Last week I wrote about the DSCSA Product Identifier on Drug Packages in the United States. Last month I wrote about shipping container/transport package identification under the Brazil National Medicine Control System (SNCM) (see “ANVISA And The SSCC Controversy”). Today I will take a look at drug package identifiers under the SNCM as regulated there by the National Agency of Sanitary Surveillance (ANVISA). Most of the factual information included here is based on the following documents:
- a Q&A document distributed by GS1 Healthcare last year to the members of their Public Policy Work Group,
- other translated documents obtained through GS1 Healthcare and provided by Interfarma (roughly the equivalent of PhRMA in the US), and
- the document “Support Guide for Codification of Medicines To comply with RDC 54/2013 (Version 2)” by GS1 Brazil,
with my additional thoughts. I am a member of GS1 Healthcare but I cannot provide you with their documents since they are for members only. Contact GS1 Healthcare for more information.
Under the SNCM, manufacturers must apply a Unique Medicine Identifier (IUM) to all drug packages imported or sold into the Brazilian supply chain and trace them through the supply chain by December 11, 2016. By December 11 of 2015, manufacturers must apply the IUM to the packages of at least three lots of drugs and trace them to the dispensing points. And you thought compliance with the DSCSA was going to be hard to achieve!
Let’s focus on the IUM. According to the English translations of various ANVISA resolutions and the GS1 Healthcare Q&A with ANVISA from last year, the IUM must be composed of, and in this specific order when printed in human readable on the drug label:
- The ANVISA medication registry number for the drug
This 13-digit code is comparable to the NDC in the US.
- A serial number
This must be 13 digits chosen by the manufacturer and must be unique to the specific package it is printed on and unique across all products marketed in Brazil by that manufacturer. This is different from the definition of a GS1 serial number because it is not tied to a specific Global Trade Item Number (GTIN) or even the medication registry number.
- The expiration date of the drug
When printed in human readable form, this must appear as “MM/YY”.
- The lot/batch number of the product
This may be up to 20 alphanumeric characters and is assigned by the manufacturer.
This data must also be encoded into a 2-dimensional (2D) Datamatrix barcode. GS1 standard encoding is acceptable, which includes the formatting of the encoded date as “YYMMDD” and the use of GS1 standard application identifiers (AIs). GS1 Brazil has confirmed that ANVISA does not mind if the barcode and the human readable includes a GS1 GTIN in addition to the four required elements listed above. However, in the human readable portion, the GTIN should appear either first, or last. That is apparently because ANVISA does not consider the GTIN to be a part of the IUM and they want to keep the IUM intact.
To encode the barcode following the GS1 General Specifications, AIs must be used to identify each data element. The GTIN is prepended with “01”, the serial number with “21”, the expiration date with a “17” and the lot/batch number with a “10”. However, the ANVISA medication registry number is treated by GS1 as a National Health Reimbursement Number (NHRN) which they began recognizing with its own AI only last year. The ANVISA medication registry number should be prepended with “713” as the AI.
GTIN-13 or GTIN-14?
According to the Support Guide from GS1 Brazil, in Brazil, a GTIN used in pharma on secondary packaging is only 13-digits long. That is the first thing the document points out and it provides the structure of a GTIN-13. The problem is, the document does not explain that the GS1 AI (01), holding a GTIN, only works with GTIN-14’s, so you must convert your GTIN-13s into GTIN-14s when encoding the Datamatrix barcode by adding a zero as the first digit. I guess the days of the GTIN-13 on secondary packaging in Brazil are over because even their examples later in the guide show GTIN-14s. The difference is mentioned in the section on Tertiary packaging (homogeneous cases), and GTIN-14 is recommended, but it still claims that a GTIN-13 will also work. I’m just not sure how you will do it without converting to a GTIN-14 because you are going to combine the GTIN with a Serial Shipping Container Code (SSCC) in a barcode using “01” for the GTIN-14 and “00” for the SSCC for the AIs. I think the only way to retain a GTIN-13 would be to use two different barcodes and using a symbology that is allowed to carry a GTIN-13. If you want only one barcode containing both data elements, you will need to convert to a GTIN-14.
