I was initially disappointed in the FDA Drug Supply Chain Security Act (DSCSA) Pilots Workshop that was held at FDA headquarters last week, but in the end, the outcome appeared to fit the need.  Going in, I knew not to expect the FDA to convey any information to the attendees, so that is not why I was disappointed.  I attended the public DSCSA workshop they held back in May of 2014 so I already knew their typical approach for workshops like these (see “The 2014 FDA DSCSA Workshop”).  I knew that the purpose of the workshop was to inform the FDA, not to inform the attendees.  I would estimate that about one out of every four attendees were expecting the opposite, and I would bet a significant percentage of those had not even read the DSCSA once.  But that’s not why I was disappointed.

I was initially disappointed because the introductory comments by the FDA were unclear about what they were trying to accomplish in the pilots and in the workshop.  I feared that the lack of clarity was an indication that the FDA has not really been putting any thought into what the DSCSA means to the industry and their own obligations in it.  My fear was that no one at the FDA has been doing any planning since the last workshop in 2014.

If I were creating the agenda for last week’s event, I would have put a high priority on establishing at the outset the statutory requirements for the pilots and other public meetings.  I would have displayed some of the exact language that Congress provided to help signal the expectations of the pilots and the workshop.  I would have specifically talked about the transition to the Enhanced Drug Distribution Security (EDDS) phase that the DSCSA defines in 2023—what is specifically mandated by the law, and what is left up to the FDA to work out with the industry.  This would serve to place boundaries around and focus on the topics that are relevant for discussion at the workshop.  Then I would have moved into the slides that the FDA started with.

Without that level-setting introduction, the FDA’s initial presentation was difficult to follow.  Now that I look at the slides they make more sense to me than they did when I was in the meeting.  In the meeting it appeared to me (and it wasn’t just me), that the FDA was not asking the audience to focus on the 2023 requirements but on today’s lot-based data exchange and other attributes instead.  Fortunately there were enough people in the audience who have been putting a lot of thought into the unknowns of the EDDS, and there were many RxTrace readers in the audience too.  These folks ensured that the discussion at most tables focused on 2023.  For the pilot(s) conducted as part of the DSCSA pilot mandate, it will take so much time to design, execute and collect the results—then act on the results by designing a supply chain data exchange architecture—that there will be barely enough time left over for solution providers to develop the systems needed and get everyone setup before November of 2023.

At my table, one of the points made by an attendee from a drug manufacturer was that they are expecting to make significant investments to meet the wholesale distributors’ demand for aggregation data prior to 2019.  They want those investments to be in alignment with whatever is necessary for 2023 so they don’t have to discard those investments.  That means they need to have a better idea than they do right now about what that 2023 data exchange architecture will be before they make those investments.  That means they need to know very soon.  They thought the workshop discussion was going to focus on various possible 2023 data exchange architectures.  I did too.

Those architectures were mentioned by several tables in their report-outs on the second day, but the number of potential pilot topics was so large that many pilots could be conducted without touching on the EDDS overall data exchange architecture.  Let’s hope that doesn’t happen.

Here is the list of “Pilot project design consideration” listed by the FDA after the first day:

• Ensure adequate mix of products and packaging levels represented
• Include all stakeholders (types and sizes) and different trading variations
• Evaluate costs and benefits
• Flexibility of pilot projects (different partners, evolving scenarios, additional use cases and products)
• Risk-based approach to determine what to pilot (e.g., target known weaknesses in the supply chain)
• Include real life business processes
• Timing of pilots, to make the useful as trading partners implement requirements
• Human factors that could lead to challenges (errors, poor business practices)

Here is the list of “Pilot project objectives” that the FDA boiled down the notes taken at each table after the first day.  I have added links to RxTrace essays that are related to some of the topics:

