Contents
On this page
- Identification of the applicable guideline
- Distinction between a protocol and a grant proposal
- ICH E6(R3): relocation of the protocol requirements
- ICH M11 and the jurisdictional dependence of its status
- SPIRIT 2025 for trial protocols
- PRISMA-P for review protocols
- Ethics committee requirements
- Protocol amendments
- Significance of protocols: the measured reporting gap
- Recommended working practices
Academy · Physician writing · Research protocols and proposals
Writing Research Protocols and Proposals: Governing Guidelines, Version Changes and Ethics Requirements
Almost every type of protocol is governed by a named guideline. The first question is which guideline applies, and three of these guidelines have recently changed location or version.
A protocol is one of the most heavily specified documents in research: almost every type is governed by a named guideline, and the first question is which guideline applies. This page addresses that question and records three version changes that may silently invalidate an existing template.
Identification of the applicable guideline
Table 1. Protocol types and the guideline for each
| Document | Guideline | Size | Notes |
|---|---|---|---|
| Randomised trial protocol | SPIRIT 2025 | 34 items | Address all items before ethics submission. Supersedes SPIRIT 2013, which should no longer be cited. |
| Randomised trial results | CONSORT 2025 | 30 items | A separate checklist for a separate document, from the same merged executive group. |
| Systematic review protocol | PRISMA-P 2015 | 17 items, 26 with sub-items | Still current; an update is in development. Not a quality-appraisal tool. |
| Clinical trial protocol, regulatory | ICH E6(R3), Appendix B | 16 topics, B.1–B.16 | Was Section 6 in E6(R2). Cite the 16 June 2026 consolidated text. |
| Clinical trial protocol, structured template | ICH M11 (CeSHarP) | — | EU effective 11 June 2026; still a draft at the FDA at the time of checking. |
| Systematic review registration | PROSPERO | — | The protocol should be completed first and submitted before data extraction begins. |
Distinction between a protocol and a grant proposal
Differences in purpose and audience rather than content
The title of this page pairs the two documents, so it is useful to distinguish them. The clearest statement identified is institutional rather than from a funder: a grant is “written to convince reviewers that your proposed project is scientifically sound”, while “a protocol is written to explain the who, what, where, when, why of your project”, and so “protocols are intended to be informative not persuasive like grants”. This is a difference of purpose and audience rather than a rule about content, and it must be stated in those terms, because WHO’s own recommended protocol format includes the budget, other support and investigators' CVs in its second part. Where the funder is also the ethics reviewer, the two documents converge.
The distinction in its source's words: a grant is “written to convince reviewers that your proposed project is scientifically sound”10, while “A protocol is written to explain the who, what, where, when, why of your project.”10 Hence: “Protocols are intended to be informative not persuasive like grants.”10 The same source names a third document worth keeping separate: “Operations manuals include very detailed information on tasks that need to be completed in a specific way”10
That source is institutional teaching material rather than a funder's policy, and it is cited as such. We could find no funder's own written statement drawing the distinction, which is itself worth knowing.
ICH E6(R3): relocation of the protocol requirements
Implications for existing templates
This structural change may silently invalidate standard operating procedures (SOPs) and templates. In ICH E6(R2) the protocol requirements were Section 6 and the Investigator's Brochure was Section 7. In E6(R3) the protocol is Appendix B and the Investigator’s Brochure is Appendix A. Any reference to “ICH E6 Section 6” for protocol contents therefore now points to the wrong location. Moreover, the current text is not the January 2025 document alone: ICH’s own cover page records Step 4 on 6 January 2025, Annex 2 reaching Step 4 on 3 June 2026, and the two consolidated into one document on 16 June 2026. This consolidated version is the one that should be cited.
E6(R3) on why the document matters: “A well-designed trial protocol is fundamental to the protection of participants”3, and “The scientific objectives of any trial should be clear and explicitly stated in the protocol.”3
The sixteen topics of Appendix B3:
Table 2. ICH E6(R3) Appendix B
| B.1 | General information |
| B.2 | Background information |
| B.3 | Trial objectives and purpose |
| B.4 | Trial design |
| B.5 | Selection of participants |
| B.6 | Discontinuation of trial intervention and participant withdrawal |
| B.7 | Treatment and interventions for participants |
| B.8 | Assessment of efficacy |
| B.9 | Assessment of safety |
| B.10 | Statistical considerations |
| B.11 | Direct access to source records |
| B.12 | Quality control and quality assurance |
| B.13 | Ethics |
| B.14 | Data handling and record keeping |
| B.15 | Financing and insurance |
| B.16 | Publication policy |
Three differences from E6(R2)
Three differences from E6(R2) are relevant to the updating of a template. The single section of R2 on “selection and withdrawal of subjects” is divided into two sections, B.5 and B.6. “Subjects” becomes “participants” throughout, and the final “Supplements” heading of R2 is removed.
