Well, hello and welcome to today's TCAA webinar. My name is Tim Murphy. I'm the chairperson of the Education Committee, and I'll be today's moderator. It's my pleasure to introduce today's presenters. Michelle Pomfrey is a recognized expert in trauma registries and was the National Course Coordinator for the American Trauma Society Trauma Registrar Course. She has also worked with the American College of Surgeons, National Trauma Data Bank, Trauma Quality Improvement Program, and established Pomfrey Consulting in 2000. Vicki Burton is the Vice President of Education for Pomfrey Consulting, and she has over 20 years experience in the state and ACS level with trauma-centered verification and re-verification. And with that, I'll turn over the program to Michelle and Vicki. Thank you, Tim. It's an honor to be here, and thank you for the privilege of being able to present a little bit of information on hospital events. It's a tremendous topic in our trauma world. You know, is this really a hospital event, and why is it so important? So today, we're going to jump in and we're going to talk about this key topic in our world. We're going to look at the electronic medical record and some areas in it in which we are to find our information about hospital events. After we go through several of them, we're going to look at a couple of case scenarios to pull out some of the finer details of the definition in hospital events. We are going to look at it from a standardization of data abstraction and being able to validate our information so that we're capturing accurate hospital events. So as we start today, I feel that our first step is to build our foundation for quality benchmarking, and to do that, we first need to understand where we've been. So when we look at trauma care from a historical perspective, we see some key milestones. The Optimal Care of the Injured Patient, that book that we rely on so much, was first published in 1976. It outlined for the first time the equipment, personnel, and infrastructure that hospitals needed to provide high quality trauma care. In 1973, Cook County Hospital developed and implemented the very first trauma registry. It allowed them the ability to measure the care and outcome of their patients internally, and it set the stage for local performance improvement. In 1986, the COT made the trauma registry an essential element in the resource manual. In 1987, the ACS Committee on Trauma, known as the COT, established the Verification Review Consultation Program, or the VRC, and this program was designed to evaluate whether a center adequately meets the criteria described in the Optimal Resource Manual for the purpose of trauma center verification. The VRC provides the public assurance that verified centers meet these criteria for their required level of trauma care that they deliver. As the VRC grew, regional and state level health authorities often used the same verification designation system to organize their trauma systems on a regional level. Now, trauma centers were limited in their performance improvement opportunities because their sole focus was internal. While looking internally answered questions about the consistency of your care from year to year, it didn't really answer the question, are we as good as we can be? And the ability to answer that question came through the development of the Trauma Score and Injury Severity Score, or better known as TRIS. TRIS allowed centers to identify patients with unexpected mortality outcomes based on their probability of survival. That was an estimated estimation by a statistical regression module, so getting heavy into the statistics at this point. Then, in 1994, the COT developed and released the National Quality Assurance Improvement Program. Now, this program was a risk adjusted model that allowed centers to externally benchmark their surgical outcomes using standardized, high quality, clinically abstracted data coupled with valid risk adjustment statistics. Each center received a report which showed observed to expected predictions for hundreds of outcomes and relevant surgeries. This allowed them to identify areas of low performance, compare it to their peers, and target quality improvement efforts in those areas. Does that kind of sound familiar? Now, the NISQIP centers enrolled were able to demonstrate improved outcomes. The longer that they were enrolled, the better their outcomes were. And this was definitive evidence that external benchmarkings took quality improvement to a higher level. Now, the NISQIP program had some advantages over TRAMA. It was developed by the COT with a single database, a single data dictionary. It had standardized training for all of their abstractors, and there was only one platform, one inclusion criteria, one set of data elements, and every hospital abided by those. TRAMA, however, up to this point, had multiple TRAMA registry software vendors on the market, as well as each center determining their own data elements, their definitions to those data elements, where they were going to find them in the electronic medical record, and even their inclusion criteria. So, in 2007, a consensus of multiple stakeholders gathered together and the national TRAMA data standards were developed. In this, we looked at each specific data field, how it was structured, as well as defining it and providing a data source hierarchy to where to locate the information in the medical record. In this work group, we looked at data elements that were currently being collected by other databases, such as Nemesis or the UB92 data. And so, in the early data dictionaries, you could see these references down at the bottom. So, this 2007 work group tried not to reinvent the