What The UDI Date Format Says About FDA’s Direction

Jay CrowleyThe U.S. FDA is poised to publish their final Unique Device Identification (UDI) rule any day now.  The publication was due earlier in the summer but has apparently been held up in the review by the Office of Management and Budget (OMB) without any kind of accurate timeline.  The organizers of this year’s UDI Conference are hoping the FDA will be able to announce its publication at their event this Thursday and Friday in Baltimore.  I understand that not everyone is looking forward to its publication, but some are kind of tired of waiting for it.  I’ll bet Jay Crowley and his team at the FDA are in that camp too, but the exact timing probably isn’t that controllable.

Mr. Crowley is a senior advisor for patient safety in the FDA’s Center for Devices and Radiological Health (CDRH) and has become, quite literally, “the face” of the FDA’s UDI program.

I find it fascinating that the most controversial part of the entire 44 page proposed UDI rule, published for comment in July of 2012, was the date format.  That’s right, the date format.  The proposed rule said:

“Proposed § 801.18 would require all dates provided on medical device labels to conform to a specified format: Month Day, Year, with the month shown as a three-letter abbreviation of the month (e.g. SEP 30, 2012). This format—Month Day, Year (SEP 30, 2012)—is the format most commonly used in the United States and is the format most familiar to patients and consumers.”

FDA UDI label example
The FDA’s example device label conforming to the proposed UDI rule from last year. The red oval is around the date code in the format originally proposed but now apparently switched to something else. Click image to enlarge.

This apparently generated lots of comments from people who were unhappy that the proposed date format would be so America-centric, among other things.  The United States is the only country whose citizens typically format dates with the sequence, “Month, Day, Year”.  Everyone else in the world apparently likes the sequence “Day, Month, Year” (which is a logical progression from least significant component to most significant component).  In that way this issue is kind of like the metric system which has been adopted by every country in the world as the official system of weights and measures except Burma, Liberia…and the United States of America.

But scientists prefer the sequence “Year, Month, Day” because it is a logical progression from most significant to least significant.  This sequence is defined in the ISO 8601 standard.  This sequence has some support in the industry.  By the way, this sequence also conforms to the GS1 General Specifications for encoding dates in barcodes, except that Application Identifiers (AI) 11, 13, 15 and 17, showing their pre-Y2K legacy, use only a two digit year instead of four digits.

However, the purpose of mandating a specific date format on medical devices is to decrease interpretation errors made by medical professionals and patients.  The FDA wants the dates on device labels to be easy to interpret and understand—for Americans.  On the other hand, industry wants a single global standard approach, which is exactly what the International Medical Device Regulators Forum (IMDRF) wants.  The U.S. FDA—and Jay Crowley himself—is one of the biggest driving forces behind IMDRF and their goals.

So what will the FDA do?  FDA News claims that the FDA has decided to adopt the “European” format of “Day, Month, Year”.  We’ll see.


I must admit that I’m really surprised that something as small as the date format was the biggest controversy in the entire proposed UDI rule.  I don’t mean to say that the date format is not important—it is—but the fact that little else struck a chord in the industry is testament to the thought and effort that went into the proposed rule’s construction.  Well done Mr. Crowley and team.

At last year’s UDI Conference, I recall that Mr. Crowley told the audience the original date code format contained in the proposed rule was an unintended oversight; that the FDA did not intend to create a new format standard in the face of existing international standards.  He acknowledged at that time that the FDA had already decided to change the date format to align with an international standard.  I don’t recall that he identified exactly which standard they would align with, but just the fact that they made this acknowledgment, I think, is significant.

It says that the FDA isn’t in the business of defining new standards for things that are already standardized internationally.  The proposed date format was an oversight because someone forgot to look into the possibility that an international standard already existed (whether de facto or de jure).  Once they were reminded that one did, they quickly decided to abandon their proposal.

In my view the FDA’s Standardized Numeric Identifier (SNI) guidance also followed this logic.  The FDA could have come up with their own way of combining the National Drug Code (NDC) with their own definition of a serial number, but instead, they embraced ISO standards that allowed them to align with the serialization standards of three different existing standards organizations (GS1, HIBCC and ICCBBA) (see “FDA Aligns with GS1 SGTIN For SNDC”).


We’ll find out if the FDA takes this approach with the pharma track & trace standard that they are currently working on when they publish their draft guidance, supposedly sometime before the end of this year (see “InBrief: FDA To Publish Track & Trace Standard By Year End“ from 2012 and “InBrief: FDA Again Says It Plans To Publish Track & Trace Guidance By Year End“ from earlier this year), but I can’t see why they would take a different approach.  If the FDA has no interest in establishing a new standard for something as minimal as the sequence of components in a date format, why would they choose to establish a new standard for something as complex as track & trace of pharmaceuticals in the face of the one that the industry has been working on for over six years now?  I’m referring to the Event-Based ePedigree (EBeP) standard (formerly called Network-Centric ePedigree, NCeP) that GS1 is working on.

Admittedly, the EBeP standard is not yet complete, but that’s just a reflection of how complex the task is.  It is being built on top of GS1’s Electronic Product Code Information Services (EPCIS), Global Trade Item Number (GTIN), Core Business Vocabulary (CBV), Tag Data Standard (TDS) and many others.  To create a track & trace standard that is robust, flexible and interoperable it takes the application of an incredible amount of technical knowledge and thought.  Building such a standard on top of a base of existing standards is the way to do it right.

By now I’m sure I sound like a broken record.  This is the same track I arrived at in the conclusion of last week’s essay (see “FDA Chooses DUNS For Unique Facility Identifier”).  Sorry.  And so I will abruptly conclude here.  See you at LogiPharma and the UDI Conference this week.


2 thoughts on “What The UDI Date Format Says About FDA’s Direction”

  1. Final FDA Rule Out for Medical Device Unique Identifiers

    “Standard Format for Dates Provided on a Device Label.
    The rule will also better ensure dates on device labels are not confusing or misleading to users thereby ensuring the safe use of devices, by requiring that dates on medical device labels conform to a standard format consistent with international standards and international practice–year-month-day (e.g., 2013-09-30).
    This will ensure dates on medical device labels are unambiguous and clearly understood by device users.”

  2. DD/MM/YY also allows extending to a Julian Date form for highly sensitive expiration such as fertility treatments. HH/DD/MM/YY/ can even prefix with a serial number, postfix with a lot number. SN/ HH/DD/MM/YY/LN and still easily be human readable (without a “decoder-ring”)

Comments are closed.