Transcript: Patient Safety and Quality Improvement

Transcript

TRIONA HENDERSON: I'm going to go ahead and get started. I know people are still coming on, but we can get the introductions out of the way. Good afternoon. My name is Triona Henderson, and I'm a clinical pathologist and the facilitator of this ECHO Model™ pilot project. I extend a warm welcome to you from the Division of Laboratory Systems at the Centers for Disease Control and Prevention in Atlanta, Georgia. Thank you for joining us.

So this topic is an interactive web discussion, the title Patient Safety and Quality Improvement. Our subject matter expert today is Dr. Yael Heher, who will present today's topic. She is an anatomic pathologist, and currently the Director of Quality and Safety for both A.P. and C.P. in the Department of Pathology at the Beth Israel Deaconess Medical Center and Harvard Medical Center in Boston, Massachusetts.

So a couple technical points for you. Before introductions and the presentation, here are some technical details. So if possible– I see a few people just on phone lines. If you either can put your name in there or, if you are speaking, introduce yourself with your name. All of your microphones are now muted. During this discussion, please unmute yourself to speak.

If you're connecting by phone only, please announce yourself by name and organization or institution to begin speaking. If you're experiencing technical difficulties during this session, please send a private message to Johanzynn Gatewood, labeled as DLS ECHO tech. And she's waving to you right now. She will do her best to respond to your issue.

And finally, we design relevant sessions based on your evaluation responses. We have two options. For those who require P.A.C.E.® credits, you fill out the survey for P.A.C.E.® credits. And for those who just want a certificate of participation, you can just fill those questions out, as well. We actually use this information to improve our sessions as we go on. If you have any additional comments, please send a private chat to Johanzynn, or you can email all of us directly at dlsecho@cdc.gov.

So how are these ECHO sessions different than most of the webinars we've been attending, and phone calls, since a lot of us are teleworking? These are different because they're expert lecturer teleconferences that discuss cases. And that's really the main feature of it, with a full discussion amongst the participants and the subject matter expert.

The subject matter expert hopes to share some of their work that may be translatable to all of you in your individual laboratories, especially the community practice laboratories. These ECHO sessions will focus exclusively on clinical laboratories in the United States and U.S. territories. Once again, we value the discussion amongst all of you that ensues and encourage you to share your experiences and challenges on this topic. We thank you for your interest and participation.

Here is an overview of the process. So the subject matter expert will present their case. I may ask clarifying questions to the subject matter expert, and then we open the floor for ideas, discussion, and shared experiences with the subject matter expert. Then we'll have some closing comments, reminders about future sessions, and then adjourn. Today's session is being recorded. If you do not wish to be recorded, please disconnect now. Closed captioning will also be provided, and you can find a link in the Zoom chat box now.

The audio and transcript will be shared on the ECHO website, where you can download them freely. And we will also send slides for this presentation via email once we've completed to those who have registered. If you have participants with you who have not registered through TRAIN, but they're sitting with you and would like the slides, please send us an email so that you can register and receive these slides.

Now to the biography of our subject matter expert. So Dr. Heher received her medical training at McGill University, and completed her residency in anatomic pathology at the Montreal General Hospital, both in Montreal, Canada. She was the Robert T. McCluskey fellow in medical renal pathology at the Massachusetts General Hospital in Boston, and a graduate of the inaugural class of the Harvard Medical School fellowship in Quality and Patient Safety, a program designed to train a cadre of physician scholars prepared to lead operational improvement efforts within the Harvard system and across the nation.

She received her master's of public health from the Harvard School of Public Health, with a thesis aimed at health systems improvement. Dr. Heher oversees quality and patient safety programs for pathology and laboratory medicine for both Beth Israel Deaconess Medical Center system and Harvard Medical School. She chairs the APC PRODS workgroup, training residents in patient safety, mandated to design and implement a nationwide patient safety curriculum for pathology trainees. Her main academic focus is on the role quality improvement and patient safety in pathology, including performance assessment, transparency, effective change, and leadership. She has a special interest in pathology error disclosure and optimization of communication with patient and treating clinicians.

Her clinical service is in medical renal and gastrointestinal pathology at Beth Israel and Boston Children's Hospital. Dr. Heher has received numerous teaching awards from Harvard Medical students for her role in education, and in 2018 and 2019, she was named the Pathologist Power List by the Pathologist Magazine, an honor awarded to inspirational and influential professionals in laboratory medicine worldwide. Now, we'll invite Dr. Heher to begin her presentation. As she speaks, please remember or consider any similar situations that you've encountered that you may apply to her experience at your institution. Dr. Heher.

YAEL HEHER: Thank you, Dr. Henderson, for that incredibly over-the-top introduction. So today, I'm going to share with you a couple of incidents from my hospital– just making sure you can see this. Everybody good with that? And then I'm going to talk about some general quality and patient safety tools that we use, and frameworks at our hospital. And then we'll leave ample time for interaction at the end. So just about 30 minutes of speaking, and then we can have an ample Q&A section at the end. QA Q&A. That's what we'll have at the end.

So I have a friend– many of you who have heard me speak before talk about the two-minute test. And I totally believe in it, especially in the era of Zoom and cellphones, which is that people take two minutes to decide whether or not they're going to listen to you or not, and then they just sort of check their email or Instagram feed or whatever it is that their thing is. So let me tell you some real stories that I hope will horrify you enough that you'll pay attention.

The first is something that will be familiar to anyone who's ever worked in an anatomic pathology lab. And that's a prostate cancer mixup that we experienced in our lab about ten years ago. So we had this patient, Mr. NB. He had this biopsy. It just looked like routine prostate adenocarcinoma. We signed it out. And then, of course, we got the specimen, and it was benign. Completely benign. We submitted the entire thing. And that's about the time my phone starts to ring. I get a little chest pain.

We went back to the grossing log from that day, and actually found his real core biopsy, which was benign. And unfortunately, the patient, who was, I think, in his early 50s– quite young– had every possible bad outcome you can have from a radical prostatectomy. And the other patient, because of course, in A.P., it's a switch, it affects two patients, developed, interestingly, anxiety, and didn't want to proceed with the therapy that was needed, and some mistrust in the health care system, totally understandably.

