It’s good to be here again actually. I was due to give this talk a year ago at the meeting that had to be cancelled and, frankly, as a result really of a combination of COVID, of things that the EU did and of Brexit, what I’m going to have to say is actually going to be an awful lot more speculative than it would have been had I been giving this talk a year ago. Things have changed and I really can’t be anything like as definitive as I would have been a year ago. So, what we’re going to talk about is where we are currently regarding the Medical Devices Regulation in GB and NI. Now, we’ve got to get used to referring to the United Kingdom as being in two bits, GB being England, Wales and Scotland, and NI being Northern Ireland, because, as we know, Northern Ireland have, as part of the arrangements that we’ve made, Northern Ireland are sticking with the EU rules in the greater part and in particular in relation to medical devices. They’re implementing the EU Medical Devices Regulation, but that’s not the case in GB, in England, Wales and Scotland. So, I’ll give you a brief overview of what the current regulations require that affect health institutions, the current regulations in GB. We’ll deal with those in the context of in-house manufactured devices, including software as a medical device and research devices and maintenance. And then I’ll do some speculation as to where we might end up, because, frankly, we don’t know at this stage and a pointer towards best practice for in-house manufacture.
So, what’s the situation at present? The UK left the EU on the 31st of January, but with an agreement for there to be an implementation period and during that implementation period there would be no legal changes to the way our law ran and what that meant was that any EU regulations, any EU directives or EU regulations that were in place either before or during that implementation period would become what they call retained EU law and, if you remember, the situation was that the new EU Medical Devices Regulation came into force in 2017, but it became fully applicable, it would have been, it was planned to become fully applicable on the 26th of May 2020 and then in very early April 2020 the European Commission decided that, well, they said for the reasons of COVID, but actually there were a lot of other reasons behind it as well in terms of having available the EUDAMED database and so on and so forth, they postponed the full application of the EU Medical Devices Regulation by exactly one year, so to the 26th of May 2021. Now, unfortunately that meant, I’d regard it as unfortunate, unfortunately that meant that the full implementation, the full application of that new EU Medical Devices Regulation fell outside of the implementation period and therefore those EU regulations have not become retained EU law.
So, where do we stand? Well, we are actually back, in England, Scotland and Wales we are back at using the UK regulations that are based on the Medical Devices Directive, the 1993 Medical Devices Directive. OK, there’s been some amendments during that period of time, but that’s where we are, we’re still working to that Medical Devices Directive. There was a set of amendments, another set of amendments that came out on the 1st of January 2021, yeah, that’s right, 2021 and I don’t know whether any of you have actually had a look at them. One of the difficulties about the way we do law in the UK is that we have a statutory instrument and it gets amended from time to time and the amendment simply says, clause x of the previous amendment is amended to say this, and so on and it’s about 200 pages this particular amendment and they have explicitly said, we are not going to produce a consolidated copy, so it’s an absolute nightmare to try and work out what the current situation is, but essentially we’re still working to the Medical Devices Directive.
Just as a quick aside, pardon me if you don’t actually already realise this, but the difference between an EU directive and an EU regulation is significant. A directive is the European Union saying to Member States, here’s an issue and here’s some legislation that deals with that issue, take that and implement it in your own legal system. So, there is an opportunity for different interpretations between different Member States and a bit of wiggle room that Member States can implement in their own legislation. An EU regulation, on the other hand, applies across the EU as its written and officially translated into the various languages, but it's directly applicable, it doesn’t have to be brought into the legal system of Germany, or the UK as was when we were in there, or France, or whatever, it is a piece of law in its own right and so there’s no local interpretation and it’s much more consistent across the whole of the European Union.