wheel, but to use standardized definitions that were already being utilized by other entities. So, in 2008 and 2009, the TQUIP pilot study opened up. There were 23 centers that participated. This provided data to the centers identifying high-performing centers so they could share their practice with the TRAMA community. In 2010, TQUIP became an open enrollment program and rapidly grew. In 2012, the TQUIP process measures were released. That was the first time that any additional information was being requested of these centers to be collected and reported to the national TRAMA data standards. Up until that point, it was the same standard data elements, data fields, that everyone in the country was to be using. Now, in 2021, there are over 850 participating centers. It has grown tremendously, and we see the value in TQUIP. So, TQUIP is successful because of three key factors. The first one is data collection. Every center that participates must adhere to the national TRAMA data standards data dictionary. This includes the patient inclusion criteria, as well as collecting all the data elements. Not only do you need to collect all of the data elements, but you need to follow the data definitions and the data source hierarchy as well. Now, with the data collection and the standardization of the patient inclusion criteria, all of this is an effort to get all of the TRAMA centers to the level playing field that the NISQIP centers were at, getting everybody to collect the exact same data on the exact same set of patients in the exact same format. That is what is going to add validity to the TQUIP program and validity to the benchmarking reports. The benchmarking reports, therefore, are how you are measuring your performance as a TRAMA center against your co-workers, co-centers down the road and across the United States. The second key factor to TQUIP success is data validation and data quality. Now, in the Optimal Care Resource Manual, they have recommended for years that we validate five to 10% of ARRAP records. This ensures that the data being submitted to the National TRAMA Data Bank and TQUIP is valid. Now, there are some misconceptions about data validation. Several years ago, the TRAMA vendors introduced a ITDX validator button, so to speak, in their software. This is the top-level data validation to ensure that the data is in the right format, but it doesn't necessarily have anything to do with the correctness of the actual data. So, for example, systolic blood pressure. The ITDX looks to see if there is a valid number in that data field. You'll receive a level two error if the number falls outside of the valid range, which is listed as 0 to 380. So, if there's something else in there outside of that range, level two error, this means that your entire data submission would fail. Now, if the number is above 220 and less than 380, you're going to receive a level three error. Now, this is a warning error that just says, is that really the right blood pressure? Are you sure that the blood pressure is 240? Now, if it is below 220, you're not going to get any errors. The ITDX will say, this is valid data. Now, if you have 80 in the systolic blood pressure field, that number is below 220, no errors, but the ITDX does not know if that blood pressure was really 80, 108, or 180. It only knows that those three numbers are within the valid range. That's where re-abstraction of 5 to 10% of your charts is needed to ensure that the data is correct and valid, ensuring or increasing the likelihood that your TQIP reports are going to be an accurate portrayal of the care that you're delivering. So, you're going to get good quality data and good quality benchmarking reports. Now, with that said, when we say 5 to 10% of your charts to be re-abstracted, that doesn't necessarily mean that you need to re-abstract every single data element. You can re-abstract the key elements of the chart that impact your benchmarking reports the most. And those fields are fields like the age of the patient, the mechanism of injury, whether it was blunt or penetrating, the initial vital signs in your hospital ED, because those vital signs are going to calculate your revised trauma score. You want to validate all injuries because that is your injury severity score. Those elements combined are going to give you your probability of survival. Now, if you've validated those elements, then your probability of survival is going to be accurate. And so, when you look at your M, your Z, and your W score internally, it's going to be based on good data. When you start to look at benchmarking externally, you want to validate your ICU days, your event days, your hospital events, and that's what we're talking about today, and your pre-existing conditions. Those are the fields that are going to have a large impact on your benchmarking reports. So, making sure that your data is valid. The third key factor that is attributed to the success of TQIP is data analysis. So, TQIP uses statistics called hierarchical linear modules, or the HLM module. This creates risk-adjusted outcome models. Now, the HMLA method produces odds ratios. That's what you see on your benchmarking report, which is represented by the OR. So, as a metrics, the OR looks at hospital performance. So, if your odds ratio is greater than one, then the odds of a hospital event occurring are higher than average compared to your peers. Likewise, if your odds ratio is less than one, the odds of a hospital event occurring are lower than average. So, let's dive into hospital events. So, what exactly is a hospital event? Now, for those of you that have been around for a while, it was previously known as a complication. Now, these complications, or