 Hi, everyone. Thank you so much for joining. It is the top of the hour, so we're going to get started to ensure we have enough time for questions at the end. My name is Jordan Gamart. I'm the program coordinator at the Patient Safety Movement Foundation, and I'm pleased to kick off the early detection and treatment of subsist webinar today. Before we get started, I'd like to briefly address some housekeeping items. I've muted all of you so that there is no background noise. If you have any questions, we do encourage you to utilize the chat feature with any questions you have during our virtual presentation. Our expert presenters will address them at the end. Our agenda today will be a five minute overview of our foundation, the groups we work with on a daily basis and our actionable patient safety solutions. Our experts will be presenting for 40 minutes and we'll be finishing today's workshop with a 15 minute question and answer segment. Just to begin, the Patient Safety Movement Foundation's vision is zero preventable patient deaths in hospitals by the year 2020. We realize that this is an audacious goal, but we believe that zero is the only acceptable goal to have because one preventable death is one too many. With that said, we strive to foster new efforts and build on existing patient safety programs through the work of our partners and their commitments to zero. We want to take a fresh approach without reinventing the wheel. So who can take action? We work with many groups. The first group we encourage to take action is our hospital and healthcare organizations. So we ask hospitals and healthcare organizations to make formal commitments around an initiative or program they're most proud of. These commitments are publicly posted on our website to create a shared learning network among all of our organizations who want to get involved and to improve patient safety. Today, we have over 3,500 hospitals across 43 countries have joined our network. And in February of this last year, we announced through their commitments that collectively saved 69,519 lives. Another group we have are committed partners. These are key key associations, societies and nonprofits. We ask our committed partners to sign a commitment to action letter, which announces their support of the patient safety movement by taking action, which is both individualized and sustainable. And this includes some major organizations that have joined us our global substance alliance. The third group we work with are healthcare technology companies. So we ask that medical device companies sign our open data pledge. This is simply an agreement that states they'll share their data openly without interference or charge. To date, we have 84 companies who have signed the open data pledge. A select few of these companies are GE, Phillips, Sturner, Drager, Massimo, Metronic, Oracle and IBM Watson to name a few. And our last group are patient and families. We ask patients who have lived to tell their story or families who may have lost a loved one because of a preventable medical error to share their stories. And these serve as not only an opportunity for those to learn from, but to work to ensure mistakes like those could never happen again. We have over 50 written stories and several patient story videos, which can be found on our website. And we film these stories each year. In our last group are policymakers. We also function as a 501c4. We do some work on Capitol Hill to increase awareness and promote patient safety legislation. So these are our apps are actionable patient solutions or apps are best practices or we like to call them recipes to the leading patient safety challenges our hospitals face today. Our apps are available for free and can be downloaded on our website. We truly value our subject matter experts like our subsist experts speaking today that help us identify the evidence based practices around each of these challenges that we have in an order to hold zero by 2017. Our next milestone is to save 150,000 lives both US and internationally. Last year we had reported over 69,000 lives saved through the commitments made by our partners and we only hope for this number to grow by 2020. Now on to today's experts we have Sarah fallen McManus, who currently serves as an advisory for her for the subsist alliance. She worked as a critical nurse care nurse for more than 20 years and recently retired from GE healthcare after 20 plus years as their clinical program manager. There is a patient safety movement foundation subsist ambassador and regional network chair. We have Dr. Ryan Arnold, who's a practicing physician special who specializes in emergency medicine and is the research director for the department of emergency medicine at the Christiana care health system in Delaware. He also serves as senior clinical investigator at the value Institute within Christiana care. Dr. And then lastly we have Dr. Chris fee. He's an associate professor of clinical emergency medicine at the University of California San Francisco. He's mentored medical students in residence and quality improvement initiatives and research and he serves as emergency medicine residency program director. And he's also the chair of our subsist work group. So let's get started with our first speaker Sarah McManus. He's going to lead us through the next slide. The video. I would like to share this brief video of Dr. Marnie go back and her son Zachary. In 2014 she found herself on the patient side of things when her healthy 11 year old son developed severe sepsis due to an aggressive infection in his femur. It was a regular Wednesday. We met Zach at the game. And after the game he started to cry. And the first thing he said was my knee hurts. I have a headache and I'm busy. Over the course of the next four days he became more and more ill. And on Saturday evening we took him to the emergency room. They immediately started antibiotics. And finally by that evening we were told that he needed to be put into a medically induced coma to be put on a respirator. And at that point the diagnosis of sepsis was confirmed. He stayed that way for 12 days and had multiple complications. I kept thinking am I going to be a mom who loses a child. It can happen to your parent. It can happen to your house. It can happen