A master protocol puts several sub-studies under one document, one screening pathway and one governance structure. This page sets out what the document contains, what the shared infrastructure has to do, and how the data capture side is organised so each sub-study keeps its own casebook.
Free sandbox · No credit card · 21 CFR Part 11 aligned
| Master level | Sub-study A | Sub-study B | Sub-study C | |
|---|---|---|---|---|
| Master protocol document | Appendix A | Appendix B | Appendix C | |
| Screening and eligibility core | plus A criteria | plus B criteria | plus C criteria | |
| Common baseline forms | ||||
| Sub-study specific forms | ||||
| Shared control data | ||||
| Own visit schedule | ||||
| Steering committee oversight |
The short answer
Definitions
Regulators and methodologists use "master protocol" for a single overarching protocol designed to answer several questions within one trial infrastructure. FDA has published guidance on master protocols, and the term is usually split into three shapes. An umbrella design tests several treatments in one disease. A basket design tests one treatment across several diseases or biomarker-defined groups. A platform design is an open-ended structure in which interventions can enter and leave over time, often against a shared control. Real programmes mix these, which is why this page is written about the document and the infrastructure rather than about one shape.
What a master protocol is not: a set of unrelated protocols that happen to share a sponsor, or a single protocol with a few extra arms bolted on at an amendment. The test is whether the common elements are written once and whether sub-studies can be added, changed or closed without rewriting the whole. If every sub-study needs the entire document re-issued, you have a family of protocols, not a master protocol, and the data capture design should reflect that.
This page deals with the master protocol document, the sub-study structure and the shared infrastructure. For the two designs that have their own data issues, see adaptive platform trial software, which covers entering and dropping interventions and shared-control analysis, and basket trial software, which covers biomarker-defined cohorts and tumour-agnostic reporting. The regulatory framing of adaptive elements is in the ICH E20 guide.
Where each thing lives
A practical split of content. Your own template and your regulator or ethics committee will have the final say on what must sit at master level.
| Content | Master protocol | Sub-study appendix |
|---|---|---|
| Overall objectives and rationale | The programme question and why it is run as one structure. | The specific question for this treatment or cohort. |
| Eligibility | Core criteria every participant meets (age, performance status, consent, general exclusions). | Additional inclusion and exclusion criteria, such as a biomarker or prior-therapy requirement. |
| Screening procedures | The shared screening pathway and the assessments done once for everyone. | Any extra assessment that decides assignment to this sub-study. |
| Intervention | Rules for how interventions are described and supplied. | Dose, schedule, preparation, concomitant-medication rules for this treatment. |
| Visit schedule | The common skeleton, such as screening, baseline, follow-up. | The on-treatment visits and assessments for this arm. |
| Endpoints | Shared endpoint definitions and assessment conventions. | The primary and secondary endpoints for this sub-study. |
| Safety reporting | Reporting definitions, timelines, stopping rules across the programme. | Treatment-specific adverse events of special interest and dose-modification rules. |
| Statistics | Principles, shared control handling, rules for adding or closing sub-studies. | Sample size, analysis populations and interim rules for this sub-study. |
| Governance | Steering committee, data monitoring committee, decision rights. | Sub-study lead and any sub-study specific review. |
| Amendments | Changes that affect everyone, with full ethics and regulatory review. | Changes that affect one sub-study, versioned on their own. |
Amendment routes differ by country and ethics body. Check how each will treat a sub-study-only change before you rely on this split.
The documents
Data capture is far easier when the document is built for it. Three habits pay off. First, give every sub-study the same section numbering, so an appendix reads like its siblings and a form specification can point at section 5.2 in any of them. Second, put all shared definitions in one place: what counts as a baseline value, how a response is assessed, how an adverse event is graded. Differences in definitions between sub-studies are the commonest source of pooled-data problems. Third, state which data are collected identically across sub-studies and which are not. That list becomes the common form set.
It helps to treat the master protocol as a small suite of linked documents with their own versions: the master, one appendix per sub-study, the statistical analysis plan, the data management plan and the monitoring plan. A data management plan template is a useful starting point, but write it once for the programme and add a short sub-study section for each, rather than keeping separate plans that drift apart. The same is true for the schedule of assessments: one shared grid with a column set per sub-study is easier to review than five separate files.
Keep an eye on version control. A sub-study appendix at version 2.0 can sit beneath a master at version 1.3, and the consent form, the eCRFs and the monitoring plan each reference a version of both. Write down the compatibility rule once (for example, "appendices are compatible with any master of the same major version") so site staff do not have to guess which forms apply to a participant.
Programme rationale and objectives
Shared screening and core eligibility
common to all sub-studies
Safety definitions and reporting
Statistical principles and shared control
SAP cross-reference
Governance and decision rights
Appendix A: sub-study A (v2.0)
Appendix B: sub-study B (v1.1)
Appendix C: sub-study C
to be added by amendment
Build the shared forms and one appendix in the free sandbox and see how the casebook looks. Every feature, no credit card, no time limit.
Setting it up
Order matters. Decisions about structure are cheap early and expensive after the first participant.
Go through every sub-study appendix and mark each form or field that is collected identically. These form the shared set. Anything that differs in wording, units or timing is a sub-study form, even if it looks similar.
Either one Capture study for the whole programme with the sub-study recorded on each participant, or one study per sub-study with a shared screening study in front. Weigh reporting needs against the study fee model: the pricing page lists a fee per live study, so model the count.
Create screening, eligibility and baseline forms once, with edit checks for the core criteria. Add a field that records the sub-study assignment and the date, so the participant's route is auditable.