There was a four-month delay between when the patients' had their diagnoses and intervention and when the error was disclosed to them, which sounds short, but it's actually a pretty long time if you think you're living with cancer or vice versa. We had a big malpractice suit. We had the Department of Public Health, the Board of Registration in Medicine, everyone visited us, which was very relaxing. And I say that with respect for the CDC and the CMS and government agencies. I'm a CAP inspector. But you never want someone to visit you after something like this. And of course, it was terrible for everyone, all of us. No one wants this to happen.

And actually, a couple of years ago, I read about a similar case in the New York Times. Exactly the same. And usually, when I lecture places, people have all experienced something like this, either with a G.I. biopsy, a breast, prostate, these high-volume biopsies that look very similar, and they're batched and often come from private offices. If you haven't had a case happen to you, then you've definitely had a near miss, where you cut deeper, and you realize there was a mislabeling.

Often around this time, people say we just need barcode labeling. All we need is barcodes. Barcodes is like manna from heaven. And I just wanted to bring your attention to this issue that actually happened to us when we published, which is, look, barcodes are stickers, right? And the stickers need to be stuck by humans onto containers.

So we had a very senior tech, who took 25 labels and just started one specimen off and had a frameshift error, and 25 patients were mislabeled. And we didn't realize it right away because our service is so big, all the specimens got disseminated amongst different pathologists, so they didn't understand the mixups. So barcode labels aren't perfect. You've got to superimpose I.T. systems on well-functioning systems. You can't just sort of put a Band-Aid on it.

So that's one case. Hopefully, your blood pressure isn't too high after the first two minutes. And the second case is interesting. So the first case was a pre-analytic problem, and the second case is a post-analytic problem. So you had this neck mass, basaloid cells. A 39-year-old who had a neck mass. It was actually a carotid mass, so maybe more like a cheek mass. And it was signed out totally appropriately, basaloid proliferation with stromal fragments. Fine. And by this, we mean pleomorphic adenoma, basal cell adenoma, or adenoid cystic carcinoma, right? But that's like inherent in what we're thinking, like telepathology to the clinician.

So the patient actually went back to his PCP, not to the surgeon who had just done the biopsy. And the PCP decided to wait and see if it grew because he thought it was probably benign from that report. And the patient came back with widely metastatic disease and died shortly after. Forty-five years old at that time, and a father of three kids. So they had a lymph node excised– sorry about my typo here, but it was adenoid cystic carcinoma, which of course we know in that list, right, but they didn't think of.

And I think it's interesting that for the first time– and I'll get to that in a moment– the communication, the appropriate communication of results is part of diagnostic error, right? And what is the role of pathologists as health care advocates? If the COVID crisis has taught us nothing, it's that pathologists have a major role in health care advocacy, and are central in the diagnostic process and in managing public health. So taking away our sort of like blinders as anatomic pathologists of like, oh, I hit release, and I'm done, or I write it for myself, or we write it for each other, to how do people interpret our tests.

So I want to bring your attention– the patient safety movement is pretty well advanced in other parts of medicine– hospital medicine, pediatrics, infection control, anesthesia, surgery, even, with checklists– but lab events are different. And I'll tell you why I think they're different. Number one, they're behind closed doors, and no one knows what the heck we're doing, right? So in order to explain to them what went wrong, you first have to explain to them what we even do.

Like, let's say you find a new lesion on deep resections when something goes out for a consult. You have to explain what a block is, what an unstained section is, what a consult is, what is this, what is that, why it made sense. I mean, it takes you the first half-hour of the disclosure to even talk about normal operations, right?

We have a production line workflow. So I think when people try to apply systems like lean or Toyota or workflow redesign, I think the lab is particularly amenable to these kinds of changes. It's very difficult to arrange patients in this way in the hospital, but specimens actually are quite amenable. We batch. We batch like crazy. Batch is our middle name. And that's very dangerous from a specimen handling perspective. Everyone knows, if you print out 40 labels, you're much more likely to pick up the wrong one, right?

Incredibly high volume. Millions and millions and millions of tests, right? So if you have one machine go down for an hour or a problem with one reagent, now, all of a sudden, you've affected 10,000 patients, right? So you've got to have some intense systems to manage that. As I said, no one knows what the heck we do, so that's a problem. And multiplicity of events affects so many people. In A.P., maybe two, maybe a few hundred. In CP, boom, thousands of patients, right? So we really need specialized systems.

And the other thing I probably should add to this slide is the incident reporting systems that are designed for hospitals have nothing to do with lab medicine, right? They have lab as a dropdown, and it'll like assign it to some operations person, but the risk manager who oversees this is used to dealing with medication error, delaying diagnosis, wrong-site surgery. They have no idea about lab systems. And so we are actually not– it's harder for us to report because it involves all kinds of explanations. So we need new systems.

So let me tell you a little bit about why I believe this is important. Why would I show you this ABC slide that everybody learns when they're 2 and 1/2 years old? Because I think we focus all of our training on analytic error. And guess what percentage of error is actually analytic, even though we obsess about it and it's like the main thing we care about? A very small percentage.

So pre-analytic is everything that happens before it gets to us. And anyone who manages or oversees a lab knows that this is the main root cause of most issues. Mislabeling, specimen lost, misprocessing, going to the wrong lab, inadequate, et cetera. Clinical information, lord knows that, once we get to systems and we can't access clinical information, we can't put things in context, it's very dangerous. And that goes for A.P.; it goes for blood bank, it goes for everybody.

Analytic, you know, for pathologists, A.P. doctors, we're like melanoma is the great masquerader, right? Of course, everyone wants to obsess about what they have control over, and it's very, very important that we get it right, but we don't focus our training on where most of the risk occurs. And the same thing with that basaloid case I told you. Look, we didn't do anything wrong, it's the right diagnosis, but the patient died. And could we have done something better? Could we have done something better? And turnaround time, of course, is a huge issue. Again, COVID crisis can teach us a lot about that in terms of patient management, cost, and just risk to public health.

So we look at all of our near misses that we can and our unsafe conditions, not just harmful events. Unsurprisingly, we and others have found that almost 90% of our events are pre-analytic. And this has been reported again and again in the literature. It's not just our group. And again, anyone who knows about that can relate.

The good news is we don't hurt that many people, right? Like, a minority of these events are harmful. So you're like, that's great. But actually, I think in a robust patient safety system, this is the kind of graph that you want. Like, in a primitive one, you're only reporting the very harmful cases because that's all you have bandwidth for. Like, you're definitely putting in a safety report for the person who died. That's for sure. But you have all of this really important information at the base of the pyramid, where you have opportunity to help and prevent error before you harm someone, right?

