Archived Content

The content on this page is provided for reference purposes only. This content has not been altered or updated since it was archived.

Medical Devices

FDA Unique Device Identification Public Workshop: February 12, 2009 - Individual Presentations Transcript

MOD: Alright, we’ve got nine presenters who are going to have their five minutes of fame here. Five minutes, remember guys, keep it on track, so that we can wrap this up and all get home in one piece. So with that here’s the nine presenters, where they’re from, and I will try to manipulate this in some useful way. Fire away.


FS: Hello everyone my name is Linda Rouse. I am Director of Federal Affairs for Premier. You heard from Joe Pleasant our CIO on an earlier panel and I’m going to bring a couple of different perspectives from the policy side of things to this that we haven’t talked about too much today. To begin we all know I think that UDI is the missing link on a lot of things that we’re trying to do in health care right now, some of which we’ve kind of talked about. We’re always out there to improve patient safety and quality that’s been said a couple times, and I think we need to keep remembering that the point of all this is to improve patient care and quality.

We can do that by several things. Improving the efficient recalls process and populating electronic health records, which was mentioned earlier a little bit that Congress as we speak has struck a deal on the economic stimulus package which does have $20 billion in it for HIT, I think $2 billion of that is direct loans and grants to providers to implement electronic health records. And if we want to prepare from the device side of things to populate those electronic health records we’re going to have a huge gap in data and accuracy which will also lead to inefficiencies in the recall process, etc.

And all that kind of means is that we need better data and better research, all about patient care. And as a side note Premier kind of took a little bit of a case study to look at can we identify patients that received a device after it was recalled? And to give you a little bit of context Premier, you’ve heard a little bit about us from Joe, we operate one of the largest clinical databases in the country. We have over 144 million patient cases that we get from over 500 hospitals. So it’s a very rich database with clinical and administrative data.

And we took a stab at looking at a few Class One recalls and could we figure out if patients were receiving those devices after they had been recalled. And we had to throw several of them away because we couldn’t definitely identify the device in all that data that we have, which itself is a disturbing little lesson and I think also strikes the need for why we need UDI.

We were able to finally to drill down on one device that we were able to get enough specificity around, because remember hospitals call the same thing all different things, we were able to finally determine one device, that was a Class One recall and we were able to look at through patient records and determine that over 50 patients in 40 hospitals got this device, implanted it several months after it was recalled.

And that took a lot of work in itself. And we’re kind of writing that up and that will be attached to our comments in writing that are due to the FDA on February 27. So there will be more to come in that. But I think that bills the fact there is a very urgent need out there for UDI, as was said earlier on the second panel, we have a responsibility to our patients and we’re not fulfilling that responsibility. And I think UDI can help us do a better job of that.

And that kind of leads up to the urgent what are we waiting for. I have to give kudos to FDA; you guys have been very diligent in this, you’ve had a very transparent process, which doesn’t always happen at the federal government level and I think it’s very much appreciated and that whatever this final proposed rule looks like. While we might not make everybody happy, you probably won’t make everybody happy, I don’t think anyone can say that their voice wasn’t heard and taken into consideration. I think Jay’s doing a great job with that.

That said, what are we waiting for? This is a little bit of a time line, and I’m trying to do this very nicely, Jay, but it’s a little bit of a time line of how much FDA’s been involved with this. We all know in 2004 was the NDC reg, and since that time FDA has been toying with the idea about a UDI for medical devices. That’s a good length of time. They’ve done a great job. There’s been public meetings. There’s been reports commissioned. All good stuff. But again what are we waiting for?

And we have this public meeting today, which again I think is appreciated. I was at the one in 2006. And I do think that while some things were the same that were said, I do think there was some movement. Unfortunately it took three years to get there and I don’t want to see for these couple increments that we get to be another three years or we’re never going to get anything in place.

With that, because things have been taking so long Premier is part of a, some of you may not be familiar with this, we’re part of a coalition of advancing patient safety, an advancing patient safety coalition, which has grown up specifically around this issue and to get UDI implemented. And you can see we have over 41 organizations.

And I think as a government affairs person for those of you that might not be as involved on the policy side of things it is very rare in D.C. that you see a broad breadth of members around one issue. There are providers, all types of provider, nurses, physicians, specialists, hospitals, as well as consumer groups. You’ve got AARP, you’ve got some disease groups with the American Hospital Associations, and you’ve got some quality folks with the Joint Commission.

And the fact that everybody is around, on board on the same page on this one issue I think is very telling. And we’ve done our work on Capitol Hill, had some champions on this issue, which is what was instrumental in getting in the FDA Amendments Act in 2007, getting language in there to require the FDA to move forward on the UDI regulation. Not because we didn’t think Jay was doing a good job, but it always helps sometimes I think to have a little push from Congress on this issue.