secondary complications, can affect patients following their initial hospital admission. So, in general, a hospital event, a complication, can require readmissions down the road. It can lead to further medical issues. Now, one big thing is that it increases hospitalization times. So, it's going to increase the length of time that they're in the ICU and in the hospital in general. That is an important factor in benchmarking when you are looking at your length of stays compared to your peers. Now, it can temporarily or permanently attribute to mortalities and even to death, and it certainly increases your hospital cost. When the event occurs, different service lines may be called in to handle the event. Let's say it's an MI. You're going to have cardiology on board now. So, additional tests, additional procedures, that's what drives the hospital cost up. So, let's look specifically at the NTDS, National Trauma Data Standards, and how they approach hospital events in general. So, the very first thing to understand is that all of these hospital events, the symptoms, the onset of the symptoms must begin after they arrive at your hospital, your ED. Now, this is a very important factor to remember. Let's say a patient developed a PE at an outside hospital. If you put that PE in as a hospital event and submit that to ACS, then it appears that that PE occurred on your watch, and therefore, the evaluation of the care you gave the patient. When in reality, that hospital event occurred outside of your doors, your staff, your care, your procedures, your protocols, your guidelines had no effect whatsoever in preventing this hospital event. So you don't want to quote unquote take credit for hospital events that do not occur inside of your hospital. Now, the second part is in, throughout the National Autonomous Data Standards, the data definitions, you will see a full description of that data element. And they always include additional information. And so many times the additional information is kind of overlooked. We go, oh, I kind of know what that is. And we don't read below the description. So we want to make sure that you understand that the additional information is a very important part of the dictionary for you to follow and be familiar with. The event needs to be documented in the EMR. This is very important to remember because let's say you go to rounds and they talk about a PE and you go back and you put it in your registry and you submit that record to the college and you have a site visit. The site visit team picks that particular chart. They see that you've listed PE as a complication and they're reviewing your records and there is no documentation in the EMR that the PE occurred. That instantly has put your hospital at a disadvantage. Number one, you have been benchmarked against other centers with that patient having a PE. Number two, the site visit team that is reviewing your medical records, finding no documentation of a PE anywhere in the chart, then will question the validity of the information in your trauma registry in general. So let's look at some hospital events. There are literally hundreds of hospital events that may occur during hospitalization. Those hospital events will impact your length of stay. It may impact your outcomes. For the National Trauma Data Standards, however, they're only looking at 23 conditions or events. So let's look at acute kidney injury. Let's start there. So acute kidney injury is documented frequently by physicians, but that doesn't necessarily mean that it meets the National Trauma Data Standards definitions. The first rule that you need to understand is that with hospital events, it needs to meet every aspect of the data element description, including the additional information. So now let's jump into that. Acute kidney injury is often found in patients with major blood loss, septic shock, or an IV contrast dialogue. All three of these conditions can lead to decreased blood flow to the kidneys and lead to acute kidney injury. But seeing AKI documented in the chart isn't enough. The KDIGO, which is the Kidney Disease Improving Global Outcome Guidelines, defines it a little bit differently. So this is the starting point. So the KDIGO, the 2012 guidelines talks about an abnormal serum creatinine level, renal replacement, decreased urine output, or aurea. Now the NTDS goes beyond that, and they give us more information. Here we find that the serum creatinine needs to increase three times the baseline, so when the patient came in, or is greater than four milligrams per deciliter. It talks about initiating renal replacement therapy there in the hospital, or a decreased urine output, and they define that as being less than 100 milliliters per day. Now, if the patient, if the patient or family refuses treatment, such as dialysis, the condition still is considered present in the combination of decreased urine output, the serum creatinine, and so forth. Now, when you read the additional information, it excludes patients with pre-existing chronic renal failure, and on renal replacement therapy at home. So if you have a dialysis patient that undergoes a trauma, they are not considered acute kidney injury patients, because they were already on dialysis prior to their hospital event. This is strictly for patients that basically have normal kidney functions prior to their trauma, and then based on their trauma, remember decreased blood flow, decreased blood volumes, things like that, dialogue, they are now in jeopardy because their kidneys are taking a hit. Now, where do you look for AKI? Well, the first thing that you're going to review is the data source hierarchy, and that's listed in the National Trauma Data Standards