E6(R2)'s glossary definitions remain the clearest short statements of what these documents are. A protocol is “A document that describes the objective(s), design, methodology, statistical considerations, and organization of a trial.”4 An amendment is “A written description of a change(s) to or formal clarification of a protocol.”4
ICH M11 and the jurisdictional dependence of its status
Specification of jurisdiction
The status of M11 currently differs between regulators, so an unqualified answer would be incorrect in one direction or the other. The EMA publishes it at Step 5, adopted at Step 4 on 19 November 2025 and in effect in the EU from 11 June 2026. The FDA, at the time of checking, still posted its corresponding guidance as a June 2025 draft marked “not for implementation”. Consequently, neither “M11 is a draft” nor “M11 is final” is correct as a general statement, and the relevant jurisdiction should be specified.
What it is: the template “is intended for interventional clinical trial protocols and is suitable for all phases”5, and it “is designed to enable modification suitable for the particular trial”5 M11 comprises a template and a technical specification, the latter recommending an open, non-proprietary standard for exchanging protocol information electronically6.
SPIRIT 2025 for trial protocols
SPIRIT 2025 “consists of an evidence based checklist of 34 minimum items to address in a trial protocol”1, and the instruction on timing is the important part: “Trial investigators and sponsors should address all SPIRIT 2025 checklist items in the protocol before REC/IRB submission.”1 — before the ethics submission, not in response to its comments.
On supersession, it is unusually direct: “The SPIRIT 2025 statement supersedes the SPIRIT 2013 statement, which should no longer be used or cited.”1
SPIRIT and CONSORT are now maintained together — “the 2 groups decided to merge into the joint SPIRIT-CONSORT executive group”1 — but they remain two checklists for two documents, and “the CONSORT 2025 statement is published separately.”1 SPIRIT 2025 has 34 items and covers the protocol; CONSORT 2025 has 30 and covers the results. Our CONSORT 2025 guide covers the second.
PRISMA-P for review protocols
“The final PRISMA-P 2015 checklist contains 17 numbered items (26 including sub-items)”2, intended “to facilitate the preparation and reporting of a robust protocol for the systematic review.”2 Its authors “encourage reviewers preparing any type of review protocol to make use of PRISMA-P as applicable.”2
Current status and function as a reporting rather than a quality tool
This point of status runs counter to the obvious analogy. PRISMA 2020 replaced PRISMA 2009 for reviews, so it might be assumed that PRISMA-P has also been replaced. It has not: PRISMA-P 2015 remains the version to use, and an update is listed as in development. Its authors also state that the checklist “has not been validated” as a means of judging whether the methods of a protocol are appropriate. It is a reporting checklist rather than a quality instrument, a distinction worth maintaining because protocol-writing guidance routinely blurs it.
Registration. “PROSPERO is an international database of prospectively registered systematic reviews in health and social care.”7, and it is “developed and is managed by the Centre for Reviews and Dissemination (CRD) at the University of York”7 “Registration on PROSPERO is free of charge.”7
Requirements of PROSPERO
This point requires precision, because it is usually stated too strongly. PROSPERO’s own guidance does not say registration is required or recommended. It states that “a full protocol should be ready before registering” and that “submissions must be made before data extraction commences”. Both statements are more useful than a general exhortation: the protocol comes before the registration, not after it, and once data extraction has begun the window for registration has closed. Scoping reviews are not accepted.
Ethics committee requirements
Two sources, one international and one a named committee. WHO's ethics review committee states that “Its mandate is to ensure WHO only supports research of the highest ethical standards.”8, and that “The approved version of the protocol must have copies of informed consent forms (ICF)”8 attached to it.
A university committee's own requirements are more concrete. “All protocols are to include the following elements.”9 “Submissions must include enough information for the convened IRB or expedited IRB Reviewer”9 to reach a decision, and “Incomplete submissions, or those that do not address the above requirements, will be returned”9 What it asks for:
- A version number, date, protocol identifier and investigator or sponsor name on every page.
- A title that matches the one entered in the submission system exactly.
- Background, significance, rationale and conceptual framework.
- Specific aims, objectives or hypotheses.
- Study design and procedures, including recruitment and data collection.
