The challenge to improve U.S. maternal health
Insights from the OptumLabs Research & Translation Forum
Over the past 25 years, maternal morbidity and mortality have increased alarmingly in the United States. A woman giving birth in the U.S. today is twice as likely to die in childbirth as her mother was.*
Even more concerning is that maternal death rates are three times higher among black women compared to Caucasian women.* And ~65% of pregnancy-related deaths are preventable.
This growing problem is a product of a multitude of health system issues, including rising Caesarian delivery rates, individual health factors such as obesity and diabetes, and structural issues. There are also large demographic and socio-economic disparities in outcomes.
Collaborative efforts across stakeholders are needed to address this crisis. Understanding the problem, learning about and implementing strategies that work, and measuring how we’re doing are all important to improving maternal health outcomes.
Understanding the U.S. maternal health crisis
Dr. Laura Riley has been analyzing available data to determine the leading causes of death to home in on preventability. She sees the need to focus not just on prenatal care, labor and delivery but on postpartum care as well, up to a year after delivery.
Learn more about the maternal mortality and morbidity landscape as Dr. Riley shares her perspective from overseeing ~20k deliveries per year at NY-Presbyterian Hospital, which services diverse populations in lower Manhattan, Queens, Brooklyn and Weill Cornell.
Speaker: Laura Riley, MD, Obstetrician-in-Chief, NewYork-Presbyterian/Weill Cornell Medical Center and Chair, Department of OBGYN, Weill Cornell Medicine
Video: Understanding the U.S. Maternal Health Crisis
Good morning. I'm gonna tell a little of my story. And then I think it actually delves in with why I'm still doing what I'm doing. I grew up in Dorchester, Massachusetts, and went to Harvard undergrad and University of Pittsburgh medical school. And then I decided that I wanted to do high-risk obstetrics. Actually really just wanted to do OB-GYN initially and do women's health. And so I did that residency. And when residency was over, I decided I wanted to do high-risk obstetrics, 'cause I really liked essentially the happy outcomes, which was all that I really focused on. And it's interesting what you think you're getting into and what you actually get into are almost always polar opposite. So I went for the happy outcomes. I did my fellowship at the Brigham. And when I finished my fellowship at the Brigham, it was the very beginning of the HIV epidemic in women in Boston. And so I went back to Boston City Hospital, nine blocks from where I grew up, for my first job. And I stayed there for about four years before I became completely burnt out. It was a tough department to be in, there weren't a lot of resources. And I think, thinking back on it, the level of care was horrible. And I jumped ship, and I went to Mass General, and I stayed at Mass General for 23 years, providing what I think was really good obstetrical care. And I didn't realize the bubble that I lived in in Boston until I went to New York. And then I got to New York City. And man, is it scary and eye opening and it is, it makes you just realize that the level of care that women receive, I suspect anybody receives, is really variable across the cities, across the country. And it just underscores how much work there is to be done. So I'm gonna show some more statistics. I'm sorry to make this stick in your faces, but I do have disclosures which I've put on the slide. But I think the issues as a provider, that are really important for us to recognize they're 65% of these pregnancy-related deaths are deemed preventable. And that's really where we need to focus our energy. I suspect once we focus our energy there we'll realize that that number is probably higher. The other issue that I worry about, and I think is really very important for the nation to worry about is that 65,000 women will have severe pregnancy complications, and have long term issues related to those pregnancy complications each year. So I'm gonna- the next probably six or seven slides are really just the statistics that come from this report that was put out by the CDC in 2018, which focused on data that was generated from nine different states looking at their maternal mortalities. Over a course of time between 2008 in some states, all the way to about 2017 in other states. And one thing about this data, I think that is really important and an eye opener to me is, to be quite frank, it wasn't really until maybe 2006 that I as a provider really started to understand that maternal morbidity was a real thing. And that maternal mortality in the US was going up and not down. And that it was as high as it was. I think it took us a long time to understand that there was a problem, and getting the data was even more difficult. Now that we have some data, I think this is an important piece of information to understand and as a provider, again, when we think about where the interventions need to be, I think what's very startling is that the timing of death in these cases that they were able to look at, about a third of them were while someone was pregnant. A whole 'nother group of women died within 42 days of the delivery. And then there are also the women who died, about 20% of women died after that, up until a year. And it doesn't add up to 100 usually in any of these slides, I'll just tell you, mostly because there's data that is missing, right? So you don't really know the timing of death. This, it's startling and it's important to remind ourselves of the very stark racial and ethnic disparity in death. I think it's also important to put out there because this is- the reality is that there is also a significant difference based on maternal age. And if you think about what the patient population is looking like, in some ways, it's a good news story that women are older and getting pregnant and how great is that, but this is also part of the bad news story as well. Because we can see that the older women do have a greater risk of morbidity as well as mortality. And then I didn't put the slide in but the most recent data from the CDC would show you that when you add race, ethnicity and age, black women at the higher ages do very, very, very poorly in this country. And that number of women is increasing. So these are the leading causes of death. The reason to show this is because this is one of the slides that we use, or one of the pieces of data that we're using currently to decide on what our priorities need to be. But I can tell you that it's not all that granular, which is part of the reason why I think despite all of the things that we're doing, we're still not making as much headway as we can and need to make. But as you can see at the top is hemorrhage, which is what Charles just talked about, that happened to Kira. And then there's this basket of cases of cardiovascular and coronary conditions. The granularity there is really tough to understand, but again, that's really an important area. And then it's infection or sepsis, which is something that we're working on right now. Cardiomyopathy, and then embolism, pulmonary embolism is getting a blood clot, goes from the leg into the lung. You can die very quickly from that. Preeclampsia and eclampsia is a pregnancy-only condition in which high blood pressure is really the hallmark. A whole lot of other things can go wrong with that. And we see it not uncommonly. It's usually benign, meaning people do okay. But obviously it's an important cause of death as well if it goes untreated or unnoticed. And then mental health conditions. So this, I think, is another interesting slide here, which I think underscores why, partly why this is so difficult to get our arms around, because you can see that the causes of death really do vary by race, ethnicity. And so it's not gonna be one fix fits all. And we need to recognize that. And many of these, certainly I would say, mental health conditions is gonna overlap with a whole lot of other things on this page as well. So the distribution of preventability. So if we're gonna hone in on preventability to start with, which I think makes sense, given the fact that none of us have endless resources. I think if you look at preventability, I think overall, it's about 63%. Again, so most of this we can fix and should fix. If you look at cardiovascular disease, as much as 68% of it is fixable. And in hemorrhage 70% is fixable. Certainly the