for Every Data Element. So for AKI, we're looking at the HNP, we're looking at physician's notes, progress notes, case management, social service notes, the nursing notes or flow sheet, we're gonna look at the triage trauma flow sheet, we're going to look at the discharge summary. Now, again, this is the starting point. Somebody's gonna raise the red flag of AKI. Now, then we need to go and look in the lab section, because we have to verify that the serum creatinine increased to three times the base limit, or at some point in time was greater than four milligrams per deciliter. One of the best ways to determine quickly if the creatinine bumps is to graft or trend the lab values. This kind of gives you, excuse me, gives you the ability to click on a lab value, graft it or trend it, and you can see over a period of time where that lab value goes. Does it stay the same? Does it go up? Does it go down? And what this will allow you to do is to quickly see what the baseline serum creatinine was, where did it jump to? Where did it peak out at? Was that three times? Was it over four? And you don't have to individually go into every serum creatinine that was drawn. So it is a tremendous time saver when you are looking for bumps in lab values. Now, where else would you look? Now, you are looking for the definition too about a decrease in urine output or no urine output for greater than 12 hours. And you're also looking for any initiation of renal replacement therapy. So here you're gonna look at physician notes, the ICU daily note, because they may document dialysis, that the dialysis nurse came in or that the patient had dialysis that day. Now, dialysis doesn't necessarily have to occur in the ICU. It may occur on the floor. So you may look at the floor progress notes and the nursing flow sheet as well. The nursing flow sheet is going to give you the urine output. And they measure that, depending on where the patient is in their admission orders, if they measured every four hours, eight hours per shift and so forth, but there will be a tally for the 24 hours. So here's where you're gonna find this information to really hone in, did the doctor that documented AKI, did this patient truly meet all of the ACS requirements? Now, if all aspects are not true, it is not an AKI, even if the physician documents as such. So this is where it's important. Just because you see it in the chart, doesn't make it true. Let's look at another one. Let's look at unplanned intubation. This is kind of a big one for trauma centers and causes a little bit of confusion. So we see that unplanned intubation is an unexpected non-planned event. It's when the patient requires mechanical or assisted ventilation. Now the patient's going to show symptoms of severe respiratory distress, which includes labored or rapid breathing, or the patient complains of being short of breath or simply says, I can't breathe. They're going to be hypoxic or have low oxygen levels. They're going to be hyper carbonate or have high levels of carbon dioxide. Now this occurs because the respiratory drive and thus the lung function of gas exchange, breathing in oxygen, breathing out carbon dioxide, isn't functioning properly. Because of this malfunction, the blood gas exchange, we're going to see a respiratory acidosis. Now this means that the blood has become too acidotic or acidic. Now, if you check your lab values, you're going to check for blood gas results. You'll see an elevated PCO2 and a decreased pH of the blood. So where are we finding all of this information? Now, as with all hospital events, this has to occur at the patient's initial stay at your facility. Now, in the additional information, we also see information that addresses patients who were intubated in the field at an outside hospital or even in your emergency department. Here, when the patient is extubated, if they require re-intubation and greater than 24 hours, it is counted as an unplanned intubation. So where do we look? Again, we're going to follow the data hierarchy guide. We're going to look at the history and physical. We're going to look at the progress notes, the physician notes, case management, flow sheets, triage, and so forth. Now, for the intubation you want to, or for unplanned intubation, you also want to look at the respiratory flow sheet or a respiratory progress note. Here, you're going to find information about how the patient presented. What were the patient's signs, symptoms? You may even find the information about the blood gases, what the blood gas results were. If this was an emergent situation, you may see an event note, such as your response team, your rapid response team of the hospital. In that event note, you'll see what the patient's complaint was, and you will see documentation of an intubation. Now, if the intubation occurred on the floor, there may be an anesthesia note. They may have been called to do the intubation. Depends on how your hospital is set up. So with that said, you're looking at all of these different places. So let's jump in and look at a specific case study. So here we have a 34-year-old female, status post head-on motor vehicle crash, has admission orders written for the ICU. Currently, there is no bed available, so she is housed in the ED. 60 minutes later, she develops respiratory distress and requires intubation. Anesthesia responds to the ED and intubates her. What value do you assign for unplanned intubation? All right. So in this particular case, we had 67% say yes and 33% say no. Now let's look at the rationale. So in this particular case, the answer is yes. And the rationale for that is, the patient had admission orders written. So if you remember in your national trauma data