And again that was in 2007 and we’re in 2009 I know we have a new administration and so hopefully, as Jay alluded to earlier, we would hopefully see this proposed rule this year. I will be around so if anybody’s got any questions, if anyone wants to join the coalition we’re always looking for more members, I’d be happy to talk to you afterwards. Did I do five minutes? Thank you. (Applause)

MOD: Thank you, Linda. I just want to mention something before Dave comes up. We have been thinking about this a lot, so don’t think that we don’t have a plan in place, we really do. And we have a very good idea of how we think this is going to work. But input is always important, so the questions that are up here, the questions that I’ve raised today are not meant to indicate our lack of understanding, insight or direction, but rather just allowing people to say their peace. So don’t think that we’re not very close to being done.


MS: Thank you, Jay. My name is David Turner. Thank you for the time Jay for Novation. Dennis very much wanted to be here today, but he had a family emergency and so I got a call at the last minute and I’ve got some notes here written on the back of some document, Federal Register something. Very quickly my title is I’m Director of Supply Chain Data Management at Novation. Very quickly who is Novation? We are a group purchasing organization. We are created and we are wholly owned by VHA and UHC. These are cooperatives of non-profit community-owned hospitals and university hospitals.

Some pretty pictures here about what I just said, and a lot of hospitals. And as I said my point is that since we were created and owned by these hospitals we do not feel in any way as a separate entity from these hospitals. As I look at these slides now I see that all this ground has been covered so I actually just want to take my moment of fame here to highlight some points here, and if anybody has any umbrage with anything I say, please feel free to send emails to Dennis and he would be glad to respond.

So as I listen and as we listen to our hospital members and have discussions about these topics, we have heard a very clear consensus and complete agreement that the concept of a universal device identifier is a good idea and that we could do it and we could do it sooner rather than later. We should keep it simple. And so we agree wholeheartedly with several views expressed earlier to start simple and go in phase.

We propose that the list of attributes be very, very short in the beginning. I do have a bit of a different opinion on one attribute that is shown up a couple of times today, which is classification. It is a thorny issue. There is some contention with suppliers on classification and I would not, my personal opinion is that we should not make that a barrier to adoption of the UDI. We recognize that UNSBSC (ph.), for example, is very, very important to our members. I actually chaired the committee back in 2000 that authored the segment of the UNSBSC and that got published back in 2001.

Then we asked that the manufacturers code their products and publish those codes to us, and we waited for five or six years and got nothing. So we went ahead and did that ourselves. And so we have actually got over 99 percent of the dollars that our members are spending classified to the UNSBSC and we are publishing that to our members already. So we do not see that as on the short list. That’s Novation’s position.

We feel that it is overdue. The current situation is actually worse, I believe than has been portrayed. We have, for example, seen 500 different ways to spell 3M, so far. We have seen 70 different products with the same catalogue number, 2001 is the champion right now, and counting. So this happens every day. And I feel that this is obvious, just too obvious to even argue with. I do have a feeling on the question of readiness that has come up several times today that I need to defend my members a little bit on this readiness question.

Though it is true that many manufacturers have bar coded their products, I believe it’s also true that if our members tried to send an electronic purchase order today with a UDI number instead of the old fashion catalogue number that it would be rejected; that the distributors and the manufacturers, even though they have bar coded their products their order fulfillment systems are not ready. So we have to do this together. How am I doing on time? Thirty seconds.

And then finally I do feel that there is a greater connection. We’ve talked about supply chain and patient safety as two different things. I do feel that they are connected. For one thing there’s a tremendous amount of waste in the supply chain that’s been very well documented. And every dollar, we’ve been talking about patient safety mostly in the context of absence of error, but if there’s absence of care altogether then that has a very big impact obviously as well, and every dollar that we waste today reconciling invoices and purchase orders, etc., is a dollar that’s not being provided as patient care. Thank you very much. (Applause)

MOD: Send your cards and letters to Dennis. Okay. Rob I think you’re next.


MS: Great, you’re not dyslexic. That does say UID. The unique identification effort began in the U.S. Department of Defense in 2002, by direction of our undersecretary. We do first of all account for all of our property in the Department to get a clean audit. And so that was kind of the kickoff, but it grows much faster than that. The focus initially of unique identification was a set of identifiers, both for items, real property, locations, sites, organizations, etc., as you see on the chart.

Obviously, the parallel to what you’re doing is item unique identification or IUID. Personal property is anything that’s not real properties. So it’s not just a can it’s an M1 tank. Okay, whatever. (Laughter) I don’t know that’s not very personal to me. Historical methods of identification. This simply says that there are communities that have been trying to identify things forever. This community is no different. It’s not surprising that this community’s cataloguing community calls it one thing, Wood Tax (ph.), somebody else calls it something else. We encountered the exact same thing.

Nomenclature’s a red herring. Don’t go there. Don’t even try to use it. So the idea was to bring these communities together and establish a single identifier that represented an item, use it permanently through life until the thing is into bits. And not electronic bits, I mean, bits of material; they’ve been destroyed, demolished, whatever. So the idea is regardless of commodity group, and that’s where it becomes critical in this community. We issued policy July 23, 2009, almost six years ago.

