Top 10 Health Technology Hazards to Watch

Paul Anderson: Welcome to Smart Healthcare Safety from ECRI, the most trusted voice in healthcare committed to advancing effective evidence- based care. I'm your host, Paul Anderson, and over the past 12 years, I've overseen our patient's safety, risk and quality membership programs here at ECRI. Tens of thousands of healthcare leaders rely on us as an independent, trusted authority to improve the safety, quality and cost- effectiveness of care across all healthcare settings worldwide. You can learn more about our unique capabilities to improve outcomes at www. ecri. org. We are recording this podcast from our respective home offices as we practice and encourage all of you to practice good social distancing to help limit the spread of COVID. Today, we're talking about ECRI's 2021 list of the top 10 health technology hazards, which was just released and is available to all of our members. It's intended to identify potential sources of danger that we believe warrant the greatest attention for the coming year, and offer practical recommendations for reducing the risks. We'll talk to the author of the report about how the list is generated, how we intend for organizations to use it, and how this year differs from past versions. To get us started, I'll ask our guest to introduce himself.

Rob Schluth: Hi, Paul. My name is Rob Schluth. I'm a principal project officer here in ECRI's device evaluation group. That's our engineering team. And one of my key responsibilities is managing the development of this list every year.

Paul Anderson: Rob, when we think about the top 10 list, obviously there are a lot more hazards than just 10 that a healthcare organization faces in a given year. How do we decide which ones we're going to consider for inclusion?

Rob Schluth: Sure. Well, first of all, just thinking about the scope of what we're trying to accomplish with this list, the word hazard just maybe define that off the top. So we think of that as a device or a system fault design feature or method of use that under certain circumstances could place patients or users at risk. So we kind of take a broad definition of what that is. The word hazard might connote imminent danger, but that isn't always what might be happening. Sometimes it really could just be a health technology management headache that if you don't deal with it, it could end up causing someone harm down the road. So we do think about it kind of broadly defined that way. And then beyond that, because the list is generated here in our engineering group, we want to make sure the topics are technology related. So it has to be something to do with a medical technology that is going to interact with a patient. And then beyond that, it needs to be generic. So one of the services we have here at ECRI is we publish alerts when there are problems related to specific models of devices. Those sorts of things, we're not going to consider for this list. So it's got to meet those criteria to hit the scope of this report. And then beyond that, when we're kind of figuring out which sort of topics to cover, we think about the impact that they can have. So how can we accomplish the most good with the topics that we discuss? So we'll consider factors like the severity of harm. Is it something that could really hurt somebody or kill somebody? That's going to be a reason why we really want to think about it, but that isn't the only reason. Maybe it has less of a harm potential, but it happens quite frequently, or it happens at a lot of different facilities. Those are sorts of things that will go into our analysis, or it may just be something that's going to get a lot of attention. Hospitals don't want to be on the front page of the local newspaper for harming somebody. So if this is the sort of thing that is starting to get a lot of attention out there in the media or from an accrediting agency or something like that, that's another reason we might consider a topic for this list as opposed to other inaudible.

Paul Anderson: So if we've got a laundry list potentially of things that we're considering for inclusion in the list, how do we narrow it down from... I mean, I've seen the draft list, right? It's many, many more than 10. How do we get it down to 10?

Rob Schluth: Yeah. Excellent. Well, the way we start out, it very much is an iterative process. So we start by soliciting nominations, mainly from ECRI staff. That's several hundred people involved in different areas of healthcare and healthcare management. So we get feedback from the experts that we deal with every day. What are they seeing? What are they reading? Our engineers, what did they find when they were testing in the lab? When they were talking to someone at a hospital, what are they experiencing? So we start to collect all those ideas. And this year, we ended up with 70 of them that fit our criteria for this list. And then from there, we go to a triage process. So we cut the list down to 20 is the number we shoot for. We think that's kind of a manageable number. After that, we take those 20 topics and we submit it to an organization- wide voting process. So we'll have everyone in the organization who is dealing with healthcare management and technology one way or the other, or risk or patient safety, vote on the 20 topics that we've considered, then take the results from that voting process and go to yet another selection process. So we'll cut the list down again. And there, we have a committee. I think it was about 50 people. So it's everyone from the CEO down to me, essentially, on that committee determining what topics we think are worth developing. And what we try to do is, at that point, get 12 of them. So then we had 12 topics that we'll develop fully through to the end. And then we get to the very end of the process and topic refinement, I call it, and draft writing and review and editing so we know exactly what we have. Then we get that same committee together again and have what's called a ranking meeting. There, we'll vote again, pick the 10 that are, we think, most impactful to be on our list, and then rank them from one to 10. And then the final stage is sending it out to the public and to our members.