standards, the minute that the orders for admission are written, that is the date and time used for ED disposition out. When the order, the date and time the orders were written. So therefore, in the eyes of the national trauma data standards, this patient is no longer an ED patient. This patient is an ICU patient, even though they're physically located in the ED. The respiratory distress was unexpected and thus is considered a hospital event. Now to further go with this one, you would check, now we've had documented respiratory distress. You would check to see what the blood gases were and see if there's any other supporting evidence just to be absolutely sure. But respiratory distress requiring intubation, this is considered a yes. Now let's try another one. Here we have a 65 year old male, status post T bone motor vehicle crash on the driver's side. GCS 12 on the scene. He was intubated and transported to your ED. On arrival, workup reveals a small subdural hematoma, bilateral rib fractures, small pulmonary contusion and a left hip fracture. The patient is admitted to the ICU. Hospital day three, he is weaned and extubated. 36 hours later, he complains of shortness of breath and is tachypneic. Blood gases reveal respiratory acidosis and the patient is intubated. What value do you assign him? All right, we have 88% say yes, 12% that said no. So the answer is yes. The rationale, this patient that was intubated in the field or the emergency department or those intubated for surgery, an unplanned intubation occurs if they require re-intubation greater than 24 hours after they were extubated. So in here, we see that on hospital day three, he was weaned and extubated. 36 hours later, he complains of shortness of breath. He was tachypneic. His blood gases reveal respiratory acidosis and therefore required intubation. So again, unplanned intubation. Now let's look at another one. Let's look at deep vein thrombosis. So here on this slide, you see the deep veins in the lower extremities. You have the popliteal, the perineal, the profunda, the common femoral, femoral anterior and posterior tibia. Now, just to understand a little bit about the blood flow in our blood vessels, we have like little one-way valves to keep everything moving in the right direction, one-way street, so to speak. So when a clot forms, the blood flow through that valve is lessened. It's narrowed, it's slowed down. So you're going to have fluid kind of accumulating, backing up. Now, if the clot breaks loose and starts to flow in that normal flow of traffic, then that's called an embolus. So with a deep vein thrombosis, it is a blood clot in the vascular system in those deep veins, typically in the lower extremities, but can also occur in upper extremities. Upper extremities are less common, but they can occur. When we are looking for clots, we're looking for venograms, ultrasounds, CT scans that are going to verify the DVT. Now, when you see these diagnostic procedures, you're going to capture them with their ICD-10 procedure code. So you're going to capture those procedures. Now, under the additional information, it says, of course, that it has to occur during the patient's initial stay at your hospital. Then it says, the patient must be treated with anticoagulation therapy and or placement of a vena cava filter or clipping of the vena cava. Now, a diagnosis of DVT must be documented in the chart and is typically confirmed with these diagnostic studies. Now, where do you look? Of course, you're going to still look at that data hierarchy guide. You're going to look in the progress notes and you're looking for the documentation of the DVT as well as the treatment. Now, you're going to check your imaging section to see, again, those venograms, dopplers, CT scans. Don't forget to put the procedure code in. So here is a venogram image of a DVT. So when they have the venogram done, the final report or the interpretation is going to list that a DVT was found and exactly where it was found. You want to always go back and make sure that the where is considered a deep vein because there may be a DVT in a superficial vessel. And if it is, that is not a DVT, deep vein thrombosis. So that is a key. You need to understand which vessels are considered deep veins. And those are the ones that we are looking for. You also want to review the medication section to see that the DVT was treated. Now, if they do a vena cava filter, an IVC, then you're going to look for that under the procedure section as well. So let's look at a case scenario. A 68-year-old female status post ground level fall, hospital day two, restarted on her home meds, which include Eloquist and Metformin. Post-op day number four, she developed swelling in the left leg. Dopplers revealed a DVT in the femoral vein. Patient treated with TEDS and STD sleeves, no change in medication. What value would you report for deep vein thrombosis? We have 42% that say yes and 58% that say no. This was a tricky one. So the rationale is no. In this scenario, the patient's home medications were started back on hospital day two prior to the DVT identification. So her home med of Eloquist, which is considered a blood thinner or clot preventing drug. After the identification, no additional anticoagulation was given, and there was no change in her home dosage. In other words, they didn't increase her Eloquist dose. It stayed the same. So therefore there was no treatment for the DVT. TED hose and STD sleeves are not considered treatment for DVTs. So treatment is a key element here. And one of the T-QIP monthly quizzes brought this out in the last year at once, if not twice, that it's one thing to have the DVT, but it's another to require the treatment. And so