And I think that your pyramid should look like this. Like if you map out your events that are reported, you'd better have some stuff at the bottom, because otherwise, you are missing incredible opportunities for safety and quality that would cost you a lot less, and share the same root causes. And I'll give you an example. Like, if you've ever fish– I want sort of a raise of hands, usually, I ask for a raise of hands of who's ever fished a specimen out of the trash. I definitely have. OK.

So anyone find it? Yes. I've found it, and I've not found it. Both. So if you found it, what did you do? Most people say nothing. Like, we're thrilled, and we move on, right? And we're relieved, and we're sweaty, and we're, like, horrified, and we take a shower or whatever. But that's the same root cause as if you didn't find it. It's just a total fluke that you found it.

So you missed an opportunity to figure out how did it get in there, how was that able to happen. And that's a very primitive example, but there are many different examples of where we're just like, phew, thereby the grace of whoever you believe in, and then we move on. And that is an opportunity for root cause analysis that you missed if you don't submit your misses to your safety system.

So other examples, I'm just going to show these briefly, but if any of them feel like you know what they are, even just the word reference lab should send you into an immediate rash of, like, all the– and this is with love and respect for the CDC as our nation's number one reference lab in our latest– I mean, the interoperability, the interface, everything about ordering, resulting, pre-analytic, everything about it is a risk for patient safety.

I mean, it's just a nightmare. And then, complex specimens from I.R., from interventional radiology, are a big problem in most labs. They get sent either to A.P. or micro. One of them puts them all on an agar plate, the other one puts them all on formalin, and now you have a person with a bone lesion or a lung lesion with no diagnosis. And this happens all the time. All the time.

So when do you do root cause analysis? When do we tell people we did something, or we just fix it and we don't tell them? Who should tell them? Like oh, well, we'll have the treating clinician tell the patient because we don't know them, but they don't know anything about it. And so when do you do root cause analysis? And I have an algorithm for this, which we can discuss in the Q&A section if there's time.

But I wanted to give you a little perspective on the patient safety movement. I think everyone knows "to err is human." It's incessantly quoted in the literature. Incessantly. Like, I kind of want to get a tattoo that says 99,000– I'm kidding. Triona's laughing. But it was a seminal paper that actually, for the first time, talked about iatrogenic harm and error, right? And talking about systems, as opposed to blaming individuals.

And around the same time, there were a bunch of very high-profile deaths. Libby Zion with a teenager in New York who was misdiagnosed and died. 18 years old. Betsy Lehman was a Boston Globe reporter who received a wrong dose of chemotherapy at the Dana Farber and died as a result of cardiotoxicity. And Linda Kenney founded a patient safety advocacy group called MITSS. She received an arterial injection of an anesthetic agent instead of venous, and she arrested when she went in for hammer toe surgery, and she became an activist. Very, very amazing woman.

So what I'll bring your attention to is the latest IOM, now called the National Academy of Medicine report, which was basically written for us. It was written for our community, but very few people know about it, which is crazy. It was a call to action for pathologists and radiologists. And actually, there were two pathologists on the committee who then wrote a follow-up call to action literally called It's Our Turn. Still, still nothing. But if you're interested, I would read that. It's a CAP Journal article.

But I think what's different for us is we basically have a paradigm shift in the lab. One of the things is around risk management. Like, let's just avoid being sued. Like, let's just save money, right? OK, fine. But now, it's like, you know what, that's actually not enough. Like, let's try and do Q.I. and quality innovation.

And I think the other thing is around quality management. Like, I remember there being, like, lab management modules, and the quality management. And basically, it was CAP stuff or Joint Commission stuff or quality control, like instrumentation and the refrigerator's the right temperature, and there's like 20 specimens when you want to validate her two, and there's variance. And it's vitally important for lab operations, but that is not quality improvement, it's not patient safety, and it's not innovation.

So I think it's super important, it's vitally important to be compliant. And I, again, respect– I am a regional CAP inspector, I'm on the Quality Practices Committee, like deep respect for the law, right? And if you think about it as the criminal law of the land, obviously, it's deeply important to maintain structure. You've got to have it. But that's not how we live our lives. We don't go out every day thinking the minimal thing I need to do is not break the law, and then I'll be a great person. Right? That doesn't happen.

You have to do that, but then you also do aspirational things to try and make things better. And I think that's what process redesign is, that's what quality safety culture is, and that's what trying to make things better for treating clinicians, for patients, for family, and to improve value. And I think financial goals are going to be a major, major part of this, certainly in the post-COVID era when we're all trying to recover financially. And anyway, it was moving in that direction. This just accelerated it.

And I think one of the things that holds us back from making these changes is we already use the words quality and safety in the lab all the time. Like, we just say it all day long. We have a quality manager; we have a safety manager, we have this, we have that. They're not actually doing quality improvement or patient safety. What they're doing is environmental safety, formalin, OSHA. It's very important, certainly important in COVID.

They're doing occupational health and safety, they're doing fire department stuff, and they're doing Q.C. and Q.A., which are, again, very important. Nothing to do with quality innovation and quality improvement, and nothing to do with identifying safety events in a reactive and proactive manner. They're purely based on regulatory compliance, which is important, but it's not a way to live.

And I think the innovation in the lab has really focused, in A.P., on molecular, and in C.P., incredible innovation recently on COVID testing. Which, bringing up testing, just amazing. And that is a critically important part of our mission, but I think the major problem is that we still have this, right? Like, everything that leads up to these fancy new tests and these, like, 15-minute serologic point-of-care testing and the amazing molecular sequencing with targeted therapy for oncology, I mean, that's amazing, but the stuff that gets there literally still comes in a plastic bag with a piece of paper that's switched around, and formalin in the wrong Tupperware, and it gets fed.

I mean, the pre-analytic and post-analytic have not changed at all. And I think, as information systems become more standardized, but also as people don't have the money to go to Epic or don't have the money to go to some standardized interface system and there's system integration all over the country, we're going to have more and more interface problems around pre and post-analytics. And you can do the fanciest test on the planet. If it's on the wrong specimen or it gets results to the wrong person, you're done. It's meaningless.

So I think the vision for quality and patient safety needs to integrate operations, but we also need to study what works, and we need to educate not just physicians, but also holistically. And I think many of our organizations have done a wonderful job of including– the lab's a unique place, right? Like, we're not just a faculty community of doctors.