That policy said that everything the Department buys, I’ll paraphrase, that’s of interest to us, $5,000 or more, serially managed (ph.), controlled items, etc., must be uniquely identified at the item level, each one distinctly and uniquely from any other. Declare the policy, go do it, use ISO 15 459 (ph.), use the 2D Data Matrix, make it happen, and no, we weren’t ready either. Neither was our supply base. But you’ve got to put a stake in the ground and say go forward and do this. So that’s what we did.

So the decisions. We established a vision. We adopted international standards. I already mentioned ISO 15 459 and others, I won’t bore you with those. But the key was interoperability across all the various commodity groups that the Department of Defense has to deal with; from aerospace, which at the time wasn’t in ISO 15 549 or 15 434, that took 39 months, heavily involved in the ISO community to make that happen. The Air Transport Association agreed to take it on.

NCMH 10 (ph.) which is the hard goods guys took it on and worked exclusively or extensively with GS1 to make sure that they created the ability to use GS1 and they produced a handbook that tells you exactly how to use GS1 to comply with DOD’s item unique identification requirement. I mentioned ISO 15 459, we didn’t want to reinvent the wheel.

This standard pre-dated us by quite a bit, it exists. Use it, be happy. You don’t have to pick. That’s the thing, you don’t have to pick a specific standard other than this interoperable standardized ISO 15 459. Here’s three examples, ATA, GS1, MH10. They all provide basically the same data. 17V in the example for data identifiers is the same as MFR in textile item (ph.) identifiers. It doesn’t matter. Pick one.

So the idea is to create intelligently a number so that you don’t create duplication and so that you can fight things like counterfeit and fraud downstream, because you know who created it, you know how it was created, you establish a mark. And from that point forward the unique item identifier, which is defined in ISO 15 459, I just love saying that, is this string of numbers you see, starting with UN. UN is for Dunn and Bradstreet; it’s the issuing agency code. That number from then on is the number for that item, through its life in all of our various systems, whether they supply logistics or whatever, that is the goal.

We started in 2003, made it mandatory January 1, 2004. Two years later we had 200,000 items. Three years later we had two million. Today we have five and half million, one year later. So it’s an aggressively progressing compliance of 40,000 items a week on average now, and steadily growing. So we expect that by the end of this year we’ll have somewhere around ten million items in our inventory that have been registered in the IUID registry at the Defense Logistics Information Service in Battle Creek, Michigan.

The data matrix is the choice. These can all be coded properly within the data matrix, no problem. What data is submitted? The data on the right is the data DOD requires, USDOD. The data above, oh, by the way, there is a NATO standard, exactly conforming to the U.S. standard, only it just requires four data elements in the database to ensure uniqueness.

We capture additional data. We call it pedigree data. It essentially creates a birth record, with some data elements in the status section primarily that indicate a change in state. Another example would be current part number. The identify remains the same, the part number rolls over life. Thank you. This is how we communicate information. The contractor through our receipt and acceptance process, standard DD250, provides it through. The standard process is we essentially shave off the data we need to store in the registry, the pedigree data.

Requirements of a DOD instruction and a parent DOD directive tell you exactly what you must uniquely identify for the Department. We want item level, systems life cycle management from the point, using that common data key always. We don’t create, in capture a whole ton of data, but we use it through all of these various systems.

Ultimately the soldier in the field should be able to scan an item, pull down from their local system data necessary to replace that item and to also query to see if there’s a warranty on that item before somebody tries to repair it, pack it and ship it, and if necessary order another and know all the information necessary to package and ship it and to obviously notify that there’s been a product quality deficiency; if it’s a failure that shouldn’t have happened when it happened, and pass that information, make it available.

Here are our statistics. We have 1,334 contractors who have delivered successfully. Out of the 100,000 that were issued contracts last year many of those were services and other kinds of things. But we already have a base of 1,334 contractors that can comply. We all share that contractor base. So whatever you do, please don’t put us in a situation where we have to add two marks because FDA has a requirement and we have a requirement.

Our medical equipment is going to come in, either as AIs, DIs, or TEIs or any of the other acceptable equivalent bin, ESN, etc. We don’t want to have to put a second mark on there if this community decides GS1 versus 15 459 or HIBIG (ph.) versus 15 549. Please use the open standard. If you adopt a symbology, make sure it’s an open standard. The data matrix is a non-proprietary standard. Anybody can use it. There’s no fees. There’s no requirements.

This community in the outside here can translate that to any other system you want. It should be driven by the business. We in the Department are prohibited from telling you you must use GS1 or you must use MH10 or you must use serial number, part number, manufacturer. We preclude ourselves from doing that by policy. You run the business. We let you serialize it uniquely.