story that you just heard should not have happened, should not happen in this country. But I can tell you it happens. So I'm at a great institution. I think that I have a fabulous department, people are working hard, I think they're very committed. But I can tell you just in the last month, my own patients had this story: She comes in fairly complicated. That doesn't matter. She has a vaginal delivery. She has severe preeclampsia. She gets a drug called magnesium sulfate, which is to prevent you from having a seizure, which you can have with preeclampsia. She has a vaginal delivery at 34 weeks. We're all excited, huge sigh of relief. She's relieved, her husband is relieved, I'm relieved. Delivery is over, she's being monitored in the intensive care unit because her blood pressure is so high. But it gets good, they move her out of the intensive care unit. She goes to the postpartum floor. Okay, she's gonna get magnesium sulfate for 24 hours after delivery because you can have a seizure up until or sometimes even beyond 24 hours, but that's the standard. The standard is also that when you've had a long delivery like she did, so she had a long labor took a little while, 'cause 34 weeks you don't really, your body doesn't really wanna do it. So we induce you. She has a long labor, she has her baby. And the interesting thing is, and I think this is a recurring theme, we're all relieved and everybody lets their guard down. But they forget this lady still had risk factors. So the next morning, she does okay overnight is what I'm told, trying to read the medical record is its own kind of thing. I won't get into my issues with that. But I go in to say hello to the patient. And she says, "Hi," and passes out. So at the time that I happen to walk in, a nurse is right next to me and she said, "I was just about to call somebody "because I think that she's hypotensive." So she puts her on the monitor and lo and behold my patient is hypotensive, she's really tachycardic and she essentially passes out. Long story short, she had a postpartum hemorrhage, a delayed postpartum hemorrhage. She had been calling overnight. Now this is at a really good hospital. She called three times and said she was gushing blood. Three times she was seen by a nurse who said it's okay. It says in the documentation that she was examined. I'm not sure what they found, it's not clear to me. But we then did what's called a rapid response because obviously this patient's checking out on us. Everybody comes, everything works out absolutely perfectly. And the patient is great. Patient is grateful and "I'm so happy you showed up," etc. But I think what is concerning to me is that when we debriefed this scenario, it's very interesting where we've come and I'll show you some of these bundles. We were so focused on the rapid response of everything that happened after I got there to see the patients when she'd already passed out. We were so focused on whether or not we had the right drugs, the right people, had done the right things. Did I call the right things? Did I document the right things? Everything worked out fine. But it was interesting to me that no one asked the question, which was my first question, "What led up to the hemorrhage?" The lady was high risk, she sat there for 12 solid hours before this happened on a unit. So I think that that's part of the work that needs to come now is that, yes, it's important to focus on the bundles and on the events, on the quick response, but we need to back that up and focus on what led us to that in the first place. So this is the near-miss situation that we need to get to now. And I'll come back to that. Again, distribution of preventability. I think the important piece here is that we are putting a tremendous amount of time, energy and resources into making sure that labor and delivery is safe. Postpartum needs to be safe, but then women go home. And in this country, when you go home, you are dropped off. You as a mom, focus on your baby, and your provider doesn't focus at all because you're not in front of us, you're at home, okay. And maybe coming in in six weeks. So that's another area of opportunity. So these are the themes, it's not really meant to be read because there's so many things in this slide, but these are the themes that came out of the recommendations from that same report that I've been talking about. There were 193 recommendations. And this is sort of boiled down out of 58 pregnancy-related deaths. I think that this has two points here. One is, is that there's a lot of work to do, that's clear. I think the other thing though, is there's a lot that can be learned when we review these cases, and really review them thoughtfully. There's many ways of thinking about this, but this is the way I think about it. This is a framework for addressing racial and ethnic disparities with any health issue. If we think about it in terms of maternal mortality and morbidity, which is what I'm focusing on here. I think that the most important thing to me, to be quite frank, is that all of these circles overlap. Which means it's not gonna be one fix, it's not just gonna be fixing the OB-GYNs in the US. That needs to happen, but that's not going to fix this problem. And we are not gonna be the only group that can fix this problem. First of all, obstetrics is a entire team sport, number one, but in addition to that, as you can imagine, everyone has a role in this problem. And everyone has to play a role, if we're gonna solve it. So healthcare systems, I'm in a big healthcare system right now at NewYork-Presbyterian. It's Weill Cornell and it's Columbia. We have eight delivery services between us on the Weill Cornell campus alone. I'm responsible for 20,000 deliveries a year. It's Lower Manhattan, it's Queens, it's Brooklyn and Weill Cornell. Those populations are so vastly diverse and different, it is mind boggling to me how to get us all on the same page and give care to so many different kinds of people all at the same time. I have to tell you, living in the bubble in Boston, I was like, why can't they get this right? This is really not that hard, 'cause we were pretty darn good in Boston. New York is a totally different beast, it really is. And I think that that's what we're seeing across the country as well. I think there are a patient factors that are important to understand and to incorporate. And this is what, honestly, when I go to Queens, it's amazing how many different languages are spoken, not even knowing or understanding the beliefs of the patients because no one's bothered to ask. And even when you do, you kind of don't know how to interpret the data. I think that things are challenging. The clinical encounter is a big black box. And I will tell you one other quick story. About two months ago, a patient came into Weill Cornell, I am not proud of this. A patient came into Weill Cornell at the last minute. She got all of her care at home with a lay midwife at home 'cause her intent was to have a home birth. And that's a tough topic in obstetrics and gynecology. It's kind of frowned upon. I have mixed emotions about it. But at any rate, this patient had a prior delivery two years earlier at home, went uncomplicated, she was happy. She planned on having a second home birth, but she had complications and the lay midwife brought her into the hospital. The experience in the hospital was nothing less than horrible for this patient. She came in for all sorts of reasons. She ended up with a Cesarean delivery, which of course, if you were planning a home birth, the last thing on God's earth you want is a C-section. She got a C-section, which I have to say was indicated based on some fetal issues. That said, the problem is convincing the patient that she needed the C-section, I'd say bordered on coercion, the way it was presented to her. And then after that, she gets the C-section. And it could have been, I feel like we should have understood that this patient was gonna push back with absolutely every medical intervention that was offered to her and recommended. And as it played out, of course, that's exactly what happened. So she refused everything for her baby. Everything. The eye ointment, things that you and I would just be like, "Oh, okay," she said no to. Well, lo and behold, New York is different than Boston. If this had happened in Boston, we would have said, "If you don't wanna put the eyedrops in your kid's eyes, "despite what we're telling you, "don't put them in, that's fine." And we would have left it alone. Well, in New York, come to find out there's a law - who