Governance
Write the consent form, publish the wording participants will read, set the study's email rules, and confirm the analysis plan before launch.
Who can open this: Principal investigator, Steering committee (full access), Biostatistics, Coordinator
Who this is for
Governance is the fourth Setup group. Its sidebar label is "Governance", and its line is "Manage participant consent, study communications and the analysis plan." The tabs are "Readiness", "Consent", "Communications", "Public page", and "Analysis plan".
Only the Principal investigator can edit Setup, and the same check publishes the consent form. Anyone else is refused with "You do not have permission to publish this form." The public page says "Read-only: only the principal investigator changes the public page." Steering committee (full access), Coordinator, and Biostatistics can read Setup. Coder, Adjudicator, and Steering confirmer do not get it. A save from a seat that cannot edit is "You do not have permission to change this."
Steps
Readiness is the summary. Consent and Communications together are the Consent & comms step, one of the ten. Analysis plan is its own step, the tenth. Public page is a tab on this screen and is not part of either step's readiness, so it does not change the count of ten. An error on Consent & comms or on Analysis plan is "Blocks launch". A warning is "Needs attention" and still counts as ready.
Table: which Governance tab Review & launch counts. The Public page badge is a different check.
| Tab | What Review & launch counts |
|---|---|
| Readiness | The summary of Consent, Communications, and Analysis plan. Not its own step. |
| Consent | Consent & comms, together with Communications |
| Communications | Consent & comms, together with Consent |
| Public page | Not counted. The badge on that page is its own check. |
| Analysis plan | Analysis plan |
Readiness. The heading is "Readiness": "Where consent, communications and the analysis plan stand." The cards are "Consent" ("Participant information and agreement.", action "Open consent"), "Communications" ("Email templates and delivery rules.", action "Open communications"), and "Analysis plan" ("Reporting standard, anonymity and what is published.", action "Open analysis plan"). "How governance is used" says "Governance sets up the content and rules. Operational sending happens in Recruitment and Rounds."
Consent. The form's sections are "Study purpose", "Voluntary participation", "Procedures", "Time commitment", "Risks and benefits", "Confidentiality", "Data use and retention", "Withdrawal", "Publication and authorship", and "Contact". Data use, Withdrawal, and Publication and authorship are filled from Ethics and from Panel design. When consent is asked is "When they accept the invitation" or "Before Round 1 opens".
Launch waits on "Choose when consent is asked.", on every empty required section, on "Add at least one required acknowledgement." and "Every acknowledgement needs wording.", on "Upload the consent form your ethics board approved.", and on "Confirm that the consent wording matches the approved consent form." A later edit asks "The consent wording changed after you confirmed it matched the approved form. Confirm again or replace the file." Withdrawal gaps are fixed on Ethics and data governance; until the terms are complete, confirmed, and still match retention, the form cannot be published.
"Consent contact does not match the study reply-to" blocks launch, and Review & launch lists that line on its own as "Blocks launch". It does not block publishing. Before the wording locks, change the Contact section to the study's reply-to on Study identity. After the wording locks, the form is not rewritten; the mismatch is flagged as needing an amendment.
"Publish for participants" says "Nobody can consent until you publish. Publishing shows participants the saved wording exactly as it is now." The button is "Publish version N". The confirmation is "Publish version N?": "From now on participants read this wording, and every consent they give is recorded against version N. A published version cannot be edited or withdrawn: to change the wording you save a new version and publish that. People who already agreed to an earlier version are not asked again automatically. Publishing does not email anyone." Success is "Version N is published. Participants now see this wording." Publishing the same wording again is "Version N is already published." Unsaved edits are "Your latest changes are not saved yet, so nothing was published. Fix the save problem, then try again." A form that is not ready is "The consent form is not ready to publish." followed by the reason.
The participant's consent page shows that published wording verbatim. Staff warnings about the form, including a Contact section that names another address, are shown on Consent and on Review & launch. They are not inserted into the participant's form.
Until a version is published, an invitation is refused: "Publish the consent form before inviting anyone. Participants agree to a published version, and none exists yet. Publish it on Consent & comms once the withdrawal terms are confirmed."
When recruitment is open, the editor locks the free-text sections. "Edit content" and "Edit agreements" are then unavailable. The refusal is "Consent is locked while recruitment is open. Changes need a new version, which may need ethics approval." Before launch the screen says "Consent locks when recruitment opens." Once the study has launched it says "Consent is locked: recruitment has opened." The server treats recruitment as open once the study has launched. That editor lock is not the same thing as the server comparing every consent-wording field on each save. On each save, the governance check refuses a rewritten consent history and a version number that goes backward. It does not compare every wording field, and it is not a check that runs only at launch. That difference stays open.