to you. So we need to know what it is and what to look for and to learn more about the symptoms. Go to sepsis.org. Thankfully Zachary survived. He spent almost a month in the hospital and six surgeries on his right leg. After leaving the hospital he spent eight additional weeks in a pediatric rehabilitation facility gaining strength and mobility. Today Zach is essentially a normal 14 year old. He has a few complications including nerve damage which causes a mild limb. But he does enjoy running and playing in the park with his friends. So what is sepsis? It is the body's toxic response to an infection. It kills about 258,000 people every year in the US. That's more people than by from breast cancer, prostate cancer, and AIDS combined. In the hospital setting sepsis is the leading cause of death and contributes to one in every two to three deaths. Of these deaths the majority of patients presented to the hospital with sepsis. When you have an infection which can be bacterial, viral, fungal, or parasitic your immune system works hard to fight it. Sometimes it can fight the infection on its own and other times it needs help with drugs like antibiotics, antifungals, or antiviral medications. For reasons not understood sometimes instead of fighting the infection your body starts to attack itself. The response to the infection is sepsis. The body develops a systemic inflammatory response which causes diffuse endothelia disruption and micro circulatory defects. This results in global tissue hypoxia and organ dysfunction. When multiple organ dysfunction and refractory hypotension occurs this is septic shock. People may refer to sepsis as blood poisoning but that term isn't accurate anymore. Sepsis is not an infection and it's not contagious. It's your body's reaction to an infection. Unfortunately it's not uncommon to hear that a patient died of complications of pneumonia or complications from an infection when in reality the person died from sepsis. It is also not rare. In fact about 1.6 million people in the U.S. develop sepsis every year. Sepsis does not discriminate and anyone of any age can get sepsis. There are those at a higher risk including people with chronic medical conditions such as diabetes, lung disease, cancer, kidney disease. Also people with weakened immune systems which can be caught taking medications like steroids or chemotherapy are also at more risk. The very young are also at higher risk because they don't have a fully developed immune system yet. On the other end of the spectrum the very old are also at a higher risk and if you've had sepsis before you may be at a higher risk to develop it again. And the myth really is it's not going to happen to my family. So some of the complications after a person does have sepsis, one of the greatest unknowns is the number of sepsis survivors. Unfortunately the treatment needed to save people from dying from sepsis can cause long-term consequences such as post-sepsis syndrome, PTSD, organs not working properly and even amputations. Post-sepsis syndrome is a condition that affects up to 50% of sepsis survivors. They are left with physical and or psychological long-term effects such as insomnia, difficulty getting to sleep and staying asleep, nightmares, hallucinations, panic attacks, disabling muscle and joint pains, extreme fatigue, poor concentration and decreased mental or cognitive functioning, loss of self-esteem and self-belief and just a change in the life that they used to have to the life that they do have. One of the survivors that I met was in grad school and had a young child at the time when she developed sepsis. She was treated in a timely manner in the emergency department and was admitted to the medical floor when her blood pressure responded so she avoided the ICU stay. She recovered and went home but the post-sepsis syndrome symptoms were debilitating. She had difficulty with memory and couldn't concentrate. She also had severe fatigue and joint pains. She was forced to drop out of school and struggled to get through her basic daily routine. Similarly, there are many survivors who are diagnosed with PTSD following the treatment. In severe cases, amputations are required after surviving sepsis. The medications given to keep the blood flow going to the major organs can cause tissue death in the extremities. Often this happens in the fingers and toes first. Many live with disabilities for the rest of their lives and feel they are alone and that no one understands how they're feeling and what they've gone through. Some of the challenges are just the fact that the symptoms are so subtle. Many times hospitals have patients present with such big symptoms and this may delay arrival to the emergency department, detection at triage, or recognition during the hospital admission process. Frequently, symptoms of sepsis mimic less severe conditions. It's not uncommon that you hear somebody that ends up with sepsis maybe was fast-tracked because they basically felt like it was something minor. The symptoms can include shivering, fever, feeling cold, extreme pain, general discomfort, pale, discolored skin. There may be a rash that won't go away when you press on it, being sleepy, difficult to arouse or confused, feeling like the person might die and shortness of breath. And these can be quite vague. Some of these, maybe you have a clue. It's when somebody has the combination of these symptoms along with an infection that that sepsis should be a concern and ask, have you considered sepsis? With sepsis, often a patient's condition escalates quite rapidly. In most cases, no one hospital department owns sepsis, which can contribute to the delays. Some communities also have limited sepsis identification and management resources in their hospital facilities. Now you'll see here the plot on the left shows an increased proportion of patients with severe sepsis. As they progress to such a shock, antibiotics are delayed. And on the right, the odds of dying among patients with septic shock are increased as antibiotics are delayed. It's found that mortality increases about 8% every hour without antibiotic administration. So what can you do? You can first find out at your own hospital, you know, where do you start? You can ask what your hospital's sepsis mortality is today. What