FDA Unique Device Identification Public Workshop: February 12, 2009 - Panel 2 Transcript
MOD: We’re going to get going and hopefully these people will file in eventually. Thank you all. That was a great first panel, great conversation, all sorts of interesting issues. And now we’re going to move on. And we touched on this a little bit and I think this will be a very interesting discussion, presentation. We’re going to talk about what goes where. We understand, okay, we’ve got a code, we’re going to develop that and we’re going to put it on the device, on the label of the device. And the question is what if the label isn’t there or maybe we have very small things, what are we going to do about that?
The kit (ph.) issue that came up before, what are we going to do about that? So we’re going to spend a bit of time talking about what goes where, maybe touch on these issues of auto ID and human readable. If we have one, do we need the other I know thing? And, you know, discuss very briefly I think, or touch on some of the standards that are out there for auto ID. So that’s what we’re going to talk about in this panel. These are our panelists. And just like last time, same drill, they’ll share their perspective and then we’ll open it up for some conversation.
And as you have, and I’ve been chastised, we have two pages of very small printed questions. I apologize. I tried to get it all onto that double sided sheet that I handed out. But there’s a lot of questions here. This is a particularly difficult issue for us, because the legislation says that the UDI goes on the label of the device, except when it doesn’t. And so we need to understand the except when it doesn’t part. And there’s a lot of devices that I think fit into this except when it doesn’t category. And it’s hard.
This is the very hard part of this. So some very good questions, hopefully you all have a chance to take a look at those. And with that I’m going to turn it over to Rick Naples from BD who is going to give us his perspective on this.
RICK NAPLES, BECTON DICKENSON
MS: Thank you, Jay. Good morning everybody. I am Rick Naples from BD. I’m the Corporate Regulatory Affairs VP and probably the first person in this lineup that doesn’t have a logistics title or is not bar coding standards expert. And believe me I’ve got a lot to learn in this area. But I can tell you that I think we all need to learn pretty quickly.
If you pick up the paper, if you watch CSPAN and watch what’s happening on Capitol Hill with peanut butter, you only need think about how that applies to the device industry, and we definitely don’t want to be there. So we want to work closely with FDA, with providers and with patient groups to ensure that we adopt an appropriate UDI standard that we can all use that makes a difference for patients.
I think everybody in this room probably knows who BD is. But if you’ve ever had your blood drawn, if you’ve ever had a lab test or if you’ve ever had a shot of penicillin chances are BD Technology helped deliver that service to you. And today BD is a very big company with lots of pieces and parts of devices.
In our catalogue alone we have over 17,000 catalogue numbers. So I think it’s probably a good opportunity for us to share some of what we’ve learned with the diversity of our products in some of the recent DOD and VA standardization initiatives as well as other standard activities that we’ve been involved in thanks to our head of Ecommerce the great Dennis Black. You’re heard that name a lot today and you’ll hear it again, because Dennis is on the third panel.
We think overall that as we’ve learned over the past three years that if this is done right that there’s a lot of common ground for what providers need, what manufacturers can provide and what FDA needs most importantly for patient safety, but also to improve supply chain efficiencies, right? I mean, if we want to talk about electronic health records and the future of health in this country it’s critical to have good data standards, the foundation on which the electronic health records will be built. And I can’t imagine them being built very well without a good UDI standard.
But the key is we need to understand what that standard should be and work together to make sure it’s reasonable, because if you go and Google the term UDI, if you did it three or four years ago you might see a few thousand hits, today there’s over 209,000 hits on UDI’s. And much of those at the top of the list are from patient groups, are from providers that say we want this. And if you add FDA and Congress to that I think that’s a mandate we have to support.
BD’s investing in this technology quite heavily for the last three years or so. And I just want to give you a few examples of it, which also will help you understand the diversity of the products we make and the diversity of the products in the device industry. Here’s a Sharp’s collector, right, to store Sharp’s in. Well, as you can see we have a 14-digit GS1, G10 primary bar code and then we have the secondary information also in bar code. And both of them are in human readable form, at least in the case level. I was told not to touch that, and you can see I did.
As we get down to the individual unit, we have a G10 bar code and then we have the additional secondary information in human readable format. Remember now we’re starting to get down into the smaller and smaller products, and so there’s less real estate. Here’s another example, 20 mil syringes. I’m not sure if any of our customers today are scanning these in the hospital, but I can tell you that we are putting, again, primary and secondary bar codes on the case pack and the shelf pack. And again at the unit level we have the G10 and then we have the production identification information, both just in human readable form in terms of lot number and expiration date.
And another example here with spinal needles, maybe these because they’re used in very important procedures may well be scanned at the point of service today. But again it’s the same model, G10s and production data shared at all packaging levels and the production data on the each’s (ph.) level in a human readable format. We think this is a good approach for UDI’s.
There was a question earlier by Dr. Susan Altae (ph.) of FDA, I don’t know, is Susan in the room right now, I don’t see her, about IVDs. And I wanted to answer that question because it was a really good one and an important one for all of us to understand. Here’s a test kit for a diagnostic. And as you can see on the kit box itself, again primary and secondary bar codes, but take a look at the size of the components in that kit, of the reagents and the ancillary test tubes. Pretty small, right?