Just like you said earlier this is the way we want to use the data. You’ve already talked about that. Our advice don’t reinvent the wheel, consider a central registry, which you’ve all talked about. Be careful what you put in the registry; it needs to be baseline data. You have a birth certificate that has your social security number, but it doesn’t have everything else about you in there. Those are available in local databases through the key, which for us is social security number. And in the Department it’s the unique item identifier.

So improvements in quality, use an automated data capture, change the processes, don’t key in stuff. We already talked about that. So there’s the summary. Thanks. (Applause)

MOD: Alright, thank you, Rob. Jeff?


MS: I’m Jeff Secunda, Vice President of Technology and Regulatory Affairs at AdvaMed. And I’d like to thank Jay and FDA for having this public meeting. AdvaMed is a trade association that represents approximately 90 percent of the medical device industry. Our members range from the largest to the smallest. Medical devices, diagnostics, software systems are all part of our membership.

I want to make it very clear that AdvaMed and the device industry clearly supports the development and implementation of UDI. There’s really no debate here. There are lots of details to work out, but I think a few years down the road a lot of the details have been expressed, we’ve been discussing them, and I think that we have a clear path forward.

The primary goal of the UDI system should be to ensure the safe use of medical devices. This goal clearly requires a commitment on the part of the device manufacturers, FDA and the device users. The device industry is extremely complex and heterogeneous. It is also characterized by rapid innovation and breakthrough advancements. Therefore, even a well crafted regulation FDA should provide a formal process for petitioning for exemption or deferral from the UDI regulation, especially for the newer devices that are not well understand.

Global harmonization. Not only have we beaten it to death, but we stabbed it to death. So it’s a good thing. Let’s do it. AdvaMed supports the exclusive use of the GS1 and HIBIG device identification standards. Both standards are well known in use throughout the health care environment. Other standards should not be allowed, as this will undermine the adoption of UDI by device users. AdvaMed supports a two element format for UDI, comprised of a static identifier, which is linked to the device manufacturer identity and product number.

FDA should also allow for the continued use of NDC and national health related information codes for manufacturer identity that have been traditionally used on certain medical devices. Devices that are designed for multiple patient use that require reprocessing, such as surgical instruments, are the only device category that should be considered for direct part marking. I say considered because some devices cannot be directly marked with the UDI due to size, material, and required processing factors.

The unit of order is generally the most appropriate location for the UDI mark, as it represents the package level from which the user draws the individual device. Convenience kits should be treated as the unitary medical device with a single UDI on the package. Single use devices that are legally reprocessed by third parties must bare in all instances the UDI of the third party re-processor. The original equipment manufacturer’s UDI must not appear on the re-processed device or its packaging.

AdvaMed supports technology neutrality for UDI carriers such as bar codes, including bar codes that are recognized by GS1 and HIBIG. RFID and other technologies may in the future be recognized by these organizations and therefore may be included in the UDI. The user’s scanning environments are extremely important in writing regulation and considering technology options.

For example, many laboratory analyzers use bar codes to identify patient samples and test components during the performance of test analysis. Conflicts between device operation bar codes and device identification bar codes in the same scanning environment could result in serious problems.

The data governance and integrity of data is extremely important. And AdvaMed looks to FDA to establish governance policies that will ensure the integrity and security of the UDI data. AdvaMed recommends that the FDA make available various data entry tools that will accommodate the needs of both large companies with thousand of UDIs as well as small companies that have only a handful of products. The web portal and business to business options that FDA has made available to electronic adverse event reporters are an excellent example of useful data input tools.

AdvaMed looks to the FDA to be the gatekeeper for access to the UDI database. Only manufacturers or their authorized agents should have access to the UDI data input and editing systems. The implementation of UDI systems is a massive undertaking. For there to be any patient safety benefit the UDI system must be seamlessly integrated into the manufacturer’s quality management system. In fact, if UDI is not properly integrated with the quality management system the improper identification and use of products may lead to serious consequences.

AdvaMed recommends a three-phased approach for UDI implementation. First, UDI integration must be integrated into the manufacturer’s quality management system. Second, establishing marking systems in the manufacturing environment; this encompasses both direct part marking and label marking at various levels of packaging. And, finally, database entry; adequate time and software tools must be available to ensure accurate UDI data entry.

In closing, the medical device industry is ready to get to work on UDI. We ask that the FDA recognize the vast differences between device types, sizes, materials and the environment in which devices are used. Adequate implementation times and tools will make the difference between an effective UDI that can have a positive impact on patient safety and one that is broken before it even starts. Thank you. (Applause)

MOD: Thank you, Jeff. Alright, Rich.


MS: First I want to say my name is Richard Eaton. I’m the Industry Manager at the Medical Imaging and Technology Alliance. We are a division of the National Electrical Manufacturers Association, NEMA. NEMA is the largest U.S. trade association which represents America’s electro industry. MITA, the medical division, is the only association which exclusively represents diagnostic imaging equipment.

