of paediatric populations, gestational diabetes, or type 1 diabetes are excluded.” |
| Risk of bias | “An appropriate tool will be used” | “Cochrane ROB 2 (Sterne et al., 2019) for RCTs. Two reviewers will assess each study independently; disagreements resolved by discussion or a third reviewer.” |
| Data extraction | “Data will be extracted from each study” | “Two reviewers will extract data independently using a piloted form. Items: study characteristics, participants, intervention details, outcomes, effect estimates and variance, funding, conflicts. Disagreements resolved by a third reviewer.” |
| Synthesis plan | “Results will be synthesised” | “Random-effects meta-analysis where at least three studies report comparable outcomes; heterogeneity assessed via I²; subgroup analyses pre-specified for intervention format and population characteristics. Narrative synthesis where pooling is not appropriate.” |
Fynman audits this style of methods-section language across every included study during data extraction, so the synthesis side of your literature review benefits from the same precision.
When to register
PROSPERO is a prospective registry. Register after your protocol is finalised but before title-and-abstract screening begins. That is the legitimate window. Researchers who register after screening has already started see their record flagged as retrospective, which weakens the methodological-rigour signal to peer reviewers.
The review process at CRD York usually takes 2 to 5 business days. Build in time: do not plan to register the morning your team starts screening. If the review comes back with revision requests, the typical turnaround for a revision is another 2 to 3 business days.
Amendments are accepted at any point after the protocol is registered. The original version stays on the public record. Plan to amend if your pilot screening reveals an unworkable inclusion criterion, or if a methodological choice (such as risk-of-bias tool) shifts after a methods review. Pre-emptive amendments are normal and do not penalise the review.
Worked example: diabetes intervention review
The widget above ships with this PICO pre-populated:
- Population: adults with type 2 diabetes
- Intervention: structured aerobic exercise programme
- Comparator: usual care
- Outcome: HbA1c reduction at 12 weeks or longer
- Study types: RCTs only → defaults RoB to ROB 2
- Synthesis: narrative + meta-analysis where feasible
- Databases: MEDLINE (PubMed), Embase, CENTRAL, CINAHL
Click “Generate 22-field draft” without changing anything. The output panel produces a complete draft. The review title is Effect of structured aerobic exercise programme on HbA1c reduction at 12 weeks or longer in adults with type 2 diabetes: A systematic review. The risk-of-bias field names ROB 2 explicitly, lists the five domains by name, and specifies dual independent assessment. The synthesis plan includes I² thresholds and pre-specified subgroup analyses. All fields stay within PROSPERO’s word limits.
Edit anything the helper got wrong for your specific review. The synthesis plan, especially, often needs your domain-specific reasoning about expected heterogeneity. The helper saves you the typing, not the methodological choices.
After registration: amendments and PROSPERO IDs in your manuscript
Once CRD York approves the registration, you get a PROSPERO ID (format: CRD42026XXXXXX). Cite it in three places in the final manuscript:
- Abstract: “This systematic review was prospectively registered on PROSPERO (
CRD42026XXXXXX).” - Methods section, first paragraph: the full ID and a note about the registration date.
- PRISMA 2020 flow diagram: the ID appears in the citation block under the diagram.
To amend the registered protocol, log into the CRD portal and use the “submit an amendment” function. Common amendments: revised inclusion criteria after pilot screening, swapped risk-of-bias tool after methods review, added databases when pilot searches return too few studies. The amendment carries its own dated record; the original protocol stays public so the audit trail is intact. Fynman keeps the PROSPERO ID and the registered protocol attached to every paper in the same review project, so the audit step (manuscript matches protocol matches included studies) takes minutes when the final draft is due.
For background on the broader systematic-review workflow this protocol sits in, the systematic literature review guide covers when to choose an SR over a scoping or narrative review, and how PROSPERO fits in.
Frequently asked questions
Frequently Asked Questions
Find answers to common questions about this topic.