Well, G10s and production data is on the IVD kit itself. And that information is linked to each one of those components via the device history record. GMPs already require us to do this, so if we ever unfortunately have a recall, and unfortunately that’s a part of life in this industry, we do, we will be able to trace back each one of those components in each one of those kits. So we know what kits and what components are linked together in our systems, even today, even without putting a UDI on those small pieces.
Another example too. This looks like a big product, but it’s not. See this? This is a safety land set (ph.) device. It fits in my hand very nicely, very ergonomic. Thanks to the human factor’s requirements from Jay Crowley (ph.). And really Jay deserves a lot of credit for that. It’s made health care a lot safer. And on the side of it we have stamped in there the lot number. And that fits in a dispensing pack. The UDI is on the dispending pack along with the production information. But there’s no room on this. There’s not even a flat surface. And I guess I would ask our providers do we really want to add a bar code to something like this, that’s sold in boxes of 100?
A couple of other examples there, at the bottom left side of the page. That is actually a tip cap. And that is the size of my little finger’s nail. So it’s very, very small. And it’s used in mass quantities. So again these are the types of depth and breadth of medical products that we need to consider how would we implement UDIs, and do we need to implement UDIs down to the unit use level?
Based on the experience we’ve had to-date we think that there are three important considerations as FDA moves forward in this area. First, as I said, one size doesn’t really fit all, so UDI solutions should vary by product type and FDA should take a risk based approach and start with the highest risk products to make sure that this system really works. And second, and just as importantly, global harmonization will be key to the adoption of these technologies. We already have some countries in the world moving on with UDIs. We need to catch up, we need to have FDA issue a good standard, and we need to continue to work with countries around the world for global harmonization.
And last, but not least, the implementation plan is key, you know, the best laid ideas often run afoul if the execution of the plan is not appropriate. And so it’s really critical that we all work with FDA to make sure the implementation is the right size fit for our patients and our health care providers.
In terms of one size doesn’t fit all, take a look at this, this is a large flow cytometer (ph.), a very sophisticated clinical lab instrument that helps diagnose cancer. It’s probably about a $250,000 piece of equipment. And we would hope that that would stay in use in a clinical lab for maybe 15, 20 years. During that time there would be replacement parts, but it would be kept up with a preventive maintenance, etc. And there’s no question in my mind that products like this should have serial numbers on them, which they do today and lot specific product data. And certainly we can bar code that information because there’s plenty of space on a piece of capital equipment like this.
Then at the other end of the spectrum, you’ve already seen some of these very tiny devices. And in this picture there’s a dime in the foreground and a penny in the background. We put it in there for a couple of reasons. First of all, that’s about the size of some of these products. But not coincidentally that’s also the price range of many of the products that you see pictured here too, so very small, very cost effective, right? So we need, we have to avoid a one size fits all approach.
Again, we need global harmonization. And if you could read Japanese, which I can’t, you would know that what this means is that JFMDA in their recent guideline, which by the way goes into effect for certain products as soon as March 9, 2009, this year. They prefer to have a GS1 128 bar code on it to ship products into Japan, and they would also prefer to have concatenated bar codes.
We’ve heard a lot about concatenation and non-concatenation. But let’s see what that would do to one of our products. The product on the left is a box about two inches wide, primary and secondary bar codes. That’s the approach that we’re taking for the United States. But now when we want to ship that product into Japan we would have to over label it with the Japanese standard concatenated bar code. And that bar code would wrap around actually and touch on three sides of the package.
Now can we move it onto another side? Could we use a different bar code? I’m sure we could in the future. But the key here is don’t we want to all control health care costs? I mean, and I think it’s critical to make sure that we avoid over labeling, if at all possible.
Again, global harmonization, the need for it. On the left you see a flush syringe with saline in it with a nice bar code on it. That’s for the U.S. market. Same product on the right side for the European market. And if you’re able to rotate that barrel what you see is there is not a single millimeter of space there, because there is different language labelings all around the barrel, so there is no room for a bar code there. So again we need global harmonization, Jay.
I’d like to offer one possible solution that comes from my days in the diagnostic industry. Thirty years ago when FDA put regulations in place for diagnostics they developed a separate set of labeling requirements to reflect the unique needs of diagnostics. You saw that kit I showed you earlier, right, with a lot of the tiny pieces and parts in it.
Well, under 21 CFR 809.10 FDA said if the immediate containers are too small to accommodate a label with all the information required in the regulation then only certain pieces of information that identify the product, its expiration date, its storage, have to be on that component and the additional information required by law can be appearing on the outer labeling container only or the kit box.
And so if we took an approach like that for the medical devices that we’ve shown here and we think about maybe putting UDIs on the next higher level of packaging, such as the dispensing pack, would that meet customers’ needs? And I think that’s the key question for us to work with our customers on and for FDA to consider.
So in conclusion we think that a global harmonized UDI system has great potential to improve patient safety, post market surveillance and enable supply chain efficiencies. We’re already enjoying the benefits of those supply chain efficiencies; we know where our products are at all times and we’ve got clean data in our system and I think it’s helping our customers too.
And so what we’d like to suggest is that as FDA considers where the UDI’s placed on the label of the device, that they should require it at the unit of sale or the dispensing pack. And why? For two reasons. First of all, that dispensing pack is usually stocked right there at the patient site of service and can be bar coded if needed. And secondly that dispensing pack is usually large enough to include a UDI without increasing the package size. If FDA wants to go further that that and look at UDI’s for the individual unit level, I think it will be key to talk about what products really require this based on risk.