Paul Anderson: That's a really, I mean, you mentioned iterative, but it's also a really cross- disciplinary, multidisciplinary process.

Rob Schluth: Oh, it is. Absolutely. And that is I think key to what we're able to do here. One thing I try to communicate to people is that this list, it's not like an accounting exercise. We're not just counting up the number of problem reports and saying," Okay, well, this one has the most, therefore it's at the top of our list." We really want this to be predictive, a forward- looking list, as opposed to retrospective. And the way we get there is by relying on the perspectives of a broad swath of people who are talking to all kinds of other people. So really just starting to get perspectives from all different fields and aspects of healthcare, and then just boiling it down through the process.

Paul Anderson: Rob, I introduced you as being the author of the top 10 tech hazards report, but we just talked about how this is not a solo effort, right? There's this multi cross- disciplinary group that's involved. What kinds of expertise do we tap into, not just during refining the list, but really developing the topics?

Rob Schluth: Right. Well, first of all, let me throw in there with the author, we actually have individual product leads for each of the topics and they end up writing stuff. I work with each of them in developing the content, so author is not really going to be the best term. I don't want to cheat them of their amazing work. So normally what we'll do is assign each topic to a project lead. Often, it'll be an engineer or a scientist within our device evaluation group. And they'll do all the research and draw on their expertise to pull information together, and we'll work together to get our draft. And then we go through a review process, again, borrowing all the expertise within ECRI. So we'll have patient safety folks and risk managers and scientists, radiologists, nurses. Whatever expertise we think the top of warrants, we'll get those people involved and they'll review the content we have. So just like we're narrowing the topics down, we'll start getting the content into the shape where it's covering the topics or covering the issues that we think need to be covered and developing useful recommendations for the people going forward.

Paul Anderson: One of the things that really differentiates I think ECRI's list like this from ones I see from other organizations and other publishers, is that we don't just list out 10 hazards, right? I've seen some of them where the whole list, the whole top 10 is about 500 words, but we really focus on providing recommendations for what somebody can do about it. How do we identify those recommendations?

Rob Schluth: Yeah. Very good question. You're exactly right in that what we want this list to be is a practical tool. We don't want it to be just a nice little," Hey, okay, here, I spent 10 minutes reading this list of 10 items." There's really a lot behind it. And the reason for that is because we want, well, going back to why we do this list in the first place, we're trying to prevent patient harm. We're trying to help people who are interacting with patients and technologies solve problems before they occur so that there aren't future incidents. To do that, we need to give them recommendations. We don't want to just say," Here's a problem, good luck. We want to say," Here's a problem. We talked to a bunch of people. We did a lot of research. We did some testing. Here are recommendations that we think can reduce the risk of that happening at your facility." And so part of this gets to how the list was developed in the first place. It's an outgrowth of two of our programs here at ECRI. One is our device evaluation program where we do the lab testing on medical devices and rate them comparatively, and the other is our problem reporting network, where we receive problem reports from healthcare facilities or device manufacturers or wherever and log them into our system. We track them. If we think there's a chance it could recur, it might be something we would investigate and publish articles on. So just through that history, we would tend to see the same sorts of things cropping up over and over again, and we'd get a little frustrated. We published about this two years ago, why is this still happening? So this list was developed as kind of a way for us to hit at some of those bigger issue type things that might repeat. And in recent years, it's done a little bit more. Also, these are things that are maybe under the radar or are coming down the pike that," Keep your eyes open. You might start seeing this," or," Set yourself up this way so you don't start running into those problems." So the work we do in testing devices and researching incidents that occur and things like that help us generate the knowledge to develop these recommendations. And part of that is talking with hospitals and device manufacturers and reading the literature and things like that.

Paul Anderson: Well, it gets back to the idea you mentioned earlier, right? Of being forward- looking rather than just recounting. I liked the way you said it. It's not just an accounting of what happened last year, it's what you might run into this coming year.

Rob Schluth: Exactly. Yeah.

Paul Anderson: So you may have heard this, I'm not sure, but there was a lot going on in 2020 that was really, I mean, obviously unlike anything we've experienced in our lifetimes, certainly in the history of doing the tech hazards list. How did 2020 shape the list differently?