We have thousands of technologists, of lab assistants, of P.A.s, of all kinds of people who need to understand– you know, send out coordinators, transcriptionists– people who need to understand the safety culture and participate, because without their engagement, we're just a piece of the puzzle. I mean, a fancy piece, but– I'm kidding. I'm joking. Doctors are very self-important. Not of Harvard, of course. Never, never at Harvard.

So I want to tell you a little bit about, very briefly, in broad strokes, some of the programs that we have. And then we'll do a Q&A session at the end. So for those of you who know high reliability, it's a concept that is well-known at the hospital level. There are a variety of places who have implemented it. HRO– Brent James and Intermountain are sort of the leaders.

And there are some principles around high reliability you can read about, which are beyond the scope of this talk. But what I will say is we have to flip the model. Right? We have to go from regulatory compliance being the gold standard and, yeah, we need to do a good job in operations, and then if something comes up, we're going to whack-a-mole and then maybe a little Q.I., because we have to because of the ACGME and our technicians, to a culture of quality and safety.

Like, coming through with what can we do better in everything we do. A preoccupation with failure, a deference to expertise, and everything around HRO. And then we need standardized tools, root cause analysis, PDSA, whatever it is that we use. Change management. So we have psychological safety, Amy Edmondson from the Business School, making people at all levels the organization's ideas valuable. And then operations, clinical, and of course, compliance. But that is the base. It's the base; it's not the top.

So we use two different dashboards, one for performance and one for patient safety. I'm going to talk about just our dashboards in M&M. And then, I will tell you, we have very robust systems around quality and process redesign, and we have a disclosure program, but I won't talk about those too much. But you can feel free to ask about them, so I include them here.

So two dashboards. Very briefly, one is our quality performance. So before, we used to sit in a basement, basically, and just read out things we were already compliant with, which were hours of my life I will never get back, and I still remain resentful about. But now, we really try and use these HRO principles. Look, I want a preoccupation with failure. I want the signal on my dashboard to be red.

Now, god help the– or whoever you believe in– help the people who are so used to the 100% compliance because they break out in a rash when they see the red. But you need to shift the culture around. Let's improve things, right? We want to look at the things that we're already performing badly on and not the things we're already performing well on. Why would we look at those?

Now, obviously you need to look at everything that's externally mandated. We have about 70% of our metrics are internally mandated, meaning we ask the frontline people what keeps you up at night, what's doing poorly. We ask the treating clinicians, what do you wish was better? And we float things on and off of the dashboard. If you're interested in it, there's a paper about it in Cancer Cytopathology called "Measuring and Assuring Quality Improvement." And we have all the info around designing a dashboard.

So if I had to use one word for the quality movement and what you need to do, it would be curiosity. I actually want to get a tattoo that says curious– no, I'm kidding. That's now allowed in my company culture. But a t-shirt maybe that says curiosity, curiosity, curiosity. So snap judgment is very human, and it's totally, you know, we want to fix things right away, so we think we understand them, and then we're done, right? But there is almost always a complicated and interesting explanation around systems of why something went wrong. And if you skip this step of trying to understand that from the bottom up, your solution's not going to work. It just won't.

So figure out why you're looking at a metric, right? Don't just do it because you're a box checker. Think about why is this important for patients for treating clinicians. Why are we looking at the target? How are people collecting it manually? Is it horrible? Should we float things on or off? Like the SOP that we just sign off on once a year, that's like very '80s. It needs to go away. It needs to be a fluid thing, where we're actually looking at performance in a dynamic and real-time way. Why are things going wrong, and what can we do to make things better? And I think this diminishes the feeling of like, oh my gosh, this is a pointless regulatory monkey situation, and we're actually helping patients and treating clinicians.

So that's a very brief foray into our quality dashboard. We use statistical process control. And I'm happy to talk in more detail about that in the Q&A period. The second kind of dashboard– and dashboard is just sort of like a buzzword. It just means visual representation of data, right, in sort of a holistic format. So you can call it whatever you want.

But our safety event dashboard looks at all safety events that are reported internal to the lab. So we look at near misses, unsafe conditions, everything at the bottom of that pyramid. I want a big bottom of the pyramid because otherwise, I feel like people aren't reporting things. And we look at details about it– what lab it occurs in, who it affected, who called it out.

And I think most important, from a cultural standpoint, is we not only monitor whether we closed the incident, did something about it, but we monitor whether we told the original person who called it out that we did something about it. Because often, we were finding, look, we would fix it, and then whoever called it out, like a med student or tech or somebody at the front line, we forgot to tell them what we did. And actually, it was making– if people feel like they're just calling things out into the darkness, they just stop, and you lose your opportunity for learning.

And I think one of the hardest things as a leader in quality is communicating back unfixable problems, either they're capital, or it's structurally difficult. And I think that's the sign of a true leader if you're able to actually value the front line enough to tell them, look, here are the current barriers. Like, we're going to try and find a way to work around it, or we're going to try and apply for whatever, but there's always limitations in life and in work. So I think the feeling of burnout and the feeling of frustration can occur if people feel like the silence means you don't care. Right? So you've got to communicate that clearly.

And that's borne out the same data that most errors are pre-analytic. This is all from our dashboard. This is where we got the data I showed you before, that we mostly don't hurt people, which is good, and that mostly there's at least two root causes, right? And we use the Eindhoven method for classification. You can use whatever method you want for classification. But I think, culturally, you should consider it's better to have more safety reports, right? So the metric of, like, we didn't have that many– like, we had zero events this month, yeah, no, you didn't. All right? Just nobody reported them. It's impossible. If humans are in a system, there were problems.

How easy is it? Like, if it takes people 20 minutes and they have to fill out 400 cells, it's not going to happen, right? So we have people send an email, and we do the RCA. Who respond, if anybody? What happens to the whiners? Do you like squish them out, or do you try and listen and sort of mine out the emotional information from the granule of importance at the center? Often, the people who are very steadfast and quiet have the same frustrations, but you don't have the opportunity to learn from them. So you have to be careful with that. And then, how do you assess risk and decide how to translate that into capital requests or budget requests or workflow redesign? And I already talked about unsolvable problems.

How are financial and efficiency concerns integrated and balanced? Like, do you do one at the expense of the other, or do you make them go together? This can't just be like a money eating operation. You have to also show your value. So be thoughtful about that. And value is also a quality goal.