There were very good reasons for revising the law on medical devices in the EU and actually for doing it as a regulation rather than a new directive. The UK actually played a pretty significant part in drafting and the development of the EU Medical Devices Regulation and a key proposal was that devices that are manufactured and used solely within the same health institution should be subject to full conformity assessment. That was what the original draft said. There was one exemption from that. You didn’t have to put a CE mark on the device and actually I’m afraid some of my colleagues hadn’t read it carefully enough and said, oh, it’s alright, it’s alright, we don’t have to CE mark it, and I had to say to people, listen, read it carefully, you don’t have to CE mark it, but there’s no exemption from anything else, full conformity assessment is what you’re going to have to do. So, a number of us went up in arms about that because we believed that would stifle innovation and there are other complications that it would do, but there are very good reasons sometimes for developing a one-off device to be used within a health institution. Modifying a surgical instrument for a particular surgeon, for example. We’ve done that on a number of occasions. That is manufacture, that would have had to have gone through the full conformity assessment process. So, we lobbied on that, we lobbied MHRA, but also through IPEM, through the Institute of Physics and Engineering in Medicine, we lobbied in Brussels as well and I went to Brussels and talked to MEPs and talked to a lady from the Commission and put the case that this was going too far. We were not saying there should be no regulation, as there is in the UK’s interpretation of the directive, we were not against the idea of having in-house devices regulated to an extent, but not having to go to full conformity assessment.
So, if we look at the next slide, so the change that that would have brought for health institutions was that as it stands a device has to be CE marked if it’s placed on the market or put into service and it was the definition of put into service in the directive that led to the UK taking the view that a device manufactured and used within the same health institution was not put into service, because that’s a defined term and it’s a defined term that talked about made available for the first time on the Union market. Well, if it’s being made in my clinical engineering department and used in my Trust’s health authority it’s not being made available on the Union market and that was the legal view that the MHRA took and therefore under the directive in the UK the interpretation was that there was no regulation of in-house manufactured and used devices. They were not covered. It wasn’t just that they were exempt, they were just not covered by the directive. The lady from the Commission that we talked to about this made it very clear that the Commission did not agree with the UK’s interpretation of the directive and you have to be tactful in these situations, but I felt like saying to her, listen lady, read the words carefully, but one is polite in those situations.
So, the new EU Medical Devices Regulation, and bear in mind that that’s not available for interpretation, it’s as written and I won’t go into the negotiations and the discussion we had about this, but the outcome was that the Medical Devices Regulation explicitly in words says, devices that are manufactured and used within a single institution are considered to be put into service, so that was definite and it then went on to say, as a result of the negotiations and the input that various people made, it then goes on to say, however, provided that this, this, this and this happen, I’m just trying to think of exactly how it’s worded, but essentially it says, full conformity assessment. That’s right, it says, are exempt from the rest of the regulation, but you’ve got some conditions to do, and MHRA have called this the health institution exemption. It’s their term, it’s not actually in the regulation and they produced draft guidance for this health institution exemption. We got to the stage, a few of us from IPEM, representing IPEM, put a lot of input into that and we got to the stage where it was in a final draft version, which we were shown the final draft version, but it was embargoed so we couldn’t distribute it round, and then all of a sudden the bombshell hit, the Commission said, we’re postponing the date of full application, so the regulation was clearly not going to come into force, sorry, not going to become fully applicable before the end of this transition period and so everything went up in smoke shall we say.
However, a few weeks later they produced, almost word for word, a few minor changes, but they actually did issue this guidance for Northern Ireland, because of course Northern Ireland are having to implement the EU regulations, so none of it now applies in GB, we’re back in effect with the old Medical Devices Directive, we can do what we like, it’s unregulated, there’s no regulations, no legal regulations at all, but I don’t think that’ll last. The Medical Devices (Amendment) (EU Exit) Regulations 2020, big title there, applicable from the 1st of January, they’re a holding operation, they’re explicit, the notes that went with them explicitly said it’s a holding operation that the MHRA, well, the Department of Health and Social Care and the MHRA are working on new regulations for the UK, which I predict will say, well, for Northern Ireland it’s the EU MDR, but for GB we don’t know and we were told that there was likely to be a draft out by the end of June, well, they’ve missed that. Consultation was promised, explicitly promised to IPEM and then we discovered that there had been some informal consultation and we hadn’t been involved, so it’s not clear what’s going on.