Rob Schluth: Sure. It definitely had a big impact. The most obvious way is just timing. The way our process normally works, in April, we're starting to collect topics, and June, July, we're starting to narrow things down. And September, we're finishing things up, and then we'll publish usually in the October timeframe in a normal year. This year, as we were getting ready to start the process, I mean, we were in the middle of a spike, right? And then later in the summer, there was another spike. So parts of the year when we normally would be dealing with key components of this list, we had to go all hands on deck dealing with COVID topics. We needed to get guidance out to the healthcare community about how to deal with some of the things that they were encountering. So we go into emergency mode, we put this on the back burner, but kind of keep it rolling through, but changed our schedule. So this year it came out in January instead of in the fall like it normally would. A second aspect is just kind of the ethos of the work we do. It was just a sense that I got this year while working on this, you could just kind of tell everybody just wanted to be engaged and move this project forward and put together good information that could help people on the front lines. We see, I mean, most of the world sees, but maybe we see more than a lot of people, just the work that healthcare workers are doing. And it's not just the doctors and the nurses. I mean, they're dealing with incredible stuff there, but materials managers, all the supply chain stuff, we deal with those guys and we know what they're going through. The clinical engineers, just administrators, everyone in the hospital really had to upend their processes on the fly to deal with a changing reality. So we kind of wanted to make sure that we were doing them justice by putting together something really useful and relevant to them. And then the final way I'd mention is maybe the most obvious, just the topics themselves. That was going to color what we do. We did have some concern going through the process that is the list is going to be all COVID stuff. And then," Well, hopefully COVID is eventually over and then this is not really relevant." So that was a concern we had. It turns out not to have happened that way, but what was kind of interesting was COVID made some topics relevant that might have slipped under the radar. It was kind of interesting the way that developed. So things that already existed or were likely coming down the pike just became more worth looking into because of COVID.

Paul Anderson: What are some examples?

Rob Schluth: Sure. Well, there were a couple of topics that were on the list this year that probably were just pretty specific to COVID. So I'll deal with those first, then I'll get into some other ones. So one topic we looked at, the N95 respirators. So those are the masks that we all read a lot about, so the personal protective equipment, the PPE shortages that healthcare facilities are dealing with. One approach to dealing with the lack of those respirators was to open up... So FDA issued an emergency use authorization which allowed respirators manufactured in other locations or that otherwise didn't meet normal US standards to be used in an emergency situation within a hospital. So one example are these KN95 respirators that are made in China. Those supposedly are tested according to a very similar standard to the N95s that NIOSH approves, so The National Institute for Occupational Safety and Health, if I'm getting that right.

Paul Anderson: Mm-hmm(affirmative). Yeah.

Rob Schluth: Thank you. So hospitals started using those KN95 masks, but there was some concern. Are these good enough? They're different from the ones we normally have. Some hospitals we're noticing," I bought two of these and they look different from one another. What's going on?" So they asked us for input. We actually went about the process of configuring ourselves to do testing of those masks. We had to buy a special device, the same equipment that NIOSH uses, and do some testing. And we did find some alarming things. So at the end of the year, I think the number was maybe 60% of the imported KN95 models that we tested did not meet the 95% titrate filtration efficiency that it was supposed to. And so what inaudible happening, some of the companies making these used to be toy manufacturers or sporting goods companies or whatever, and they just quick jumped into the market to make this because the world needs this stuff, right? But hospitals here in the US were a little concerned, and it turns out those concerns were reasonably well- founded. So one of the things that we mentioned in that report is these KN95s, that you really don't want to rely on some of those unless they've been tested if you're dealing with a COVID positive patient during an aerosol generating procedure, but if you have these masks, they still may be better than a surgical mask. So if you would normally wear a surgical mask for something else, maybe this would work. So it's not a total loss, they just have to be careful. So that was one topic that was very COVID specific I think. Another one was the remote operation of medical devices. So say like an infusion pump might normally be at the patient's bedside. Now, hospitals started moving them out into the hallway. So a nurse didn't have to gown up and go in and check the pump, they could kind of do it from the hallway, but then that changes some aspects of how the device functions, so the issues related to that. Same thing with ventilators. We were starting to see ventilator controls might be in the hallway and the ventilator near the patient. That raises some issues. You can't see the patient as well when you're interacting with the machine. So there were some hazards related to that. So those are two probably COVID specific ones, but then there were a few other topics that, like I mentioned, these are things that already existed, but they took on new relevance in this environment of unprecedented stress and rapid technological change. So the number one topic on our list this year deals with FDA's emergency use authorizations, and hospitals now have a lot of these devices in their inventory. These EUAs, as they're called, have existed for a long time, so hospitals maybe are used to dealing with it, but it's not the scale they're dealing with it now. And what can happen is once the emergency ends, hospitals need to know what to do with that equipment. They need to have a process set up. So it's just a different scale to what they're used to dealing with. Other things are telehealth. I've had experience with that. You maybe have.