So that's our two dashboards. The next thing I'm going to talk about very briefly in my last six minutes is around M&M rounds. So we instituted a novel pathology M&M round. And we did that because I noticed that we really loved going to other people's M&M rounds and telling them what they missed and then leaving. That was really fun. Like, going on rounds and saying like, ta-da, and then sort of walking off into the sunset. But we actually never publicly looked at our own errors in the lab.

So we instituted this about eight years ago. It's very similar to traditional M&Ms, except with a completely different twist. Instead of being sort of the old shame-and-blame surgical model that was very provider-centric, it's a systemic M&M that's focused on a robust root cause analysis, sort of a just culture, and a safety culture. So we look at RCA, we identify approaches altogether, we promote open discussion, transparent culture, and we focus on quality and safety.

The other thing that's amazing is everyone in the department comes. So the transcriptionists, the frontline people, the chairman. And I think that's vitally important because, often, we invite– I'll get to it in a minute, but we invite people outside the department, as well. We have a standardized model. We have a checklist. We have a conference guide that's been well-developed with feedback from presenters over the years. And there are differences between the traditional M&M and our path M&M. And it's really around systems learning and, again, curiosity, curiosity, curiosity. Like, not who is this idiot who made this mistake, and like, well, how did this really happen?

And I'm not saying that there aren't people who need coaching or there are never H.R. concerns, right? Everybody who's ever worked anywhere knows that, occasionally, there's a person who has repetitive bad behavior or bad performance, and it's a real concern. So that exists, and that kind of review is not in an M&M forum. So this is for systemic issues that recur, and that pass what we call the substitution test, meaning could a similar person with a similar training have made the same mistake in the same situation.

And the example I give is, look, if you had me at the checkout desk in histology all day long, trying to look at those sequential, like, tiny 800 numbers, like S-20, 1, 2, 3, 4, 5, S-20, 1, 2, 3, 5, 4, like, I definitely, definitely would do a terrible job, all right? Like, we're just setting people up for– and they match the profile, but it's like five little pieces of nothing. Like, how are you supposed to do that? So I have deep, deep respect for those folks, and I think they're set up for failure. So that's a big problem.

Everyone's heard of the Swiss cheese model like everything has to line up for it to go wrong. The example I give for that is we once– and I think I have a– yeah, I have a cause map for this. This actually happened. We gave someone with a brain met– or not a brain met a brain lesion, who had a history of lung adenocarcinoma a diagnosis of metastatic cancer when it was a benign lesion. What happened was, right before that patient, there was another person with lung adenocarcinoma. A resident cut a bunch of sections, didn't label them because they were, like, troubleshooting and trying to pick their favorite. Then they got distracted and stained only some of them.

The next resident came along, picked up a slide from the top of the microtome, thinking it was a new slide, but it actually had lung adenoma on it. Cut the brain on it, and then looked at the lung adenoma. That patient had a history of lung adenocarcinoma. Swiss cheese. Like, if it had been a uterus, everyone would have known right away something was wrong.

So we use just culture, which is the balancing of systems thinking and personal accountability. Set clear expectations for people, but also try not to set them up for failure. Like, don't leave on stained slides on the microfilm. Bad idea. We use either fishbones– this is another real case that happened in our institution, you're going to start thinking it's a terrible place– or cause maps. If you're interested in learning more about that, there's a short review [INAUDIBLE] called "A Brief Guide to Root Cause Analysis" that's free and downloadable. And we award risk management credit. We encourage frontline workers and trainees to present. We have this just culture.

And I will say, very, very important, inclusion of non-pathology stakeholders is key. So if we have an issue with a fetal autopsy, we include O.B., we include social work, we include all those folks because the thing about pathology is a lot of it is around ordering and handoffs and expectation and resulting. If you focus only on the part we have control over, it's never going to be fixed, right?

Or even the solution you design has all these downstream problems you don't even realize. So you really have to be comfortable with inviting people over to the family dinner table and talking about your family problems with guests. Like, just get used to it. It's OK. Everyone has problems, and you need their help, all right? So do it. We have this conference guide. It's very defined. Happy to share it. And it's been sort of changed over time.

I think, really importantly, process, focus on the process. Obviously, HIPAA. Try and be patient. People get upset when problems happen. They get worried. They're anxious. They think it could happen to them, and all kinds of emotions come up. But I think once you have this settled up, people understand the tone, and it gets better. So changing culture takes time, right? A long time.

So start small. If you want to do this, I'm happy to help. Reach out to me. I'm sure Dr. Henderson can share my email, or you can find me on Twitter or elsewhere. The residents have created an account for me and taught me how to use it. We have about 80 to 100 people. And actually, in the time of Zoom, we have a lot more, from all different places. And we livestream all over the place.

So I'm going to not talk about these last two things explicitly, but if you have questions, we can use it in the Q&A. I will show you this one brief diagram, which we developed for disclosure of pathology error. If you're interested in discussing that, I'm happy to talk about it in the discussion period. And if you want to read more about it, the reference is at the bottom. It was with Suzie Dintzis, who's a wonderful, wonderful person with lots of research experience around disclosure. And she's based in Seattle at the University of Washington.

So we also have a newsletter. Happy to share that. And my take-homes for you are we really have to redefine quality and safety to join patient safety. It's not Q.A. and Q.C. and environmental safety. It's quality improvement, process redesign, quality innovation, and patient safety. We have to expand our responsibility to all phases of testing. We don't just think, oh, it's nursing who has to label, or it's the I.T. people who have to result. Our community needs standardized tools that are specific for us, beyond what is just required for the criminal law, right? I'm kidding, but regulatory compliance. And there's an immense unexplored safety risk and value opportunity.

A broad vision includes an integration of operations, education, and scholarship. Often, operations is divorced from education and scholarship in academic medical centers, and it's dangerous. We, MD or PhD, are leaders. I think I might drop the M.D., and just any competent human who's interested should be trained on how to oversee and study in this emerging discipline. So with that, thank you. This is my team. And also, a plug for the next ECHO session that I will leave to Dr. Henderson.

TRIONA HENDERSON: Perfect. Thank you so much. I really do have to say, even before we get into the discussion, thank you for pushing quality improvement and patient safety in the laboratory. I also found you on Twitter way, way, way back in the beginning, and it really gave me the fuel to be able to allow the change culture. My first job was in a laboratory that was deep in the boondocks, totally community. I mean, chickens and pigs outside literally.