is your hospital blood cultures, antibiotics, fluids, vasopressors, and repeat-lactic? Also, does your hospital utilize an integrated electronic medical record for sepsis identification, surveillance, or in with the clinical workflow? The incidence of sepsis is startling, but up to half of sepsis deaths could be prevented by timely recognition and treatment, and many complications could be avoided. The actual patient safety solutions or apps, early detection and treatment of sepsis is a key tool that all hospitals can use to improve patient outcomes globally. The components of the apps are listed with the executive summary checklist, performance gap, leadership commitment, practice and technology plans, patient and family engagement along with the metrics. And you'll see here where you can download it to get your own copy. And this is currently being updated, and we should have a version to replace it by the end of the year. Now, Dr. Arnold will mention some of the controversies that exist today. Thanks, Sarah. And I think one of the challenges that many that are on the front line of detecting and treating sepsis is realizing that there are discrepancies between even internal agreements among the treatment nurses and physicians on when the patient actually has criteria and has sepsis. Many of the initial definitions include service criteria and inflammatory response and how those relate to the current criteria. And with many of the then once identify what is the treatment recommendations to go forward, as many are aware of the recent trials that have addressed early gold directed therapy and dictated that perhaps there was not a benefit there. However, the challenge and I think the interpretation of those trials is important to understand that this is all in the context of a much more aggressive and attentive clinical team. So the next slide is to address this. I want to show that we have a ways to utilize this where we are now focusing our detection and messaging towards different bodies within the hospital system. So messages that are targeting our providers on the front line may be different than those to our administrators as well as different to our families and the patients themselves. And this is important, though, to have targeted information to each one of these groups as it gets you different results. The importance of moving towards providers understanding this is a more common agreement upon its currents and as well the disease severity. Whereas when hospital administrators are aware that you'll understand the sepsis burden within your system and better able to address how much resources can go towards this. When patients and families are aware now they're better able to detect and sometimes even make clinical teams aware of a patients deterioration and things more subtle such as mental status. And perhaps respiratory effort, things that families are oftentimes more attuned to than even our clinical teams. And as we move on you'll see that the concept of incorporating this to these groups will help your team understand and provide better resources. As you step back from all the discrepancies and disagreements and definition, one of the concepts we put together here is the idea of looking at sepsis on a spectrum and understanding that anyone with an infection at any time Please enter your access codes at risk of developing sepsis. And once your infection that a patient has whether it's viral or bacterial translates and transfers into organ dysfunction. Now you'll have you've met the criteria even at early subtle times for sepsis as an entity. So you'll whatever you your system is used you'll you'll notice that you can advance the slide will I'll show you that there's you have different aspects of organ to systems dysfunction. And when those happen, you now have incorporated different criteria with which your system identifies as patient has crossed that line and that could be an altered mental status that could be lab values of new auction requirements and new respiratory distress. As a family this idea of moving from the infection to sepsis is more better made aware when when everyone agrees and understands what those thresholds are. But the challenges for us on the front line is you have to have you might have different agreement on who has it so the next slide you'll see just some comparisons. When you might target a system for example compliance with the new CMS guidelines and what those identified, but you'll notice that that will omit by definition a certain criteria certain group of people. And moving on you'll see that maybe perhaps you're going with the new sepsis three definition and and using that as your definition however again understanding you will be omitting a certain group of people. There are also systems in place such as this that utilize sooner and they have their own such as alerting tool again which is useful and helpful but can be omitting and I think the message here is to understand that as a system you have to be aware of who you are targeting and what happens when you omit that group. So that's just where if you are targeting complete compliance with CMS you might be omitting a very important population that you still want your system to be aware of even if CMS was not as interested quote unquote in that group. So now moving on to what we've done in our system give you as an example on the provider side to improve this we've designed a sepsis alert process and this is driven as you'll see here by criteria that the clinical team once identified will initiate and what this does is we focus in on the alert process on the next slide. You'll see that we have a threshold to identify that when any clinician orders an antibiotic or antiviral you now have assessments that are guided for any degree of organ dysfunction. Once this occurs then you've met the criteria in our system for a sepsis alert. And what this does is on the next slide you'll see the response that this generates is a bedside huddle and this huddle is similar to huddles you guys may have experienced in your systems for a trauma, a heart attack or a stroke in which we dictate bedside discussions for various aspects important to sepsis care. Is there adequate IV access, what fluids are we going to be giving, is there adequate