Manage Blinding and Permissions covers the randomization side. Blinding fails elsewhere: a kit number in a form, a listing in Reports, a template applied without its restrictions. Go through these five places before go-live.

🔑 Prerequisite: Settings › Roles (Edit) and eCRF Builder (Edit). The study in Draft status.

1. Roles

Three permissions decide who sees the arm.

Permission	Gives
Blinded Access	The randomization result without the arm
Unblind Access	The arm of every subject. For the unblinded pharmacist or statistician
Emergency Unblind	The right to reveal one subject's arm, with password and reason

Give Unblind Access to named roles created for the purpose ("Unblinded pharmacist"), never to a general role such as "Study coordinator". List those roles in your blinding plan.

2. Forms

Any question that holds unblinding information needs a restriction: kit content, drug accountability by product name, PK concentrations, a lab value that betrays the active arm.

On a question: Advanced tab › Role-Based Visibility, and select the roles that can see it.
On a whole form or visit: Exclude roles, and select the roles that cannot open it.

⚠️ Warning: a Calcul question of type System variable with {TREATMENTARMCODE} prints the arm in the form. Put Role-Based Visibility on it, or leave it out of a blinded study.

3. Exports

Do not rely on Role-Based Visibility or excluded roles to blind an export. Those settings govern what a role sees on screen. Treat the export permission as an unblinded permission.

Give Export on eCRF records to unblinded roles alone. A blinded data manager who needs data receives a file prepared by an unblinded colleague.
Exclude from data exports on a question removes it from every export, for every role. Use it for unblinding variables that the unblinded statistician receives through another channel.
Check it: export a test subject with each role that can export, and search the workbook for the arm and for each restricted variable.

4. Reports

A user with access to Study › Reports sees every variable placed in a table, even a variable their role cannot see in the eCRF. Keep restricted variables out of report tables, or keep blinded roles out of Reports.

5. Templates

A template drops Role-Based Visibility, Role-Based Edit and excluded roles. When you build a blinded study from a template, every restriction is gone and nothing on screen says so. Set them again, then test.

Test the blind

🛠️ Step by step guide

Create one test user per blinded role
Randomize a test subject
Log in as each blinded user and open the record, the Inclusions list, Reports and an export
Search each for the arm, the kit content and the restricted questions
File the result in your validation report

What emergency unblinding reveals

An investigator with Emergency Unblind opens the subject's menu, selects Unblind, types their password and a Reason for unblinding.

The arm of that one subject becomes visible to every user whose role holds the appropriate blinding permissions. Other subjects stay blinded.
The record shows a banner with the user's name and the UTC date.
The study audit trail records the action Emergency unblind with the reason.
The action cannot be undone.

💡 Tip: add an automation on Subject Randomized for the unblinded pharmacist, and ask investigators to call the sponsor's medical monitor before an emergency unblinding when the protocol allows the delay. Datacapt does not send an alert on unblinding by itself, so agree on the communication path in advance.