Major Release 2025.2.0.0 |  Release Dates:  1st-2nd July 2025 

New features and / or  enhancements

System Access | New Browser & Operating Support Policy

Study teams (clinicians, researchers etc.) accessing CDMS with an unsupported browser or operating system (OS) will now be blocked from logging in and shown a clear error message. This ensures a reliable experience by enforcing access only through allowlisted environments. This change is applicable to all CDMS users that are required to log into the system.

Two new messages have been introduced as part of Castor's Browser and Operating System Policy enforcement for study teams.  If both the OS and browser are unsupported, the OS message takes priority.

Unsupported Browser Message:

Unsupported OS Message:

Note: The above messages apply only to clinician-facing CDMS interfaces

For study participants accessing CDMS functionality (web-surveys), dedicated new messages will only be active by default for new studies created on or after July 1st, 2025. This means that, while the policy is still applicable, participants from existing studies will not encounter these upgrade messages regardless of their browser/OS combination. Upon request, our Support team can activate the warning screen functionality for existing studies.

Supported and validated versions for v2025.2.0.0

Clinicians:

• Chrome 125 or newer

• Microsoft Edge 124 or newer

• Firefox 127 or newer

• Safari 17 or newer (macOS)

• Safari on iOS 17 or newer

Participants:

• Chrome 125 or newer

• Microsoft Edge 124 or newer

• Firefox 127 or newer

• Safari 17 or newer (macOS)

• Safari on iOS 17 or newer

• Samsung Internet 27 or newer

Validations Management

A new "Validation" permission has been introduced, to give study admins more granular control over who can manually update validation statuses, without granting full query rights. This permission is now visible and assignable in the Roles and Permissions overviews, with dedicated tooltips for clarity. For all existing studies, roles with the “Query” permission have automatically received the “Validation” permission to ensure a seamless transition. This update improves compliance and role precision across both legacy and the modern Validations logic. 
All changes are logged in the Audit Trail.

We have added a new "Confirm validation updates" setting in the ‘Good Clinical Practices’ tab. Enabling this study setting will require a reason for manual validation status edits and will add a static system generated reason for any automatic changes of the validations status (i.e. when a validation is auto-closed after a value update event). This setting is going to be enabled by default for all new studies, but kept disabled for existing studies to prevent any workflow disruptions. Existing studies can opt in by simply toggling it on, as it enhances traceability.

To strengthen audit traceability and compliance, users updating validation statuses in studies with the “Confirm validation updates” setting enabled will now be prompted to enter a mandatory reason for change. 
This reason is logged in the Audit Trail. 

For system-triggered status changes, the static note "Automatically updated by the system" is recorded automatically. For studies where the “Confirm validation updates” setting is disabled, the status updates can be completed without inputting a reason.

Users can now update the status of Information and Warning Validations directly within the CRF views in modern Data Entry, eliminating the need to switch to the Validations overview. If the user does not have 'Validations' permissions, the status button will be read-only. If Monitoring is disabled, thin this will not shown in Data Entry.

Status changes are permission-based, logged in the Audit Trail, and reflected instantly across all views with confirmation toasts. If Monitoring is not enabled for the study, validation statuses remain read-only. This update streamlines workflows for data managers and monitors by reducing clicks and context switching.

Applicable only for studies using the newly released using the updated Validations logic (CDMS v2024.4.0.0) 

Validation status updates now include a full change history within the status update modal. Users can view the status change history including who made each change, when it was made, and the reason for it, if the “Confirm validation updates” setting is enabled. For studies without the setting enabled, reasons are omitted or marked as “N/A.” 

Entries are listed from newest to oldest, providing transparency and auditability directly within the workflow. 

A new button-based trigger replaces the old dropdown, aligning with the improved user experience.

For studies that have not adopted the new Validations logic, the validations status change flow remains as before. Only some visual changes apply.

Mid-Study Updates with Form Sync

We’ve expanded the Interim Data Impact Report for studies using the new Validations logic to show how structural changes might affect participant data validations. A new column, data_validation_impact, has been added to the structural_changes.csv file, and detailed participant-level impacts are available in a new data-validations-impact-per-participant.csv file. 