makes laws like that? But that's- whatever- there's a law that you have to do it because of whatever. So she then gets into literally a fight with the pediatricians. And she... they go back and forth and they involve the ethics committee and the this and the that. Well, let me just tell you something: at the end of the day, I don't know who's right and who's wrong, but I can tell you when that woman left the hospital, she will never ever go back to any hospital in the country again. It was such a horrendous experience for her. She now has a cesarean delivery. She now has significant risk with her next pregnancy. And you and I both know, she's gonna have a home birth. Okay so here, we've already set up a disaster. And so I just think, and it's interesting when it was presented to the ethics committee, the number of people sitting around the table, not even recognizing that this was an important thing to talk about, was even more frightening. And I just leave that with you just to think about because I think that these are not conversations that you're gonna see in the data. These are not things that you're gonna pull out of all your fancy data and big data sets. But that's actually what's going on and that does lead to some of the issues that we have. So I think that the clinical encounter and the provider factors are just big black boxes as far as we know. So what's happened so far? So it's not all bad news. It's not like we haven't done anything 'cause we, it took us a while to figure it out. I think you'll see that these are lots of publications that have come out both from ACOG the Society for Maternal-Fetal Medicine, which is the high risk obstetricians. We've done a lot to try and get the word out. And I think that that has helped in some small way to start the conversations across the country. But it's more than just getting the word out, it's what are you gonna do with the information once it's out there? And so these are the things that have come about. There's no question that this is a crazy busy slide. But the reason that it's crazy and busy is because all of the different specialties that touch women's health are involved in this effort now. And that was really important. And the biggest effort that people have focused on and it's been a priority has been these bundles. And these bundles, the principle behind it is so that for that hemorrhage situation that I mentioned to you, it's important that every single hospital in New York City has, at the very least, all of the tools to treat hemorrhage. And that all of the people who are doing the treatment or recognizing the patient have the same level of education. And that they are doing periodic drills and rehearsals and whatever, and education, so that when those rare events happen, which don't seem to be all that rare. But they are rare over the course of many, many deliveries, that people are are ready and prepared. But that's the, that's actually the bottom, right. That's not really where we wanna be but that's at a bare minimum, this is what needs to happen. So that's what is the basis for these bundles, and more and more bundles have come out. I think that the information is available. But I can tell you now that I'm in charge of four different delivery services. Figuring out how to get that to all four delivery services, and to actually be able to monitor whether or not people are doing what they say they're doing. And doing more than checking off the boxes, that is a whole 'nother beast. It is way harder than you think. I'm figuring that out now, over the course of the last year. But I think that ACOG has led the charge and that's the American College of Obstetricians and Gynecologists. Society for Maternal-Fetal Medicine has done the same, recognizing that there's lots of high risk patients out there. The California Collaborative, there are many collaboratives now, thankfully. And California has been able to show that when you put these bundles into place in all the hospitals in California, you can see a decrease in mortality. And that's encouraging because I think there's nothing worse than putting in the time and the energy and realizing that you're making no progress. There is progress, but it's incrementally small. Other state initiatives, this focuses again, these are bundles and ways to practice and giving people the information. It's all surrounding health care, it's all surrounding being in the hospital and doing the right thing. And as I mentioned to you, a lot of the issues and problems happen outside of that encounter. So, just to let you know that some progress has been made, I think that this is from the Society for Maternal-Fetal Medicine, which put out the first call out. And we have made some headway in terms of educating providers. Clinical care, we've worked on, at the levels of maternity care whether or not that really works and whether or not regionalizing healthcare is really where we think we're gonna go. I think the problem with that is, it's wonderful if you're in Boston, it's wonderful maybe if you're in New York City. But if you're in a rural community where they only do 600 deliveries a year, it's harder to do all this stuff. They have limited resources. And it's easy to say, "You know what, "if you do so few deliveries, "you shouldn't be doing deliveries." But if you take that one hospital away that's doing 600, where are those women gonna go? They're gonna drive four, six, eight, 10 hours to get to a hospital. So we need to think about what that means for the whole country. And then this goes back to my original thought, which is, this looks at morbidity and those near misses. This is where we need to go next because waiting and just looking and debriefing the deaths is really too far. We need to be able to back it up. So this is where I think having the big data, figuring out what is in the big data and helping us risk stratify patients, maybe we can back it up further, right. So if we can risk stratify patients and teach them to risk stratify themselves so they can speak up and tell us, I'm not getting the care I need, I know I'm high risk. And then they can go and look for whoever they need to look for, that may help us as well. And then finally, this is the elephant in the room right now, is this whole issue of equity, of dignity, of taking good care of people, of equity in care. California is hoping to put this bundle into place and I kinda feel like it's a bit of a hope. I can tell you what I'm doing in my own institution. We're opening a new hospital, I used it as an opportunity to have the entire service, so the secretary, the cleaning people, the doctors, the nurses all do implicit bias training. Do I actually think that that's gonna make people less racist? No, I don't. But I do think that it will give people the ability to at least start talking about racism and how it adversely affects our care. So that when we do our debriefs, we can at least look at that as an aspect of why we have the outcomes that we have. Because if we don't start to get to that piece, we'll never understand what's happening in the clinical encounter. And that's a really important issue. So I end with this: I think that we need to start looking at these structural factors. Again, I think when we're thinking about the big data, I venture to say that what's in that red box is not well represented in big data. And that definitely biases the information that comes out of that big data. So I think that that in of itself is gonna be a challenge. But I think it's something that all of us have to recognize, all these things. Because if we don't have legislation, if we aren't able to give pregnant women, who then deliver, any kind of health care after their six week visit, we're never gonna take care of those high-risk women who continue to have high-risk conditions that don't get responded to. We'll never get where we need to be. So we need everyone's help in this issue. That's my last slide.
Central to understanding and preventing maternal mortality is obtaining and using insights from strong, comprehensive and standardized data to identify and act on opportunities to prevent maternal deaths. Rear Admiral Wanda Barfield, MD, has made that a focus at the Centers for Disease Control and Prevention (CDC).
Learning about strategies that work
Find out how CDC has created a common data system that informs programs designed to tackle the underlying causes of adverse events and preventable maternal deaths. Dr. Barfield also shares insights on tools that help mothers and babies get care at the right place and time.
