When Will The DSCSA Ever Require Investments In Aggregation?

Packaging Hierarchy.  Drawing by Omega Design
Packaging Hierarchy. Drawing by Omega Design

I have been outspoken on the question of whether or not the Drug Supply Chain Security Act (DSCSA) requires companies in the supply chain to provide their customers with serial number-based aggregation data prior to 2023.  In my view, it does not, but others disagree, saying that there are requirements in the law that lead to the need for aggregation data during that time.  I do not agree with that either.  If you would like to review those arguments and find out exactly what “aggregation data” is, here is a list of RxTrace essays you should read:

Date Published Title/link
March 26th, 2012 Pharma Aggregation: How Companies Are Achieving Perfection Today
November 22nd, 2013 DQSA: Will U.S. Pharma Distributors Mandate Aggregation Data In Phase 1?
February 10th, 2014 Does The DQSA Require Manufacturers To Provide Aggregation Data? Survey Says…
June 9th, 2014 The Aggregation Hoax and PIA


The DSCSA requires the FDA to conduct at least 5 topical public meetings between now and about 2021 to collect stakeholder input on topics that will help the FDA define how the so-called “Enhanced Drug Distribution Security” (EDDS) phase will work.  That phase is to begin in November of 2023, which is quite a few years from now, but the law contrains details about how that phase is to work.  One of the topics Congress instructed the FDA to collect stakeholder input on in preparation for the EDDS is the systems and processes needed to utilize the serialized product identifiers to enhance tracing at the package level, including allowing for aggregation and inference.  I don’t think they are likely to hold that public meeting for quite a few years (see “Interoperability And The DSCSA”).

One of the few things the law says must occur as part of the EDDS is that all members of the pharma supply chain must include the unit-level serial numbers in the Transaction Information (TI) that they provide to their customers in each shipment.  The EDDS will be the beginning of the fully serialized pharma supply chain in the U.S..  The DSCSA mandates very little use of serial numbers prior to that date.

But after that date, for anyone selling drugs within the supply chain, they will need to know exactly which unit-level serial numbers are contained in their shipments so they can include those serial numbers in their TIs.  To do that, they will need their own serial number-based packaging hierarchy data—that is, they will need their own aggregation data—whether or not they provide that data to their customers.  Anyone using a third-party logistics provider (3PL) will also need to provide that aggregation data to them as well so they can properly document the manufacturer’s shipments.  Odds are, customers will justifiably demand to receive that same data after 2023 to assist them with their receiving verification process, even if the FDA does not require the seller to provide it.

Given all of that, I think it is very clear that the operation of the EDDS that Congress intended will necessitate—if not mandate—the capturing and sharing of aggregation data by manufacturers, repackagers and wholesale distributors.  At that time, everyone will view aggregation data as essential to the smooth operation of a secure supply chain.  Fortunately, the technology to ensure the accurate capture of aggregation data has been available for several years now (see “Pharma Aggregation: How Companies Are Achieving Perfection Today”).

Last week an industry friend of mine told me about a new theory going around about aggregation—this one, for the EDDS after 2023.  As the theory goes, the FDA would become convinced through the public meetings that manufacturers and repackagers would not be capable of capturing “certifiably accurate” aggregation data and so errors would be inevitable, which would lead to ineffective operation of the EDDS.   As a result, the theory goes, the FDA would invoke a provision in the DSCSA that allows them to establish “alternative methods” for any EDDS requirement (see DSCSA Section 582[g][2][B]), and the FDA would choose to replace the need for seller aggregation data capture by requiring the recipient to capture all of the serial numbers they receive.   That’s the theory anyway.


If that happens, it would effectively eliminate the need for manufacturers and repackagers to capture and pass aggregation data.  Problem solved, right?  Not really, because it would simply push that burden onto the buyer who would now need to open all shipping containers and cases upon receipt and scan all of the unit-level barcodes.  Because 87% of the drugs passing through the U.S. supply chain go through only three wholesale distributors, that means these three companies would bear the burden that was removed from many hundreds of manufacturers and repackagers.  Rather than distributing the problem it would intensely concentrate it.  That’s not a solution.

But there are other problems with this theory.  What happens when there is a cargo theft during transit between the manufacturer and the wholesale distributor?  Since the manufacturer would have no idea which unit-level serial numbers were in the truck, they would not be able to update their verification database with the units that were stolen.  That means stolen product would continue to be verified by the manufacturer as legitimate product.  Whoops.  Bad idea.