And also we need to collaborate even then for a very flexible approach that allows us to meet a standard either with bar code information or human readable or some combination thereof. So thank you very much. (Approach)
MOD: Thank you. Rick. Tom Werthwine.
TOM WERTHWINE, JOHNSON & JOHNSON
MS: Thank you. Good morning everyone. When Jay called me he kept stressing to me you only have ten minutes, so at some point he’ll yank me off, but I apologize for that. Jay started with showing a couple of Johnson & Johnson labels, so the tactic I want to take in the next few slides is not to talk about Johnson & Johnson. What I’m going to preface is this is not a Johnson & Johnson presentation. I had the privilege of working with Jack Yoken (ph.) who will be speaking in a later panel and (unint.) leading a AFAMED (ph.) working group to write a written response to the FDA UDI docket.
What I’m going to do is present some of the opinions and some of the issues that have been raised in our teleconferences. In our last teleconference we had over 40 member companies talking about UDI. So again this is not a Johnson & Johnson or this is not the world according to Tom Werthwine, but it really is kind of an industry snapshot. This is an operating theater. This is an orthopedic surgery going on. We have many, many more medical devices than there are pharmaceuticals. So when we begin to talk and we’ve heard a number of words today, such as implant, or such as IVD or such as instrument, or such as kit, these broad sweeping terms almost become meaningless for us. They acquire a number of devices within their scope. What we don’t need from Agency is a broad sweeping, all implants shall have or all instruments shall have, and then say to the individual manufacturer ask for an exclusion.
With pharmaceutical hospital unit dose rule we had that, we had a paper exclusion, so we could come in and say within this NDC number we can’t meet the need, you know, because of the size of the container or whatever and then where was a paper response that said, okay, your drug is excluded. The number of devices that we have and the number of exclusions we’d be asking for would only put a burden not only on the manufacturer but put a burden on the Agency to answer. So that’s one of the points that come up.
Direct part marking. And again all of these, and it’s embarrassing because Jay did show J&J products, I’m not showing any J&J products in this presentation, trying to give us an industry approach. And these are some devices that I received permission to show from Alcon Laboratories. They are instruments I believe used in ophthalmologic surgery. Again, very similar to the BD rule; presentation, too small. We have to look at these things and realize that fit, form and function may be altered by the inclusion of a UDI, whether it’s a data matrix bar code or something.
So it’s a whole different area; it’s not just affecting package and engineering, but truly this becomes a research and development and a quality issue for the manufacturers. I didn’t use any (unint.), Alcon used a ruler. (Laughter) In terms of packaging placement, there again you have a number of form factors. Not all devices are the same size, you can have one UDI for one device that’s on a palette or you could have one UDI for tens of thousands devices that we would like to sell in palettes.
So what we’re saying is the placement, where on the label should be determined by the manufacturer. It has to do with better understanding, how our product flows through the logistics chain, what our customers need and what our own manufacturing and labeling and packaging guidelines can provide. The industry does have guidelines. And we’ve heard about HIBIG (ph.) and we’ve heard about GS1. There’s HIDA, and there’s HDMA. And I apologize for all the letters, but there are already outlined guidelines out there for packaging and logistic labeling.
This I believe is in the public domain. But this is a combie (ph.) stopper from B. Prawn (ph.). They use a form, fill and seal machine to package this product. So that’s the sterile barrier. There again looking at the inclusion of some devices; if you have an infusion pump it makes sense to have UDI, if you have a full IV set and you’re trying to determine the difference between a plain IV set and a set that’s used to dispense nitroglycerin it makes sense to have a UDI, but a combie stopper, a lure lock injection site, something like that you may not need to label at the unit of use. Again, with VD you may want to go higher up, you know, in the packages of 500 or 1,000 or whatnot.
Packs and kits. There again the word is sneaky. You’ll hear people talk about convenience kits. And so these are things that are kind of assembled per the surgeon’s instructions. What our kit manufacturers have said to us, they should be treated just like any other device, not mandate that you have a bill of materials of UDI and have everything listed on the labeling. And especially in the database this becomes very problematic with substitution. And that information is there because of the GMP process also.
This is a central venous calcular (ph.) kit in the lower right hand corner. And there again just to give some, the idea is that you have some of these products, they’re very, very critical, right, they’re going into the vein underneath, the suplavion (ph.). So you can understand where that information is very, very needed and required. Some of the other accessory items there you really question where you would put your UDI or if it was needed.
Technology. We’ve heard from GS1, we’ve heard from HIBIG, we know that NDC is being considered for some devices that are already labeled with such. The good thing about using industry standards is that there’s a vetting process. I was aware of GS1, for instance, received a request from Baxter to use Data Matrix for health care products. And that went through a vetting of the GSMB process and it was approved. A similar thing happened with HIBIG. At one point member organizations came to HIBIG and said we want to use Data Matrix. They went through their ANCI (ph.) approval process and it was included.
So the Agency doesn’t necessarily have to specify what symbol or what frequency, leave that to the manufacturers and the providers that work together and make the decision, you know, when does the symbology (ph.) change or when does the frequency change and move from there. That’s the value that the industry organizations can bring us.