UnitedHealthcare and Optum use data and analytics to evaluate promising programs to improve maternal health. Learn more from Dr. Janice Huckaby about how they invest in programs that support patients and physicians via safety bundles, educational programs and more.
Speakers: Rear Admiral Wanda Barfield, MD, MPH, FAAP, Director, Division of Reproductive Health, CDC and Janice Huckaby, MD, Chief Medical Officer, Maternal-Child Health, Optum Population Health Services
Video: Strategies To Tackle Underlying Causes of Adverse Events, Preventable Maternal Deaths
I am a nurse. I was a bedside labor and delivery nurse for 15 years out in Washington State. I had other patient stories; I never lost a mom but got close. And those memories, those experiences are part of what drove me into working on more system level change. So while I was still in Washington, I joined the hospital association there, which is also the quality collaborative for the perinatal world. And I worked on the 60 delivering hospitals in Washington State, really rolling out bundles and quality measures across the state. So when I met Neil Shaw, he talked me into moving across the country. So I joined Ariadne Labs a year ago. Boston is new to me, it is a wonderful area, but I am new to the area and I've been in Ariadne now for about a year. So, I'm going to mess with the presentation a little bit because it's called Research and Translation. But Chris and I decided to flip it and we're gonna talk translation first and then we'll give you the data. So I know you probably are a lot of data nerds in this room, myself included. But we're gonna mix things up and talk about this huge challenge of once we know that we have a problem in our data, what do we do with it? And how do we actually transform healthcare at scale? Which is what everybody is stuck on and we know it's a huge challenge. So Ariadne Labs, it was started by Atul Gawande and Bill Berry. Many of you probably are familiar with those folks. But if you're not, and you've had surgery in the last few decades, you probably benefited from the surgical safety checklist. So that was sort of the original project at Ariadne Labs. Their vision is basically to reduce suffering and improve lives and well being to people everywhere. So it's not just a United States based company, it is global. Don't worry, I won't read any of those to you. But I want you to get an idea that Ariadne is working on projects in the United States but also around the world. We do have five main programs, they span basically from birth to life. And in the middle we focus on surgery and also in primary care. The way we think about scale at Ariadne Labs is through this arc; it's Design, Test and Spread. So in the design phase, we're really looking at this data, we're doing a lot of human factors design, how can we create solutions that are gonna have impact at scale? Testing: obviously, we have to make sure it works. And then if it does work and we can prove effectiveness, we do work towards scaling these innovations. Those little yellow lines are representing separate programs at Ariadne Labs. So the one that's the furthest around is that surgical safety checklist I was mentioning to you. The smallest one is the Delivery Decisions Initiative, which is the program that's led by Dr. Neil Shaw, which is the program I work on as well. Our vision is that every person can choose to grow their family with dignity. And the way we see dignity, this is a really important point, we don't see that once you have a safe delivery, then therefore, you know, the next best thing is a dignified experience, we see dignity as being a precursor to safety. So if patients are really listened to and heard and appropriate shared decision making happens, we believe that actually leads to greater safety. So that is why that is specifically in our vision. Okay, has anyone seen this before? C-section rates? Yeah, few people. This is crazy. For those of you born before 1970, your mom's chance of a C-section or risk was approximately 5%. When I was born, it increased about 17-18%. Now you can guess how old I am. And my four children, I have four, that's a little crazy as well. But by the time they were born, my risk for a C-section was around 30%. That's in one generation, that is wild. And I know that we hear about it and it's in the news. But I also don't just want to blur past this because I think it is a moment that we need to stop. It's a signal to us, not the C-sections are inherently bad, they save lives, but it is a signal of something that is going on in our healthcare. So what this is, is a 500% increase C-section rates in one generation. I don't think this is ever- any other medical procedure, I would love if someone could find data on it, that would match this. I do believe this is the kind of craziest extreme in such a short period of time. Now the problem with those C-section rates is, we would love it if more moms were surviving and more babies are surviving, 'cause then we could justify all these additional resources and this overutilization, but we can't. There's actually no shown improvement in fetal outcomes. We actually have 50% chance higher for maternal mortality in this generation than we did in the generation ago. I won't go into much more data because I know that's what Chris is gonna do. But I do think it's important that we all are on the same page and acknowledging that we do have a problem. C-section rates are being done too often, and intervention for moms are being done not often enough when they really need it. So I'll talk a little bit more about how we actually design at Ariadne Labs. So we have principles that are our design principles. And they come around simplicity and teamwork. So any of you who work in healthcare or have worked in healthcare know when the electronic record came about, it was a promise that your lives is gonna be so much easier, you would have less time documenting. But the reality of that is it got more difficult. And with a lot of other technology advances, the life of a clinician actually has gotten more challenging and more burdensome. And now in health care and the news, we're hearing a lot about clinician burnout. So it's a real thing. So we really tried to focus on simplicity of making the right thing, the easy thing to do and not adding significant burden to clinicians. So that was the first principle. The second principle that we focused on was teamwork. And why we feel like this is important, is somewhat based on Dr. Edmondson's work, Amy Edmondson, where she writes and researches about high functioning teams. Are you guys anyone familiar with her work in here? Yes. Okay. But one of the things she talks about is that, what's needed for high functioning teams are basically two main functions: one is psychological safety, and the second is some structured or expectations around communication. So if we implement all these bundles and all these different checklists, but you don't have safety on your unit, and the nurse can't call the doctor in the middle of the night, is it going to help? I don't know, might not. So we view teamwork and elevating all the expertise on the team to be a core principle of how we design. So, we started with those data sets of what the signals were. We started with our core principles of teamwork and simplicity. And this is what we designed. Now if you're severely underimpressed, I was too the first time I saw it, 'cause it doesn't look fancy. But how it works is really amazing. It's a whiteboard. Most of your hospitals that you've been in or worked in have whiteboards. They're typically areas where people write things to remember on them. But we tried to change that into an actual shared decision making board, shared planning board, that the patient has full transparency to at all times during their birth. We have four main sections that go on it. First is the team. So all members of the team are listed here, including the mom and whoever her support person, because we feel like it's very important that they be named, that they have that psychological safety, which is both the opportunity and the invitation to speak up. Second thing, preferences. So this can be that "I want to breastfeed," but this can also be "It's really important to me "that the first two minutes of birth, everybody's quiet." It can be cultural, it can be anything. But it's acknowledging that patients do have preferences and that we shouldn't be assigning them, and that preferences can change over time. When you show up at the hospital and you expect a certain kind of delivery, it may not go that way and that your preferences need to be revisited and need to continue to be an actual co-creator in your birth experience. This next thing is the plan. We purposely separate this out by mom, baby and labor progress. For those of you in the room that are clinicians, when you're on labor and delivery, you're