And having to make this change with– and I see some of the questions coming in– with staff who were scared that there would be retaliation because they reported that an error happened, even if it was a near miss and the patient wasn't affected or having your leadership and administration say, OK, maybe this is something that you should look into in your department. And so one of the first questions, which I can completely relate to, is how do you persuade your staff to report problems and barriers?

YAEL HEHER: Yes. I think there's a couple different things. It kind of depends what the barriers are, right? So here are some examples of barriers, but you'd have to go and ask them what are the barriers, right? So if the barriers are that it's too hard to report like you make me fill out 17 forms and I have no time, you already cut my staff in half, we're all on furlough because of COVID, and you want me to fill out a 17-page form when nothing even happened to the patient?

Ain't going to happen, right? Like, you have to make it really easy for them. And so the way you would design that is dependent on your system. That's why we just use emails, and we use Excel. I mean, not that I have any shares in Microsoft, but that's what we use because it's the easiest, cheapest, cheap and cheerful way to document errors. That's one barrier.

Another barrier is people feel like, well, you're not going to do anything about it anyway. Like, I reported one last year, and nobody ever got back to me. So you have to be very thoughtful about how you close the loop with people who call things out. Even if you can't fix something, you have to have the respect and dignity and sort of responsibility to give them some follow-up. That's what you would want if you called something out.

The other thing you can do is you can incentivize it. Like, some people have like a good catch or a shout out sort of thing, where they sort of– they have like a lunch once a month, where they have a shout– like Colleen, shout out for catching this issue, and here's how we fixed it. So validate. And then the other thing is, if there are barriers at the leadership level, like out of fear or out of an old culture, you need to dissect that out, for lack of a better term, and figure out what the actual concern is because all of this is peer-protected.

It's within Massachusetts, it's within laws of Q.A. review, and it's not discoverable. And so if that's a concern, you can clarify that. And make sure you do that in a format that's legally not concerning for your leadership. So like with anything, how do you persuade people to do things? Figure out why they don't want to do it and fix that.

TRIONA HENDERSON: And even to add to that, we modeled after a lot of what you said, the good catch, but also I even had the hospital Q.I. system and patient safety come into the lab. They're not used to interacting with technologists. Neither are the technologists. They didn't even know who these people were. And actually had a sit-down session to let them know that you are the frontline workers like you need to own this. And they were so empowered to believe that them reporting a near miss was so important to the system.

And so the first time– I think after the first two months of doing this, they got the good catch because they had 100 reported near misses, and they were like, oh my goodness. So they kept on reporting. And so it's really important to empower the people on the front line to be involved in these, and not just the top leadership.

If you're focusing– so I think I added that– if you're focusing on things that are not working, might the hospital leadership think that you have specific problems just in your area?

YAEL HEHER: Sorry. So I think that's a really common fear, right? Like, if I go on a date and I tell them about all my bad parts, they're never going to want to see me again, right? But it turns out, if you get married, like, people already know there are problems, right? So I would make a similar algorithm with the hospital. Like, the people who use your services, they have their frustrations. Like, they know when things get lost and when things get misprocessed. So I would encourage you, it's all about how you communicate what you're doing to the hospital leadership.

A lot of this work that I showed you today is around internal quality improvement. Like, I wouldn't bring the granularity of my safety event dashboard to the board, right? Like, what you want to bring to the board is that you have a safety event dashboard, and that your turnaround time for closing events is two weeks, and that you close 97% of your events, and that you did a survey and people are very satisfied, and that you've done Q.I. on 50% of the events.

So you need to– you know, crickets is not a sign of quality, right? Like, you need to show that you have robust systems, and you're addressing service, you're addressing satisfaction. I mean, it depends on who your audience is, but if you're thinking about hospital leadership from a quality perspective, like the CQO, then you can put things on your dashboard that matter to them, like how are you doing your C. diff testing? Is that affecting their CMS reimbursement? How are you looking at central line infections? Is the way you're doing reflex culture testing of urine affecting the CAUTI rate?

You can put things on your dashboard where you're partnering with them, and they will see what such an ally and they will come to you when there are issues they think the lab can intervene on. So I actually think that it's not about like a confession, where you just go in and tell them everything that's wrong. It's about you having robust systems to make things better before you harm patients.

And those systems can be shared from a schematic sort of like a way, where you give an overview of like, look, we're accredited. We're in good standing. There are 3,000 CAP checklists. We're accredited by ASHI. We're accredited by FDA. We're accredited by this, that, the other, Joint Commission, whatever you use. Or if you're international, whatever international system you use.

But also, we felt that wasn't enough. So we developed our own internal system to look for risk before it affects patient care. And here are all the events we found, and we fixed them, or we fixed 70%. I mean, that's pretty impressive, and I think that actually instills confidence in people, rather than concern. And if you think you're going to hide things from people by not talking about them, I think that's sort of a losing management strategy, and people who are interacting with you at a clinical level already know.

TRIONA HENDERSON: That's perfect. Another question, do you or your system incentivize identification of pre-analytical errors by individuals outside of the laboratory? And then how do you improve the quality of pre-analytical issues that may seem to be out of lab's direct control?

YAEL HEHER: Please hold. Hold on. Am I still sharing my screen?

TRIONA HENDERSON: Not yet, no.

YAEL HEHER: OK. I'm going to show you something. Is that OK?

TRIONA HENDERSON: Yes.

YAEL HEHER: Is that permitted? So we do that. We actually have– you know, it's a 30-minute talk, but we have dashboards that we created, and we share with frontline providers. And let me just show you this. I'll give you an example. We use Confluence. And I have, again, no shares in Confluence, it's just cheap, and we don't have a lot of money, and that's what we use. So let me just show you as an example collection errors, which is a major pre-analytic error.

Does that work? Does this work? Yes? OK. So this is an example of one of our dashboards that we use for collection errors. And we do it by– you know, we have all kinds of data here, but we classify the pre-analytic error here, QNS or are unlabeled. And then the other thing that we do to– I think the question was around engaging them– is we actually say where they're from if you see here.

So this is 36% came from ambulatory, 25% came from BCS, 10 from E.D. And then we actually tell them what floors did they come from. And all the floors input their information into here, and they can see how they compare to other floors. So here are the specific floors, right? And they can see where they are over time. And no one wants to be the worst. I'll tell you; it's like a human thing.

So this is a real-time dashboard we use. They pull information and input it directly. It's bi-directional, from the lab and from the front line. And it goes to nursing, in this case. And they like it. I mean, they like it, no one likes to see– but actually, they at least know where they stand. And they also get frustrated when they feel like things are QNS, and they want to know that we're tracking it, and we're not just rejecting things willy nilly.