Depending on the device, and again inclusion versus exclusion, you may say some instances where your UDI can be human readable only, your UDI can be, if you look at the two elements, you have one that’s a product identifier and one that’s a production identifier, your product identifier could be bar coded, your production identifiers could be human readable. You may have a more critical device where you want it scanned and it goes into the electronic health record. You would want both elements to be there.
So again it’s not one size fits all, it’s not you can’t say all implants or all kits, we need to look at that very carefully. And AdvaMed and industry leaders like BD are very willing to work with Agency to really specify and get into granularity of what would work at what level.
Time lines. I think this is interesting and this is where I’m going to divert a little bit and really talk about some personal issue. I’ve had the privilege of calling on a number of the hospitals actually who are in the audience. And one of the things that I’ve learned is that there’s value in all three. There’s value in direct part marking for some devices, there’s value in package labeling for some devices, and there’s value in the data. I’ve had a number of hospitals that said to me the data for me may be more important than the bar coding, because I use an automatic dispensing machine.
So what I would suggest there, and even the challenges of implementing these three portions of UDI, in a company could be handled by three different departments. Packaging, engineering would do the packaging, research and development and manufacturing engineering could do the direct part marking, and IT and regulatory affairs could do the UDI database. So consider three different time lines. And interesting, okay. Thank you. (Applause)
MOD: Corwin? And thank you Corwin for substituting at the last minute. I appreciate it. Mark Coil (ph.) who was going to be with us could not make it. So thank you.
CORWIN HEE, COVIDIEN
MS: Hello, my name is Corwin Hee. I’m the Director of Ecommerce for Covidien. I’ve been with the company, this is my 15 th year with Covidien. My thoughts and prayers go out to Mark Coil for his safe recovery. The reason why I only have one slide is because again I’m coming at this from a very different perspective than Mark and many of the other specialists who work on a daily basis with the GS1s or other labeling standards.
And so what I wanted to do was to get beyond sort of the minutia of exactly how do we mark this particular part and talk about where are we trying to head with this? And so that’s the reason for my general commentary here. And that, realizing that this is going to be a very long process, this is not as if we suddenly have a UDI database that we will ultimately have all of the things that we need, this is really the first step. And so where the power of this initiative comes in to play is in the database itself.
And so the marking of the individual parts is really the first step; it’s the key to get into the database so that you know what you’re talking about. And from there you build out from that. And it’s vitally important that we get the information that’s in that database to the end user, because if you don’t have some mechanism to do that then really we have not met the needs of the patient.
And so in order to do that the best way that we can do that is to have the greatest positive impact on patient safety it’s critical to use a risk based approach. And by that what we mean is we’ve got a finite limit on our resources available to these types of initiatives. And so we should use that in the most cost effective manner possible. And by that what we have to do is we have to balance the needs of what it is we’re trying to do with what we are able to do with what we are able to do, because the more that we ask of any database the more difficult it will be to build that database and then to execute upon that and to deliver that information.
So again this is why I stress that we do need a risk based approach and we also need an iterative approach; a phased approach, so that we can learn lessons. For many of the most critical products today there is an enormous regulatory burden to begin with. And so there’s a corresponding amount of information available for those products. We could use that as a starting point for our database such that we could learn lessons and then apply that to lower and lower risk products.
Now this does not obviate the need for barcing of all products, at some point in time, but we do recognize that we have to focus our efforts in an effective manner. And with that I’ll close and look forward to the discussion in the panel three which is to my mind the most critical piece of this. (Applause)
QUESTION AND ANSWER SESSION
MOD: Okay, now for the fun. We will get to the database and it is interesting and difficult as well, but there’s still a lot of issues here that we need to understand. So we talked about it a little bit before and so I really open it up to the audience to come forth on what issues and questions you have about what goes where, how are we going to deal with this issue.
Rick mentioned and we’ve talked about this quite a bit, he used a couple of terms, which are all terms of bard (ph.), I think. We have dispensing pack, outer carton, the shelf pack. And what makes this so difficult for us is that we don’t have any regulatory construct for this. I can’t find shelf pack in the regulations. So I can’t point to it and say put the UDI on the shelf pack or put it on the dispensing pack. And so I’m really looking for some help in how were we going to talk about this exception because it is often an exception.
The legislation talks about, as I mentioned already, the UDI going on the label of the device. So if you think about the syringe example that you brought up, the label is the thing on that syringe. And so if we’re not going to put the UDI on that syringe and we’re going to put it on whatever term you want to use, shelf pack, how are we going to talk about that? And that’s the challenge that we have. So I’d like to open it up to questions. I mean, we talked about kits a little bit and I think maybe we should get back into that as well.
I have been asked that when you do come up to talk to the mike that you talk to the mike and not to the audience so that we actually capture what you’re saying. So questions? Comments?
FS: From a provider perspective and looking at the risk and the costs involved I just don’t see us needing to have, and when I think about this, I think about it from a recall purpose, I don’t see us having the need to have that UDI on the syringe or what I refer to as do-dads, the little bitty items, you know. (Laughter)
MOD: Thank you, Jean.
FS: Sure. I just think that we need to be concerned with the products. I mean, if a needle doesn’t work it doesn’t work they get another one or if the do-dad doesn’t work they get another one. It’s three cents. We don’t, you know, we’ve got these economic issues going on right now, and even aside from that, I just don’t think in my opinion it’s the right thing to do to increase that nickel item to eight cents or ten cents when we use gads of them, scads of them.