taking care of a mom and the baby together, and you have to make decisions about both of them. And sometimes those variables get confounded. So we purposely split them out so that you're making very clear decisions, and that the patient really understands what is happening and why. And then this last thing, is next assessment. Again, a really simple concept. But it's just when the team has decided they're gonna come together again and meet as a team. We call them huddles, when they're gonna huddle again. And the best sort of comparison to this experience is, anytime you've been on a plane and you're like supposed to depart 30 minutes ago, but you're still in your seat, and nobody has told you anything, and you have no idea what's happening, that's what laboring women feel like often. They don't know what's happening next. They've may have had a baby or not, but they don't know when is the next check in point. So this is a point where we at least set a clear expectation with all the care team about when they're gonna come back together again. Okay, so it doesn't look very impressive, but we just completed a pilot on it in four hospitals around the country and have some amazing results. In our pilot, we tested for feasibility, acceptability, and whether or not clinicians could really implement this with high fidelity. The reason we chose to test this first, is because there are many, many, many things that have been shown to be effective but have failed to go to scale. Because they're not feasible, they're not acceptable to clinicians and families. And that's where that research and translation drops off. So we decided to test this first. Basically, like an FDA drug or device trial tolerance test. Could clinicians tolerate having a whiteboard, communicating this way? Do patients like it? So like I said, we just concluded the pilot, and we expected clinicians not to like it. Again, for the clinicians in the room with burnout and quality improvement and my own experience with clinicians, they don't tend to like change or any other new project they have to do. Shockingly, our clinicians- I can't give you the exact numbers 'cause we are working on the manuscript, but reported really high rates of recommending team birth to other hospitals. We also had really high acceptability within patients. That they actually liked being involved in their care, which is not shocking. I know I said "actually," but we all want this level of transparency. And as myself when I get health care, I demand this of level transparency, but it's 'cause I know that I can do that. So we're trying to offer this to patients. So again, this problem of transforming at scale. So a couple things need to happen. One, I wrote scale again, because when we think of scale, we actually design for scale. So in the very beginning when we're thinking about our data, or what the problem is, we design for scale in mind, which actually creates a much different design than if you were just designing for one local hospital. And the second thing that we think about with scale is advocacy. So people like Charles and others that go around and they drive demand. They let families know that they deserve transparency, they let families know what the data says, what the risks are for them. And we believe that you can create the best product out there, but if there's no demand, it's not going to implement at scale. So one of the things that we're doing on our team is, we do have a project called Expecting More. It's a website, expectingmore.org. If you also put in com, it'll direct you there. These are highlighting pieces that are mainly from Dr. Shaw, but also other members of our team. So if you look in the bottom left there, there's me when I was younger and I was still wearing scrubs. The point of this movement is really for patients and families to expect more, to know that they deserve more in their care and help to drive solutions and come to the table to solve this maternal mortality and morbidity crisis. So, I'm gonna explain a little bit more just about the other projects at Ariadne and how we think again about this, how do we translate research into scale, and then Chris is gonna come up and talk about a specific project that we worked on together and the opportunities that we all have around it. So this first area, we think of innovation. So in our work, the whiteboard I talked to you about was TeamBirth. So we just finished a pilot and we're looking to start an effectiveness trial for that in an entire state or region in 2020. But we also have other projects, we call them R&D. This is where a lot of our data work goes. And one of them is called Nurse Influence. A lot of the projects in this bucket is us attempting to measure things that are unmeasurable or solve for problems that people haven't solved for before. So nursing, it's the largest workforce in a hospital or healthcare system. And they do a lot of things that we have no idea what they're doing. They don't bill, so we have no way of actually looking at that data. It's actually fascinating when you think about it, because the nurse spends most of the time at the bedside, the obstetricians tend to come and go. So we're looking at how to measure and nurse influence on cesarean rates. And we're doing this in a cohort of hospitals in Washington State and also a couple hospitals here in Boston. It's fascinating. Nurses have a wide variation just like individual physicians do. Another one of the projects we're looking at, is how to measure what's happening in prenatal care. What we know right now is currently how many visits there are. But what is the quality of visits? What's the care that's being experienced by the patients? And so this is a new project that- Rebecca, raise your hand. Rebecca is leading this project. It's with Optimum Labs and Robert Wood Johnson Foundation of really trying to understand what is happening during that prenatal care period. So I'm excited because these are gonna come up with some great products. But again, if we don't have anybody who wants them or is interested in them or cares about them, they will go nowhere. So this is where we think about market. Again, at the top there, is that Expecting More, that's what we were talking about with that website of really trying to drive public narrative, but we also have couple other projects. One are Safe Childbirth Cities. So Ariadne Labs is working on creating a scorecard/dashboard-- to be determined--of how to measure city-based support for growing families. When a family leaves the hospital and goes home, support drops off. And so what is happening at the city level where families are having support systems or lack thereof? Another project is called March for Moms, and this is led by Katie Barrett. And this is again, really working in the policy space of trying to drive change and the national narrative around policy. So all of this is necessary for scale. And we put this at the bottom of Community Practice, 'cause what that means to us, is that everybody in this room is a part of bringing change to scale. And that can't be done individually, we need data, we need really cool frameworks, which is, Chris gonna talk about in just a second, but we also need market demand. We need patients to know they deserve better and that their race should not determine how likely they are to survive childbirth. I met the teacher who was in Minnesota and she teaches at a school for pregnant teens. And she contacted us and she said, "My girls are so upset. "One of their friends died in childbirth." And these young women were all black, and they wanna know what they can do to not die. These are teenage girls worried about this. Like, it blows my mind. It's something I never thought about when I was pregnant. But this is a real issue that we have to come together on. And one of the ways that we are working together with OptumLabs is with this framework around maternal morbidity. So for any one case that a mom dies, there's a whole lot of other near misses or opportunities for improvement within how moms receive care in the healthcare system. And unfortunately, the statistics that we currently use which are based on the CDC measure, the severe maternal morbidity, is a really, really good start. But it's a really good start for a population-based measure. But it's really hard to measure to look at and actually know what to do differently. It doesn't easily lead to quality improvement. And so Chris will come up and talk about an idea that we're working on of how to use this measure to promote change at the hospital level or the healthcare system level. And not just in the moment of childbirth but afterwards, because two-thirds of those deaths actually happened once the patient left the hospital. So if we're just thinking about this narrow inpatient stay, we're missing so many opportunities to save maternal lives. So, I'll hand it over to you.