Each sub-study gets its own visit schedule and its own forms for assessments that are specific to it. Approved forms are locked for live use, so a change to one sub-study is made in a draft and released without touching the others.
Decide who sees which sub-study, and who is blinded to which intervention. Blinded roles do not receive treatment values, which matters where one sub-study is blinded and another open label.
Document how participant numbers are assigned, how a participant moves between sub-studies if the protocol allows it, and how shared control participants are flagged. Put these in the data management plan.
Decide the pooled dataset and the per-sub-study datasets before you go live. Capture exports CSV and Excel with a data dictionary, can filter by site, and produces SDTM datasets as SAS XPT with Define-XML, so statisticians can work on each sub-study or on the combined data.
Per-sub-study casebooks
A casebook is the set of forms, by visit, that a participant is expected to complete. In a master protocol it is the main way a sub-study keeps its identity. The shared forms appear in every casebook, the sub-study forms only in the relevant ones, and the visit schedule differs by sub-study. A participant in sub-study B never sees a form that exists only for sub-study A.
The visual below is an invented schedule for two sub-studies of a demo programme. The first three columns are common; the on-treatment visits differ. Notice that sub-study B adds a week 6 imaging form that A does not have. In a vendor-built system that is a change order; in a self-serve build it is a draft form attached to one sub-study's visit, tested and approved by your team.
Check the form versions that each participant used. When a sub-study appendix is amended, participants already in the study may finish on the old version while new participants start on the new one. The audit trail records who changed what and why, and the form version used for each record should be reviewable at the time of analysis. See eCRF versioning for mid-study changes for how to document it.
| Assessment | Screen | Baseline | Wk 2 | Wk 6 | Wk 12 | Follow-up |
|---|---|---|---|---|---|---|
| Consent and core eligibility (shared) | ||||||
| Baseline forms (shared) | ||||||
| Adverse events (shared) | ||||||
| A: dosing log | ||||||
| B: dosing log | ||||||
| B only: week 6 imaging | ||||||
| A only: biomarker panel |
x = form expected at this visit. Rows labelled A or B appear only in that casebook.
Governance
A master protocol concentrates decisions that a single trial would spread across one team. A steering committee decides which sub-studies open, close or change. A data monitoring committee reviews safety and efficacy across them. Sub-study leads run their appendix. The sponsor, or the sponsor-investigator, carries regulatory responsibility for the whole. If these roles are not written down with clear decision rights, the first conflict (for example, closing a sub-study while another is mid-recruitment) will be resolved in email.
In the data system, governance shows as access. Committee members need read access to the views they review, and unblinded statisticians or pharmacists should not be the same accounts as blinded investigators. Capture has role-based access with researcher, site coordinator and admin roles, row-level security, and blinded roles that never receive treatment-arm values. See blinded and unblinded role management for the pattern. A DSMB charter template helps document what the committee sees and when.
Safety is the area where programmes most often cause confusion. A serious adverse event in one sub-study might matter to all of them if it relates to a shared component, such as a common backbone treatment. Decide in advance who sees pooled safety data and who is notified when something in one sub-study may affect another, and write that rule down. The reporting basics are in adverse event reporting software for clinical trials.
Scope
Be clear about the boundary before you commit to a design.
Built in the platform, with draft-to-approved lifecycle and a field-level audit trail.
Built once and used by all sub-studies, with edit checks that raise auto-queries when a rule is broken.
Capture supports blinded and open-label randomization with stratification. Confirm on a demo how your allocation scheme across sub-studies maps to it.
Stay in your statistical analysis plan. Capture does not include built-in adaptive algorithms.
Define your own participant identifier convention and document it. Ask on a demo how it works with your chosen build.
Done by your team. Capture is not connected to a trial registry.
It is the data capture system that supports a trial run under a master protocol: a shared screening pathway, common forms, separate sub-study casebooks and one audit trail. It records the data; the design logic stays in the protocol and statistical plan.
It depends on how much you need to report across sub-studies and how the study fee model affects cost. One study with a sub-study field keeps pooled reporting simple; separate studies isolate changes more. Model both against the pricing page before you decide.
Version the appendix on its own, change only the relevant forms in draft, test them and approve. Approved forms are locked for live use. Check with your ethics committee how they treat a sub-study-only amendment.
Yes in design terms: use a shared screening form set and record the assigned sub-study and date. How assignment is decided (a biomarker rule, randomization or an investigator choice) is defined by your protocol.
Capture records the data and lets you flag participants by group. Which control participants are eligible for which comparison is a statistical rule in your analysis plan, not an automatic calculation in the platform.
You can build a new casebook and forms in a draft state while other sub-studies run. The design question is whether your master protocol and approvals allow a new appendix without a full re-review.
Capture is suitable for Phase 1, 2 and 3. The right question is whether your team can specify the structure and testing; the phase does not change that.
Exports are available as CSV and Excel with a data dictionary and can be filtered by site and date. SDTM datasets can be produced as SAS XPT with Define-XML. Plan the sub-study identifier as a variable so your statisticians can split the data.
Keep exploring
Adaptive platform trial software
Interventions that enter and leave over time.
Basket trial software
One treatment across disease or biomarker cohorts.
ICH E20 adaptive designs guide
Regulatory framing of adaptive elements.
Blinded and unblinded role management
Access when sub-studies differ in blinding.
Multi-site clinical trial management
Running many sites under one study.
Data management plan template
A starting point for the programme DMP.
Open the free sandbox, create the screening pathway and a single sub-study casebook, and test it as a site coordinator would. No credit card, and you pay only once you go live.