they honed in on the treatment element. So again, once you see DVT documented in your chart, you're going to look for how it was treated. So you may look for medication section of the chart. You want to go back and see what their home meds were. So if you're collecting your data correctly and you see under pre-existing conditions, you have captured anticoagulation therapy, then you know the patient was on some type of anticoagulation when they arrived at your hospital. Okay. Now in the national trauma data standards, there is a wonderful chart of what are the anticoagulations. It shows which are antiplatelets, which are the different categories. Now, it is a great idea to be familiar with what the brand name and what the official names are, the generic and the brand names, because this is going to help you look for these drugs as you're abstracting charts. So if you see that the patient has AFib, for example, you're going to be looking at how are they treating the AFib. And so a lot of patients are on Eloquist for AFib. So that is anticoagulation therapy. So understanding which drugs are anticoagulation is going to help you capture your pre-existing conditions correctly, but it's also going to give you this added information if the patient does develop a DVT and had been started back on their home meds. Now, let's look at how some of these hospital events can be intertwined. So remember the patient with the DVT. Now, again, it's formed a clot. And if part of that clot breaks loose and becomes free, it's known as an embolus. That embolus is then in the blood flow of traffic, so to speak, and the blood's going different places. So if that embolus is traveling through the bloodstream, when it reaches a vessel that is smaller or narrow that it cannot pass, then that's where it's going to create another problem. Here, if that happens in the lung, it is considered a pulmonary embolus. And if it goes on to the head, then it is known as a stroke or a CVA. So this one blood clot can lead to three or two additional hospital events. So let's look at that. Let's look at the PE definition. So here we see that it's a clot in the pulmonary artery, and it usually originates from a clot in a deep vein in the lower extremities or in the pelvic venous system. Okay. For this one, the additional information is vital to capturing the PE. Here, the very first thing that it says is that it must occur during the patient's initial stay at your hospital. This is not to be confused with a pre-existing condition, which they packed up and brought with them, quote unquote, to their hospital event, to their trauma or to your hospital. So basically what we're saying here, pre-existing conditions are things typically that the patient packs up in their suitcase and brings with them to their car accident, their slip and fall, whatever their traumatic event is. Meaning it is something that they have been living with prior to the trauma. Things such as hypertension, diabetes, dementia, COPD. These are all things that the patient is dealing with from a health standpoint prior to their trauma. There are patients that have PEs prior to their trauma. For example, lung cancer patients have a higher rate of PEs. So making sure that the event happens during the hospital stay. Now, now if the PE is seen on initial scans in the ED workup, those are not counted. Those are PEs that they brought with them. They may not have been dealing with them for months, but depending on what their traumatic event is, a blood clot can form in the time between the injury event and the time that they arrive at your hospital. So we are not going to count those. They were present on arrival. They didn't occur during their hospital stay. Now, another additional information thing here, we're going to consider it positive if they have a VQ scan. We're going to see it as interpreted as high likelihood probability, or it's a positive CT angiogram. Again, these are the tests that definitively diagnose the PE. And we're going to put in these tests as procedure codes. Okay. Now, one of the biggest missed additional information factors for PE is this one. It excludes sub-segmental PEs. And radiology is really good with identifying which ones are sub-segmental. Okay. So where do you look? Of course, you're going to first go to that standard data hierarchy. Then we're going to look in the progress notes. Now, in the progress note, we should see documentation around an acute onset, something happened. So with a PE, it's going to typically be a rapid onset of shortness of breath, difficulty breathing, could be pain in the chest. It's going to be an acute event. Now, we're going to look for follow-up imaging from this complaint. They're going to diagnose, was this a PE? So we're looking for these VQ scans, a CTPA angiogram. Now, because it's an acute event, you may have your rapid response team called. They're going to leave a note in the chart. Now, if this occurs on the floor, you may also be looking at an unplanned admission to the ICU. Depending on how bad the shortness of breath is, if they develop a respiratory acidosis, you may also be looking at an unplanned intubation. So again, how these things are intertwined, looking at what happened, this acute event, and then what stems from it, a PE that may cause them to be intubated, that may actually cause them to go into cardiac arrest, it may be an unplanned admission to the ICU. So looking at the puzzle pieces of how they all are intertwined. Here is a CTPA that shows bilateral PEs. Now, let's look at some others that are a little bit intertwined, and those are the infections. And as you know, an infection can extend the patient's length of stay. It also increases their