- Thanks, Amber. So, I am going to really take a deeper dive now into the data. So for all of those data people here like me, no bodily fluids, here. We're gonna look at numbers and rates and data. What we wanted to do is think about, how can we build this framework where we can identify the areas of opportunity? What are our opportunities to really go in and make a difference? So, currently there is no comprehensive set of measures that we can look at things from pregnancy right through delivery and then into the postpartum period. The center- the CDC's metrics that they publish on their website are a really great start. They look at the delivery inpatient stay, that's the admit date to the discharge date. And a lot happens after that discharge date. They also look at hospital discharge data, so they're looking at EMR data. So we thought this was a great opportunity for us to go into our claims data. We can look at a longer period of time and see how we can address this. So building on those CDC measures, we looked in that inpatient delivery episode, plus six weeks on a limited set of measures. We didn't do them all. They're 18 of the CDC's, we did a limited set, and we focused on the ones that we thought were the most preventable. We sought and, even more importantly, received input from a lot of really smart people. So first, we were very fortunate to bump into Amber. I called an old friend at Ariadne and she said, "We just hired Amber, she came from Seattle." And so we got Amber, we talked to Dr. Barfield, others at the CDC, we talked to the people in the California Collaborative, got great input. So we talked to clinicians, we talked to obstetricians and neonatologists. So we developed these metrics that are based on ICD-10s. And our data set that we looked at this is delivery episodes from March 2016 through 2018. And we looked at our claims data which is commercial and Medicare Advantage, but overwhelmingly commercial. We don't have Medicaid data here. So this was our starting point. This is the CDC metrics. And again, we picked a subset of what we thought would be the most actionable to start this almost as a proof of concept we were gonna do a limited set to see if we could see something. So these are our metrics, the asterisks indicate something we added that was not in the CDC measure. So the CDC had blood transfusion, that's really hard to see in claims, there's not always a claim for it. So we looked at hemorrhage instead to see if we could get better indication of what was happening there. We added in some categories of preeclampsia, and then we made a composite variable of the preeclampsia or eclampsia. For the cardiovascular risks, we also added in congenital heart defects. And then for our composite, we made it with and without those defects, we looked at sepsis. And then for embolism, we also made a composite measure. So this was our starting point. This is the CDC metric, existing metrics on inpatient episodes, admit date to discharge date, on health record data. So this is ours, it's that same inpatient to discharge date, using claims. One other thing we did here is we added an indicator for whether the condition was present on admission, so we could differentiate what happens in the hospital versus what the patient came in with. Our cohort for that was about 357,000 delivery episodes. We then expanded it to include the six-week postpartum period. We define that as admit date plus 41 days for that six weeks. And this is inclusive. So we didn't separate the two, it's inclusive over that time period. We again had that indicator for present on admission. And our cohorts a little smaller here, because we lost some people due to the continuous enrollment criteria for that six weeks. We also evaluated all of the episodes by whether they were vaginal or C-section. So this is just a little bit of information on what our cohort looked like. So it was 99.8% commercially insured, very few Medicare Advantage births in here. It was a little bit of an older cohort, although 84% was between 25 and 39, 94% of the inpatient stays were between two and five days, and we had just over a third were C-section deliveries. But we don't have time today to go into all of the metrics. So I'm just gonna pick a few to show you and talk about some potential opportunities. This is a confusing table, so I'll tell you the orange, pinkish section is the inpatients stay. The gray is that inpatient stay plus the six week postpartum. We have a number, the rate per thousand. And then that gray line on the bottom is whether or not it was present on admission. So for the hemorrhages, obstetric hemorrhage, it was higher in C-sections. The majority of those happened during the IP stay, and 24% of those hemorrhage events were present on admission for the IP stay. So what might be an area of opportunity here is perhaps related to the AIM bundles and making sure that the hospitals are prepared and trained and everybody is educated on those AIM bundles. And just to note, we didn't apply any severity measure to these hemorrhages. So we're assuming they're probably the more serious because a milder bleed may not have been coded. This is our composite variable for the cardiovascular at rates per thousand. So I've listed here what goes into that. So those are the CDC metrics around the cardiovascular. So the cardiovascular was highest when we looked at the inpatient plus that postpartum period together, and it was four times more likely in women who had a C-section. So, is there an opportunity here perhaps for screening? So, although very rare, an underlying cardiomyopathy is very serious and it could be picked up by perhaps screening for biomarkers like BNP or troponin. And just as a point of comparison, all pregnant women in the US are screened for syphilis, which has a much lower rate. This is sepsis. The sepsis is- the difference in sepsis between the inpatient stay and inpatient plus postpartum is so striking I felt the need to put it in a separate bar chart for you to see. The sepsis is happening primarily in that post discharge, postpartum. It is more common in C-sections. And then I've also on the map showed, there's really great geographic differences in that sepsis rate as well. So there might be a lot of opportunities here for sepsis. But one other thing this framework is allowing us to do, is once we have this framework and we start looking at these measures, we can then move backwards through it and start looking at What are the pathways? What led people up to this SMM event that has occurred? So this is a very confusing alluvial flow of 28 different patterns of sepsis onset that we identified in our sample. But I just want you to just for a second, just focus on that green bar that starts at the top and goes down. That's vaginal deliveries that were discharged and then they had a later admission for sepsis. Okay? So that could be a real opportunity, is that, you know, an opportunity for educating women on the signs and symptoms of infection? There was actually a trial published in Lancet a couple of months ago, a clinical trial where all device-assisted deliveries, so forceps or suction, we're given a one-time antibiotic post delivery and it cut the rates of infection including severe infection like sepsis, it cut that rate in half. So there's opportunities here. Then just the geographic variation, these are local problems. So these are our national maps for eclampsia, embolism, hemorrhage and cardiovascular, and you can see they have very different patterns. So eclampsia seem to be highest in Hawaii, Mississippi, New Mexico and Oregon. That thromboembolic SMMs were highest in West Virginia, Idaho and Mississippi. Very different patterns, could be different risk factors within those populations. So the ways to address this is gonna be local. So just to summarize. And I wanna start with the last bullet here because I think it's the most important. These assessments we've done, we've demonstrated it can be done in claims, and we've demonstrated more than we could show you here, some pretty striking results. I say here 'could', I'd say it 'should' be extended back into the pregnancy, and further out into the postpartum period to pick up both what's often called that fourth trimester of three months postpartum, as well up to one full year postpartum, post delivery. So I think that's a really important step. But this KPI framework has demonstrated, first of all that claims are a really useful tool here. We can use claims that we can benchmark, across systems, across insurers, across platforms. It's hard to address a problem if you don't know where you are. And it's hard to know what's working, if you can't see where you've been. Our findings underscore the importance of the postpartum period, including the postpartum period, and also on identifying whether that condition was present on admission or it's something that happened in the hospital. We also see this great geographic variation which could be based on local risk factors as well as training and facility level variables. We had a few limitations here. So we looked at commercial and Medicare Advantage, we do not have access to Medicaid data, so these results might not be as comparable in a Medicaid population. Our results suggested that some measures, for example, blood transfusions, may be very inconsistently coded, so they're hard to track in claims. And some of our state-level results were based on a fairly small number of deliveries and rare events, so they could be unstable. Thank you.