We have 95 percent of the market in terms of our manufacturers; they are small and large. And as you can see the types of devices that we’re involved with are mostly, for the most part, large. We don’t have too much of a space problem, like a lot of people have, in terms of very small devices. What we’re characterizing, our device is pretty much capital equipment, which is the major re-usable, large devices like a CAT scanner, like an MR scanner, but also includes patient monitors, diagnostic ultrasound scanners, and not, for the most part, implantables, disposables. The interesting thing about our devices is that I don’t know if anybody has ever lost track of a CT scanner. I would doubt that.

I just want to outline a few of the key points that are of interest to us. With regard to UDI minimum data set, we think that that should be comprised of an enterprise ID and a production ID. So the enterprise ID and a production ID. In other words, name of manufacturer, model number, part number and serial number or lot number.

Now for our equipment we believe that UDI should be applied to systems and only those system components which are finished devices. Finished devices are defined in 21 CFR. This makes it a lot easier to track in terms of what should have a UDI and what should not. With respect to software we believe that a UDI should be either embedded into the software or in the user manual.

One of the major problems that we face, a lot of our businesses are involved not only in production of new devices, but also in refurbishment and re-manufacturing of devices. And the question we had to wrestle with is when should these devices have a new UDI and when should they be able to carry the same UDI. We’re defining re-manufactured devices as a device that’s been modified but has new performance or safety specifications that differ from the original device. That we think should carry a new UDI.

Secondly, if a company which is not the original manufacturer puts its name on a device which it has reconditioned we believe a new UDI is warranted. With respect to a re-furbished device; that is a device which may have been reconditioned, but has the same specifications as the original device, we believe the same UDI would be adequate for that purpose.

The problem with these terms, refurbish and re-manufacturer, is that they’re not interpreted the same way in every jurisdiction. In Europe there’s a different interpretation of these words, and a different interpretation, depending on who you talk to in this country. It would be very helpful for global harmonization of these definitions so that we’re all talking the same language. For example, in Europe they talk about fully refurbish, but fully refurbished in Europe is like our re-manufacturer, so hence the confusion.

We believe the UDI should be human readable and only machine readable if appropriate. And in making that decision one should consider the environment of use of the device. We also feel that FDA should identify automatic identification standards instead of automatic identification technologies because auto ID technologies tend to evolve and auto ID may create RFI issues for medical device manufacturers in the clinical environment. And this is something a number of our manufacturers have experienced.

With regard to the database, we believe that should include manufacturer and product information, but not the serial number or the lot number, because that is usually very sensitive information and does not belong on the database. We feel that the development, operation and maintenance cost should be shared by users and the FDA.

Now finally on global medical device nomenclature, MITA supports the use of GMDM, but it needs to be globally adopted. But prior to its adoption the structure of GMDM needs to be cleaned up, it needs to be clarified and simplified. There are many instances in which it is difficult or cumbersome to identify specific device using GMDM and that diminishes its utility as a tool. We believe that GMDM information should be placed in a database and not on the product label.

And in conclusion, first of all, let me thank Jay and the FDA for sponsoring this workshop, we’re glad we can share our views with you, we look forward to sharing our comments with you later this month, and finally is a plea that you’ve heard all afternoon, we need to proceed in a deliberate and careful and incremental way. Changing to a UDI system is going to cost manufacturers a great deal of money no matter what way we proceed and there’s going to be a lot of re-tooling that’s going to need to be done. We support FDA and its movement toward more patient safety. We want to work with you in that regard. But the devil’s in the details and we need to be very careful about how we proceed. Thank you. (Applause)

MOD: Thank you, Rich. Brad?


MS: Good afternoon. My name is Brad Sokol. I’m an independent consultant, health care consultant and you can call me Joe Six Pack, not really. I’m here to talk about the UDI and software, and I think the UDI should be included in the software. But what we’re really talking about here is we’re talking about network monitoring, risk management, liability management, mitigation of that risk in hospital networks.

How many of you out there have tried to put some type of peripheral on your computer and it just didn’t work? I understand; I feel your pain. I’ve done that as well. And this is just you, at home. And then how many of you out there have now gone one step further and it says contact your system administrator? And it happens to be at your house. So you can understand I want to really make this very simple, this UDI. I want to make it so simple that we just pair up the UDI to the network. That’s how simple I think we should begin. So we can get some success out there and mitigate the risk for the provider.

The questions I’m going to directly answer are the alternative placement questions 4B, C and D. And once again the UDI’s placement should be in the software and firmware, on the electronic medical devices, and we should enable the network to monitor the device. So the four basic questions are. How can the UDI be used to address the network and patient safety issues, which I just gave you an example of. Second, how can the UDI be extended to address clinical network issues in medical devices that can properly be networked? So in other words, can you match the UDI to the proper network address?

Third, how can the UDI help the clinical enterprise networks deliver quality of service and service level assurance, thus reducing risk and liability in the hospital, because this is your root-cause analysis that you do in a hospital. And I think somebody said before that you really should base this on risk analysis.