Personal view, and maybe I’m displaying my other views that I have, but my personal view is that I think they’re probably under political pressure, I think the people drafting this are under political pressure to make sure that it doesn’t look like an EU regulation, that it’s as far removed from looking like an EU regulation as possible. I may be wrong, I may be being cynical, but I’ve got a bit more cynical over the last few years I have to say, but I very strongly believe that in-house manufacturing use will not remain regulation free. I think it would be very strange if that happened, it was recognised as a loophole, it’s not the right thing to happen, frankly, and if you put that in the context of the Cumberlege Report and the whole issue of patient safety and so on, I really do not think it is conceivable that we will get a situation in which the existing unregulated development and putting into use of devices will continue. I also think that, and hope, that we won’t go to a situation where full conformity assessment is required, because we’ll have to start the lobbying all over again and I think we had convinced, we’d certainly convinced MHRA that that was not a sensible approach and we’d convinced MEPs actually.
Just as an aside, the way EU legislation works is that the European Parliament appoints what they call a rapporteur for a new piece of legislation and we were hugely fortunate in this particular instance that the person that the European Parliament had appointed was a UK MEP called Glenis Willmott, who’s Nottingham East or something. So, when we got the opportunity to go and talk to her we didn’t have to start explaining the NHS to her and how it worked and so on, she was well on board straightaway and we made a convincing case that there needed to be this exemption from full conformity assessment and then she asked me to go back a few weeks later because she’d drawn together a group of MEPs from the healthcare leads from the different parties in the Parliament and so I had the opportunity of making the case again to a wider group of MEPs, which was very useful. And the Parliament, in their comments on the draft legislation, went along with the approach of having an exemption provided certain conditions were met and these, in a sense, are the conditions that you have to meet or you would have had to meet and I think, and again I apologise but I’m speculating, had it been a year ago I would have been able to say, this is it, this is what you have to do, but the conditions are that the devices are not transferred to another legal entity. Well, that’s placing on the market, there’s no question of that and one of the biggest discussions around this has been, what about software, what about sharing software between one institution and another, and there may be possibly some ways, some wiggle room for that, but transferring of a physical device, saying, here you are, we’ve made it, you use it, is placing on the market and so that’s the first proviso.
The next one was that manufacture and use occur under appropriate quality management systems. Let me just, a quick show of hands, how many of your departments actually work under a formal quality management system? Goodly number. That’s good. Now, next question. How many of you are using ISO 13485? Many less. OK. Bear that in mind. I think, depending on what you do, 13485, and 13485 is based on the previous version of ISO 9001, and also if you look at the scope of it it’s perfectly reasonable to run the clinical engineering services under a 13485 system rather than an ISO 9001 system. Provision of technical support and so on is included in the scope of 13485 and we’ve been doing quite a lot of work in Wales in converting departments that have got ISO 9001, they’re converting across to 13485 for these sort of reasons.
The third point was you need to justify not using an available CE marked device. One of the discussions around that has been, is cost a justification? I think it can be, but not just on, I mean supposing what you need is a single parameter monitoring device for a particular purpose but the only way you can buy it is actually in a multiparameter monitor that you’re only going to use one fifth of it, that might be justified. Information is provided on request to the MHRA. The health institution appoints an appropriate person responsible for regulatory compliance, because you see not all of this sort of work goes on within clinical engineering medical physics departments, as you’ll see in a minute, so that there needs to be somebody overall within the organisation that understands the technology, understands the legislation and can sign off on the next point, no, the next but one point, making available a declaration of the fact that the organisation has put into use in-house manufactured and used devices, and in my view that represents a really good opportunity for clinical engineers to take on that role of being the appropriate person to sign these things off. Point number six is really important, design and risk management is documented in a technical file, and then point number eight, reviewing experience and correcting if necessary, so you’ve got that whole cycle and you can see how that fits into a quality management system and into the whole issue of post-deployment reporting and monitoring and so on.
