Tuesday, April 06, 2010

Reality Check on Supply Chain Security

Last Thursday, The New York Times published Are You Buying Illegal Drugs?, a scary-sounding editorial by Katherine Eban (author of Dangerous Doses) and Aaron Graham (ex-Purdue Pharma). The authors explain the economic logic behind prescription drug theft and advocate a national track-and-trace technology infrastructure.

Yes, we do need a national system for tracking and tracing pharmaceuticals, although the risks are much lower today than 10 years ago. However, I wonder about the real-world potential for track-and-trace technologies.

Tag all you want, but if no one checks the tag, it's a waste of money. "Don't ask, don't tell" is the mantra of people who buy diverted products from unsavory resellers. Pedigree laws and track-and-trace technologies only work if pharmacy buyers refuse to buy outside legitimate channels and agree to authenticate (scan) an electronic tag.

Adding insult to potential injury, the FDA’s recent guidelines for serialization will do little to address our crazy patchwork of state rules and regulations for the supply chain.


The Times article is correct about the potential dangers in our complex drug distribution system. Counterfeit drugs can slip into the system when someone buys from an unreliable secondary supplier who intentionally counterfeits or inadvertently mishandles an authentic medicine. This process, a.k.a., diversion, has been the entry point for almost every case investigated by the FDA in which a counterfeit or adulterated drug ended up in a local pharmacy.

However, the article comes across as scare mongering in today's environment. The gray market is not dead, but secondary trading has been dramatically curtailed since the incidents in Ms. Eban’s book. I recommend you read Dangerous Doses for its fascinating historical perspective on diversion and counterfeiting activity circa 2002.

Eban and Graham point the finger at pharma companies (naturally), suggesting: “Drug companies are apparently reluctant to pay the nominal cost of tagging pills and bottles.” Unfortunately, this perspective is woefully outdated given the industry’s evolution. Since 2002:
  • Manufacturers signed Inventory Management Agreements (IMAs) and Fee-for-Service agreements, thereby limiting product leakage into the gray market and closing a significant entry point for counterfeiters
  • Wholesalers have renounced secondary market sourcing
  • The HDMA has tightened its membership requirements
  • Major pharmacy chains have committed to secure sourcing
Don't forget that more than 7 out of 10 prescriptions today are for low-cost generics, which are rarely stolen or counterfeited.


Two years ago in my op-ed Securing America’s Pharmaceutical Supply Chain, I wrote:
“A national system for tracking and tracing pharmaceuticals would close these remaining gaps by requiring greater documentation across the supply chain for all purchases, significantly reducing the chance for fake drugs to sneak in. This system would also create a chain of custody history using unique serial numbers on a drug package. This technology is available today and would allow both consumers and regulators to validate that a pharmacy follows safe business practices. It would even be possible for a consumer to check up on their local pharmacy by entering the serial number on a web site.”
Pharmacies, the presumed beneficiaries of pedigree laws, oppose national legislation. Why? I speculate that pharmacies resent the inevitable transparency into their business along with the actual operational costs. See Pharmacists Haggle over Pedigree Costs for more perspective. The NACDS and NCPA even funded a study that dramatically inflated the costs of a track-and-trace system. Savor Accenture's Track-and-Trace Straw Man, one of my personal favorite posts from the Drug Channels archives.

Last week, the NACDS sprang into action by ignoring the content of the New York Times editorial and instead warning consumers not to buy from unlicensed online pharmacies. Read the NACDS’ response.


The Food and Drug Administration (FDA) took a baby step forward last week with its guidance on their Standardized Numeric Identifier (SNI) for prescription drug packages. Dirk Rodgers at RxTrace provides an outstanding analysis of the technology implications in FDA Aligns with GS1 SGTIN For SNDC, so I won't rehash what he says. (BTW, add Dirk’s excellent blog to your must-read list if you have any interest in pharmaceutical supply chain technology.)

Alas, U.S. states have the greatest influence over the wholesale and retail distribution of drugs, presenting practical hurdles for any manufacturer, wholesaler, or pharmacy that operates in multiple states. Say hello to 50 different pedigree requirements!

As I wrote in my 2008 op-ed:
“[I]ndividual state pharmacy boards retain the authority to establish unique documentation requirements for manufacturers, wholesalers and pharmacies operating within each state. This approach has created a disparate patchwork of inconsistent regulations for tracking pharmaceuticals in the U.S. supply chain. In the absence of federal standards, ambitious local politicians around the country are establishing their own incompatible systems, raising costs, reducing product availability, and lowering safety.”
Still true, I’m afraid. Read Federal Pedigree Sand Trap for the rest of the story.