This enhancement helps study designers anticipate and minimize validation disruptions before merging updates to production, supporting better study quality and decision-making.

In this new report file, we preserve the granularity of the detected changes and their predicted impact as before. This means that for each change to validation configurations, the system evaluates the impact at the most granular level.

Since this is a preview report, it shows the individual impact of each change independently. In reality, once both changes are merged together into production, and validations are fully re-evaluated, the system might retain or reset the validation differently based on the combined final expression.

Form Builder 

We have standardized the date format in Form Builder field previews to ISO format (YYYY-MM-DD) for clarity and consistency. A tooltip now informs users that data entry formats may vary by site settings.
 

We have made additional changes to the Form Builder page to make better use of the available space, given longer field labels and/or smaller screen sizes. Returned results in the search list are ordered alphabetically within each section.

We have extended the interactive preview in the Form Builder to also cover template-based fields (i.e. calculation, summary, QR). Users can now view the configured template and test it with dummy values directly in the field preview, in the same way that is available in the Calculation helper tool.

• When expanding a field card, you can now immediately see the configured template, input dummy variables (up to six), and preview the resulting calculation outcome, helping you validate templates faster, stay in the form builder flow, and reduce reliance on extra test participants.

• If a template has more than six variables, a message will inform you that the template is too complex to preview interactively.

• Please note that previews for template-based fields (like Calculations, Summaries, and QR Codes) currently do not carry specific field configurations, and date and time fields are shown as plain text in the preview, which may lead to differences compared to live data entry where dates are treated as Date objects.

• A warning message is shown in the preview to highlight this. 

In future updates, we plan to connect previews to actual field types and apply the same data entry logic to further improve accuracy.

Empty CRF export in PDF format

We have added a new export type: you can now request a PDF export of an empty CRF from the Study Structure, Forms, and Option Group pages. When generating the export, you can customize:

• Page size and layout, including title page, table of contents, and whether each form starts on a new page

• Which Visit, Repeating Data, and Survey forms to include

• What information to display, such as hidden calculation fields, field details, and calculation templates

Once the export is ready, it will be available for download from the Exports page.

Requesting and downloading exports via the API

We’ve introduced three new API endpoints to request, monitor, and download exports. More information can be found in the API documentation.

• POST /study/{study_id}/export: This endpoint allows you to submit export requests by specifying the desired export types, formats, filters, and options.
  Please note that this endpoint is currently in beta and will be improved over the coming releases. Implementation details may change.

• GET /study/{study_id}/export: This endpoint allows you to view the status of previously requested exports.

• GET /study/{study_id}/export/{id}/download: This endpoint allows you to download completed exports.

System defects fixes

Validations Logic & Behaviour

• We have fixed an issue where decimal values (i.e. 1.1 or 1.10) were incorrectly treated as equal in field validation expression checks. The system now correctly distinguishes between such values.

• We have fixed a defect, where triggering an Exclusion validation message on a Repeating Data or a Survey field incorrectly marked the participant's progress as excluded in the Participants Overview, which indicated that data collection is no longer possible on that participant. Now, participant exclusion status is only updated when exclusion validations are triggered on Visit fields, as intended. 
  As per the current CDMS logic, exclusion validations messages that have been triggered on a Repeating Data or a Web- Survey field will only prevent further data entry on the instance where they occurred. 

 Form Builder & Data Entry 

• We have fixed an issue on the search input in the Form Builder. Long field names, labels, or variables are now fully rendered within the search box, without being cut off.

• We have fixed a layout issue, where long label descriptions on horizontal slider fields caused the preview to break on smaller screens. The slider now adjusts correctly.

• We have improved error handling when users attempt to use system reserved variable names like record_id or randomize_id. The system now displays a clear validation message instead of [object Object].

• We have reintroduced the missing warning text shown in the Delete modal when removing a custom Repeating Data field. Users are now informed if and when deleting the field will also remove its column from the Repeating Data Overview grid.

• We have fixed an inconsistency where the order of study elements in the SDV Plan Tree View did not match the order defined in the Study Structure tab. Users managing SDV plans will now see elements displayed in the correct, expected order, improving clarity and reducing confusion during configuration.