- The study sample and data source, with eligibility criteria and expected enrolment.
- Ethical considerations: compensation, the consent process, and privacy authorisation or its waiver.
- Risks, benefits, and privacy and confidentiality protections.
- A statistical plan including sample-size justification.
- Data handling and record keeping, including security and storage.
- Study registration and publication plans.
The first of those is the detail most protocol guidance omits and most committees enforce: a version number, date and identifier on every page. It pairs with E6(R3)'s requirement that any amendment also bear its number and date, and it is what makes it possible to tell later which version a participant actually consented under. Note too that the title “Protocol title, which must match the title entered into IRB e-system.”9
Protocol amendments
Triggers, approval and notification requirements for amendments
Three questions arise regarding an amendment, and each has a sourced answer. Triggers for an amendment: under US regulation, a new study not covered by a protocol already in the application, or a change that significantly affects participant safety, the scope of the investigation or the scientific quality of the study — with a dose increase or “any significant change in the design of a protocol” given as examples. ICH E6(R3) adds the hazard trigger: the sponsor “should consider whether the protocol requires amendment in response to an immediate hazard”. Approving body: the ethics committee, in advance — “no deviations from or changes to the protocol should be initiated without prior documented IRB/IEC approval”, except to eliminate an immediate hazard or for logistical and administrative changes. Reporting requirements: reporting is required to the committee and, where applicable, the regulator, and in the US the amendment must be submitted before implementation.
The regulatory triggers, in the regulation's words: “any change in a Phase 1 protocol that significantly affects the safety of subjects”11, with examples including “Any increase in drug dosage or duration of exposure of individual subjects”11 beyond the protocol and “Any significant change in the design of a protocol”11 such as adding or dropping a control group. Timing: an amendment must be submitted “before its implementation”11, though amendments that only add investigators “may be grouped and submitted at 30-day intervals”11.
The ethics requirement, from E6(R3): “no deviations from or changes to the protocol should be initiated without prior documented IRB/IEC approval”3 — with exceptions to eliminate an immediate hazard, and for purely logistical or administrative changes. On hazards: “The sponsor should consider whether the protocol requires amendment in response to an immediate hazard.”3 The committee's standing remit includes reviewing “Protocol and amendments;”3
Significance of protocols: the measured reporting gap
Consequences of failing to adhere to a protocol
This is the strongest practical argument for writing the protocol rigorously, and it is based on measurement rather than assertion. A prospective study of 67 trials across five journals found that 19.4% had at least one pre-specified primary outcome that was never reported, that only 55.1% of 818 pre-specified secondary outcomes were reported, and that 365 outcomes appeared that had never been pre-specified (a mean of 5.4 per trial, of which only 13.7% were declared as such). Fifty-eight of the 67 trials warranted a correction letter, and 40% of those letters were published. The protocol is what makes such discrepancies detectable at all.
Table 3. Protocol against publication, in 67 trials
| Measure | Figure | Of |
|---|---|---|
| Trials assessed | 67 | Across five journals |
| Pre-specified primary outcomes reported correctly | 76.3% | Of 97 |
| Pre-specified secondary outcomes reported | 55.1% | Of 818 |
| Trials with an unreported pre-specified primary outcome | 19.4% | — |
| Novel outcomes reported without declaration | 365 | A mean of 5.4 per trial |
| Novel outcomes correctly declared | 13.7% | Of 365 |
| Trials needing a correction letter | 58 of 67 | 87% |
| Correction letters published | 23 of 58 | 40%, range 0–100% by journal |
The study's own conclusion: “Outcome reporting was poor overall and there was wide variation between journals on pre-specified primary outcomes”12 Specifically, “19.4% of trials had at least one unreported pre-specified primary outcome.”12 and “365 novel outcomes were reported without declaration in total, a mean of 5.4 per trial.”12 Of the correction letters the investigators sent, “Twenty-three letters were published (40%) with extensive variation between journals (range 0–100%).”12
Our pages on the benefits of an experimental plan and biostatistics in clinical research cover the design decisions a protocol fixes.
One more use for a protocol: it may be publishable in its own right. BMJ Open lists among the study types it accepts “study pre-protocols; protocols; pilot studies.”13, and states that “All research study types are considered, from study protocols through phase I trials to meta-analyses.”13
We could not reach that journal's detailed author requirements for protocol submissions, so none are stated here. Check the instructions to authors for word limits, registration and ethics requirements before preparing one.
Recommended working practices
Practical application
- The guideline governing the protocol should be identified before drafting begins, and its checklist should be followed.