So, where does all of this leave us? With a reasonably safe system that could be slightly improved at an enormous effort and expense for everyone (not just manufacturers).

Apparently, H.R. 5839 Safeguarding America’s Pharmaceuticals Act of 2008 will be reintroduced in Congress soon. Stay tuned, but keep your expectations low.


  1. Adam,
    Thanks for the great plug for www.RxTrace.com. As usual, I agree with all of your major points. In my view, they are a strong indicator that "inference" will have to be an essential part of our national approach to drug traceability. But it's not a fool-proof technology...more of a double-edge sword perhaps. I'll have more to say about it in a future post.


  2. One reader pointed out to me that NACDS' rebuttal appeared as a letter to the editor in the Times yesterday: Keeping Medications Safe.


  3. In response to Anonymous above, regarding the use of case serialization instead of unit-level serialization. There is a very good reason that regulations focus on the unit-level while ignoring the case level. Most cases remain intact and in use only from the manufacturer to the first wholesaler where they are opened and orders filled at the unit-level. To be able to track/trace drugs through the supply chain from manufacturer to the dispenser, unit-level serialization is the only level that works.

    As you point out, simple audit trails would be insufficient to protect patients. That's one reason why the Florida and California pedigree laws require supply chain recipients to validate the pedigrees and the products they receive before they can sell the drugs further. See "What Are Pedigree Laws Trying To Accomplish Anyway?" http://www.rxtrace.com/2009/11/what-are-us-pedigree-laws-trying-to.html for more of my thoughts on this.


  4. If someone is going to counterfeit or divert - do they usually do it at the bottle level?

    I would encourage you to read Katherine Eban's book - Dangerous Doses to get an idea.


  5. Adam,
    Hmmm-Are we ready for bottles?
    FDA Delays PREDICT Expansion Indefinitely
    An expected nationwide rollout this summer of the FDA’s PREDICT electronic ranking system, designed to help uncover adulterated and misbranded imports, has been delayed indefinitely. The delay is due mostly to IT infrastructure problems, such as server crashes and overloads, FDA spokeswoman Ayse Yeaton said. She noted that fixing the problems, which are affecting FDA field data systems nationwide, is a high priority issue. “As soon as these problems are fixed, deployment of PREDICT can resume,” she added.

  6. Adam:
    As background, I was the founder of two bar code companies have 3 US Patents on bar code technology and have been in the Auto ID sector since 1999.
    I am now a consultant in the bar code and digital watermark area. There is another alternative to the expensive process of bar code serialization and/or RFID, and that is digital watermarking. Applied at the artwork level, they are  invisible to the naked eye, but decoded with an everyday camera equipped smartphone, as well as standard CCD scanners in the supply chain.  The benefits are many. No expensive set up. No production line hardware. No production line downtime. Imbedded watermark information can be linked to Lot/Exp and/or NDC for validation. Easy field check for counterfeit or diverted goods with camera equipped phone, which can record GPS information among other data. ROI's include: track and trace, anti-counterfeiting, diversion protection, re-call assistance, return validation among others. And, as an added benefit, additional watermarks can be applied for consumer/patient interactivity or for clinical trial participation. At the OTC level watermarks can be used for information, discounts, rewards and contests. Backend can support many levels of data and encryption, and of course can be blind to the data as it flows through to the Pharma.
    I hope this provides information that might be useful to you and your readers.

  7. Actually, I have read "Dangerous Doses." And I wonder if we couldn't come up with a better and more efficient solution than unit-serialized pedigree. When you look at what is actually diverted / counterfeited out of the "normal" supply chain, is it worth the hundreds of millions of compliance dollars compliance would cost or could we devise a better model? This expenditure would surely be passed to consumers (insurance) in one way or another and would not do anything for gray market "out-of-normal chain)drugs.

    The case level thought would come complete with a parent/child relationship on repackaged cases (in the database, referenced in a pedigree record). This way, should an incident occur downstream, track and trace - under this model - would be in effect back to the manufactured case as well as the repackaged one(s). I should have explained that in the original comment.

    The point is - before the industry spends huge capital on this idea - perhaps it would be well worth the effort to consider other ideas (authentication systems, etc.) which may do as good a job for lower cost.