• We have aligned the public API behavior with the CDMS UI when it comes to the required permissions to create Repeating Data Instances (RDIs).  Users with only the “Edit” permission can now create Repeating Data Instances (RDIs) via the API in a consistent manner with the UI. Previously both "Add" and "Edit" were required to fully complete the action.

• We have improved the configuration of fields by ensuring consistent default width behavior when no field length is specified. Fields now default to the maximum width allowed, preventing layout inconsistencies.

• We've addressed an issue where the Data Entry interface would attempt to open a form that had just been hidden by an automation, leading to blank pages and confusing system errors. Now, if a hidden form is targeted for display, users will see an error.

• We have fixed a defect in the Add/Edit Field modal where it was possible to assign a field to a domain, without entering a domain variable.

• We have fixed a small issue in the Add/Edit Field modal where error messages for domain variables were not aligned correctly.

• We have fixed a defect where an error message was not shown upon importing a survey with a name that already exists in the study.

• We have added an additional check for uploading images in image fields. Uploads of webp files are not permitted anymore due to compatibility issues.

Exports

• We have fixed an issue where in the Participants exports grouped by domain and "Export missing values in separate columns", upload fields marked as User Missing would not appear in the correct column.

• We have fixed a defect where, in some cases, the download export button on the Exports page required multiple clicks before the export was successfully downloaded.

• We have fixed a defect in the variable list and domain mapping files for the Participant Export grouped by domain, where rows were duplicated. This defect caused the SAS field types for radio- and dropdown-fields in Domain-Based Exports to be misrepresented as numeric fields, while these should always be exported as textual fields.

Miscellaneous

• We have fixed an issue that rendered query automations to be triggered incorrectly, when the field's value was being marked as "value missing", specifically when an automation that was set on a value-based condition of "equals zero".

• We have resolved an issue where duplicate "Form locked" events were logged in the Audit Trail when signing and locking a Repeating Data Instance with already locked forms. Now, redundant events are no longer recorded.

• We have fixed an issue in Modern Data Entry when using new Date(); in a calculation field. Pages now load correctly and calculations are updated as expected.

• We have aligned the public API behavior with the CDMS UI when it comes to the required permissions to create Repeating Data Instances (RDIs).  Users with the “Edit” permission can now create Repeating Data Instances (RDIs) via the API in a consistent manner with the UI.  A sunset notice will be included in API responses to guide users through the transition, as per the API developer policy.

• We have fixed an issue, where the native browser shortcuts using the 'Alt' key were not working in CDMS. These shortcuts now function as expected.

• We have fixed an issue where dates were not displayed or accepted in data entry when the browser language was changed. Date fields now render and function correctly across supported browser language settings.

• We have resolved an issue where incomplete Grid field configurations (caused by missing field type selections) could be saved and later break data export. The system now prevents saving unless all required field types are selected, and newly added columns or rows will have "Text" auto-selected by default to ensure export integrity.

• We’ve fixed an issue where studies could not be set Live if stale data was submitted from another browser tab. A versioning and locking mechanism has been added to prevent overwriting newer changes. Now, if outdated study settings are submitted, users will receive a clear message prompting them to refresh before proceeding.

• We've corrected the description for the {queryMessage} tag in the Notifications configuration panel for "New Query Created" and "Query Updated" event types. It now accurately reads: "the actual message entered when the query was created or updated", instead of incorrectly duplicating the description for {fieldLabel}. This helps clarify the purpose of each tag when setting up email notifications.

• We have resolved an issue where users on Firefox v133.0 were blocked from importing CSV files, due to an incorrect file validation error. Users can now successfully upload and preview CSV imports for participant data.

Hotfix Release 2025.2.0.1 | Release Date: July 2nd 

• We’ve extended support to include the last 3 major versions of macOS, up from the previous 2. 

This change reflects our ongoing monitoring of platform stability and ensures broader access for users on slightly older Mac devices, without compromising performance or security.

• We’ve resolved an issue that caused excessive backend error logs when automated tools or scanners accessed invalid API routes. 

While these errors didn’t affect functionality, they created noise in our monitoring systems. 
This fix improves system efficiency and stability, ensuring smoother performance behind the scenes. 

