PROSPERO Registration Guide: How to Register Your Protocol

PROSPERO is the international registry for systematic review protocols, maintained by the University of York's Centre for Reviews and Dissemination, and registration is now expected or required by a growing number of journals before they will consider a systematic review for publication. Despite how routine it has become, the registration process demands genuine methodological specificity, and rushing it is one of the more common ways reviewers end up needing to make undisclosed protocol deviations later.

What PROSPERO actually accepts

PROSPERO registers protocols for systematic reviews with a health-related outcome, including reviews in health and social care, welfare, public health, education, crime, justice, and international development where there is relevance to human health. It does not accept reviews already completed at the time of registration, reviews without a health-related outcome, or protocols for scoping reviews, which have their own separate registry, the Open Science Framework, as a more common option.

Timing: register before you start screening

PROSPERO registration should happen after your protocol is finalized but before you begin screening studies for inclusion. This timing is the entire point of registration -- it creates a timestamped, public record of your planned methodology before your results could have influenced it. Registering partway through screening, or worse, after screening is complete, undermines the credibility benefit registration is meant to provide, and experienced reviewers will notice registration dates that postdate a review's apparent start.

The fields that require real specificity

The review title field should follow the same convention expected in your final manuscript, explicitly identifying the work as a systematic review. The research question fields require full PICO or PICOS specification -- population, intervention, comparator, outcome, and study design -- written with enough precision that another researcher could independently judge whether a given study would be eligible.

The search strategy field expects the databases you intend to search and, ideally, a representative search string, not just a general statement that "a comprehensive search will be conducted." The eligibility criteria fields require explicit inclusion and exclusion criteria, not a restatement of the PICO framework in different words. The data extraction and risk of bias fields require you to name the specific tool you will use -- RoB 2, ROBINS-I, or another named instrument -- rather than a vague statement that studies will be "critically appraised."

Common reasons registrations get returned or flagged

Vague eligibility criteria that would require judgment calls not specified in the protocol itself is the single most common issue. A search strategy field that states only "PubMed, Embase, and other relevant databases will be searched" without any specificity about search terms or filters is another frequent gap. Missing or unclear specification of which risk-of-bias tool will be applied, particularly in reviews that anticipate including both randomized and non-randomized studies, is a recurring issue given how often RoB 2 and ROBINS-I get conflated at the planning stage.

Registering a review that has already started screening, evident from a registration date that postdates other public information about the review's timeline, will not necessarily be rejected outright, but it substantially weakens the credibility benefit of registration and should be disclosed transparently in your eventual manuscript rather than presented as if timing were standard.

Amendments after registration

PROSPERO allows protocol amendments after initial registration, and this is a normal, expected part of the process -- searches sometimes need adjustment once you see initial results, and eligibility criteria occasionally need clarification once screening surfaces edge cases you had not anticipated. What matters is that amendments are logged with a clear date and rationale, creating an auditable trail showing what changed, when, and why. Undisclosed deviations between your registered protocol and your final published methods are what draw reviewer objections -- documented, justified amendments generally do not.

What to do once you are registered

Your PROSPERO registration number should appear in your final manuscript, typically in the abstract and methods section, along with the registration date. This allows readers and reviewers to directly compare your registered protocol against your reported methods, which is exactly the transparency mechanism registration is designed to enable.

Treat the registration record itself as a working reference throughout your review, not a box checked once and forgotten.

What happens after you submit

PROSPERO registrations typically undergo a brief administrative check before appearing publicly, generally completed within a few working days, checking for basic completeness rather than methodological quality -- registration is not itself a form of peer review or endorsement -- it confirms your submission is complete and appropriately scoped for the registry, not that your methodology has been independently evaluated for quality. This distinction matters because some early-career researchers mistakenly treat a successful registration as a signal of methodological soundness, when it is genuinely just a timestamped record of your stated intentions.

If your registration is returned with a request for clarification, this is a normal part of the process rather than a rejection, and usually reflects one of the common gaps covered above -- most often insufficiently specific eligibility criteria or an incomplete search strategy description. Responding promptly with the requested detail is generally straightforward once you know what the registry team is actually asking for, and it rarely delays your overall timeline by more than a handful of working days in most cases. Once live, your record is publicly searchable, and you can log back in at any time to submit an amendment if your methodology genuinely needs to change. Keep your registration number and the direct PROSPERO link on hand; you will need to cite both in your final manuscript, typically in the abstract and again in the methods section.

Treat the registration record itself as a working reference throughout your review, not a box checked once and forgotten. Returning to it at each major stage -- after finalizing your search strategy, after completing screening, before finalizing your analysis plan -- is a genuinely effective way to catch unintentional protocol drift before it becomes a discrepancy a reviewer has to flag for you.