Like, we're looking at why it was rejected, what the problem was, and we're trying to help them fix it, too. So that's what we do. Does that help answer your question for– we also have blood waste feedback to the OR when they don't send us stuff back timely, and we have to throw it away, or it's not on ice or whatever. Usually, after a massive transfusion, it's expensive and et cetera. So we feed all these things. Critical values, we feed back to providers like they didn't call back or whatever. So we have a lot of bi-directional dashboards for high-risk areas because you really do need to engage partners. Does that help?

TRIONA HENDERSON: Definitely. And even to add to that, we had– exactly, we'd share this information, and it did go to nursing, and even in the E.D., we had a blood culture contamination issue because there was no specific phlebotomy team and there was so much turnover. And you really empower the people, again, on the front line, this time being nursing and nursing education, that they didn't want to be the team on the dashboard that had the highest peak of errors and issues.

And so they had not necessarily a wall of shame, but it was where they displayed this information so that everybody in every shift would see how can we get these numbers down, how can we do better for our patients. And not just data thrown at them, but they really immersed themselves in the process. And that really improved the quality of the work they were doing. And then for leadership, way above just the regular line physicians, and it went up to the board, they were really impressed, and then highlighted this team that they had such a great overall improvement in the last quarter. So it's really about empowering and showing the data.

YAEL HEHER: And I think it's also– like to get back to a cultural element of breaking down silos, we have this tendency to be like, oh, the nurses never send enough– they don't draw enough blood. And then the nurses are like, well, they always reject our NICU specimens because– look, we can't even get more than 1 CC. Like, what do they think? The patient weighs a pound. And it's like this crazy thing where everybody wants the same thing, but they're fighting.

And so to get people actually in a room or to collaborate on something like this, where we actually understand each other's problems, and there are faces like everyone knows Yael from the lab is trying to help and, you know, Carrie from the blood bank and, you know, so and so, it becomes like all of a sudden it's a real person, and you're trying to solve problems instead of just getting really frustrated about somebody you don't know, even in a large system like ours. We have a lot of hospitals and a big, big network.

TRIONA HENDERSON: So another question, are your dashboards made from new cloth, from the ground up, or is it something that's commercially available? Since we have so many community laboratories on the line, how can they translate it? I know in our community, we just use Excel. We just copy–

YAEL HEHER: We just use Excel. We're also poor. But if you email me, I'm happy– we will share everything. And actually, we're working with one of the journal's archives, the CAP Journal, at the moment to find a way to publish an article about it, and then make the templates free and downloadable. So that's my goal, is to make the templates free and downloadable.

Obviously, how the dashboard is populated needs to be specific to your– like, ideally, it gets pulled from your LAS. The other option is people input things manually, but that's obviously dangerous and time-consuming. But if you have no other choice. But at least you'll get an idea of how we did it, what our targets are, and the metrics. So yeah, that's our goal.

TRIONA HENDERSON: So shifting a little bit, I know you've done a lot of work with residents in residency programs, and it's an ACGME requirement for them to have their PSQI project before graduation.

YAEL HEHER: Yeah.

TRIONA HENDERSON: How do you motivate them– so you and I naturally have a passion for Q.I. and patient safety. How do you motivate them beyond the checkbox of I've completed my Q.I. project for graduation as a pathologist, so that they can carry it over into their actual work life?

YAEL HEHER: Yeah. So this is what you're talking about, the new– can you see that still? The new Q.I. requirements of the ACGME, they're the milestone requirements. They went live in 2019. And there's a lot of different parts, not just around Q.I., but there are parts around root cause analysis, disclosure, safety culture, and et cetera. And I think what's interesting around– this is the new workgroup that you talked about at the beginning, Dr. Henderson, and that were working on an RCA tool, but I think what motivates humans?

Like, if you force people to do things because it's regulatory mandated, like anyone who takes those soul-sucking PowerPoints at the beginning of every year so that they can keep working knows that that is not the way to learn. And I'm, like, a very compliant person, and I still scroll through them while I'm doing something else. Don't tell. Don't tell them that.

So I think you can't mandate. People don't get excited about that. What you have to do is find something they care about. Right? And this is where the crossover occurs. Like, we do some collaboration with people at the B School or otherwise because it's around change management, and it's around leadership, and it's around getting people to do what you think is right.

And so with residents, I think there's a very unique mindset and a very unique experience that really, really, if you tap into properly, can greatly benefit the quality movement. They're very frontline, they don't have preset notions, and they are naturally– how can I put this– sort of self-centered in the way they look at the process, and free to criticize and free to– we seek their feedback often. So they often are kind of like canaries in the coal mine for identifying problems, in terms of like being stuck trying to follow up on a critical value or like being stuck trying to solve a mess that's a big quality problem.

So I think, in order to involve them, you really need to get them engaged in selecting the projects they're working on. So you don't just like hand them something, you ask them what they think is important and what bothers them. And they need some guidance, of course, because they don't always have the right perspective or the right context, and also sometimes they want to solve unsolvable things or boil the ocean. You have to make sure the scope is reasonable. But I think they have great ideas, and they notice things we don't because we're in a totally different part of the workflow.

So I think if you engage them in the actual event, and then I think the other thing is they need standardized tools. And that gets back to my first point of, like, you can't just tell them do Q.I. and then come back. You need to give them a template. I mean, they need some help. Even grown-ups need help. We're all learning how to disclose and how to do everything.

So I think the development and implementation of these tools, and then integrating feedback of actual trainees are going to be critical for the success and for the culture change that's going to have to occur in the next generation with regards to making quality and safety a part of everything we do, as opposed to just like a box we check or like one person in the department who does that.

TRIONA HENDERSON: Perfect. I have another question that will definitely pull on your years of wisdom and experience.

YAEL HEHER: Sorry, Triona, can I add one more thing? I also think–

TRIONA HENDERSON: Yeah.

YAEL HEHER: –it's really important if you're working in an academic lab center, which you probably are if you have trainees, to encourage your trainees to write up their Q.I. project. So design your Q.I. products in a way where you can publish them. And I just have some examples that they've done here. And I think that motivates them, too, even if it's a case report like this. That's fine.

It doesn't have to be, like, flyer quality improvement that they redesign the whole universe. I think it can be really important. And there are journals that are really accepting things in pathology. And we track quality and safety scholarship in our department. We reward it. We have a whole separate sort of division where we look at that. And I think it's very important to put value on it, not just from an operational perspective. So, sorry.

