UDI representation in CDM needs improvement

[linikujp]

Currently the CDM only represent the UDI as a full string and the size of the string is too small (50 characters) to capture a full UDI string in many cases.

UDI should be parsed before being represented in OMOP CDM. The UDI has multiple parts - UDI-DI (device identifier of UDI) and UDI-PI (production identifier of UDI) and UDI-PI has sub-parts - lot, serial, expiration date, manufacturer date, distinct identification code. See https://www.fda.gov/media/96648/download for more details.

• per email between Asiyah Lin, Terrie Reed, and Andrew Williams.

[vojtechhuser]

+1

[AEW0330]

+1

[clairblacketer]

Thank you - I added a proposal tag so we can keep track of this for future versions.

[linikujp]

Thank you, Clair!

On Tue, Jan 5, 2021 at 3:06 PM clairblacketer notifications@github.com wrote:

Thank you - I added a proposal tag so we can keep track of this for future versions.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/OHDSI/CommonDataModel/issues/353#issuecomment-754868793, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACCPEPN6YIBMJGGQSZOCDGTSYNWLLANCNFSM4OR3ZZ4Q .

[clairblacketer]

Separate this into UDI_DI and UDI_PI

[aandryc]

Recommendation from MD working group is to separate UDI into UDI-DI and UDI-PI instead of expanding the single field.

[linikujp]

This topic was discussed on the MD WG meeting on 5/27/2021. UDI-PI will have the potential to make the patient information identifiable, and often the study may only require the UDI-DI, so it makes sense to separate.

[vojtechhuser]

To make a full proposal: MODIFY: unique_device_id _(rename to deviceid ?) TYPE: string LENGTH: 255

• change user guide to say DI portion of UDI
• remove talking about FDA and make sure wording is in world wide scope (use FDA only as example, add EU example)

ADD FIELD called production_id (optional), (mark this as possible PII) TYPE: string LENGTH:255

_(well, we can prefix the same way and say unique_productionid (or remove unique from the above field)

• User guidance: This is production identifier for the device. For example for FDA regulated device, it is the production ID portion of UDI
• ETL guidance: null

For vocabulary improvements - see this issue https://github.com/OHDSI/Vocabulary-v5.0/issues/501

LINKS

• https://www.fda.gov/medical-devices/unique-device-identification-system-udi-system/udi-basics

• https://ohdsi.github.io/CommonDataModel/cdm531.html#DEVICE_EXPOSURE

existing text for the first field is :

_This is the Unique Device Identification number for devices regulated by the FDA, if given.

_For medical devices that are regulated by the FDA, a Unique Device Identification (UDI) is provided if available in the data source and is recorded in the UNIQUE_DEVICE_ID field.__

[UDIis-worthit]

Sorry to have missed the 5/27 meeting. 255 seems like a reasonable length for UDI-DI and each UDI-PI but, the most important thing (and to retain UDI as global standard) is to document that UDI is a regulatory requirement on manufacturers that work with issuing agencies/entities to meet ISO standards. It is not a number generated out of that context. That said, I wanted to make sure that it is clear that UDI-DI and UDI-PI are not FDA terms. They are terms used globally by US, EU, Canada, Saudi Arabia, Korea, Singapore, Japan etc. based upon IMDRF/UDI WG/N48 Final: 2019, a voluntary agreement by 13 jurisdiction that is being adopted even beyond those represented in the original workgroup. Note that the issuing agencies (entities) specified in this document are the ones that must meet ISO standards for unique assignment of the ID and that, to date, only 4 groups - GS1, HIBCC, ICCBBA, and IFA - have been officially designated as agencies authorized to assign UDIs. GS1, HIBCC, ICCBBA are used most commonly. EU added IFA to the list but the agencies.

[Arnaud-CH31]

My colleague left the company, please unsubscribe this email address thanks Lilla Papp Dr. Project Manager & Legal Counsel

Avenue de Rumine 20 1005 Lausanne Switzerland T: +41 78 679 6309 E: @. @.> www.medows.ch http://medows.ch/

CONFIDENTIAL This message and any attachments (the "Message") are confidential and intended solely for the addressee(s). If you are not the intended recipient, any use, copying or dissemination is strictly prohibited. If you are not the intended recipient, please notify the sender immediately by return and delete this original Message and any copies from your system. E-mails are susceptible to alteration. Medows Sàrl and its subsidiaries and other affiliates shall not be liable if the Message is altered, changed or falsified.

This is an environment friendly email. Please do not print it unless it is really necessary.

On 4 Jun 2021, at 15:38, UDIis-worthit @.***> wrote:

Sorry to have missed the 5/27 meeting. 255 seems like a reasonable length for UDI-DI and each UDI-PI but, the most important thing (and to retain UDI as global standard) is to document that UDI is a regulatory requirement on manufacturers that work with issuing agencies/entities to meet ISO standards. It is not a number generated out of that context. That said, I wanted to make sure that it is clear that UDI-DI and UDI-PI are not FDA terms. They are terms used globally by US, EU, Canada, Saudi Arabia, Korea, Singapore, Japan etc. based upon IMDRF/UDI WG/N48 Final: 2019 http://imdrf.org/docs/imdrf/final/technical/imdrf-tech-190321-udi-sag.pdf, a voluntary agreement by 13 jurisdiction that is being adopted even beyond those represented in the original workgroup. Note that the issuing agencies (entities) specified in this document are the ones that must meet ISO standards for unique assignment of the ID and that, to date, only 4 groups - GS1, HIBCC, ICCBBA, and IFA - have been officially designated as agencies authorized to assign UDIs. GS1, HIBCC, ICCBBA are used most commonly. EU added IFA to the list but the agencies.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/OHDSI/CommonDataModel/issues/353#issuecomment-854731019, or unsubscribe https://github.com/notifications/unsubscribe-auth/APM24JASLAA7MMAEYTMISG3TRDJOJANCNFSM4OR3ZZ4Q.