MOD: Lots of do-dads.
FS: Gads of do-dads, yes. So it amounts to a lot of money, you know, it’s not like it’s a stint that’s $1,500, yeah, one time that’s a lot of money, but when you really look at what overall cost is for these products and you multiply that out it becomes biggish.
MS: I strongly disagree with the previous comment, because if you have an infection inside of a hospital then you want to know which needle and which one, where it came from. So I don’t think it’s fair to make that decision at this point. So I don’t believe that is quite right, because it should be done by actual people who are doing clinical studies. I don’t believe you should decide it by somebody who does material management or supply chain issues. I think those are two definitely different issues. Thank you.
MOD: Thank you. Questions, comments?
FS: You know once the package, once the item is used the package is thrown away, if we have it on the shelf pack, you know, if it’s on the box and it’s a box of needles, then we’re going to know at least to that level that, yes, we did receive that lot number, that there’s a recall for it. But again if you go back to the simple types of items, like the needles, I don’t see, from a cost perspective maybe not now, maybe as we get into this, and we can get to that point in five years from now, seven years from now, whatever that timeframe is, but from a cost perspective and from tracking we’re not going to be, the end users are not going to be tracking to that point. The nursing staff is not going to be tracking down to that point. So I just think we need to keep that in mind.
MOD: Great, thank you. Panel do you want to react to this discussion at all? (Laughter) Oh, come on. No? Alright, Jim.
MS: Jim, from Eckery (ph.) Institute. A couple of points on the individual labeling. Definitely there is value in having information on a device when you’re doing, for example, an investigation. So that is helpful. But I wanted to bring up another point on another class of technology that hasn’t been brought up much yet today. It’s the software based products. And we saw the BD analyzer and that’s going to have software revisions in it, over those 10 to 15 years. I was surprised by the 10 to 15 years, given the way that technology changes, but software revisions, patches, like security patches that go onto the devices.
And then also devices that have modules, like capital equipment that have modules built into them. And there are challenges as to whether or not you have the UDI on the individual modules and then just like with a kit does the parent have a UDI for the mix of modules that are in it today and then how does that parent product, the larger device that has the sub modules in it get changed as you pull out one module and put another module in. So it’s just a complexity that needs to be addressed and I know that you’re well aware of it, Jay, but I just wanted to bring that up today.
MOD: Thank you. Panel?
MS: Yeah. Jim you bring up a good question and first of all I’d like to say that the 15, 20 years I was talking about that was a wish list on my part. You’re right technology changes very quickly. But I can say that certainly if we have a serial number and a UDI in the instrument itself there’s probably ways to use UDI’s plus some type of variant code for the key components that change on a regular basis that we need to track.
MOD: Corwin did you want to...
MS: I just wanted to reiterate that I think that it’s difficult for us to say that we have all the answers right now. I think an earlier comment was that is this a work in progress. And it absolutely is. I don’t think anybody should be under the illusion that we know everything there is to know. I think that this is where there are standards out there that are used in other industries; however, that we cannot just assume that health care can operate in the exact same model. It does not.
And clearly there are going to have to be changes to, or modifications to standards in order for it to become widely used. And I think it’s important for us to remember that we have to start with a base line. And I think this is why I’d advocate for a risk based and for an iterative approach, because I think that by doing that what we can then do is we can work on the worst cases first and then move out from there. And I think that’s really the only way to bring this to the industry.
MOD: You are now.
MS: I am now. Okay, thank you.
MS: Just to comment on the software question both the GS1 and the HIBIG system and indeed the NDC system were designed to label physical objects. And so I think if you look at software as a product, as a regulated product, there may be other standards out there and other standards bodies like HIMS (ph.) or HL7 or DICOM (ph.) that may give you more utility and more features. So just don’t assume again, as Corwin said one size fits all or one system works for everything.
MOD: Great, thank you.
MS: Perry Manalapolis. While there are three points now, (unint.) discussion. First of all, HIBIG standards do accommodate and serve as not only the number that identifies the product, but for more revisions, software revisions of what’s installed on the equipment. So that’s in place today. Secondly, the point I want to make about what to mark, I think we need some fairly strong guidance because today it’s really at the discretion of what the manufacturers feel that they should mark, and oftentimes it’s the unit of shipment.
We’ll do the cases, we’ll do the outer box, but we don’t think there’s any value or there’s a lot of cost to marking the individual items of use. And I think that just leaving it that way causes a lot of problems; providers are re-packaging items when they see value and benefit and identifying the individual products for patient safety, for inventory, for a lot of reasons. So I think we need a good discussion on that and good guidance on that and not just let it be discretionary by the manufacturers.
MOD: That was two. I think there was a third wasn’t there?
MS: Yeah. And that third one was...
MOD: We’ll come back. Panel, any reaction?
MS: Much like the gentleman sitting to my left here I was kind of thrown into this at the last minute. As far as representing what some of the providers would be thinking about, first of all I have to say that manufacturers do a really good job in marking their products with lot numbers and being able to look at sterilization numbers and things like that, so I think that I’m not saying that they should regulate themselves, but I think if we could take what they’re currently using now and adapt it into what we’re proposing to use, I think that we’ll have a pretty good starting point.