But the biggest reason this theory is nonsense, in my opinion, is that it is based on a misreading of the DSCSA itself.  That section on “alternative methods” that is the basis for the theory does not apply to all participants in the supply chain.  It only provides the FDA with the latitude to change provisions that impact small dispensers—defined as dispensers with 25 or fewer full-time employees—and (possibly) other dispensers where the provision results in undue economic hardship.  So maybe small dispensers will not need to receive and make use of aggregation data, but that’s about the only effect that provision could potentially have on aggregation.  In my opinion, it just does not apply to manufacturers, repackagers or wholesale distributors.


The 2015 RxTrace U.S. Traceability Survey Analysis, sponsored by Frequentz will not be available for a few more weeks, but here is a preview.  One of the findings is that a large percentage of respondents—including a large number who work for drug manufacturers—are planning to capture aggregation data even prior to 2023.  So whether or not the law requires those companies to capture and share aggregation data after 2023, they are already planning to capture it prior to that date.  And they will be doing it during the time that the FDA will likely hold that public meeting to discuss aggregation and inference.  This seems to make what companies will do after 2023 a moot point, at least for the manufacturers represented in this survey.  But the fact that so many companies will be capturing and using aggregation data prior to 2023 is going to make it very hard for companies trying to fight aggregation as a requirement to claim that it cannot be done.

Disagree?  Leave a comment below.


7 thoughts on “When Will The DSCSA Ever Require Investments In Aggregation?”

  1. Smart businesses wanting to provide service beyond the script directly to patients have already understood the commercial value in understanding exactly what patient has what product. Serialization enables service oppprtunities. Service opportunities present especially with specialty pharma. Specialty is growing. For many reasons specialty is moving toward home administration. Serialization of connected drug delivery devices is already a reality. SEE: Internet of things and first mover advantages.

  2. “But after that date, for anyone selling drugs within the supply chain, they will need to know exactly which unit-level serial numbers are contained in their shipments so they can include those serial numbers in their TIs.”

    At what point does the inferred drugs get verified?

    1. Anonymous,
      We won’t know how and when inference will be allowed by the FDA until they make a lot of progress on their understanding of aggregation and inference and their implications. But in my view, all use of inference should be resolved, with the units involved in the inference being verified, at some point before a patient is dispensed the drug. This might happen when the wholesale distributor ships the product, or it might occur when a large dispenser unpacks a manufacturer’s case. Whichever company opens the manufacturer-packed case should be responsible for “closing” the inference that has occurred prior to that point in the supply chain by verifying that the aggregation data provided matches the contents of the shipment. But, again, this will need to be the subject of industry negotiation with the FDA as part of the public meeting(s) surrounding this important topic.


  3. Sorry Dirk, but this statement is incorrect.
    “But the biggest reason this theory is nonsense is that it is based on a misreading of the DSCSA itself.”

    That section reads
    alternative methods of compliance with “any” of the requirements set forth in paragraph (1), including establishing timelines for small businesses…..

    Thank you for your perspective.

    1. Anonymous,
      Your interpretation of just the few words you cut and pasted from the law is probably correct. But you have to read the whole section to realize that Congress was specifically talking about the alternative methods only for small pharmacies. I’m referring to the fact that the provision requires the FDA to take into consideration the assessment required in subsection 582(g)(3) when they consider those “alternative methods”. You will find that the assessment is to determine the feasibility of small pharmacies conducting interoperable, electronic tracing of products at the package level. I suppose it is at least possible that the FDA could interpret the section you quote to mean that they can allow alternative methods for any provision of the EDDS requirements, but it would be a real stretch, and I don’t think they are likely to use this provision that way.


  4. I have always wondered how aggregated hierarchical data will be tracked in the supply chain with so many logistical complexities/hurdles. All I hear is vendors/providers talking about the “Happy Path” scenarios. Unless there is a clear understanding on inference and how the data will be captured upon shipment and receipt, this is all a consulting ‘cash cow’.

    I am also an active member of the track and trace solution implementer, but it always bothers me when the end game is not in sight.

    Authentication at the time of dispensing makes the most sense to me with what we know so far. Without proper manpower/technology, tracking in the supply chain seems unrealistic.

    I would like to hear industry comments on this from the experts.


    1. Sunil,
      Thanks for commenting. I share your concern about how the industry groups in GS1 and elsewhere always seem to stop after working on the “happy path”. Obviously, if an industry solution only works when nothing goes wrong, or, if that’s the only well-documented situation, we’re in for a catastrophe when that system is implemented. I know GS1 US and HDMA have both been spending a lot of time working on exception handling and my hope is that they have worked out how to deal with all possible exceptions, not just the happy one.


Comments are closed.