But what’s happened is is that once a medical device manufacturer, electronic that is, puts the electronic device in a hospital, and they put the software in the hospital, now it gets on the network, the risk is shifted from the medical device manufacturer to the hospital. I think what we should really consider and concern ourselves with, the providers especially about that risk shift. So who validates this network and the medical devices?

The key here is what we’ve been talking about today is the supply chain. I’m used to doing this here, but it’s not working too well. So basically the supply chain you’re looking at through basic inferences such as last manufacturer, last maintenance, last sterilization, last instrument and last location. However, the UDI does not truly gain value to the clinician, to the provider until you reach the top of the pyramid where you can actually identify all the things that had come before as well as now, the last user, and tying that in with the last procedure and the last patient into the PHR which is the patient healthcare record.

So if you look at the very bottom, what you have is you have an associative environment where the device is married to the patient to the location, procedure, the clinician, hopefully the infection in the future, and the patient record. That’s where you get the true value for the clinician and the provider to adopt a UDI. That’s why I’m setting the bar so low; I want to make sure that the medical device and the network pair off together.

Finally, I’d like to talk about the UDI and the handshake. The UDI can now become an affirmative handshake and it can be the first step matching the device to the network. Have any of you ever worked with mailing machines and you had to change the date? You know how you had to lift up that door to make sure that it worked? Well, why can’t we do that with a UDI when we update the software, you lift up a door, you press a button to make sure that works? It’s that simple. It’s been around for ages. It’s not that different. It’s not that different.

Number two, connectivity is one of the most misunderstood areas in the market. And you know very simply put whenever a device manufacturer assures that the network is working in a hospital and there are no conflicts for that device in the medical device manufacturer for that network and the provider, that medical device manufacturer is not guaranteeing the network does not have any conflict at all, the provider’s network doesn’t have any conflict at all.

How this conflict works. You basically, the medical device manufacturer issues the UDI to the one, that’s number one. Number two, the new UDI is placed manually or in the network into the medical device. Number three, the clinician first to use it opens up the cover or pushes a button to say, yes, I have received this UDI. Number four, the network has a small software utility driver that actually is loaded on the client that falls under the MBDS guideline and says I can now monitor this particular medical device.

Number five is as these things change and as the state changes of the medical device you can now report that to the hospital directly as well as the medical device manufacturer and the FDA. Where are the gaps? Here are the gaps currently today. I did an analysis of a lot of the current standards and regulations out there and what would happen is the first part of the UDI between the software and the network would be corrected by a UDI handshake and something acknowledging it manually or by pushing a button.

The second part would be to enable a network monitoring device or network monitoring software and that would reduce between the network monitoring device would reduce between the network and the hospital. That would reduce liability and risk. And finally you would also now be able to close up the gap by the network monitoring device and the UDI and directly report the adverse event. That’s it. Thank you very much. (Applause)

MOD: Who’s next?


MS: thank you very much for allowing me to share some of the experiences we have. I basically work in a space of ED, neurosurgery most of the time. And so I do occasionally see some of the patients not surviving, which can be quite emotional many times. And I guess I do tend to be sometimes emotional. So we’ve been looking at this issue for some time. So it appears that the type of problem that you’re talking about is supply chain and it’s close look problems and open sort of problems. And I think when you talk about international harmonization and whatnot that’s open sort of problems.

However, sometimes I think we don’t even have a solution for close look problems. And for me that’s more of an interest because that’s where our patients live, not in open look, way yonder beyond Japan or China somewhere. So that’s what I’ve been more interested in. And I also noticed that the people talk about these problems and solutions based on their backgrounds, supply chain folks a certain way, IT folks talk about it a certain way, and clinical folks talk about it a certain way.

And unfortunately many times I don’t see too many clinical folks involved, maybe we are probably (unint.) that may be the reason. But anyway, so the concern we have is that we would like to make sure that there’s a clinical benefit and improve the patient safety. Everybody talks about patient safety; sometimes I wonder what that means. Maybe we talk different languages. But certainly some of the things I have heard it does not ring a bell as a patient safety screening.

And I think the other thing is adoption is very important to us, because if you are in an ER the patients come and go, and let’s say there’s a horrible accident on the highway, you may see six, seven patients terribly damaged, you know, the time is essence. So if we don’t have all the items on the same sort of a management this would create a huge problem. And even though we hate to talk about cost, still the cost is important, because if you ask me if you have $100,000 what would you do? I would buy ultrasound, because I know I can (unint.) your patients, and I know what the outcome is.

And also the local environment. We had a hospital at Los Angeles when the Katrina hit and you know what sort of things we went through. In reality there’s no network, no computers, the only thing which worked during that time was the handheld units with the batteries. So whatever we design we need to keep that in mind. Maybe all you guys live in a world where nothing happens like that, that’s fine, but some of us do live in slightly different environment.

