Nielsen and England: FDA’s 1980s Approach to 21st Century Import Risk

Feb. 8, 2006
The Agency has supply chain risk all wrong and needs to dramatically improve its IT infrastructure, say two former FDA experts.

Anyone who cares about our nation’s drug importation policy has an opinion about what needs to be done to “fix” it. Two people who should really know are Carl Nielsen and Benjamin England, who worked together at FDA for 15 years and oversaw the drafting of the Agency's first Import Strategic Plan. Nielsen ended his government career as director of ORA’s Division of Import Operations and Policy, England as regulatory counsel to the Associate Commissioner for Regulatory Affairs.

Now on the “outside” — Nielsen as a private consultant, England as a partner with Rodriguez O'Donnell Ross Fuerst Gonzalez Williams & England P.C. — the two can better assess what’s wrong with our import system and what FDA must do to ensure the safety of drug shipments. The Agency needs to trust industry, they say, and it needs to shed its antiquated understanding of supply chain risk.

Pharmaceutical Manufacturing spoke with Nielsen and England recently to get their vision of a reformed FDA and a safer import system.

P.M.: What’s wrong with FDA’s approach to import safety?B.E.: As a whole, FDA’s import program isn’t actually doing anything. It isn’t improving consumer confidence in the industry. Basically, what the FDA is doing is just operating on a transactional level. Before product is permitted into the country, there are a whole lot of transactions that happen — either commercial, regulatory, relating to tariffs and quotas, etc. The Agency just figures out if it wants to look at something or not based upon those transactions, or if it thinks that it has enough information to suspect that there’s a problem or not. If it does, it will engage in its administrative process.

The problem is that the resources are so slight in contrast to the volume that 98% of the time product just comes through. The information is transmitted, but it’s not information that means anything.

C.N.: Essentially what they’re trying to do is make a safety decision based upon an invoice.B.E.: FDA is operating on the general risk associated with a product rather than the risk associated with that particular product in that particular container that came from that particular foreign manufacturer. Its approach is not specific until it finds a problem, and then it becomes ultra-specific. This becomes a nightmare for people who find themselves caught in this percentage of the 2% of shipments that FDA looks at.

This worked in the 1980s, because so much of what came in was bulk. FDA could always go to the domestic manufacturer and do a GMP [inspection]. It wasn’t as much of a threat. In that environment rose up counterfeit bulk drugs because there was so little scrutiny at the borders.

So, the Agency’s view of the problem arose out of this 1980s environment. It has still really never changed its paradigm, and here we are in 2006. But now, the majority of the products coming in are not just raw materials. You have a lot of finished products as well. GMP questions can only be answered at the foreign facility.

P.M.: Why weren’t you able to fix this approach?B.E.: The government has a way of creating a rubber-band effect. You can make progress in an area, but the tension is so strong to stay the same that if you don’t maintain that pressure it will eventually snap back.

In a government setting, you have to make up your mind that you’re going to break the rubber band (or you’re going to kill yourself trying!), or you’re just going to go along with the pack and become a bureaucrat. The system is simply not designed to change itself, so there’s always a need for change agents.

In some circumstances, I think we broke the rubber band. In other circumstances, not. Consequently, a number of things have reverted back.

P.M.: What were you able to accomplish?B.E.: We started to make some changes, especially after 9-11, and put together our Import Strategic Plan. We brought in people from all of the centers and internal stakeholders, the field, legal, legislative, we must have had up to 200 people at any time to identify how the Agency could better leverage resources to assess risk, and then focus on areas of true risk.

The key was to focus resources away from those areas where the risk had been mitigated already. A plan was put together to that end, and it has had some success. But the problem the FDA has always had is that it doesn’t know how to deal with industry very well. The people who know the most about their products are the people who make them, and FDA hasn’t allowed industry to participate in creating change.

P.M.: What is the Agency’s current thinking about the supply chain, and how does it need to look at supply chain risk differently?B.E.: They generally get the idea of controlling risk in the supply chain. The problem is that supply chains can be as varied as the products in them. Visibility is the important thing. You have to be able to see into supply chains and track the product through to the market. A lot of this is out of FDA’s jurisdiction, and that’s part of the problem. FDA probably doesn’t have a whole lot of jurisdiction over the security element, even though it thinks it does.

The Agency needs to come to the realization that the majority of the industry has the same interest as the government does in securing the product in the supply chain. One of the government’s concerns is that the interests of the CFO and COO and others lie with the shareholders. That’s true, but the shareholder is damaged if the product is tampered with. There’s really not competing interest.

C.N.: The key is for the government not to be a barrier, but to be a facilitator. What the Agency needs to do is start declaring what real risks are, at least on the import side—not just those inherent in the product. Different entities have the capability to mitigate specifics risks at different points along the supply chain. The current FDA import model tends to think of the supply chain as the transaction at the border delivering goods to an importer or consignee. The Agency, however, does have the capability to develop and implement risk-based systems considering the entire import lifecycle—that is, from the foreign manufacturer to the U.S. retail level.B.E.: Even a product that’s considered high-risk may be to the point that the risk has been managed. Then it’s not high-risk anymore. FDA needs to recognize this, and get the government to recognize this. Once it’s decided that a product is not high risk anymore, FDA ought to take credit for this fact. Congress is still looking at how many shipments the Agency has looked at, rather than a logical justification as to why they’re not looking at certain shipments because they may be low-risk.P.M.: How is this limiting what industry can do?C.N.: Part of the dilemma for industry is, how do you invest in different systems if it’s not going to be recognized by the Agency? If the systems cannot be recognized and evaluated within a risk-based paradigm by FDA that facilitates movement of low-risk shipments, then there is little incentive to participate in partnership activities with FDA.