- For a randomised trial, all 34 SPIRIT 2025 items should be addressed before ethics submission, and SPIRIT 2013 should no longer be cited.
- Any template or standard operating procedure that cites “ICH E6 Section 6” for protocol contents should be updated, since these contents are located in Appendix B of E6(R3).
- The consolidated E6(R3) of 16 June 2026 should be cited, rather than the January 2025 document alone.
- The relevant jurisdiction should be specified whenever the status of ICH M11 is mentioned.
- For a systematic review, the protocol should be written first and registered before data extraction begins. PRISMA-P 2015 remains the current checklist.
- A version number, date, protocol identifier and investigator name should appear on every page. Ethics committees enforce this requirement, and it enables consent to be traced to a specific version.
- Every primary and secondary outcome should be pre-specified and the list retained. Nearly one fifth of trials in one study left a pre-specified primary outcome unreported.
- Any outcome added after the protocol should be declared as such. In the same study, only 13.7% of novel outcomes were declared as such.
- An amendment should be submitted before a change is implemented, and documented ethics approval should be obtained first unless the change is made to eliminate an immediate hazard.
Key points
Every type of protocol is governed by a guideline: SPIRIT 2025 (34 items) for trial protocols, PRISMA-P 2015 (17 items) for review protocols, and ICH E6(R3) Appendix B for regulatory trial protocols, with ICH M11 as the structured template. Three version-related errors should be avoided: the protocol requirements moved from Section 6 to Appendix B in E6(R3), the current E6(R3) text is the consolidation of 16 June 2026, and M11 is in force in the EU but remains a draft at the FDA. PROSPERO requires the protocol to be complete before registration and before data extraction. The importance of these requirements is supported by measurement: 19.4% of trials in one study never reported a pre-specified primary outcome.
The applicable guideline depends on the type of protocol: SPIRIT 2025 for randomised trial protocols (34 items); PRISMA-P 2015 for systematic review protocols (17 items, 26 including sub-items); and ICH E6(R3) Appendix B for a regulatory clinical trial protocol, with ICH M11 providing a structured template. The first question for any protocol is which of these governs it.
Appendix B in E6(R3), and they were Section 6 in E6(R2). The Investigator’s Brochure has also moved, from Section 7 to Appendix A. Consequently, an SOP or template that cites “ICH E6 Section 6” for protocol contents now refers to the wrong location.
The consolidated text of 16 June 2026. ICH’s own cover page records Step 4 on 6 January 2025, Annex 2 reaching Step 4 on 3 June 2026, and the two consolidated into a single document on 16 June 2026. Citing the January 2025 document alone is incomplete.
The answer depends on the jurisdiction, and an unqualified answer is therefore incorrect. The EMA publishes it at Step 5, adopted at Step 4 on 19 November 2025 and in effect in the EU from 11 June 2026. The FDA, at the time of checking, still posted its corresponding guidance as a June 2025 draft marked “not for implementation”.
Yes. SPIRIT 2025 states that it “supersedes the SPIRIT 2013 statement, which should no longer be used or cited”. It contains 34 items, and investigators “should address all SPIRIT 2025 checklist items in the protocol before REC/IRB submission”.
Not at present. PRISMA 2020 replaced PRISMA 2009 for reviews, but PRISMA-P 2015 remains current for protocols, and an update is listed as in development. Its authors state that the checklist “has not been validated” as a means of judging whether the methods of a protocol are appropriate; it is a reporting checklist rather than a quality tool.
PROSPERO’s own guidance does not state that registration is required or recommended. Its statements are more useful: “a full protocol should be ready before registering” and “submissions must be made before data extraction commences”. The protocol therefore comes first, and the window for registration closes once extraction begins. Scoping reviews are not accepted.
They differ in purpose and audience. A grant is “written to convince reviewers that your proposed project is scientifically sound”; “a protocol is written to explain the who, what, where, when, why of your project”, so “protocols are intended to be informative not persuasive like grants”. This is a tendency rather than a rule: WHO’s own protocol format includes the budget and investigators’ CVs.
An amendment is required under US regulation for a study not covered by an existing protocol, or for a change that significantly affects participant safety, the scope of the investigation or the scientific quality, with a dose increase or “any significant change in the design of a protocol” given as examples. ICH E6(R3) adds that a sponsor should consider amendment in response to an immediate hazard.