And the kind of information content we are interested in is eventually is the EMR. So we want to feed all this information into the EMR. And so the kind of information we have to track is staff badges, patient wrist band, IV medication, non-IVs, medical device identifications, blood products and all these things. And so we have been looking at some of the documents and so some of these things we have looked at closely.

And we’ve been focused on cost structures. What is a cost structure? There’s the set up cost obviously, that everybody has to pay. And there’s the recurring cost; memberships, you know, printing costs and all these things, and there’s a huge liability cost. I come from Florida so south Florida has huge liability problems.

And then there’s monopoly costs, which is sort of hidden, that comes from patents and all these things. And so monopoly is something which we are very much concerned about; that is, I agree 100 percent should have intellectual property rights and they should make a living for that, but it’s a source of unknown cost. And laws, rules and regulations spawn off liabilities, lawsuits and whatnot. And the standards are something that industry and government sponsored, gives us a guideline. But it can also become an uncontrolled cost if it’s not managed properly.

And the experience at the hospitals, we have done many of our code, passive RFID, 2D bar code and active RFID. So active RFID and 2D bar code, one of the reasons we got into 2D bar code is that when we did, we were doing asset management and rental control and all these things in the hospital. The biggest problem was the database. 3M has 820 different variations.

And we have basically the same problem. And part of the reason we discovered was that the person we hired to type the database in was we were paying $8 an hour, and she just doesn’t care or he doesn’t care. So there are lots of errors. And I was told that traditionally there’s one error every hundred keystrokes. But we found maybe one error every 20 keystrokes. So that was part of the problem. So what we thought is that if the manufacturer would give us 2D bar code information, then we’ll just read in it. We’ll be very happy.

And then this is work mostly done at Mass General. And Harry who has spoken before is behind this. There are some of these bar codes, 2D bar codes we are using for patient care at controlling devices. And these are very important to us. Now supply chain folks we may not be, but at least certainly to us. And so the comment is that we would like to see vertical monopoly removed, more rational patent laws so we don’t get impacted every day, and maximize competition innovation because we still have lots and lots of work to do. And improve the adoption using DROI.

Now supply chain guys recently told me I’m the guy sitting in that runway they want to land the airplane. However, there are some non-believers who think that we are going there. But my problem is that I don’t want this to happen during the process. And I live in this environment. This is Katrina and I’m somewhere down there. Thank you very much. (Applause)


MS: Thanks, Jay. My name is Mark Leahey. I’m the President of the Medical Device Manufactures Association. We represent hundreds of smaller medical technology companies across the country. First, I want to thank Jay and his team at FDA for putting this forum together. And again this has been a long process, as one of the speakers alluded to, many, many years. And again I’m here to say that like AdvaMed and MITA, you know, MDMA looks forward to working with FDA to develop an effective and efficient UDI system that focuses on patient safety. I think that’s the key that we can’t lose sight of.

Although BD and J&J and Medronic are not members of MDMA many of the points that they raised in their presentations are shared by our members as far as the implementation. I wanted to take two minutes to talk about some of the general perspectives on this issue and perhaps on today’s meeting.

Again, there’s a lot spoken about patient safety, recalls, but for me I keep on getting a sense, and particularly from some of the speaker today there was a lot about supply chain management, ecommerce. I’m not saying those aren’t worthy goals and that we shouldn’t try to take cost out of the system, but I think again from FDA’s focus it really needs to be first and foremost on patient safety.

If there are other benefits that are derived from it, or I think as Dr. Plurrough mentioned linking this to payment, again those are all laudable goals from a UDI system, but from our perspective first we need to make sure that we put something in place that does take a risk based approach to the implementation that is iterative in nature. And as I mentioned earlier I think it’s critically important that FDA maintains control of this database because our members do have experience in working with GPOs and not being able to penetrate some of the markets.

And the one thing we do not want to have is UDI to be a toll used by some folks who can drive compliance to preferred vendors and thereby really impact, negatively impacting patient care by preventing physicians and patients from having access to those technologies. Some of the words I heard today, efficiencies, margins, cost effectiveness, value predictable, utilization. Simplify, again all laudable goals, but I think we need to drill back down and focus on patient safety.

Again, I know we’re coming close so I’ll just close in saying you know we heard a little bit from large manufacturers that they’ve already invested millions of dollars in trying to bring this to fruition. Obviously they have many more products then some of the smaller players as well, but there will be significant costs I think incurred here. So to the extent that we can agree on the approach we take, have a proper time line for implementation, these are all things that will mitigate the development of this process.

And as I said in closing I think there are a lot of additional benefits to UDI, whether it’s linking it to post market surveillance from FDA’s perspective, as Dr. Plurrough talked about, coverage of evidence development, that’s something that we’re seeing more of, and obviously UDI can make that more efficient for CMS to collect that data, integration, into electronic health records, comparative effectiveness research, which is likely to really get a shot in the arm after the stimulus package.