Hotfix Release 2025.2.0.2 | Release Date: July 3rd 

• Resolved an issue where invitation emails associated with Castor Connect survey packages were sent with default text, rather than configured text

Minor Release 2025.2.1.0 | Release Date: July 28th 

New features and / or  enhancements

On-Site ePRO

As part of release 2025.2.1.0, Castor will be making On-Site ePRO available as a paid additional feature for studies using surveys. Enabling On-Site ePRO adds an additional option - either via direct URL or through a dedicated function on the participant list screen - for a site clinician to administer an ePRO survey on a site-tablet or similar device.

Clinicians with the send survey and manage participants permissions will be able to select a patient, visit and survey and administer a survey package invitation directly from the device in the primary language the participant needs.

At launch, 3 options will be included as part of on-site ePRO:

• On-device completion: The clinician proceeds to hand over the site device, is logged out of on-site mode and the participant can complete directly on the device. Once completed, the participant hands the device back, the clinician logs back in to on-site mode

• QR completion: The clinician selects the QR code button - displaying a scannable QR code on-device. The participant can the complete the on-site assessment there on their personal device

• Remote completion: Whenever a clinician elects to administer an on-site ePRO with the ‘Continue’ button - a back-up copy of the assessment, identical to Castor’s standard web survey offering, is sent directly to the participant for completion remotely. This ensures that, should the participant need to leave site or otherwise be unable to complete during their time on-site, they can still submit data from home.

Participants can effectively switch between these options as is convenient and per study design. Imagine a participant starts completing on their personal device via QR code but their personal device then runs out of battery - they are now able to switch to the on-device completion mode, exactly where they were previously. Next, the participant is running out of time on-site and has to leave before completing - but they can still pick up using their remote option once they get home. This ensures data can be captured from anywhere and maximises flexibility in study design. As long as lock on completion is enabled for the relevant survey package - once completed, regardless of completion method, the survey package will be locked for any further edits or resubmission by the participant.

How do you determine which participants and surveys are selectable for On-Site mode?

To be able to be administered in on-site mode participants must be:

• Active - I.E. not locked, archived or already completed

• Assigned to a site the logged-in clinician has the aforementioned permissions for

  • For example, if you are responsible for two sites, but only have ‘send survey’ and ‘manage participants’ permissions for one site - participants for that singular site will be listed in on-site mode

To be able to be administered in on-site mode survey packages must be:

• Active - I.E. not locked, archived or already completed

• Assigned to a visit for the selected participant

Can on-site ePROs only be administered at certain times, similar to ‘scheduling’ a survey package?

No - we have made the assumption that if a participant is on-site, even if that happens to be a little earlier or later than expected - clinicians may still want to administer an on-site assessment. We don’t want data capture to be blocked, just because a participant arrives at site a day late (within protocol tolerances.) To that end, when you administer an on-site ePRO:

• If the survey package didn’t already have a scheduled date to be made available: then it is scheduled for ‘right now’

• If the survey package was already scheduled in the past: the original scheduled date is retained for auditing, but the survey package is effectively rescheduled for ‘right now’

• If the survey package was scheduled to be available in the future: it is effectively rescheduled to ‘right now.’

Does this include Castor Connect surveys?

At launch we will not be offering on-site equivalents of Castor Connect surveys. However, if you want to use Castor Connect in combination with on-site ePRO, you would be able to by configuring specific ‘On-site’ versions of the relevant survey packages and assigning them to participants.

When I select a language for administering an on-site ePRO, which text changes?

The language select option controls the display text of non-assessment text presented to the participant e.g. Next, Back, Select. It does not change the language of configured assessment content - which must be configured and selected accordingly, consistent with our existing web survey feature.

Miscellaneous

• We added pagination to the "Exports" page, so that more than 25 exports can now be reviewed and downloaded.

• We have added an additional check for uploading images in upload fields. Uploads of web files are not permitted anymore due to compatibility issues.

System defects fixes

• We’ve corrected a data import issue where decimal values with trailing zeros (i.e. 60.50) were not being imported as expected when uploading repeating data via CSV. Previously, the system would incorrectly store these values as 60.5, leading to validation errors in fields configured for two decimal places. This fix ensures that imported data now retains its full precision, preserving consistency and accuracy across datasets

•  We fixed an issue where exports could hang in a queued or 'in progress' state for an indeterminate amount of time. All exports are now automatically marked as failed after 12 hours of waiting.

