“Except for certain circumstances, you do have to add a linear barcode, even if a 2D code is provided. We hope manufacturers were paying attention to that,” said Christopher Smith, Director of Federal Public Policy at the National Association of Chain Drug Stores (NACDS), when he addressed the 11th Pharma Packaging and Labeling East Coast 2019 conference.
Though pharmacists don't really have to start transacting with serialized product until 2020, the NACDSis looking at the practical aspects of serialization to ensure patients maintain access to their medications.
Practical data transfer
How do pharmacists know if a drug is legitimate grandfathered product and that the manufacturer met their obligation? Smith said that at this point, pharmacies are relying on the Transaction Statement (TS).
There are still many lingering questions, he noted, such as:
-
What if the product arrives at the pharmacy before the data does?
-
What about delays caused when the 3 Ts—Transaction Information (TI), Transaction History (TH) and TS—aren’t received by a pharmacy directly but at a centralized location or third party?
If pharmacists are not able to verify the data right there where the product is, and data must be verified through a third party, it can be burdensome and slow down the supply chain.
Barcode issues
Smith said that pharmacists are continuing to see problems with barcodes. Issues arise when manufacturers do not communicate product changes quickly enough for pharmacies to be prepared. “If there’s an unexpected barcode—a company switches without notification—it can slow down your automated processes,” he explained. Additionally:
-
Not all dispensers have 2D-scanning technology employed yet.
-
Manufacturers introduce GS-1 stacked barcodes where there isn’t enough room on the package.
-
Linear NDCs are being left off in favor of 2D codes. Based on FDA’s Q&A draft guidance, this should not be occurring.
-
Color choices can cause problems when Product Identifier barcodes blend in with marketing information text.
-
If two barcodes are printed too close together, scanners may pick up the wrong one.
-
Smudging can result during the laser ablation process.
-
Outserts can block part of the barcode so that it is unreadable or difficult to read.
Smith also referenced the 2018 GS-1 assessment that found that only 6.6% of packages scanned in the study contained usable barcodes in 2017, and that in 2018, only 20.7% of packages had usable barcodes with all four data elements. “This is a major problem and if those numbers don't change drastically… we’ll have a serious problem,” he said.
An audience member asked if, in the case of very small drug packages, it is still acceptable if the UPC code goes slightly below the GS-1 standard of 80% magnification factor, to 78 or 79% for example. Smith said NACDS members want to make surethey can read codes, but that they are not so concerned with whether the codes technically passes standards, i.e. if you shrink the code a small amount, but your trader partners are not complaining about readability, it should be acceptable.
FDA draft guidance
Smith pointed out that his association was generally happy with the Q&A draft guidance, particularly where it required that a linear barcode be included, even if providing a 2D code (except for certain circumstances). The NACDS is also in favor of the recommendation that the barcode be placed near the human-readable portion, if space permits.
Manufacturers and repackagers can opt to place the PI on individual units, smaller than the saleable unit, if they choose to. For example, five vials of a given drug may be the smallest saleable unit, but pharmacists typically only need one vial. “It’s not required certainly, but the manufacturer may serialize the unit that the pharmacy deals with,” Smith said. “It’s a business-to-business decision based on the trading partners.”
Aggregation and inference
Smith noted that aggregation is a critical piece of the puzzle, as pharmacies can’t scan each individual unit given the massive amount of products that move through the supply chain. They will have to rely on scanning the parent unit to get the child unit information.
Looking ahead, the NACDS has asked FDA for more guidance on how errors on inference and aggregation should be handled. With estimates of 4.5 billion units passing through the supply chain annually, even high accuracy rates will result in many errors.
What happens when the data doesn't match the product, for example where there’s a list of serial numbers and one product is missing, the serial number is wrong for a product, or there’s an extra product present that’s not on the list?
The association’s general view is that stakeholders should resolve these issues amongst themselves, leaving it to the businesses how to deal with errors. Does every error, even if it’s clerical or administrative, need to be flagged as suspect product? If so, this has the potential to slow down the whole supply chain and potentially cause patient access problems.
Smith also noted that the FDA is running a pilot project, accepting applications through Mar. 11. This may provide additional clarity, once complete, on some of the practical concerns. The industry is still waiting on wholesaler licensing standards from the FDA.