antibiotics in place and whether diagnostic tests are necessary. This huddle can be quick and reproduced with a very minimal effort as opposed to some of the other efforts we know that can take some time. The nice thing about this is it also alerts and labels the patient and makes everyone aware on the team from the nurse to the docs and the residents as well as to the family and the patient themselves. They are being labeled with sepsis and that actually helps in many ways or sometimes even hard to quantify in just the overall system wide awareness for this event. We also have an assist, you'll on the next slide you'll see if you can advance to a code status within our system and that dictates a higher level of severity. In this scenario you'll see the patients who have been provided initial care who persist to show abnormalities that are high severity, high risk of mortality such as low blood pressures and high lactates after fluids will dictate a repeat of that bedside huddle and a code will be established that brings now our ICU team in contact with a more direct comprehensive assessment for fluids that are at a level of interventions that we need to be given. This process again is one that is a much smaller select group of patients but that repeat bedside huddle is now further identifying to the team and the patient that the severity is much higher and dictates that clinical care that we think is so important. On the next slide you'll be seeing we'll be moving towards now a developing and this is something that can be happening both on paper as order higher fidelity within your EHR but we are working on systems where we are pushing information to our clinicians. Next slide please I'm sorry that we're developing where you are informing box that the patient needs criteria one more slide I apologize there one more. There we go that you'll see here that now in the context of giving patients antibiotics you are letting them know that your patient has these at these. Presence of organ dysfunction and that you know kind of pushing to them the idea that did you know your patient meet substance criteria can we label them for you. Now this push mechanism is take some cooperation and collaboration with it and design folks but you'll find that it drives compliance with a lot of these measures because you are now requiring a clinician to now put on on their order said yes I'm treating substance and here's a criteria I've met allowing them to be much more compliant and this is we've found a driven a lot of improvement in our in our cap for these patients. And then at the end you'll see the next slide that you can then go on to actually dictating what you want them to do with this information give fluids order blood cultures etc. And then to find to finalize here I just want to show you quick two quick examples that we've done have been integral in our cultural and health improvement here which is feedback so the next slide you'll see that part of the immediate feedback we've done is is starting to incorporate reports to our clinicians clinical team. These reports show even the next slide yeah this is a good example in this immediate feedback where you can be given on the day of your treatment. Examples to how your patient matter criteria and what things were performed or maybe we're indicated to be performed and we're not. This will include our, you know, a collective message to both our nurse our docs and our residents have involved that allows the team to understand that they are you kind of what was expected of them and where they maybe could have improved next time. And you'll find that case based reviews are integral for this kind of adoption. And on the next slide you'll see the one of a exciting new product we've started as a forecasting model, where we're trying to improve the awareness and capture the, the team kind of dynamic of understanding how sick these patients can be. And we're doing this in the way of a forecasting model. The objectives on the next slide I'll show you will elicit that this quality improvement effort is targeting an improvement in the in the system wide awareness but also doing that through gamification where we actually have the teams predict outcomes and then give them feedback as of those patients outcomes down the line. Now what's interesting about sepsis outcomes is they are generally delayed by several days to weeks, as you'll find with your sepsis patients when you track them. And you'll notice that on the next, you know, page that we've had that the resources we needed for this have been an extensive group of helping either sepsis nurses or our volunteer research team. But you'll see that on the next slide an example of what we asked them to forecast specifically targeting like what their patients how long they'll be there and the outcomes of those patients to include will they go to an ICU and have a higher risk of mortality. And then what the report that we generate on the next slide you'll see gives a very concrete feedback to the clinicians as to what they predicted and what was observed. And what is interesting what we found is that on the bottom you'll see that we have this reporting of the average scores of our nurses and docs and that they all kind of really enjoy comparing how they're doing to each other and to themselves. But it's also they really appreciate we found the feedback of knowing what they're what happened to their patient as those outcomes can be delayed by five to 15 days at times. So this has been a very integral piece for our success and something that we feel is very helpful in in driving this change. So we hand it off now to Dr. Fee who can go and describe some of the efforts they've had on their success over the higher fidelity tool at UCSF. Thanks. Thanks Ryan. Yeah, I thought it would be beneficial perhaps to some in the audience who have not yet instituted a sepsis screen through use of an EMR their facility who may be interested in doing so and learning a little bit from both our failures as well as our successes and ongoing tweaks to the system to continuously improve it. And I'm going to start back about 15 years ago when I first started as a clinician at UCSF and shortly after arriving here. It was clear to me and to others that we needed to develop a bit