Normally drugs dispensed by American pharmacies are repackaged into the “standard” amber vial. The pharmacy places their own label on that bottle so the patient normally doesn’t receive the manufacturer’s package that would have the new 2D barcode mandated by the Drug Supply Chain Security Act (DSCSA). There are a few drugs that are not repackaged by US pharmacies, like most things in an inhaler and drugs in “compliance packaging” like birth control pills. But even then, the pharmacy puts their label on the package
One of the few drugs I take daily is a statin, which has always been put into an amber bottle by my pharmacy. A few months ago I received a three months supply, and to my surprise, the pharmacy dispensed the prescription in the manufacturer’s original 90-count bottle, and there was the DSCSA 2D barcode on the label. The pharmacy label was positioned so that it formed a “flag” and did not cover the 2D barcode, and the pharmacy label was easily removed to expose the entire manufacturer’s label. This drug was made by Lupin Pharmaceuticals, a generic drug manufacturer base in India. How did they do?
Actually, not bad at all. In my opinion, they would almost certainly comply, with only a couple of fairly minor things I would have done differently. Here are my observations:
The linear barcode is still present, fulfilling the requirement of the FDA’s 2006 Barcode Final Rule
- The separation between the linear and 2D barcode is excellent so that wholesale distributors and dispensers will not have any trouble scanning one or the other
- I don’t have a barcode quality analyzer but the barcode looks visually good. I would bet it is a grade B or better. I’m using the Cognex barcode app on my Droid smartphone to read it and the read is very snappy. That’s not the best indicator. The image looks like it is printed with an ink jet printer and it is very square. The individual datamatrix cells are clearly square with very sharp edges.
- The only negative thing I can say about the Datamatrix barcode is that it is not quite black. Comparing the color to the linear barcode, which is offset printed as part of the static information on the label, it is a little grey, but that shouldn’t be a big deal.
The encoding of the data looks perfect to me. The expiration date in the human readable is the same date encode. Both have the day of the month shown/encoded. There is only one FNC1 character encoded and it is properly at the end of the lot number, preceding the AI 21 for the serial number. The serial number is the last field in the barcode. That conforms to GS1 standards and if anyone in the supply chain complains about that, I will get good laugh out of it. I think everyone is on-board now with not imposing any additional restrictions beyond the GS1 General Specifications for the encoding of the barcode, but we’ll see. (There was a time when a few powerful companies were going to try to demand a certain sequence of the data in the encoding.)
- The GTIN encoded in the linear barcode is the same GTIN encoded in the 2D barcode, as it should be.
Lupin has chosen to show only the text abbreviations for the data elements rather than including the goofy GS1 Application Identifiers, even when the encoding of the barcode matches the printed data format (see “I’ve Changed My Mind, Follow GS1’s HRI Specification” and the earlier essay “The DSCSA Product Identifier On Drug Packages”). In my view, Lupin has made the right decision because it eliminates potential confusion by the healthcare professionals who will read this data. But as someone who has lots of friends who work for GS1, I am obligated to tell you that GS1 still recommends that you include the AI’s on healthcare items when you can. It’s just a matter of time before they finally agree that no one is ever going to use those AIs in the human readable so they should never appear there.
- But now we come to the part that they have chosen to follow GS1 standards in a way that I don’t advise. They have included the GTIN in human readable and have not included the National Drug Code (NDC) next to the barcode. This is not a big deal because they have included the NDC elsewhere on the package, which some feel is all you need to do to cover the DSCSA requirement (see “Human Readable Text Controversy”). Lots of companies—probably most companies—are taking this approach. What I recommend is to print both the GTIN and the NDC in the human readable next to the barcode. That way you cover yourself in case the FDA ever issues guidance saying you have to print the NDC in the human readable, like the law seems to say, and you will keep GS1 happy. Of course, I think GS1 agrees with my ultimate recommendation: to get rid of the NDC system and replace it with one like the Unique Device Identification (UDI) approach where GS1 GTINs would be one type of identifier that could be used (see”An Open Letter To The FDA, RE: Please Fix The National Drug Code Soon” and “NDC Nearing Its End, Afflicted by ‘Identifier Failure’”). But the FDA doesn’t seem to see much urgency in that, since they think they can just add another digit to the Labeler Code and that will solve the problem (it won’t, that would be a catastrophe for the industry).
I think that’s it. Well done Lupin. You’ve passed my test. Unfortunately for you, that doesn’t mean much. I still recommend that you, and others, send your labels to the “Big-3” wholesale distributors to have them give you their evaluation (see “Wholesalers Find Troubling Results In DSCSA Barcode Assessment”). Let me know if they don’t agree with my assessment here.