•   Resolved an issue where surveys containing exclusion calculations and validations as a discrete form separate from their data points could refer participants back to the datapoints form despite having triggered the exclusion validation.

  • Please see Hotfix release 2025.2.1.1 for the latest

•   We fixed an issue in the Source Data Verification CSV export where the visit and form names of fields would not be rendered. Additionally, full usernames are now displayed in the export instead of user first names only. 

• We fixed an issue where gridfields were not rendered correctly in the Empty CRF export.

• We fixed several issues with the rendering of slider fields in the Empty CRF exports.

• We fixed an issue where users without Manage Forms permissions could not export an Empty CRF PDF via the Participants Overview page.

• Resolved an issue where the web survey access code system was expiring after 4 minutes instead of 20.

• Resolved an issue where, when updating a slider to use the updated MUI slider (typically for EQ-5D) the value displayed in the slider picker/number indicator was including decimals when viewing previously entered responses, potentially confusing or concerning users.

Hotfix Release 2025.2.1.1 | Release Date: August 22nd

• Reverted a change in the previous release that prevented participants being referred back to another form despite having triggered exclusion on reloading the page. Though this solved the problem for some studies/customers - it prompted issues in the workflow of others. The team will assess options for long-term resolution that works for the full spread of studies.

Minor Release 2025.2.2.0 | Release Date: September 8th 

New features and / or  enhancements

Participant Casebooks

• A new and improved implementation of the Participant Casebooks is now available upon request. A Participant Casebook is a PDF-type export that shows all data for the participant in a form-based overview, with audit trail and query tables available for each form. The PDFs are compliant with the FDA’s PDF Specifications. The feature is currently available upon request from our Support team and is due to performance constraints currently limited to 10 Participant Casebooks per batch.

New supported languages for web surveys

• A number of additional languages have been made available for participants completing surveys using Castor's web surveys solution, as well as 'locales' or country variations of languages. New languages/variants include:
  • 简体中文 (CN) / Simplified Chinese
  • Dansk (DK) / Danish for Denmark
  • Nederlands (BE) / Dutch for Belgium (Flemish)
  • Français (CH) / French for Switzerland
  • Français (BE) / French for Belgium
  • Deutsch (BE) / German for Belgium
  • Deutsch (CH) / German for Switzerland
  • Italiano (CH) / Italian for Switzerland
  • Splitting of Spanish into: Español (ES) / Spanish for Spain and Español (US) / US Spanish
  • Ελληνικά (GR) / Greek for Greece
  • Slovenčina (SK) / Slovak for Slovakia

Updates for users of Castor Identity with CDMS

• Users migrated to Castor Identity can now manage their account via the new My Account panel, accessible from the side menu in CDMS. This provides access to updated account settings in a centralized location. Users who have not yet been migrated are not affected.
• CDMS now enforces a temporary 12-hour lockout after 5 failed login attempts for users migrated to Castor Identity. The lock can be lifted by resetting the password or waiting for it to expire.
• Users without two factor authentication enabled attempting to access studies that require it will now be redirected to set it up before being permitted to proceed. 

System defects fixes

Data Entry

• We have resolved an issue in Modern (React) Data Entry, where summary fields did not display the selected values, if the source field used an option group with more than 250 options. 

Mid Study updates with Form Sync

• A defect was resolved where the latest version of the Interim Data Impact Report would fail to generate for studies using it, but without also having the 'Validations V2' feature flag enabled.  With this fix, interim reports are now correctly generated by omitting the validations part when the study has opted out from enabling the new Validations logic, ensuring reports can be produced without errors in mixed-feature flag setups.
• We've fixed an error in the Monitoring module that caused Form Sync Data Impact reports to fail when an SDV Plan had no selected fields. The system now correctly treats null values as empty lists, allowing reports to generate successfully even when no fields are selected in the SDV Plan.

Export

• We fixed an issue where for some older studies SPSS data could not be exported when they contained a specific slider configuration.