Right OK, well, what counts as manufacturing? It can be quite broad really. So, putting together a medical device from raw materials or component parts. That’s obvious, isn’t it? Completely rebuilding of an existing medical device, and linking that with the next one, making a new medical device from used devices. I don’t know how often that might be done. It’s the sort of thing that might happen perhaps in rehab engineering with wheelchairs and there’s bound to be a bit of wiggle room, certainly in the second of those as to, well, when does repair become manufacture, and it is important to realise that we are not talking about regulating repair under this, we’re talking about creating a new asset, if you like, and fully refurbishing. That’s a defined term in the regulations and it does explicitly talk about giving a thing a new identity, so doing a really good job of refurbishing a wheelchair is not fully refurbishing unless you decide, well, this is now a new device and we’re going to give it a new asset number and so on. That might count. Developing software that meets the definition of a medical device. I could give a talk on issues around software. Thank you, John. Putting together combinations of medical devices or other equipment. Significant deviations from instructions for use that alter the function, performance or purpose of a medical device. That needs some careful thinking about. The word significant is important there. Using an existing medical device for a different purpose from that intended by the manufacturer or modifying a device for a new purpose, function or performance. What I haven’t listed there is the issue that occasionally happens of using a non-medical device for a medical purpose, that needs to be added to that list as well, and then the bottom condition, where that action is not explicit in the manufacturer’s intended purpose.
Which departments are likely to be affected? Well, when we started, I ended up nearly two years ago doing some work for the Welsh government and NHS Wales and this is when it looked like it was all going to come into being and we started realising it was much wider than just medical physician clinical engineering, so you’ve got wheelchair and seating services, orthotic and prosthetic services, podiatry, occupational therapy, max fax, labs, medical physician clinical engineering, oncology sometimes, particularly with software stuff, dental labs, CSSD, HDU departments, endoscopy departments, pathology may be under a different bit of the regulation, possibly information technology departments if they’re developing apps that actually meet the definition of a medical device, and then the whole issue of departments and individuals developing software that meets the definition.
One of the problems about software is an awful lot of people think they can write software, you know, doctors writing apps and that sort of thing and some of them think they can do electronics as well, but you usually open the box up and you can tell it’s been done by an amateur. You can’t necessarily do that with software, unless it’s examined thoroughly by someone who really understands software it’s not as explicitly easy to spot. So, if what we get as regulations for England, Wales and Scotland, if what we get is a set of, there’s likely to be, I think, and again this now is speculation, there’s likely to be a set of general safety and performance requirements as there is in the Medical Devices Regulation, there’s likely to be a set of rules to classify the risk level, there’s likely to be a requirement for a clinical evaluation of new devices and there is likely to be requirements for post-market follow-up.
Just very quickly then, minimal guidance from MHRA is there. Look at your quality of management system and think in terms of updating to the 13485. IPEM have very recently published a best practice guidance for the in-house manufacture of medical devices and non-medical devices, including software in both cases, for use within the same health institution and that is available as a free download on the relative bit of the IPEM webpage. Let’s skip that because I’ve said a bit about that. Can I go on a little bit, John? Two minutes. Right OK.
Repair and maintenance of a device in line with the manufacturer’s instruction is not considered a modification of manufacturing. Health institutions which follow manufacturer’s instructions in providing a repair and maintenance service, this was in particular in relation to the idea of providing a repair and maintenance service to another organisation, that’s not manufacturing and they don’t need to apply the requirements of the exemption, but health institutions which provide a repair and maintenance service which do not follow manufacturer’s instructions are potentially modifying the device and would therefore need to apply the requirements. That’s contentious in my view, because the whole issue of - let me get this to move on - skip that one. John, are these slides going to be available for people?
PRESENTER
Yes.
JUSTIN MCCARTHY
Yes. That’s fine, OK, so you might want to have a look at some of these later. The whole issue of alternative maintenance provisions is now well understood as being acceptable and so on, even in the most litigious country in the world, the United States. Research devices, a research device usually does not have a strict medical purpose, but if you get to the stage where you’ve done the proof of concept and you all of a sudden think, actually this might have commercial possibilities, then the regulations, what I’m suggesting are likely to be the regulations will kick in.
Justin McCarthy's presentation at the EBME Expo The Impact of the New Medical Device Regulations on NHS Services