of a more robust interaction between us and our critical care. Colleagues in terms of implementing and improving the management of patients with sepsis that came through our emergency department and I think it's important to know that in our facility and this is not unique to UCSF is it's certainly true of a number of facilities but about 70 to 80% of all patients in this medical center who present. I should say who are ultimately diagnosed with severe sepsis or septic shock first present in the emergency department so if we're really talking about these critical interventions that have a significant impact on mortality. And if they're time based ie that the quicker you can intervene that the better your outcome will be then you really need to include your emergency department personnel. In these discussions and in these guideline development strategies so I was fortunate to walk into a good situation here with that was recognized and had great colleagues within the critical care world here. And I would say it failed. So we implemented a process whereby if a patient was identified in the emergency department as having septic shock I think we all pat ourselves on the back thinking that we did a good job of initiating fluids and giving antibiotics after getting cultures and initiating pressures and ventilation strategies if needed. It's easy patients to identify those are the sickest of the sick and they are fairly easy to identify it's the people who are kind of teetering on the edge sometimes or have somewhat more subtle presentations that can be tricky and those are the ones I think you actually to be honest with you have a better chance of rescuing if you will. But it's wholly dependent upon identifying them early on. And I say that 15 years ago when we had this joint program between us and ICU that it's somewhat failed and that it was difficult to in our system anyway to to really delineate who was responsible for what at what points should the patient be transferred to the to the ICU who was going to place the central line if it was necessary etc. It kind of swirled around that transition of care issue that we all know is a dangerous point for patients. So we kind of teetered along with that for a little while and then several years after that there was increased scrutiny in the institution from our own within, as well as from some external bodies we were participating in a demonstration project that had financial repercussions but it was really about improving the care for patients and that spurred on a new degree of vigor within the institution to really demonstrate that we could do something here and make a difference. And the really important thing here I think that I can't emphasize enough is that the, the, the institution at the highest levels are C suite or CMO or C EO our chief technology officer etc all back this 100% that means that they were supporting us with resources and not just lip service so they were financial resources as well as personnel that were dedicated to this project and I don't think we could have been successful without that. We, the institution implemented a big gathering of medicine in emergency medicine critical care etc, and brought in an outside consultant and that that last slide at the very bottom said a small test of change I'll never forget sitting in this room and they said you know what are you going to do. In the next several weeks as a small test of change to try to screen patients in your emergency department for sepsis, and they are tossing around ideas like, you know, on a paper based screen can you check for serves criteria at triage on 5% of your patients have your triage nurses do that. And here's where I think another lesson learned can't be emphasized enough and that is that you need to know your individual unit and I don't mean hospital or medical center I mean down to the individual unit so for me it was the emergency department. You need to know the culture of that unit. And I could tell this group that this small test of change and implementing a 5% patient volume screen on paper was not going to work. It was ad hoc. It was a change in workflow to introduce a new form. People were going to forget it, or they would fill it out and it would go, you know, sit on a desk somewhere. What we decided to do and you can go to the next slide was to take advantage of a homegrown very rudimentary ED electronic surveillance system. So our chairman at the time created a, like I said a very rudimentary electronic medical record, which was really just file maker probe based database which those of you know, those of you who know file maker probe might recognize some of the format here. This is just a database that included, you know, sort of typical stuff up at triage chief complaint triage vitals etc. But what we recognized was that the system itself we could take advantage of the fact that a computer is likely going to be smarter than us or at least more facile and not distractible the way we are. And we had the computer set up such that if you met two or more serves criteria at tree at triage which means all the vital sign based things a heart rate, temperature and respiratory rate that it would flag the triage nurse and say hey this patient meets serves criteria. Do you think the patient might have an infection if so please notify the faculty. And that's what our first step was. And this is what it looked like it was nothing fancy it was very kind of basics as your patient has two or more serves criteria and a suspected infection if that were true. And it kind of give some guidelines or guidance on what should be done and draw labs that are appropriate potentially blood cultures and consider and organ dysfunction looking for the other lab parameters and or exam findings that might confirm that and then you know consider giving fluids etc. So this is the version that popped up when the physician opened the note. You can go to the next slide. The problem with this system was that it was very rudimentary and it only included our triage vital signs so subsequent vital signs that that took place or recorded after the patient was moved into the room would not be captured by the system we it wasn't smart enough to do that. It was also the alert would pop up at triage only and when the