API access

• We fixed an issue where API access via the OAuth2 Authorization Code flow would be blocked due to the authorization code's expiration date being evaluated against the wrong timezone, leading to issues when requesting access tokens on the AU cluster.

Survey packages | ePro

• We have resolved an issue where newly created Survey Package Instances, in studies without secure web survey access codes enabled, could not be opened when using On-Site ePRO. This issue occurred even for users with the correct permissions and affected both clinician and participant access. SPIs can now be opened as expected.
• A defect was fixed where clinicians opening a completed, but unlocked Survey Package Instance (SPI) would incorrectly trigger an “Survey package invitation completed” audit trail event and update the `updated_on` timestamp, even though no changes were made. With this fix, no audit event is logged and the “updated on” value remains unchanged, unless an actual update to the survey data occurs.
• We have resolved an issue where newly created Survey Package Instances, in studies without secure web survey access codes enabled, could not be opened when using On-Site ePRO. This issue occurred even for users with the correct permissions and affected both clinician and participant access. SPIs can now be opened as expected.
• Participants are no longer prompted for a “Reason for change” when removing uploaded files in a file upload field in a survey.
• Resolved an issue where in a subset of circumstances the finished on timestamp for a completed survey was being updated when a clinician viewed it in-platform.

Maintenance Release 2025.2.3.0 |  Release Date: September 9th 

System defects fixes

Survey packages | ePro

• We have addressed an issue where web surveys could sometimes get stuck in a “Processing” state when calculations with #allowempty# configuration were rendered before the calculation service had fully started. Calculations will now only run once the service is properly initialized, preventing surveys from being blocked.

Castor Identity

• We’ve started moving users to our new login screen, Castor Identity. This update makes it easier and more secure to access Castor: you’ll use a single set of credentials across all Castor products and data regions.
  Migrations are happening gradually, on a study-by-study basis. We’ll reach out directly to study's main contact when their study is scheduled.
  You can already read what this change means for you here: Moving CDMS users to Castor Identity.

Hotfix Release 2025.2.3.1 | Release Date: September 15th 

• We deployed updated SSL CA certificates for our database connections, as a preparation for a change by our hosting provider.

Maintenance Release 2025.2.4.0 |  Release Date: September 23rd 

New features and / or  enhancements

Participant Experience | Web Surveys ePro

• Resolved an issue where the 20-minute timeout for web survey access codes was being applied based on initial access, rather than activity. It now refreshes based on user activity, preventing a hard cutoff and improving completion rates.

Empty CRF Export 

• Added a configuration to hide Validations in the Empty CRF export.
• Added a configuration to hide Field Dependency information in the Empty CRF export.
• Updated the Export Empty CRF dialog to display all formatting options in the viewport by default — no scrolling needed. Reduces likelihood of missing a setting.
• The Empty CRF export now includes a hyperlink to the related Repeating Data form for Repeated Measure fields, improving navigation and context.

Study related notification emails

• We have updated the reply-to email header for study-specific notification configuration emails. Previously, these notifications were sent with a reply-to address set to support@castoredc.com, which risked misdirecting participant replies to our support team. Going forward, these emails will use no-reply@castoredc.com as the reply-to address. If a recipient attempts to reply, the message will bounce, making it clear that the mailbox is not monitored.

System defects fixes

Data Entry | EDC

• Fixed an issue where the final digit in the year value was cut off in certain date formats during data entry. Now displays full year consistently.

CRF Export 

• Fixed a bug where Empty CRF exports failed to render when Option Group names included trailing whitespace. This resolves a recurring LaTeX compilation failure.

Participant Experience | ePro

• Addressed an issue where surveys could freeze in a “Processing” state if calculation fields using ##allowempty## were triggered before the backend service was ready. Now calculations wait for proper initialization.
• Resolved an issue where participants whose access code for a web survey had expired were being redirected to an error page, rather than the access code screen which is more helpful

SSO | Access

• Resolved an issue where URLS for redirecting were being cleared after authenticating with 2FA, disrupting the user flow

Maintenance Release 2025.2.5.0 |  Release Date: December 11th 2025

• We updated one of the system's underlying libraries related to queue handling in order to improve queue handling reliability.