• Product Identifier Issues
  • How serial numbers are issued & managed (including CMO’s role) [see “Should You Off-Load Your DSCSA Obligations To Your Contract Partners?” and “DSCSA ‘Serial Numbers’”—Dirk]
  • Compare different standards for the identifier (10 or 11 digit NDC to be used in the SNI, 14-digit GTIN) [for my proposal to address this, don’t miss last week’s RxTrace essay “An Open Letter To The FDA, RE: Please Fix The National Drug Code Soon” and also see “Anatomy Of The National Drug Code”—Dirk]
• Bar Code Quality Issues
  • Readability of bar coded printed or affixed (environmental and human factors)
  • Convergence of linear and 2D barcodes on product [see “Will The FDA Eliminate The Linear Barcode On Drugs?” and “Why NOW Is The Time To Move Away From Linear Barcodes”—Dirk]
    • Distinguishing which barcode to read/use and when
  • Determine minimum acceptable grade for bar code quality
  • Test various readers/scanner capabilities and variability
• Interoperability [see “Interoperability And The DSCSA”—Dirk]
  • Process and Technical challenges due to variety of solutions expected
    • Central database vs. peer-to-peer (decentralized)
      • [see “InBrief: The Rx-360 Traceability Data Exchange Architecture White Paper”, “DQSA: How Should Transaction Data Be Exchanged?”, “The Viability of Global Track & Trace Models”, “InBrief: ePedigree Models and Points of Failure”, “How To Make The Semi-Centralized Track & Trace Model A Reality” and “A Semi-Centralized, Semi-Distributed Pedigree System Idea”—Dirk]
    • Trading partners with systems vs. others with little to no systems or using someone else’s system
    • Maintaining visibility of the serialized product throughout the distribution supply chain
    • What to do when a trading partner goes out of business
    • Evaluate the use of EDI and EPCIS separately [see “Will EPCIS Event Exchange Replace EDI ASNs for DSCSA Someday?”, “Will GS1’s EPCIS Be Used Widely For DSCSA Data Exchange?” and “The Coming Transition To Serialized Data”—Dirk]
  • Data Issues
    • Use of technical data standards that data attributes to enable interoperable transfers [see “Will EPCIS Event Exchange Replace EDI ASNs for DSCSA Someday?” and “Will GS1’s EPCIS Be Used Widely For DSCSA Data Exchange?”—Dirk]
    • Test different methods to handle the Master data vs. transaction data separately
      • Feasibility and acceptability of sending master data only once per shipment [see “Pedigree Models and Supply Chain Master Data”—Dirk]
      • Controlling master data to minimize redundancy [see “Master Data, Supply Chain Master Data and Instance Data” and “Supply Chain Data Synchronization and Patient Safety”—Dirk]
    • Integration into individual/company data systems [see “The Future of Traceability Repositories and Inventory Management Systems”—Dirk]
    • Evaluate data format or processes for data transfer [see “DQSA: How Should Transaction Data Be Exchanged?”—Dirk]
    • Performance measures (e.g., how to evaluate data from beginning to end of the product lifecycle, and vice versa, can you ascertain the actual change of ownership and transaction flows when examining data extracts)
    • Management of the system or data: use a consortium
    • Maintain data integrity/accuracy through distribution [see “InBrief: ePedigree Models and Points of Failure”—Dirk]
    • Performance of the database when full or partially loaded [see “Could It Be The Cloud? More Thoughts On IBM’s Divestiture Of Its EPCIS And E-Pedigree Suite”—Dirk]
  • Database/System Issues
    • Controlled/limited access to data by trading partners, FDA or other federal or state officials (data governance) [see “Who owns supply chain visibility data?”, “Data Ownership In The Track & Trace Cloud” and “Global Traceability Data Exchange: Troubled Waters Ahead”—Dirk]
    • Status of product at all levels (each, case, pallet): g., expired, illegitimate, data error, associated decommissioned product identifier [see “The Coming Battle Over Decommissioning At The Pharmacy” and “Should Pharmacies Decommission EPCs Upon Dispense?”—Dirk]
    • Process for redaction of data (may not need to provide all data downstream)
  • Aggregation/Disaggregation
    • Evaluate processes for product flow and data flow [see “Aggregation –> Chargeback Accuracy –> ROI”—Dirk]
    • Identify gaps in data or errors, accuracy of data, particularly downstream when searching or examining the data; how can errors be corrected [see “Pharma Aggregation: How Companies Are Achieving Perfection Today”—Dirk]
    • Impacts when inference is used vs. when inference is not used; impact on trading partners [see “How Should Inference Work?”—Dirk]
    • When in the distribution inference no longer needed
    • Test multiple levels of adoption of inference, by different trading partners
  • Verification scenarios [see “Drug Verification: EU Vs US”—Dirk]
    • Using 2D barcode at the dispenser level (for verification or other purposes, determine training of personnel or equipment needed)
    • Process for investigation of suspect or illegitimate product (including all applicable trading partners), including testing boundaries of the system
  • Re-packager Scenarios – how to effectively and reliably link newly-issued product identifier back to original manufacturer product identifier [see “Who Is A DSCSA Repackager?” and “Repackaging Drugs Under A Serialization Regulation”—Dirk]
  • Special Scenarios – data and product do not necessarily move together, which changes data governance and where data goes (drop shipments, 340B, investigational drugs) [see “Vendor Managed Inventory Under the DSCSA”—Dirk]
  • Notification Scenarios
    • Communication to brand owner when a suspect product is found or when illegitimate product is found and reported to FDA [see “The FDA’s Draft Guidance on Suspect Product, and Farewell Columbus”—Dirk]
    • Capabilities of the supply chain and data exchange mechanisms to achieve the statutory reporting timelines due to security, access to data, personal availability
    • How to link or leverage notifications to patient
    • How to support forensic or lab analysis (up and down supply chain) when illegitimate product is confirmed
    • How do we ensure the current ‘authorized partners’ processes are in place and actually help to prevent illegitimate product from entering the supply chain
  • Exception Handling/Errors/Inconsistencies [see “DSCSA Exception Handling: A Preview of Your Next Surprise Headache”—Dirk]
    • Focus on ‘honest mistakes and errors’ (includes aggregation error)
    • What triggers a suspect product or makes it non-saleable, vs. the ‘honest errors’
    • Scenarios with “mixed product” (e.g., product with product subject to grandfathering, or that was subject to waiver/exception/exemption)
    • Fixing over/under shipments (when more data needed or more product needed)

If you were not able to attend this FDA workshop—or even if you did but you don’t think you got your message through—don’t fret.  There is another, perhaps better way to get your message through to the FDA DSCSA team.  Submit your comments to the FDA docket tied to the workshop.  Go to this FDA webpage for instructions on submitting before the deadline of April 21, 2016.  I will try to monitor the submissions and comment here on RxTrace on anything that is of interest beyond the topics that were covered at the workshop.

My initial fears were probably a little overblown, but I still think they were at least partly justified.  Fortunately there were enough people in the audience with the necessary long-term serialization and traceability knowledge and experience to ensure a successful outcome.

Regardless, during her summary comments, Dr. Ilisa Bernstein told the audience that the FDA is now in “pilot mode” and are working to get pilots up and running ASAP.  That’s good news.  We should see some action very soon (although it would be nice to get those four overdue guidance documents soon too [see “Who Is Being Harmed By Four Overdue FDA DSCSA Guidances?”, also see “The Many Faces Of The FDA“]).

Dirk.