physician first opened the note but again, as lab value data came in, like the white blood cell count or creatinine or liver function that could potentially indicate subsequent and organ dysfunction that wasn't recognized up front there was no way to set up the alert within that system to to alert our providers that that was the case. It identified people early if they presented with the signs at triage but we know that people don't come in with a sign on their forehead that says I have sepsis they may have normal vital signs or or not meet the criteria at our one but they may meet it at our two, three or four. And those patients had the potential to slips as it cracks. And it only as I mentioned it only appeared once that alerts so subsequent providers if there was a sign out or again if the critical of the clinical status had changed, we wouldn't be aware of it. So move on to the next slide please. Now we knew that when we were developing the system it was going to be on a trial basis because approximately three months after we initiated that. screening process which again as opposed to doing a 5% screen at triage on paper we went to the next day, changing the electronic medical record and screened 100% of patients and have been ever since. But three months after that we went to a fully integrated the MR and the fully integrated the MR had some power that our rudimentary version did not and that the main one is that it had the ability to screen not only the triage vital signs but all these subsequent vital signs that were inputted into the system so serial heart rates, blood pressures, temperatures, respiratory rates, etc. And today, take it one step further could also screen the labs as they were coming in. And so we had now had a system that we could build. Again with the support of our medical center and it folks that would excuse me have a continuous surveillance in the background. And so anytime a nurse physician or pharmacist in the ED entered the chart of one of our patients. If at any point during that stay in the ED, the person met criteria for two or more serves criteria, this alert would come up. And again, it would tell you what the, what the concerns were that your patient met service criteria if you felt that they had an infection it would suggest further lab values that you might want to draw as well as the interventions that you should put in place. This was purely if they had service criteria and you can move on to the next slide which shows what the alert looks like a slightly more obnoxious color, going from the, you know, the low level of alert, and the yellow color to the more significant. So the patient's really sick alert color, which has now not only do you meet service criteria but your patient has end organ dysfunction. So now you're talking about low blood pressure or, you know, creatinine abnormalities and elevated lactate, etc. And again, this will pop up, not only for our residents but also faculty nurses and pharmacists and, and the attend, you know, the attending as I mentioned. So, if somebody else saw it and they weren't the attending, they were instructed to go ahead and alert the attending right away. This was, this was not something that could be sitting and waiting to be addressed. You can move on to the next slide. As you might have picked up on the problem with this system, was it really dependent upon somebody going into that patient's chart. Now we're fortunate in the emergency department to be going into patients charts fairly commonly as we're checking on incoming lab values and checking on, or adding a new note to update status, etc. But the problem is if the clinical status changed, you know, vital signs, which again could be subtle, suddenly met criteria or if a white blood cell count count came back but nobody looked at it, and nobody got into the into the chart. They could a patient theoretically could meet criteria, but nobody knew it because they weren't looking into the chart. And so to address that issue we created a column in our in our sort of our virtual track board of where patients are, which you can see to the far right there if it had that yellow star in it that indicated that person met service criteria. If they had the severe alert, ie they had and organized function that star would turn orange. And if they were in, if the provider indicated that that person truly had severe sepsis or septic shock by clicking yes this person has an infection it would turn red. It would allow everybody in the department to know exactly the status of that patient. If the provider felt that the service criteria, or and and organized function was not due to an infection so I even not severe sepsis or septic shock. They would click on it and interact with that alert. And rather than having the star on the track board disappear it would stay gray, indicating to everybody in the department that yes we know this person has vital sign and lab abnormalities. It's due to infection but this person's still sick. And so it should be on our radar screen. We didn't want it to drop off entirely. So moving on to the next slide. You can see what has happened since we've implemented our electronic surveillance that the blue line there trend line is our compliance with the sepsis bundle of care. You can see how it has dramatically increased and it was in late 2011 where we implemented our screen, or I guess early 2012 where we implemented the, the more robust electronic screen you can see sort of the jump in our compliance level there. And then our subsequent or core correlated decrease in mortality in our patients and this is these, these data represent all of the patients at UCSF and I think it's really important to note that we in the emergency department were the first unit to go live with our screen. I mean, we have subsequently adapted our screening model to all of the units in the entire hospital. So the various wards, the ICU, etc. And just as I mentioned earlier how critically important it was to adapt to our low or our local culture as to what we knew our nurses would would and wouldn't do or like to do or how they would respond and fit into their workflow. And as well as our fish physicians and pharmacists, when we implemented and adapted the screening system into the other units in the