You know this is going to be something that needs to evolve. We can’t attack all or the issues at once. So we need to come up with a starting point, and that’s kind of why we’re here because we’re hoping that the FDA will help us develop that starting point. I’ve been working with GS1 for ... sorry, 2008 was the year of the acronyms, did you ever notice that? When I first started getting some of the information from GS1 and so forth it was like a study course to try to figure out where all these acronyms were coming from.
But GS1, we’ve been with them for about nine months and we got involved in it because Steve Huckabee has been part of this incentive for the past ten years. So Steve asked me to come in and at least sit in for him so that I could rely some of the issues that we’re having. Our biggest issue is patient safety. And there’s some wonderful technology in this room from a lot of manufacturers. And technology really has evolved quite quickly over the last ten years, you know, when you think back on what we used to do during an open heart to what we do now.
Our responsibility in patient care has also increased ten fold today. The technology that’s available to us in patient care is evolving very quickly. And the problem that we run into is that from a supply chain management, because our responsibility is to purchase the items, make sure they’re at the point of use, and also deal with things like recalls. I know that from my own experience sitting in positions that required me to go out and follow up on some of these recalls that not having a UDI to reference to made it a very time consuming effort and not very effective.
A lot of times we would get notifications where manufacturers were trying to blanket the customer base and sending out general notifications of recall. And, of course, trying to comply with that, because we understand what the ramifications of not fulfilling a recall notification, we would go out and look and start searching through our systems, trying to find some type of index to relate the products that we purchased to what we’ve distributed onto our nursing units or into our patients. And it makes it very difficult.
The UDI numbers are something that manufacturing fields outside of health care have had in place for a very long time. And GS1 has helped them develop the standards that they need, so that if you go to Wal-Mart or K-Mart or a grocery store and you scan one of their items that grocery store except for a price difference can identify that item from store to store. In some cases, if you take something back to say K-Mart and you originally bought it at Wal-Mart they’ll be able to tell you that that’s exactly where it came from and turn you away at the return counter. (Laughter)
But we need that same type of technology in health care, because as our technologies advanced our responsibility to the patient has advanced. And we’re really holding ourselves at risk and we’re holding our patients at risk. And we’re really doing them a disservice if we put this wonderful technology in them and use it to save their life where at the same time if we get a notification from a manufacturer because of an issue that they had in the production of it, we can’t go back in many cases and find out what that product was and where it was put into a patient at.
Could you imagine if you were working at Wal-Mart and the FDA issued a recall and they mandated that Wal-Mart contact every customer that had ever purchased a stroller and notify them of a recall? What do the manufacturers do now? Typically they’ll maybe put an advertisement in the newspaper or as you’re leaving or entering a store they’ll have it posted up there for you to see.
Our responsibility really goes much further than that. If it’s an implantable device or we’ve used it in a procedure as a tissue transplant or something like that, it really is our responsibility not only to search within our own facilities to see if we have any of that affected product, but actually go back to the patient and notify them, so that if they have to take any type of reactionary approach to what the recall involved they can do that.
Without a UDI it makes it almost impossible. We’re really doing ourselves and the industry a disservice by not adapting a UDI format so that we can utilize these tools to enhance our patient care.
MS: Napoleon Monroe, I’m a consultant to the industry and I often ask myself agreeing with Dr. Moon, but in this particular case I’d like to point out that some years ago Congress in their wisdom passed a law that said every part, every medical device must be marked with the name of the manufacturer. FDA had to toll that, they couldn’t enforce it. So marking every part with a UDI I think will be impractical. Just a comment.
MOD: Great, thank you.
FS: Susan Altae (ph.), FDA. We are discussing in this panel where to put this UDI and on what devices and what’s the common sense and when do we stop labeling things, whether they’re do-dads and we use them in gads or whether they’re a specific heart valve and we use them in one specific patient. So being from the FDA we’re a risk based agency. We recall on the risk base, we approve on the risk base, why don’t we label on a risk base? Just a thought.
MOD: It’s a good thought.
MS: This is (unint.) FDA. I was wondering, we talked a lot about bar codes and their role in the UDI. I don’t know if this is the right panel or not, but are we going to be discussing RFID’s role in this UDI process?
MOD: Sure. Go ahead. (Laughter) It’s an open topic and we can talk about it here or we can talk about it in panel four. What would you like to talk about?
MS: Well, we’ve been discussing you know different standards and different ways to implement bar codes. But we haven’t discussed any of the RFID standards. I don’t know if we’re talking about implementing them. Just trying to open a discussion.
MOD: Right. I think they’re on the table. There’s obviously a number of technologies. There we go... Do we identify specific technologies or not? And how broad or inclusive do we want to be? And what are those technologies or do we want to limit them in any way? I mean, I think that’s one of the topics that we do need to talk about, and if we are going to identify standards, which ones, and how specific do we want to be? Yes?
MS: I have a comment. And again this is a more personal observation. But I’ve seen some excellent RFID supply cabinetry or supply chain management that is out in hospitals now. Out in what I would call the bread and butter hospitals, so this is not just the cutting edge educational institution. So it is happening. It’s effective now. So I don’t think you would want to exclude it. But I think again there’s this question when we talked about do we have a UDI that’s human readable no matter what, no matter where it’s marked, or how it’s marked, there’s a question of redundancy.