DO NOT USE GS1’s HRI
I am very, very, very surprised that ANVISA apparently (according to GS1 Brazil anyway) will accept the human readable presentation of this data on the drug package with the AIs—known as GS1 barcode Human Readable Interpretation (HRI)—rather than the true field names. Very surprised. It seems to be a contradiction to the ANVISA goal as stated in their Technical Note 01/2015:
“The printing of 2D code information in human readable data must be made in an order and location that allows the clear identification of the IUM by the user.”
This translation uses the word “user”, which I assume includes healthcare professionals and patients, and I’m not sure if those folks are going to “clearly identify” what “(713)”, “(21)”, “(17)”, and “(10)” mean. In my view, if you are trying to ensure clear identification of the IUM by those not familiar with GS1 standards (probably more than 99.999% of the population in Brazil), you should only use words to identify the data elements. Do not use the GS1 HRI if you want the IUM to be clearly identified, no matter what they claim. I gave the same advice last week for compliance with the DSCSA
In fact, you cannot use the HRI and remain compliant because the expiration date would not be formatted correctly. GS1 Brazil includes the following figure to show how a GS1-compliant IUM might be shown on a package:
There are several things to note about this figure. First, they are using the GS1 HRI—a bad idea as I just said. The expiration date in the human readable is formatted improperly. It should be in the form MM/YY according to ANVISA. I’m not sure why they lost the month and day in the rightmost example, but neither example complies with the regulations as I read them.
The example on the right has a box around the data fields that represent the regulated IUM. I cannot tell for sure if the authors are simply showing the reader which data elements are part of the IUM, or if they are recommending that manufacturers include a graphical box on their labels just like that image. In my opinion, adding the non-regulated “GTIN” above or below the regulated IUM fields will interfere with “clear identification of the IUM by the user” and so I think adding the box is a good idea. Use your own judgment.
The GS1 Brazil document does not provide example images that do not use the HRI, but be very careful here when you replicate the description just below where the above image appears. The document translator has made a very small error that could cause your drug packages to fail compliance with ANVISA. The English translation says:
“In the human readable part, it is possible to use the AI header (e.g. (10)123ABC) or to specify the information (e.g. Batch: 123ABC).”
But according to the GS1 Brazil Q&A with ANVISA, if you choose not to use GS1’s HRI in the human readable portion, the text next to the data elements must be in Portuguese, the official language. This is very logical and not surprising, so the word “Batch” in the GS1 Brazil guide should not have been translated into English. The proper field name for “Batch” in the human readable portion would be “Número do lote” (from the untranslated ANVISA Resolution 54/2013), or perhaps just “Lote”. Likewise, the other data elements of the IUM would be in Portuguese, as in my example at the top of this essay.
Speaking of translations, I am in the process of revamping some of the formatting and infrastructure of the RxTrace.com website. Please bear with me as it might take a few weeks to get everything working right. I am now using a more up-to-date WordPress theme and I have just begun to add translations into multiple languages—Portuguese first, for our friends in Brazil (pass the word). RxTrace is read throughout the world but until now, only English readers were able to find it useful. Now I plan to use Google Translate to provide translations of new RxTrace essays from now on in lots of different languages. Over time, I will also go back and add translations of most essays from the last two years, and then selected essays older than that. Just use the “Language” selector in the pulldown box at the top of the left sidebar to select your language. If your browser is set for a given language, and if the essay you find has a matching translation, the essay should come up in that language. Let me know if it does not work.
Be aware that I am not multi-lingual. If you write to me in your native language, I will use Google Translate to understand what you have written, and I will use it to help me reply. I look forward to hearing from you!