hospital over time. It was critically important that we adapted the screening tools. For example, we have a large liver transplant service here and if we were using things like for example, to use as a screen for end organ dysfunction, they were going to get a tremendous number of false alarms or false positives. So we adapted some of the metrics or the screens that were being used to the individual units to try to optimize the sensitivity and specificity and to reduce the false positive alerts. The interesting thing and I'll end on this anecdote is that when we adapted these screening tools to our different units in the hospital, we oftentimes would run the system in the background but not have it live to the providers, meaning that we could see when people were seeing the various criteria and run reports and see how often that would happen and get down to the individual patient level and see exactly when the alert would turn positive. At the same time we could cross reference that patient and look in the records to see when a rapid response team was notified about a patient's deterioration and compare the time from when our alert would have picked it up to when clinically it was addressed. And it was a dramatic difference. We saw times of anywhere from a few hours up to 12 or even more hours earlier when our alert went off compared to when the rapid response team was notified. When the alerts went live on the other units, we saw a similar increase in compliance and a similar kind of stepwise decrease in our mortality from a medical center wide standpoint. So we think that there's a correlation there's hard to prove that it was the one and only thing because it wasn't. We had a lot of education efforts and review like Ryan was talking about. I think it's critically important, particularly early on when you are adapting and implementing a system like this that you provide feedback to your frontline providers so that they understand why the alert went off and understand how to interact with it. Next slide please and then I'm almost done. So the lessons learned that as I mentioned institutional support, excuse me for us was critical having the resources it and personnel time to address what we wanted to do was was absolutely critical we wouldn't have been able to have the successes we had without it. The initial education case review that Ryan had mentioned the timely feedback is really critical early on if you don't have that piece. People think you're implementing a system just to annoy them. You know what are these alerts why they keep going off. Why did my patient have this trigger. And if you can give them the immediate feedback pull the vital sign data in the lab data and say here's why it went off and you get the aha moment. It raises the level of attentiveness among your entire staff. For us a paper based screening tool on the wards as I mentioned wasn't particularly helpful but when we adapted the EMR surveillance system across the board so that it was taken out of the hands of providers who are busy and distracted. These little subtle changes in in clinical status could be more readily identified and sooner. And then again of course the importance of adapting to your local culture what works in our system may not work for you what works in our ED doesn't work necessarily on the floors. And then the next slide please. And this is just to summarize what we've discussed the steps is common obviously very costly, extremely high mortality with significant sequelae for those survivors and their family members and those who love them. Early identification and intervention saves lives we know this from a number of studies. It's really critical that local champions in not only the institution but in each department that touches these patients continue to cheer lead and to educate and to provide feedback to the providers otherwise your program will be destined to fail. And again one size does not fit all you need to adapt it to your local culture and what resources you have available in your setting I realize what we have done won't necessarily work at every hospital. And with that I will turn it back over to Ryan. Thanks Chris. And just wanted to highlight some of the partnerships and an exciting one with substance alliance for those that are unaware of this group. This is a nonprofit patient advocacy group that really helps with identification and awareness for for patients and families about not only science to be to detect and to be aware of substance when it's occurring but also resources that help them understand that that the community is much larger I think that people are aware previously as well as some of the sequelae in the recovery period that Sarah went through today. On the next slide you'll see some of the resources that sepsis alliance has and why this is such a useful tool and I'll tell you that I personally incorporated this in some of my patient interactions with these very handouts that are geared towards patient audiences with infographics that will be helpful within your department for providers but also the information guides and how sepsis relates specifically to different syndromes and conditions. Such high risk situations such as patients with cancer immune immune kind of suppressed states such as HIV. Another common scenario is the COPD patients and many others that you can see there. On the next slide you'll see some of the references and the opportunities within this including the video that Sarah played today is one of many that are very helpful for setting the stage for the importance of this and the importance of how this has spoken to our patients and families to understand what they have. You'll see on the next slide that there are some examples of some of these infographics and what they look like and they are very simple and really helpful for the patient and the family to understand why this label is being applied to them and how this is useful. What's interesting is that some of the feedback I've had is when a family member taking care of the patient who has pneumonia and they get told that now actually the patient has sepsis even though it's the exact same scenario that the patient was aware of now that that label has been applied to them they are much more tuned and aware of just being