And you will see this in some of the DOD standards that are written for RFID. If you have RFID and the chip doesn’t work where does the user get the data? Do they get the data from the bar code? And then if the bar code can’t be scanned, do you get the data from the human readable. So if you want to employ a technology like RFID then I think you need to consider the backup systems and the redundancy and that’s been well documented in some of the DOD work.
MS: I guess I need to clarify what I have said. My point is we are talking about patient safety. Personally I don’t believe supply chain folks know what really happens at the floor by clinicians. So I believe we should listen to what the clinician says, what they need. As an example in Britain they currently track all surgical tool, because of mad cow disease related issues. That we did not see before. So it is premature and I think it would be extremely unfair to allocations in the future to allow certain decisions made by people who really doesn’t do anything in that area.
And I feel very strongly it should be clinician who makes some of these decisions. And in terms of a cost savings you know there are cheaper ways to do UDI. You know, GS1 is not necessarily most cheapest or most cost effective, I hear. So maybe if you’re really interested in cost maybe you should look at the total cost, see whether there are any other options.
MOD: Thank you.
MS: Just a comment on the point. I believe that we all share the fundamental goal of patient safety. I don’t think anybody in the room is against patient safety. I think we all have different ideas about how perhaps to get there. And one of the things that we have to remember is that we have to get a little bit beyond the actual carrier of the number, because again that is just the starting point to the databases. And it’s really the databases talking to the different databases. And that is where you, the power of having these standards. Without that you have nothing.
And really if we don’t, I cannot underscore that enough that it’s what information, that that number, the UPC code only tells me, it’s meaningless in and of itself, I have to go back to a database to find the information that tells me the data. So that is where I think much of the work maybe needs to be centered. So, for example, to your point about patient safety, what are the fields that make most sense for us to include in that?
And that is, I think, where some of the discussion needs to be focused and get away from some of the carrier discussion and get away from some of that, because I don’t think it necessarily matters if we have the right information and we’re passing it to the right way through the system.
MS: And Rick may be able to comment on this much better than I. But I almost envision a system where if we’ve developed a new medical device, you know, something brand new, something not made before, and you do a pre-market PMA or you do a 510K or whatever the regulatory submission that you need to do, at that point you have a section of that submission that says, and this is our UDI and this is how we intend to display it or we intend to use it. And then it does become part of the clinical review.
If you go in and you do clinical trials, you know, the person managing the trial can come back and say, you know, I can’t read it or I want it here or I want it this big or whatever. So I think there can be mechanisms in place that brings the clinical review to this. And I think Dr. Moon you’re right I think it’s invaluable and could be part of the approval process.
MOD: And just to clarify so that you all come back after lunch this is what we’re going to talk about after lunch and we’re going to get into the database and the fields and what fields are necessary for different products. So stay tuned. Rick, did you want to comment?
MS: Tom’s right, I mean, there’s lots of ways to handle some type of risk based review. And Susan made an excellent point, FDA’s mission is to regulate devices appropriately to ensure patient safety based on risk, but also to promote innovation. And so I think it’s a huge challenge for the Agency, it’s why they’re in front of Congress so many times, but it’s really critical that we get this right in terms of making sure patient safety comes first. And I think we can do that with a flexible UDI approach.
MS: I’d like to talk to the second bullet again, which is the data carrier. And I have to strongly disagree with you relative to that it doesn’t matter about the data carrier, because the longer we rely on these and these to get the information into a system, the longer we’ll continue to make mistakes. So when the USDOD developed its IUID policy it said we’re going to establish a minimum machine readable requirement.
In our case we checked the Data Matrix because the ECC 200 Data Matrix has built in error correction. It is applied by any technology available to mark. Dot matrix can be applied by dot peaning (ph.), laser edge bonding, etc., so you need to look at the applications that you’re going to do, whether it’s print or anything else, you’re going to have to look at how to improve the data quality going in, because as long as you rely on these kinds of input mechanisms the longer you’re going to have bad data and you’re going to make wrong decisions on bad data.
When we’ve got 600 parts marked in an F22 engine, if I am going to write that down and I get it wrong and a plane comes down that’s a problem. In terms of the medical devices you have just as critical at the individual level, just as critical decision making. So I think you have to go and look at pushing the state-of-the-art in terms of the marking methods. Now I do not think we’re ready prime time for RFID at the item level, but I think you have to look at using vision based methods like the ECC200 Data Matrix; you need to move on beyond laser.
And believe me, personal experience you’re going to get a lot of resistance. But in this community you’re already moving that way; you see them every day. You see them, just look at your mail and you’ll see postal, you’ll see financial statements, you’ll see Johnson & Johnson products and others that are already mark on aga foods (ph.) with Data Matrix symbols. So that’s a pervasive symbol. I’m not saying you have to choose that one, but just trying to explain why we chose it. And I really urge you to think about the quality of the data and what carrier will give you good quality data. Thanks.
MS: To amplify that the carrier’s important based on the application. Where you were talking about Data Matrix and what I found with customers is that Aztec (ph.) works better for many of the curved or regular surfaces, particular if you’re labeling a syringe barrel, then Data Matrix, I’ll spare you the technical reasons, but the point is I’ve heard people ask about RFID when it’s a very costly item. Whether it’s a very costly medication, we want to know where it’s at, but it’s not necessarily you know cost effective for the do-dads, so to speak.
