Dictionaries, uploaded once and available to every study of the tenant

Mappings, to declare which eCRF answers must be coded and against which dictionary

A coder workspace, with dictionary lookup and code assignment

A full audit trail on dictionaries, mappings and codes

Click Upload dictionary, name it, drop the package exactly as delivered by MedDRA and enter its password.

Version and language are read directly from the package.

The row shows Uploading, then Active, or Upload failed with the reason.

Several versions can be Active at once, which long studies need since MedDRA publishes about two releases a year.

Deprecate a version to stop it being offered in new mappings. Codes already assigned never change.

The package must be uploaded exactly as delivered. Unzipped, renamed or repackaged files fail processing.

One dictionary = one language × one version. Uniqueness is checked on type + version + language.

A dictionary used in at least one mapping cannot be deleted. Deprecate it instead.

A mapping is always one item to code, up to 10 supporting items, one dictionary.

Only free-text questions can be coded. Items inside repeated measures work; items inside a matrix cannot.

Supporting items can be of any type, from the same subsection. They give the coder context and are never coded.

Creation is retroactive: answers already filled are pulled into the coder's queue, not only new ones.

The item is deleted, or is no longer free text → Broken mapping (red). Collection stops and the verbatims collected so far and their codes are deleted.

The title, the variable or a supporting item changes → Modified mapping (yellow). Collection continues and codes are kept and a flag is displayed on the mapping.

The item is moved elsewhere → no flag, no impact.

One mapping per item to code and per dictionary. The same item can be mapped twice only against another dictionary.

Mappings cannot be edited, and deleting one deletes codes associated to verbatim, with no way back. A warning modal says so before confirmation.

One tab per mapping, alphabetical, with the Mappings tab last, and two views: Uncoded, the working queue, and All.

Columns are Verbatim, Code, Status, Subject, then one per supporting item. Verbatim and Code stay visible while scrolling sideways.

Clicking a verbatim opens a side panel with the full text and its supporting items, open by default.

From a verbatim. Clicking its text pre-fills the search bar, in the dictionary of its mapping, and stays editable.

From the dictionary. Find the term first, then pick a verbatim using that same dictionary.

The coder picks an LLT, displayed with its full path (LLT → PT → HLT → HLGT → SOC), so the choice can be checked before it is applied.

Search matches LLTs only, in contains mode, so partial words are enough: “ven recan proce” finds “Venous recanalization procedure”. Retired terms are excluded.

Terms on several branches show all their paths, blue for primary and white for secondary. The path selected is the path stored.

A verbatim can carry several codes, one per medical concept (“headache and vomiting” gets two). The status covers the whole set, and the verbatim stays one row.

The list reflects the state at load time. eCRF corrections appear after a reload or a tab switch.

Coding stops at Coded. Review, approval and export come in a next iteration.

Blinded studies are out of scope for now.

No filtering beyond the two views and no re-sorting. Use the search bar to narrow a long list.

Subjects Database centers: when a subject is added via an EDC study, their Subjects Database record now takes the study's center as the participant's global center.

Filters (Subject Database & Recruitment): after adding many filters, they could not be scrolled or added (Subject Database), and filter labels were partly cut off (Recruitment).

eConsent menu: some eConsent menu entries were renamed and reordered (e.g. "Dashboard" → "eConsent list") for clarity.

Randomization & permissions: a display issue and incorrect behaviour when configuring roles so that users (e.g. investigators) couldn’t see the randomization list without seeing the randomization settings.

eCRF review: sometimes a review was not showing in the visit view.

eCRF inclusion: form progress was not refreshed in real time (especially when answering quickly or switching sections).

CTMS payments: decimal amounts could not be entered in the payment settingd by study, the number was turned into an integer (e.g. 9.9 became 99).

CTMS payments: inconsistent behaviour when entering payment values (decimals handled differently in payment by visits / payment confirmation).

ePRO QR code: the QR link worked only for the first occurrence — after completing it, the participant could not complete the next ePRO occurrences.

ePRO dropdown: ePRO titles were not fully displayed in the selection dropdown, with no tooltip.

Translations (French, ePRO): a small wording inconsistency in the ePRO list menu.

Fix an issue when a radio-button question with an "Other" option was selected and not displayed to the user on the interface. The "Other" option is now displayed correctly.

Fix a bug, where the date/time format of an edited eCRF answer differed between the new-value and old-value fields in audit trails.

Section and subsection visibility permissions no longer apply inside the Builder. The user who restricts a section keeps access to it when editing the form.

Removing the source question of a Repeated Measures rule no longer breaks the conditional logic of the whole form. The remaining rules stay editable and can be replaced.

Duplicating a section or a subsection, or adding one from a template, now recalculates the conditional logic of the new section. The records shown were previously inherited from the original section.

Question order now holds after a page reload. Blocks could reappear in a different order than the one saved.

Updating a calculation created a duplicate entry, displayed out of order. A single entry is now recorded, in the right position.

The reason and comment entered for a Missing / Unknown answer are now recorded in the study audit trail.

Uploading a file to a file-upload question added a second entry a few seconds later that looked like a deletion, both in the Subject Database and in the eCRF. Only the upload is recorded now.

Reminders stop once the target date and time have passed, even when the questionnaire is still uncompleted. Participants no longer receive reminders for a survey they can no longer answer.

When an automation creates an ePRO for a participant with no email address, the ePRO is created but not sent, so a follow-up "ePRO sent" automation is no longer triggered.

A scheduled occurrence could be viewed and edited before sending but could not be opened from its QR code or its link. Both now open the occurrence.

Sending an eConsent to a subject with no first or last name returned a misleading message. The error now states that a first name and a last name are required.

Calculations are recalculated after importing data and exporting the list to refresh it. The imported values were kept as they were.

Returning to the participant list while it was sorted by the Availability column no longer displays an error message.

The Full Name column stays visible when the table displays only four variables.

An OPTIONS request on the public API returned a 500 error. It now returns Method Not Allowed, as expected.

❕ Platform - Draft & Suspended Studies
Study status is now much more visible while working inside a study. Draft and Suspended EDC studies now display a status chip and banner throughout the study, while Draft Recruitment Studies display the corresponding Draft indicators.

The banner is collapsed by default and can be expanded when additional information is needed. Live studies do not display a banner.

When adding participants to a Draft EDC study, a "Test subject" badge is also displayed to make it clear that participants created at this stage are test subjects.

In Recruitment Study settings, "Recruitment details" has also been renamed "Landing page" for greater clarity.

Center access is now clearer in Platform Settings > Centers. The action is displayed as Edit or View according to the user's permissions, and users with view-only access now see a read-only page.

Role changes now display both the previous role and the new role in the Study and Platform Audit Trails, providing clearer traceability of permission changes.

Fixed an issue that could generate an error when opening the Audit Trail for a Matrix question.

When an inclusion is opened from a Subject page, closing the eCRF now correctly returns to that Subject page rather than the Inclusion List.

In studies with more than 50 subjects, newly created subjects now open correctly after creation.

File removals are now recorded consistently in the Audit Trail. Previously, removing an uploaded file from an eCRF could be recorded as another upload.

Corrected discrepancies in eConsent exports where some information, including Randomization columns, could be missing depending on the selected records.

Resolved an issue where filtering a report by Center could return an empty report despite matching data being available.

Subject Database questions using conditional logic are now represented correctly in Recruitment exports instead of being displayed as ".hidden".

When a user is assigned as an administrator, access to all centers is now applied correctly.