monitoring of how that patient does and asking harder questions which I think is very useful in a clinical setting because of that label now being applied to them. And I think it's been a useful tool that kind of puts some of these and if you think of it in the context of a cancer patient going through cancer and knowing the path forward for chemotherapy. Sometimes these are tools that help them understand that it can be a long road of both recovery and next steps in surviving. So some of these tools on the next slide you'll see just a few of the examples here that are very useful that really speak to specific instances about why these labels might be applied. For example, a patient with appendicitis and they might meet criteria for sepsis and sometimes that is a surprise in many of these patients. And so it's very helpful for them to see this and I've found a very positive interaction with my patients and their families with this. And next slide I think there's an upcoming. There's also this is one of the exciting, I'm sorry, one more slide back I apologize. This just showing you that the focus on the life after sepsis this has been a collaboration sepsis alliance with the CDC. And really emphasizing an aspect that doesn't often get talked about within the clinical setting because we are so focused on the immediacy and the response of the current condition. But really setting the stage and expectations for the patients of the recovery period for sepsis and how long this can take and what can be involved there. I think we're starting to understand better now, but it's something that has been overlooked. And I think the assumption has been that once you're discharged previously you're back to a normal state and that's far from true we're seeing now. And just for all the audience that are on the next slide you'll see an upcoming important webinar that is being discussed looking at antibiotic stewardship as a big advocacy from the CDC. And how do you align that with your sepsis efforts? Because remember the message of stewardship is the appropriate use of antibiotics with the timely de-escalation and not initiation in cases where they're not necessary. But how do you march that in combination with your sepsis efforts to be brought broadly applied to the patients when you identify them early? So it's a very useful and helpful tool and I think it's something that is worthy to head on in your institution because you'll have the efforts should be aligned and aware of each other's work because you don't want to have one counteracting the efforts to the other side. So I think that's it for today's slides and we'll now open up for questions and any specifics. So here's the website for sepsis alliance for those looking for those free patient advocacy items you can print out and give to them and our great tools as well to get access there. So welcome to questions now. Thanks Dr. Arnold. Just as a reminder please type your questions through the public chat interface. I know if you have if you join by by phone that'll be difficult for you but I'm trying to avoid putting everyone. I don't want to have any background noise if I unmute to everybody on the line. So if you could utilize that feature that would be great. We do have one question and how much I think this is to Dr. Fee. How much of the decrease in mortality at USF is due to an increased denominator increased identification of people with less severe subsist. Yeah, thanks for the question, Miguel. I think that's a really important and astute question. We certainly asked ourselves the same thing and when we looked at the, the overall morbidity within our medical center we did stratify and look at various ways of trying to risk adjust. And with all the models that we did mostly through our quality improvement team but also looking at the data that we report out to the. The UHC, the University Health Consortium. It looks like it's not a delusional factor that I suspect that there is some degree of that it's hard to tease out in the end. But it looks like even among the sickest of the sick cohort, the septic shock patients that our mortality did decline. So I think it's a little bit of a mixed picture that there's probably a delusional effect with identifying some patients who may not have been in the high mortality group. We have certainly made an impact on both and I think the other thing I'll just add to that is I don't care if it's a little delusional because obviously we want to pick up on those people who aren't quite as sick and prevent them from getting to that point. I think the question is a good one and it's a little hard to answer with any specific metrics, but in every way we've looked at it, it looks like the trend line is real. Great. Thank you. So we're going to go ahead and unmute all of the callers that have called in. So if you don't have any questions for the chat, we will give you the opportunity to ask your questions now. But if you don't have anything to say, please mute yourself so that we don't hear any background noise. You have one question from the chat. At what point does a septic hemorrhage rash appear in some patients with sepsis? I can address it to start. What's interesting about the rashes you'll see with the patiki or perpara, the hemorrhagic rashes that when you push them, they don't blanch, they don't go away, is that they can be anywhere in the course and they can sometimes see them at the end of a critically ill patient. But the awareness of them early on, when patients maybe don't have other manifestations or clear cut signs of it when you see that rash, I think that's the biggest utility when they push the awareness of those rashes are there. So while it can be variable in many patients and never occur in some patients, the importance is, as I understand it, if you see that in an early case of infection, a medical evaluation, a very, very quick evaluation would be very important because that could be a very early sign of a highly severe, highly deadly illness stage at that point. Great, thank you. Yeah, I would just echo what Ryan said there that unfortunately, the majority of patients won't develop that rash. So it's one of those things where it's lack of a presence shouldn't reassure you, but if you see it, it should be quite concerning.