To improve maternal outcomes, strong metrics are needed to assess what is working and what is not. Currently, there is no comprehensive set of measures that can evaluate issues from the prenatal phase through delivery and into the postpartum period.
Measuring program effectiveness
Amber Weiseth understands the challenge of using insights from data to develop programs that can improve maternal health at scale. Learn how the innovation process works at Ariadne Labs through a focus on simplicity, teamwork and design, and about the role that measures play.
OptumLabs has collaborated with Ariadne Labs and others to develop a key performance indicator framework to help improve maternal morbidity. Learn more from Chris Chaisson about the impact that claims-based measures focused on the delivery episode and beyond could have.
Speakers: Amber Weiseth, DNP, RNC-OB, Associate Director for the Delivery Decisions Initiative, Ariadne Labs, Chris Chaisson, MPH, Sr. Director, Translational Research, OptumLabs, Moderator: Darshak Sanghavi, MD, Chief Medical Officer, UnitedHealthcare Medicare & Retirement
Rear Admiral Wanda Barfield, MD, MPH, FAAP and Janice Huckaby, MD speaking
At CDC we're working on raising the awareness and building the support for the efforts to really impact and reduce pregnancy-related deaths, and our work is in understanding and preventing maternal mortality that falls into two buckets, making sure that we have strong data and at the same time ensuring that women have access to quality care during pregnancy, delivery, and postpartum, and the bottom line is that stronger data can save mothers' lives. Now robust comprehensive standardized data are essential because they help us to understand the opportunities in terms of preventing deaths amongst mothers, as well as designing effective interventions. Maternal Mortality Review Committees, or MMRCs, get the most detailed, complete data on maternal deaths and then they translate that data into effective programs that address the factors that are contributing to maternal deaths. We currently have 41 existing reviews, including in states and two cities. Now this leaves 11 states without maternal mortality reviews and most of these areas are considered rural. However, we have contacts in every state, and we have more momentum and potential than ever before. All states are now in the planning stages of implementing Maternal Mortality Review Committees that will give them better data in terms of priorities for action. One key tool that has been provided is called the Maternal Mortality Review Information Application, or we call it MARIA. Now MARIA is a data system designed to empower the Maternal Mortality Review Committees to then create action with a common data language, and CDC supports the implementation of this standardized system by review committees targeted with technical assistance, training, and promoting their successes. We currently have 33 jurisdictionals that are using MARIA data, and they're starting to create a standardized, comparable information that includes the committee decisions, the pregnancy relatedness of their findings, underlying causes of death as well as preventability, and then they're able to also collectively think about those recommendations for action in every case that they're reviewing. I want to emphasize that there are also opportunities for interstate collaboration, and this is through common data, so this common language is really important to allow regions to work together, and to facilitate their work. Now CDC also funds 13 state perinatal quality collaboratives. These are state-based initiatives that aim to improve the health and quality of care of both mothers and babies as they receive their care, but we also partner with many states that we haven't funded and it includes a total of 47 jurisdictions that are involved in what's called the National Network of Perinatal Quality Collaboratives, and PQCs play a critical role in translating the information coming from MMRCs into quality standardized care for women and infants. And then I also want to talk a bit about CDC LOCATe, and this is the Levels of Care Assessment Tool, and what CDC LOCATe does is it allows states and other jurisdictions to think about risk-appropriate care. We need to make sure that women are getting care at the right place and the right time, and unfortunately states aren't able to do facility-specific interventions at the public health level, but what this tool does it shows them the distribution of staff, services, as well as competencies and capabilities at both maternal and neonatal levels in birthing facilities. In addition, CDC has recently made 24 awards, supporting 25 states on enhancing reviews and surveillance to eliminate maternal mortality, we call it ERASE MM. And this program directly supports agencies and organizations that coordinate and manage Maternal Mortality Review Committees so that they can identify, review, and characterize maternal death, and then identify prevention opportunities. So they're gonna facilitate the understanding of the drivers of maternal mortality, and the complications of pregnancy in order to better understand the associated disparities. They're also determining what interventions at the patient level, at the provider level, at the facility, system, and community level that is going to have the most effect, and they're going to inform the implementation of initiatives at the right place for families and communities that need them the most. So why does better data matter? It's because it gives us more information so that we can have better programs and interventions. For instance, the proportion of deaths due to hemorrhage and hypertension have actually decreased since 1987, while the proportion of deaths due to heart disease and other medical conditions have increased. Knowing this type of information provides a deeper understanding of the circumstances that are surrounding death, and helps us to identify actionable recommendations, and interventions in order for us to prevent future deaths. At this time, I'll say thank you, and we'll have a chance later for discussion.