And the second part is that I remembered number three so I want to pose the question. If we were to consider having something in the UDI that designates a complete unit, such as the medical device, it’s a machine, versus a component that says I’m a part that could be used in one, two or three machines, is that beneficial or confusing. And also contrast that with I like that idea, but I’m not sure it would work for kits that are put together because the individual items that construct a kit could have other uses and aren’t necessarily always going to be in a kit that somebody makes up. So I’d like to hear your thoughts on that.
MS: Interesting point. If you go back into the consumer world and you look at laptops and laptop batteries, if you had a parent/child relationship and every laptop manufacturer identified their battery, their unique way, when there was a problem with batteries you wouldn’t be able to trace that commonality of you know where are all these batteries used from? So it is a complex issue, this whole parent/child relationship. We do see the parent/child relationship in the DOD Data Matrix marking. But it is something that we really need to look at and understand the utility of that. There’s arguments to go either way.
MS: Thank you, Jay. My name’s Dennis Cooper. I’m with the Cooper Group in North Carolina. I just want to bring down a, get us back in focus a little bit. All of our health care providers take an oath when they join the health care field and that’s first do no harm. Today unfortunately our providers do a lot of harm. They do that without meaning to and they do that without their own knowledge. The health care system is the number one cause of death today, with 100,000 deaths related to medical device errors or medical errors.
A lot of those are related to the devices. We have another 90,000 deaths related to infections. And that’s out of the 1.7 million hospital-acquired infections a year. These things are very significant to our cause today. We look at, we have about half a million products going into the hospitals each day, each work day. That’s over 500,000 units being distributed and they’re going down to the point of use. Not going to the shelf pack; they’re going down to the point of use.
These products are getting in the hands of the providers and they’re being used on the patients. And they’re causing harm. And given the fact that hospitals don’t get reimbursed for these hospital-acquired infections there’s a substantial revenue loss to these health care providers today that we have to address. And we’re looking at how do we label and what do we really focus on, it’s putting an electronic barrier between that patient and that product.
That’s the key to success with this. So that we can scan the product at whatever level that we’re using it and stop it from getting to that patient if it’s been recalled or if it’s expired. We have to keep that utmost in mind as we go through these standards. Let’s err on the side of the patient and on protecting them. And there’s a lot of work to be done. There’s a lot of issues coming up. You may want to look at doing some study groups that focus on some of these specific issues if we’re going to get this done before 2020. It’s a big issue. Thank you.
MOD: Thank you. We’ve got time for about two more, so please.
FS: Thank you. I really appreciate the discussion on kits given that that’s something we do. The other topic that really hasn’t been brought up yet is implants. I don’t think you want to have a recall and then have to go back in the patient and dig it out, especially when we implant 100 of our devices in every patient. And trying to find that particular one I don’t think it would be very cost effective or appreciated by the patient. So it’s just something to consider. And try to keep it on a surface level. Peripherals I understand, maybe, the needles and whatnot, but implants I think there’s a limit to what we identify. Thank you.
FS: My name is Carol Stone and I think the discussion about the patient safety is a really critical one. I think ultimately that’s why we’re all in this industry. It’s about the patient. It’s about doing things the right way. So I want to lay that on the table first. And I think you’d be hard pressed to find anybody in the room that didn’t agree with that. Having said that, however, for those of us who have also been in this industry for a long time, and I hate to admit how long I’ve been in it, we know how long it’s taken us to get to this point to have this kind of interactive discussion amongst trading partners in the industry, and it’s taken a long time just to get to this point.
If we go after, right now, the ultimate solution, I’m not sure that it’s ever really, I mean, I’m not sure that we’re going to be able to get there realistically. If we don’t use an iterative approach and a risk-based approach I think it’s going to be very difficult to actually see true implementation of this. If we can be reasonable and say you know let’s at least try to get a standard numbering system implemented with data behind it that gives us good solid data behind our products, use some of the approaches that are already being used today in terms of labeling, don’t try to press a more difficult labeling approach than we have today, but maybe set some dates or some targets for future review of those more difficult issues, whether it be the symbology, the technologies or whether it be, you know, the level of packaging, those types of things.
You know, maybe the next level thing that we go after and we set a date of I don’t know, make up a date, and you say let’s go back and re-address the level of packaging, you know, let the manufacturers do what they can today and then go back and try to force it down to a lower level, once they’ve gotten over the shock of having to spend all the money to just use a standard identification number. And then maybe you get to the next level of really pushing for more sophisticated, more dependable, symobologies and technologies. So that’s, for what it’s worth.
[END OF SIDE]
MS: Keep it super simple. But you’re absolutely right. This is where we need to start now, working together on where do we start, what’s our foundational work? Let’s go with the highest risk and highest value implementation first and then as technology evolves let’s embrace that technology and drive it down to lower levels working together.
MOD: Alright, that brings us...
FS: It’s a short comment to amplify. The FDA could think about maybe and the panel could comment on a couple of different risk based approaches if you decided to go that way. Obviously the FDA is already gone through massive analyses of all the device categories and there are things that are 510K exempt, QSR exempt. It could be dealt with categorically so it wouldn’t be such an onerous exercise for industry or the FDA. That would be one model that maybe the panel or the FDA could comment on.
Another would be the way the FDA handles software to allow a manufacturer to make an analysis of risk level according to criteria the FDA would set forth and deal with it that way.
MOD: Okay. Thank you.
[END OF FILE]