TRIONA HENDERSON: No, that's OK. That's perfect. I absolutely agree. I think my last institution, they even started incentivizing publications for quality in the medical school, so not just for residency.

YAEL HEHER: Well, and they've changed the promotion track to include innovation, which helps with this new emerging area.

TRIONA HENDERSON: So another question. "I have recently started a new position. Several of your ideas would be great to implement. We have a very green quality dashboard and an ineffective M&M, mostly about blame. Do you have suggestions on how to get people onboard with the changes?"

YAEL HEHER: Yeah. Well, the M&M I would say I'm happy to share our conference guide and all of our resources about that, but I would say you have to have a strong moderator for the M&M, and you have to help frame people's expectations. So the moderator at the beginning needs to talk about what the goals of the sessions are, right, and remind people that it's about systems, it's about being respectful, it's not about H.R., it's not about shaming people, it's not provider-centric.

And one way you can do that is you can batch events. So like often, when you have an event that's a problem, it doesn't happen once. It happens a lot of times. So if you batch events together and use a fishbone, it becomes less personal. Like, it's not that Triona did something wrong or Yael did something wrong, it's that this has happened seven times, or three times. Clearly, no one can argue with it, then, that it's a systemic problem. And so start there.

Or I would say start your M&M in one small division of your lab, like a group that you think are a little more passionate, a little less argumentative. And then, you can grow it once you get comfortable with the format. You don't need to do it, like, on a broad scale overnight. We grew ours over time.

TRIONA HENDERSON: Perfect. Thank you. Specific question from CDC for you. "In your opinion, what can or should CDC do to help promote the ideas that you've discussed today? What CDC guidance would you appreciate, or do you think would be appreciated by smaller community hospitals?"

YAEL HEHER: That's a great question. I feel like, in order to answer that question correctly or well, I'd have to understand what the current state is. So maybe we can talk more offline. That seems like a– because, in order to make a recommendation of what you could do better, I'd have to understand a little bit more about what you're doing now.

TRIONA HENDERSON: Good. Any other questions? We're almost at the end. Yes, please remember to unmute yourself if you'd like to speak over the microphone, or you also have the option to type your question in.

YAEL HEHER: I see another question. Are you moderating around incentivizing people outside the lab to report things?

TRIONA HENDERSON: Yes.

YAEL HEHER: We definitely do that. So we take complaints from whoever wants to complain to us. And I think there is an amazing wealth of knowledge from people out in the community who know things we don't even know that go wrong. And I– sorry, again, it's sort of like a short talk, but let me just show you this. We even monitor that. So let me show you that. Can you see this?

TRIONA HENDERSON: Yep.

YAEL HEHER: So we look at who's reporting, people within the lab and without, and then we also look at– once we've done a request, sometimes it's actually not a lab issue. So we monitor– we want incidents to be called out about half and half. We want people at the front line to feel comfortable. And you know, they don't always know where to send it. Sometimes just whoever in the lab they send it to will flip it to our dashboard, so that we don't mandate that they fill out a form, because then they won't. So we are open to learning from anyone who's willing to feedback to us that they've had a problem. And we do equal RCAs, and then we do root cause analysis on all of them. And we track those data.

TRIONA HENDERSON: Perfect. Thank you so, so much for this discussion today. Can we advance to the final slide?

YAEL HEHER: Yes.

TRIONA HENDERSON: Thank you.

YAEL HEHER: Be bold. [LAUGHS]

There we go. I want to attend the next thing. It looks awesome.

TRIONA HENDERSON: This has absolutely been an amazing session. And by the number of questions and even the number of people participating today, it's definitely well received. So next month, July, we will send an update with the date and the time. We're going to have Dr. Tony Tran from the District of Columbia Public Health Laboratory in Washington, DC.

And this is catered to– not catered to, but about the public health laboratories, and really the integration that needs to happen with clinical labs. And so his title is going to be From the Dark Ages to the Renaissance: Implementing Electronic Laboratory Data Exchange During a Pandemic Response. So please visit the DLS ECHO website to register for this session and to view all future sessions, and then the past recordings and transcripts from the last sessions.

Thank you for taking part in our discussion today. We hope that you find it valuable and important in your work that you're engaged with your individual laboratories. We look forward to your participation in the future, and we'll keep you posted on any changes that may occur. Now we will adjourn. Thank you, and have a great day.

YAEL HEHER: Thank you for this opportunity.

TRIONA HENDERSON: Thank you. Thank you for taking time out of your busy schedule.

Additional resources and related publications

1. Badrick T, Gay S, Mackay M, Sikaris K. The key incident monitoring and management system – history and role in quality improvement. Clinical chemistry and laboratory medicine. 2018;56(2):264-272.
2. Badrick T, Gay S, McCaughey EJ, Georgiou A. External Quality Assessment beyond the analytical phase: an Australian perspective. Biochemia medica. 2017;27(1):73-80.
3. Tosuner Z, Gücin Z, Kiran T et al. A Six Sigma Trial For Reduction of Error Rates in Pathology Laboratory. Turk patoloji dergisi. 2016;32(3):171-7.
4. Rao S, Masilamani S, Sundaram S, Duvuru P, Swaminathan R. Quality Measures in Pre-Analytical Phase of Tissue Processing: Understanding Its Value in Histopathology. Journal of clinical and diagnostic research : JCDR. 2016;10(1):Ec07-11
5. Improving Diagnosis in Health Care. Military medicine. 2016;181(3):183-5.
6. Njoroge SW, Nichols JH. Risk management in the clinical laboratory. Annals of laboratory medicine. 2014;34(4):274-8.
7. Thammasitboon S, Thammasitboon S, Singhal G. System-related factors contributing to diagnostic errors. Current problems in pediatric and adolescent health care. 2013;43(9):242-7.
8. Rakha EA, Clark D, Chohan BS et al. Efficacy of an incident-reporting system in cellular pathology: a practical experience. Journal of clinical pathology. 2012;65(7):643-8.
9. Plebani M. Exploring the iceberg of errors in laboratory medicine. Clinica chimica acta; international journal of clinical chemistry. 2009;404(1):16-23.
10. McCay L, Lemer C, Wu AW. Laboratory safety and the WHO World Alliance for Patient Safety. Clinica chimica acta; international journal of clinical chemistry. 2009;404(1):6-11.