- So I have the opportunity to speak to you a little bit today from a different perspective maybe than some of the other folks who've approached the issue. We've heard about the great work from Ariadne, from the CDC, and we at UHG United Health Group, where awareness of this problem became almost impossible to ignore. We thought how can we, or what niche can we fill, as a payer, with all the other good work that's going on? Organizations like National Healthy Start, March of Dimes and all are already on the ground, engaged in their communities doing work, so what can we do as a payer to supplement and assist in that? One of the reasons that this is important is of the roughly 300,000 babies that were born in this country last year, about one in 10 of them had a parent with the United Healthcare card in their pocket. So obviously there is scope and scale that this crisis of maternal morbidity and mortality impacts our members. So if we look at approximately 700 maternal deaths each year we imagine that about 70 of our members or about six or seven of these tables have been impacted by this. And I was really glad to hear Dr. Riley talking about maternal morbidity being part of the issue, because we looked at that as well, and know that from our data, it mirrors the national 1.5-2% of our women are having near misses. So obviously this is an issue that we care about very much for the benefit of our members. So in September of 2018 there was a multidisciplinary group that was drawn together to help address this. And yes, there were clinicians, and I'm one of them. Several of my colleagues who are here are others, but it also included Mike Curry, who is our Chief Health Equity Officer, because there absolutely cannot be a discussion of this without looking at the racial disparities and health inequities that are driving part of the problem. We too did some data analysis using claims, we were able to have Medicaid data in our poll, and we looked for areas that I'll just call hotspots for lack of a better term. It probably won't surprise anybody that a number of these were located in the south, which as you can tell from my accent is where I'm from, and we saw opportunities in Mississippi, Louisiana, Tennessee, New Jersey, and the greater Atlanta area. So with, again, our goal to being not to duplicate, but to support local initiatives, we went on the road, actually, and met with some of the hospitals where we had a high volume of patients, and where we saw a high instance of morbidity and mortality. Again, not with the idea that this is a punitive, here's something you're doing wrong, but the goal to see what we could do and help them to support the many things that they were doing right. So I'm gonna just list here some of the partnerships that we had after the road tour and we were listening. Healthy Beginnings is an organization that's committed to health and safety of mothers and newborns. They were already working with one of our Medicaid businesses, and we expanded that pilot into Louisiana. These programs provide support for women during the prenatal period and then afterwards. I did want to mention the work with the Joint Commission, because there's been discussion around the safety bundles, and through some of our advocacy, and the efforts of many people in this room the Joint Commission has engaged on that topic, and there wasn't the big splash I thought there might be but there was an announcement that starting in July of 2020 hospitals will have to implement to maintain their accreditation the hypertension and hemorrhage safety bundles. Now these are not the full safety bundle, these are skinnied down a little bit. For hemorrhage, it involves having a hemorrhage cart, and simulations or trainings. For hypertension it involves having the standards at which you'll start intervening is at 160 over 100 or what, and the treatment course with that, but I think that this will have an impact on those events that obviously occur in the hospital setting so we're very appreciative of that. As we at United Health Group look at how we can reinforce that, I mean, the hospitals want to be accredited, because if you're not accredited, you don't get money from Medicare and Medicaid, so that's a pretty good incentive, but we could also have the opportunity to reinforce that in our value-based contracting where we have what we call a quality gate, certain metrics that you must achieve before you could share savings or have any kind of an escalator in your fee schedule or something like that. I mentioned that we did the road tour with the hospitals, and we are developing partnerships in some of these targeted states. One of our hospitals in Tennessee actually, where they have a robust cardiovascular program would benefit from some of the remote monitoring of blood pressure. Because they're a referral center, it's often difficult for women to come in for followup care so remote monitoring is something that we're investigating. We also want to see how we can better support patients and physicians. Some of that may be through educational programs that we can offer to both patients and members and physicians, so there'll be more to come on some of that kind of support as well. We're also developing training on implicit bias, and Mike Curry and I were talking about this earlier. United Health Group just unveiled an implicit bias training module that's available to anybody. For physicians, you can get three and a half hours of CME, this is desirable, but we also want to develop one that is specific to maternal morbidity and mortality. One of the tragedies, I think, of many of the stories that we've heard, not only this morning, but that we've heard in the lay press and all has been that women simply don't feel that they're being heard. Families don't feel like they are being heard. And often that's because of, deficiencies in how the system is set up, but sometimes it is they don't know the words, they don't know the terms, they don't know how to advocate for themselves, and so that is something that we are going to, as I said with the patient support, be building out, but also with the idea that we have to be sure that it is culturally sensitive and appropriate, and empowering. We have accountabilities as an organization, UHG does, to American College, to AIM, to CDC, and to HRSA as well. We appreciate they recently attended a followup meeting that we had to talk further about some of the things that we at United Health Group could do. We heard loud and clear from them that the remote monitoring was one thing, but we also heard about the call for advocacy, to advocate for longer periods of coverage. Many women, especially on Medicaid, may lose their coverage after six weeks, and so we want to- if they've had a complication in pregnancy, they may never really get the followup that they need for it. So there's lots of opportunities in advocacy as well that we are exploring. There were some things that we wanted to do, I mentioned the external things, but in parallel to that, to get our own house in order. We have a Women's Health Dashboard that we're able to use the commercial and Medicaid data on, and so we enhance some of the capabilities so that we were calling out severe maternal morbidity as well. We do have, as most payers do, a quality of care process where, if there's been an untoward event, anyone can refer a case in. A patient can call in and do it, a nurse at a hospital, one of our internal staff could do it as well, but we wanted to enhance that, so that people were more aware of some of the signals that we'd look for. I mean, obviously a maternal death, and I have reviewed plenty of those, is gonna rise to the top, but the woman who maybe has overwhelming sepsis from pyelonephritis, who's in the unit, who's 28 weeks pregnant might not necessarily. So we did some education around those types of criteria and all that we were using to refer cases in. We also formed a National Maternal Mortality Review Committee. It's an offshoot, and sort of under the protections, if you will, of the quality of care, and now when there's a maternal untoward event, it will be reviewed by an obstetrician gynecologist. And we've started some of that work. I think that there's gonna be opportunity to go deeper with that. Sometimes it may not be a "quality of care" problem, you know, all the right things, all the right boxes might have been checked, but there were problems before the patient ever arrived at the hospital and so we want to take a little more global look at the impact that external factors prior to admission may cause as well as the things within the hospital. One thing that we are also working on, and you'll be hearing more about, is what we're calling an after-action checklist. Dr. Migliori has said he wants us to be the Maternity NTSB, and it took me awhile to kind of get my head wrapped around that a little bit, but you know, the idea that, even within the safety bundles, there are actions there, but I think what we've seen, and I know what Dr. Riley was saying about some of their after-action reports, the way that these cases may be looked at may be very good in some hospitals. I'm sure they're great at Weill Cornell, but may not be so great at Smith General Hospital, or something like this, and if there were a standardized checklist or something that people could use. I mean, it's not legislated, but something that is a tool, we think that that would be helpful and we're going to be working on that more in 2020. We also found within our own house, we needed to do some executive level education and support for the initiative, and I'm happy to report that the response has been overwhelmingly positive. So with that, I will stop.
* Centers for Disease Control Monthly Morbidity and Mortality Report, May 7, 2019