One of the main areas of breakdown is at the border transaction. Generally speaking, industry has its controls that go beyond information contained in manifests, bill of ladings, and invoices. Industry exerts physical security measures on shipments based largely on internal notions of risk that are not fully considered by the government for admissibility based on risk-management principles. FDA and industry must partner to be most effective and efficient.

Leveraging is not just avoiding the duplication of work, but also looking at existing business practices and related records designed to assure the selling and receipt of safe and secure products and leverage. If industry is able to maintain a low-risk status of its products and chain of supply from the manufacturing source to retail, then the Agency should be better enabled to spend resources on higher-risk shipments posing greater potential threats to public health.

P.M.: What should FDA be doing to gain control of imports if it, admittedly, has scant resources?B.E.: In many ways, FDA could stop doing 50% of what it’s doing about imports and it really wouldn’t matter. The only impact would be the lack of presence of FDA at the border. Having people at the border might make people feel safer but it doesn’t really do anything.

A vast majority of FDA’s detentions on drugs and devices are things like “failure to list,” because someone didn’t put the right information into the electronic system, even though the drug or device is probably listed and in FDA’s database. They’ll have thousands and thousands of detentions that require man-hours to do the paperwork just to refuse or detain the shipment that has absolutely nothing to do with GMPs or approval, absolutely nothing! This is an FDA compliance officer spending hours and hours checking to see if a drug or device is listed. It doesn’t do anything with regard to the risk to the patient.

You could take half of those resources and cycle them through training, about what it is the officer can see in documents that demonstrates that a manufacturer has a grasp of GMPs and that the supply chain is under control. That information is available. I know of many companies that are prepared to give to FDA results of third-party inspections of their supply chains. They do it because their customers want us to demonstrate a certain quality of products. Their specs are higher than the FDA’s.

C.N.: When people say that FDA just needs resources, they think that it just needs warm bodies at the ports. But the consensus we heard was that IT systems were what was really needed. Current FDA import systems have just electrified historical paper systems. Take, for example, a typical FDA import alert that it sends to its field officers for guidance to take action on certain shipments. In order for the alert to be used effectively in the field, the entry reviewer literally has to read it to determine its relevance and application to a specific entry. Even for the international mail facilities, there’s nothing electronic; it’s all paper. So an alert is just not practical for execution even if the contents are helpful.

The biggest leverage for FDA human resources is investment in the whole IT architecture of the Agency. Risk management is information — some inside the Agency, some outside. Inside, much is still paper. Significant IT overhaul of the legacy systems is the only way you’re going to have a system that automatically identifies the greatest risks for drug products entering the United States and minimizes delays of low-risk shipments. Collectively, the Agency processes 14-15 million entries of regulated products annually.

P.M.: Why isn’t the IT system being overhauled?B.E.: On the IT front, FDA was given the mandate to do “prior notice” [the requirement that food importers alert the government in advance of shipments due to arrive], which is ridiculous. You’ve just frame-shifted the problem. If everybody gives you notice, you still can’t get to them all. But the Agency was not given the resources to set up the electronic system. Consequently, all the resources that they were going to put into other upgrades, they put into prior notice. So in some respects it’s not the Agency’s fault.

Ironically, in the same legislation that authorized prior notice, FDA was given the mandate to establish entry processing systems that facilitate the importation of, at least, food that complies with the FDCA. Exactly the opposite happened. Although that wise congressional mandate did not carry over to drugs, biologics and devices, those industries actually provide better opportunity to leverage advancing processing and manufacturing technologies that can demonstrate a facility is operating within GMPs or QSRs.

If FDA would focus its IT resources on facilitating the importation of drugs and devices that are safe and effective with something like an automated pre-shipment screening process or third-party inspection regime and then audit that process off line (after entry), it could use its precious HumINT and inspection resources on demonstrably higher-risk shipments or true unknowns. The effect would be that companies would want to be in the “facilitated entry” column rather than the unknown or high-risk columns. FDA activity at the borders and ports of entry would actually impact public and patient safety, not to mention biosecurity.

And FDA could justifiably take credit with Congress for shipments it didn’t inspect because another risk-management regime was assessing and mitigating those risks. If a foreign manufacturer can demonstrate that its products are approved, properly labeled, manufactured within the limits of proper processing controls, as certified by a trusted third party, why would FDA ever want to examine its shipments except on an audit basis to make sure the manufacturer is doing what it said it could do?

I might add, regarding the IT problem, that FDA has been given resources for IT before and hasn’t proven itself competent in coming up with forward-thinking, outcome-based systems that make sense. The tendency of government when it gets IT resources is just to electrify existing processes, which doesn’t work.

C.N.: The other Agency failure is that it stays within itself. Industry knows how to do it. So improving IT systems needs not only internal champions at FDA, but also external champions.B.E.: The biggest role the government can play is to establish the standard to permit connection of systems that verify product, whether they’re electronic or otherwise. The industry is moving forward, with EPC and projects such as those at MIT. They’re essentially establishing what the standards are going to be to control products and shipments.

But the government has to eventually make up its mind that the standard for this particular technology is going to by “XYZ” for this particular use. The problem is, which one do you pick? Basically, what’s happening is that a lot of the technology is developing that is interchangeable, and maybe that’s a better way of doing this.

But standardization is going to become an issue, because you’re going to run into it with, for example, trans-Atlantic cargo. Maybe the Germans and French are going to operate on a different platform. These are issues that I don’t think people have figured out even how to approach yet.

About the Author

Paul Thomas | Managing Editor