Paul Anderson: Yeah.

Rob Schluth: That's obviously grown a lot in this past year, but the rapid implementation of those systems really needs to be looked at now once we start getting a little bit more time, and maybe you very quickly rolled out a technology that's not going to serve your needs long- term, you might've cut some corners, not done some analysis processes in picking your technologies or implementing things. You're going to want to go back and do those steps, or you're going to want to make sure you train your users how to use it, or maybe you need to switch to a different system that better fits your workflow. You had to roll this out in a couple of week's time this past year, but you really don't want to build that for the future. Similarly, ultraviolet disinfection, you see a bit about that in the news sometimes. That's getting deployed in a lot of new areas where it hadn't been before and by users who never used it before. So that's a situation where it's just ripe for issues to occur. First of all, the ultraviolet radiation itself could be dangerous to a bystander if they're not well- protected and depending on the wavelength and things like that. But beyond that, it may not do what you think it's going to do. It may not disinfect the surface. So you may be leaving infectious organisms around if you're not doing it correctly. And then the final thing I'll mention is one of our topics deals with consumer grade medical devices. These aren't formally medical devices. They haven't been approved by FDA, haven't been through the normal processes. During the pandemic, those sorts of devices were getting used a little more frequently and sometimes medical decisions were being based off of those. So a home blood pressure cuff or pulse oximeter that you can get at your local pharmacy, it might be good enough for kind of keeping track of trends, but might not be the sort of thing a medical decision should be based off of. And in crisis situations, sometimes those were being used in critical care units to check a patient's blood glucose levels maybe, and that's not necessarily going to give you the best reading. So we talk about that in one of our topics as well.

Paul Anderson: I mean, I see so many of these being related to each other, right? You talked about the supply chain issues and that leads to the KN95s, that leads to using the consumer grade devices. The telehealth leads to me, right, trying to get my toddler on my bathroom scale to tell the doctor how much he weighs, which is just not going to be as sensitive or accurate as bringing him into the office. They're all related.

Rob Schluth: Yeah. Yeah, it's very interesting. And the fact that the information is relevant outside of a COVID world was what I found very interesting. These were trends that were developing, especially telehealth. That was happening and it was going to happen, it just rushed at everybody. And there could be consequences from that.

Paul Anderson: Something unique about this year's list I think is that every topic is new. And in past years, something like alarms safety or another topic might show up multiple years because it's one of those problems you sort of referenced, right? That these problems are not going away. But this year, everything's brand new. How do we get to that point? I mean, obviously I don't think it's because all of those old problems have been miraculously solved and we can just sort of wipe our hands of them now.

Rob Schluth: Yeah. Yeah. It's another excellent point to make all the time and a question that comes up every year. Just because a topic isn't on the list and it had been last year or the year before, it doesn't mean the problem's gone. A lot of these problems unfortunately they do persist. Now, we hope improvements have been made. Some facilities maybe have, but others haven't, and they inaudible need to hear, and some of them, so alarm safety is a great topic. It's complex and it gets into everything and everywhere. So it's not an easy topic to address. So that one in particular, that had been on the list, this is our 14th edition, it was on the 13 previous lists and was not on this list this year because there's so many other things to talk about. But with those old topics, sometimes what we'll do is deal with a new angle, give like a bite- size inaudible of information. So make an improvement this far this year, and then maybe down the road, there are more recommendations you can implement. But the way we get to the new topics this year, first of all, every year we start fresh. So we take last year's list and we throw it out. We'll throw those topics into the the 70 that we consider. So we'll also look at what was there last year and have that conversation, is this something worth including again, but the decision comes down to, and again, we're talking about impact, right, where can we do the most good? So sometimes that may be by repeating a topic, having it on the list a couple of years in a row. Talking about a different aspect of it might help move healthcare forward to really minimize the risks of that one. And eventually, you'll start to see some real changes to that. But other times, and this year proved to be the case, we get a greater impact by talking about something different, raising awareness about a new issue. And that's what we ended up with this year. So as we were discussing earlier, the pandemic brought a lot of new wishes to the forefront. So some of those we talked about already, and it wasn't everything though, right? So medication safety is a topic that we do broadly cover in a lot of our lists, and that was on the list this year too. So maybe I'm lying a little bit when I say everything is completely new, but getting now specific, the medication safety topic that we covered this year is unique to our list. We worked with our partners at the Institute for Safe Medication Practices on an issue that they've really been harping on for a while. Kind of like autocorrect, When you're texting something, you start typing in a couple of letters and a suggestion shows up, well, that happens with drug names and drug delivery systems, store systems, things like that. And there have been fatal instances of you type in a couple of letters, you select the wrong drug, and a patient is severely harmed. So we did cover medication safety, but in a different way. Cybersecurity, again, that's been on the list in recent years a lot. The new topic we covered this year was related to third- party software that's within a medical device. So the easiest example to think of is say a Windows operating system that might be used in some sort of device. With a medical device, you're not really interacting with that software, so you may not even know it's there. If that software has a vulnerability, there are issues for the manufacturers in terms of how to deal with that and how quickly they can help hospitals. So because of the practical realities of that, it can take a while to get information to the hospital. Hospitals need to be prepared to deal with that information in advance to make changes. And then there are issues that hospitals have to worry about when making those sort of patches or whatever to address that. So we hit those topics as well, but then there were some new areas that popped up on the list. Two really interesting ones were artificial intelligence, and the first time I really started taking a look at that on our list specifically related to diagnostic imaging. And that kind of had a COVID angle as well that kind of moved us in that direction because, this past year, AI has been used in imaging to help recognize lung lesions that might be indicative of COVID-19. So kind of had an angle, but we talk about it a different way. And 3D printing patient specific medical devices. That's the one topic on the list that we haven't talked yet, so I feel like I should mention that. It's kind of interesting. So maybe like a surgical guide, if you're getting a knee implant or something like that, to help the surgeon position things, or 3D models of the anatomy where physicians can practice a procedure before they do it, all kinds of uses of this kind of weird, in my mind, 3D printing stuff. So it's likely going to become bigger and bigger as this stuff gets used more often, but our list is starting to look in that direction.