These are all areas where I think UDI can be helpful. And actually there may be a cost benefit for companies to place UDIs on these types of products with the post market surveillance. When we’re talking about consumables, I think then the cost benefit from the industry’s perspective is a little bit more questionable. And that seems to be focused or the focus seems to be more on supply chain efficiency not on patient safety.

So again we look forward to working with FDA, Jay and the team to move forward to put together an effective and efficient UDI system. (Applause)


MS: Thank you, Jay. I think it was 1973 when Eddie Bunch (ph.) picked me 12 th on the basketball team and we went 18 and 0 in regular season and I think Jay hats off to you and everybody here because I think in the end you’re going to get what you’re looking for. I represent a group of companies, 700 of them worldwide within an organization called AIM. It’s a trade association, many of the member companies are outside today. We wouldn’t exist if it wasn’t for folks like you looking for ways to do a better job of collecting data.

We’re really, really good at what we do. And what we do is educate people on how to use computers. As I said there’s 700 companies worldwide on that original slide it said AIM North America, I’m actually an AIM Global member, but as an AIM Global member they also asked me to participate in AIM North America and we’re going to talk about that in a minute.

Jay also asked us in an email that he sent a few days before I arrived not to waste your time talking about things that have been covered. But it would be wrong of me not to tell you the amount of effort that’s gone into making a bar code work. It’s not an easy thing to do. And those standards groups that I participated in (unint.) and I don’t think you’ll see a bar code in the world that our member companies didn’t create.

We’re the people that invented them. We’re the people that invented the technology that allowed you to scan that bar code and get useful information out of it. It hasn’t always worked well. And if it wasn’t for the effort of the companies and the members within those companies sitting through meetings a whole lot worse than this one today, they just wouldn’t work. They just simply wouldn’t work. So the reason you can scan a bar code and get the right information is because of our member companies.

But it would also be wrong of me not to mention the fact that RFID has kind of taken over our group, because it is the next wave. There’s a lot of benefit to be gained from RFID. And where you folks talk about the fact that you can’t actually get, I forget the term the lady used, but I loved it, the widgets, we call them widgets in our industry, but to get the gadgets identified, do-dads, because they’re under a penny, you can do that. You can actually do that.

In the DOD they developed a strategy so that the items that couldn’t actually be physically marked are virtually identified, where you talk of your database, just as long as you can identify it somebody can acknowledge that we use that do-dad in that emergency situation later or during the event, you’ve got your data. That’s what you really need. So think in terms of virtual, where you can’t put a mark. Use your heads.

Mobile computing. I sat through a group of meetings, got myself in trouble on stage or actually I was in the audience and asked a question. This gentleman from China came in and showed us how RFID and cell phone technology was being used to gather where they put an RFID antennae on a Sim card and they could identify all the security people at the Olympics. I was so impressed with that I said to the audience, so should we sell out stock in Symbol (ph.), which is now Motorola? And I got attacked by the guys from handheld products, which is now Honeywell.

But the point being I think you’ll see today when something like a cell phone and an imager, you’ve got a little camera on yours, I don’t. We have the cheap ones. But you’ll be able to go out and scan and capture that same information if it’s available or read it off of an RFID tag. Mobile computing is really the biggest part of our industry today. I happen to represent a software company that makes the marks.

There was another point I wanted to make and now it escapes me. The folks from AIM don’t want to be forgotten, and we beg you Jay and the rest of you to include us in any efforts as you continue to move forward. And now I just remembered what my point was as I stalled. One of the things the DOD failed to do, and I think Rob will agree with this, is to put not just a mandate forward that says you have to market this way, but to put some teeth with that mandate in terms of the minds of most of the people involved teeth that said if you don’t do it we’re going to punish you.

But the thing you need to think about especially in today’s market or today’s political climate is that teeth could also include a smile in terms of a stimulus. The DOD could have gone to the manufacturers and said here’s the system to do this with and you might have an opportunity to actually do that. And that was the other point. And thank you very much. (Applause)

MOD: Look at that, two minutes to spare. Perfect. Thank you, presenters for everything. That brings us to the end of the meeting. Thank you all very much for coming. As you’ve heard the docket associated with this meeting, which is the docket, because if you don’t know where you send your comments is open until the end of the month, I think, give or take a couple of days.

As I mentioned before you know we have been thinking about this for a long time and we really do know we want to do, we do appreciate your help in us fine tuning some of these last points. You have my contact information. If you have any questions, comments, concerns, please feel free to contact me. And otherwise I will see you all at the next meeting. Thank you very much. (Applause)


Page Last Updated: 02/12/2015
Note: If you need help accessing information in different file formats, see Instructions for Downloading Viewers and Players.
Language Assistance Available: Español | 繁體中文 | Tiếng Việt | 한국어 | Tagalog | Русский | العربية | Kreyòl Ayisyen | Français | Polski | Português | Italiano | Deutsch | 日本語 | فارسی | English