"If wording changes" is not that confirmation, and publishing does not apply it. The choices are "Ask again before the next round", "Ask everyone again now", and "Only ask people who haven't agreed". Saving a new version after launch shows that choice on the save confirmation: "People who already agreed will be asked again before the next round.", "Everyone who already agreed will be asked again now.", or "Only people who have not agreed yet will see the new wording. Invitations will refuse to ask anyone who already agreed." Before launch the same confirmation says "Nobody has agreed yet, so no one is asked again." The setting does not email anyone, and it does not ask them by itself. After launch, the "If wording changes" choice stays editable. The locked free-text sections do not. "Only ask people who haven't agreed" makes a later ask from Invitations refuse: "The consent settings say only people who have not agreed yet see the new wording, so nobody is asked again. Change "If wording changes" on Consent & comms first."
Communications. The sections are "Sender and branding" ("Who the emails come from and what is in the header. Set once in Identity and used by every template."), "Message templates" ("The wording of each type of email, edited in Invitations. Nothing is sent until you invite someone or open a round."), "Delivery rules" ("Set when and how messages are sent. Recruitment and each round show these before you confirm a send."), and "Message history".
Launch waits on "Set the sender name in Identity.", "Set a valid reply-to address in Identity.", "1 email template needs attention in Invitations." (or "2 email templates need attention in Invitations."), "Choose the study time zone.", "Quiet hours need a start and an end that differ.", and "Include the consent and contact-preference footer in every email." The time zone is one zone for the study.
Public page. "Public page" is "A page anyone can read without a link or a login, to learn about the study and apply to join the panel. Applications come to Panelists for screening; consent still comes after the invitation." The switch is "Show this page publicly": "Off, the address answers “not found”, the same as for a study that does not exist. Turning it off takes the page down at once." Until it can be seen, the page says "Not public yet. The page goes public when it is switched on, the study is launched and the consent form is published." The sections are "Opening", "Purpose", "Timeline", "What taking part involves", "Study team", "Overview video", "References", and "Applications and results". "Overview video" is "Optional. The page links to where the video is hosted and prints its transcript, so nobody has to watch it to know what it says." The badge on this page uses the same words as the other Setup screens, including "Blocks launch", and that badge is this page's own check. It reads "Blocks launch" when the page is not the untouched empty page and one of its own errors applies: "Show this page publicly" is on and the summary is empty ("Write the summary a visitor reads first."), or a video has no transcript ("Add the video's transcript, so everyone can read what it says."). Switched off, with an empty summary, the badge reads "Not started" instead. Review & launch does not use this badge. It counts Consent & comms from the consent form and communications only, so this page does not change the count of ten and does not stop launch. "Accept applications (“Join the panel”)" is an application. "Applications and results" says "What a visitor can send. An application is screened in Panelists before anyone is invited; nobody joins or consents from this page." A sample study says "A sample study has no public page. Create a study to write one."
Analysis plan. The heading is "Analysis plan": "What is fixed before the data arrives: how ratings are summarized, and what the study promises about privacy. Both lock at launch." Launch waits on "Confirm the analysis plan before launch." The button is "Confirm the analysis plan". A later edit asks "The plan changed after it was confirmed. Confirm it again." "The response-rate warning level must be at or above the block level." An anonymity the organization does not allow blocks launch. After launch the page says "The analysis plan is locked: the study has launched. Changes need a controlled amendment."
On Review & launch, a study whose method records structural rules for the steering committee to apply offline, while the analysis plan promises aggregate results only, shows a warning. That warning does not change the privacy promise.
What the app will not do
The participant form shows the published wording verbatim. Staff warnings are not inserted into it.
Publishing does not email anyone, and it does not ask people who already agreed. The confirmation says they are not asked again automatically. "If wording changes" does not send that ask either. A published version cannot be edited or withdrawn. Publishing a later version is how participants see new wording. It is not a way to edit free text the editor has locked.
This screen does not offer a model-generated suggestion, and it has no approval step for generated text. A panelist's own wording proposal is a different kind of suggestion. The control is "Suggest different wording" on that person's statement. It is not consent text.
No invitation can be queued while no consent version is published.
Switched off, the public page's address answers not found. "Join the panel" is an application, and nobody consents from that page. The page is not one of the ten steps.
Communications does not send mail. "Nothing is sent until you invite someone or open a round."
A save that lost a race is refused: "This section was changed somewhere else since you opened it. Reload to see the latest version."
Related
Checked against the app on 2026-10-31.