References
- Chan A-W, Boutron I, Hopewell S, et al. SPIRIT 2025 Statement: updated guideline for protocols of randomized trials. JAMA. 2025;334(5):435–443. doi:10.1001/jama.2025.4486 read 15 October 2026. View on JAMA34 items. Published simultaneously in five journals, and it tells authors to stop citing SPIRIT 2013 rather than merely superseding it.
- Moher D, Shamseer L, Clarke M, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Systematic Reviews. 2015;4:1. doi:10.1186/2046-4053-4-1 read 15 October 2026. View on Systematic Reviews17 numbered items, 26 with sub-items. Still current, unlike PRISMA 2009 - though an update is in development. Its authors state it has not been validated as a quality-appraisal tool.
- ICH E6(R3): Good Clinical Practice. Step 4, 6 January 2025; Annex 2 Step 4, 3 June 2026; consolidated 16 June 2026. read 15 October 2026. View on ICHThe current authoritative text is the 16 June 2026 consolidated version, not the January 2025 document alone. Appendix B holds the protocol requirements and Appendix A the Investigator's Brochure.
- ICH E6(R2): Guideline for Good Clinical Practice. EMA/CHMP/ICH/135/1995, 1 December 2016; effective 14 June 2017. read 15 October 2026. View on European Medicines AgencyCited for the previous structure and for the glossary definitions of a protocol and a protocol amendment, which R3 does not restate in the same form.
- ICH M11: Clinical electronic Structured Harmonised Protocol (CeSHarP) — Template. EMA/CHMP/ICH/778801/2022, Step 5. Step 4 adopted 19 November 2025; EU effective 11 June 2026. read 15 October 2026. View on European Medicines AgencyA structured protocol template plus a technical specification. Its status differs by regulator, which the page states rather than smoothing over.
- U.S. Food and Drug Administration. M11 Template: Clinical Electronic Structured Harmonised Protocol (CeSHarP). Draft guidance, June 2025; docket FDA-2022-D-3054. read 15 October 2026. View on U.S. Food and Drug AdministrationThe same guideline, still posted as a draft marked not for implementation at the time we checked. This is why the page gives M11's status by jurisdiction.
- Centre for Reviews and Dissemination, University of York. Registering a review on PROSPERO. Undated. read 15 October 2026. View on PROSPERO (CRD, University of York)Cited for the timing requirements, which are the practically important part: the protocol must be complete before registration, and registration must precede data extraction.
- World Health Organization Research Ethics Review Committee. Recommended format for a research protocol. read 15 October 2026. View on World Health OrganizationA funder's and ethics committee's own content list. Notable because its Part 2 includes budget and investigator CVs, which complicates any clean separation of protocol from grant proposal.
- University of Arkansas for Medical Sciences Institutional Review Board. Policy 10.3: Protocol Content and IRB Submissions. Effective 24 January 2011; last revised 24 June 2025. read 15 October 2026. View on UAMS IRBA named committee's actual submission requirements, including the per-page versioning rule that most protocol guidance omits.
- Oregon Clinical and Translational Research Institute. Transforming your Grant into a Protocol: Study Guide. OHSU Research Forum, February 2024. read 15 October 2026. View on Oregon Health & Science UniversityInstitutional teaching material rather than a funder's policy, and cited as such. We could find no funder's own written statement distinguishing the two documents.
- Protocol amendments, 21 CFR § 312.30. read 15 October 2026. View on Cornell Legal Information InstituteThe US regulatory triggers for amending a protocol, and the requirement to submit before implementation. Read from a mirror of the Code of Federal Regulations rather than the eCFR itself.
- Goldacre B, Drysdale H, Dale A, et al. COMPare: a prospective cohort study correcting and monitoring 58 misreported trials in real time. Trials. 2019;20:118. doi:10.1186/s13063-019-3173-2 read 15 October 2026. View on TrialsThe hard numbers on the gap between a registered protocol and what gets published. Read on the publisher's own site.
- BMJ Open. Research aims and scope. read 15 October 2026. View on BMJ OpenCited only for the fact that protocols are a publishable article type. We could not reach the journal's detailed author requirements, so none are stated.
Pubrica AcademyWritten and maintained by Pubrica’s subject matter experts. Articles are revised when the underlying standards change, and each article states the date on which it was last reviewed.
Related: CONSORT 2025 · the benefits of an experimental plan · biostatistics in clinical research.
How Pubrica can help
Pubrica’s regulatory and medical writing teams prepare protocols in accordance with the governing guideline (SPIRIT 2025 for trials, PRISMA-P for reviews, and ICH E6(R3) Appendix B and the M11 template for regulatory submissions), including the statistical sections, the outcome list and the version control that ethics committees check first.