Paul Anderson: Yeah, gosh, I could think of a half dozen lists of, I mean, even before 3D printing was really available to consumers, you'd always hear stories about physicians who are also tinkerers and just using... I mean, I remember it's even a couple of years ago now, but we talked about them using like community maker spaces to go fiddle with something. I mean, literally fiddle with something and then come into the hospital and start using it. And now, you have to worry about not only the safety, but if you cross the line into becoming a device manufacturer legally.

Rob Schluth: Yeah. And with these patients specific 3D printed devices, the surgeon or whomever is pretty closely involved or needs to be involved in the development of this. Usually it'll work you'll have some sort of say CT scan or something of the anatomy and then all kinds of crazy complicated software to build a 3D printed model of it, but the physician needs to be in that track the whole time looking at that to make sure the size is correct and it's going to match their particular patient. So yeah, this mesh of manufacturers and clinicians gets even closer with that competence. It's very interesting.

Paul Anderson: Yeah, that is. And then, I mean, to your point earlier, talking about forward- looking, right, this is where we're going to be looking. Yeah. So I always try to wrap up the podcast with this question, what's something our listeners can do today to start advancing safety in their organizations. And with regard to the tech hazards list, right, I don't anticipate that anybody's going to solve all 10 of these problems before dinnertime tonight, but where can they start?

Rob Schluth: Sure. It is tough. It's a big ask, 10 topics. Our executive brief, like you indicated earlier, it's a pretty succinct discussion of things, but then there's a full report behind it. So we might have a 200, 250 word summary of it in our executive brief and then a 3000 word report with detailed recommendations for how you're going to deal with it. So it'd just be a hard time reading it all by dinner, let alone solving all the problems. But I think a good way to start is to assign that executive brief as a reading. Schedule a meeting. Bring together various people from the organization, the clinical engineers, IT folks, administrators, medical people, nursing, nurse managers, a good representative from your organization, people who interact with medical technologies. Have them read through the list, get people together, and start looking at things that might apply. We know all 10 topics are not going to be relevant in every healthcare facility. Certainly we hope not, but that's been the reality up until this point, but we do hope that maybe two or three would be worth tackling each year. So if you get your committee together, look through the list and see," Hey, which ones do we have a good handle on and which ones do we need to spend some time with?" And from there, you can assign staff to go look at the full report and get recommendations and start implementing measures. But I think just looking through the list and assessing it in your environment is really the most useful way to use this information.

Paul Anderson: Sounds great. Rob, thanks so much for joining us.

Rob Schluth: It's been great. I appreciate it.

Paul Anderson: You can register to download the report's executive brief on ECRI'S website, and members of all ECRI programs can see the full report and its associated solutions kit by logging in. Be sure to subscribe to Smart Healthcare Safety on Spotify, iTunes, Google Play, or wherever you get your podcasts to get our latest episodes. We welcome your feedback. Visit us at ecri. org or email us at ecri- podcasts @ ecri. org.