mortality. So depending on what the infection is. So let's look at this. I love this drawing. You'll always remember it from here. Bugs versus drugs. Well, it's actually bugs and drugs. So this is a great resource for you. Most of the EMR systems on the market will have some type of antimicrobial stewardship report. Okay. This is an example from Epic. It's taken from an article repurposing antimicrobial stewardship tools in the electronic medical record for the management of COVID-19 patients. So it's not really a trauma patient, but it is just a good visual example. Here, this report gives a snapshot of the patient's temperature. Lab work, such as a WBC of white count. As you know, with infections, your white count goes up. It's going to also give you microbiology specimens that were collected. Was it a wound culture from the surgical site? Was it bronchial washings? Was it a deep surgical site swab? It's also going to give you the results. What grew out? Is it from their central lawn site? Is it a urine culture? What grew out then? What bug? And it's going to give you the name of the bug and usually or typically the colony forming units. So when you look at all of these infections, if you follow the algorithm, therefore the infection, it's going to tell you sometimes a colony forming unit number, because what they want to make sure is that it's not normal flora or it's not a mild infection, such as, you know, a low colony forming unit number. So you're going to see what those results are. Then you're going to also see in this snapshot what antibiotic was prescribed and when it was given. So this is a one-stop kind of shopping that's going to give you a lot of information related to infections. You know, their spike in temperature. Did their WBC go up? When was the culture drawn? This would be important for say your VAPs. If it was drawn the day they went on the ventilator, then that doesn't count because for a VAP, you have to be on the ventilator for greater than I believe two days. So when was the culture drawn, collected versus when did they start on the vent? When was the Foley inserted? When was the central line inserted? And so forth. There's some time elements in these data definitions that we need to follow, but this is just a one-stop shopping. So if you don't already know about this report, contact somebody in your EMR department, maybe even in infectious disease, and see what the report is called and how you can gain access to the report. Most of these reports are in the EMR and can be run by anybody or pulled up by anybody. It doesn't really require special access as much as knowing what the report is called. Because if you think about every department and every special report that's been written for each one of those departments and service lines, you have hundreds and hundreds of reports that have already been created within your facility. So it's learning what those reports are and then understanding how to use them and where to gain, you know, how to pull them up and when to look at them. Now, remember the slide from earlier, the success of PQIP is based on the three key factors. Well, we've kind of talked about the data collection end of it. We've looked for how the elements are linked. We've kind of talked about where to find them in the EMR. So now we really need to talk about that second key element, the data validity part. And so understanding in your daily process that you implement some type of data validation at a minimum when you close out the month. So what I mean by closing out the month, that's when you run that report, all the charts are closed, you've run the ITDX for that month, and you have no level 1 errors, you have no level 2 errors, you've looked at your level 3s, they make sense, and you kind of go, OK, everything's in a good format. Now what do we need to do to make sure that the data that's in good format is also good data to begin with? So let's look at data validation because it is one of the most important things as you start to look at your benchmarking. If you are reporting data that is untrue data, then you are being benchmarked with that false picture. Now that false picture can make you look really, really good or it can make you look really, really bad. But begin with your last benchmarking report. Are you a high outlier in any particular area? DVT, PE, VAT? If so, that is your starting point. Now we've talked about PEs. How many of you are a high outlier for PEs? Now think to yourself, how did you answer those questions as we were coming along about PEs? Do you put in PEs that were identified on the initial scans, meaning they showed up at your hospital with the PE? If so, you may be a high outlier because you have put those patients in. Have you included PEs that were sub-segmental? If you have, that may be the reason that you are a high outlier. In other words, it is not based on the care that you provide to the patient to prevent the PEs. It's based on you inaccurately putting in PEs that do not meet every aspect of the definition description and or the additional information. So how do you look at this on a monthly basis? Well, you can run a monthly report and see what your hospital events are. You can see every patient that had a DVT, a PE, a VAP, or any of the other complications or hospital events. So when you run that report, you want to verify that those hospital events meet every aspect of the definition. DVT, was it treated? The VAP, did it meet all of the parameters in that algorithm that the college provides for you? This means going back to the EMR and validating that information. That's going back, OK, when was the PE identified? If it was identified on arrival, take it out. If it was sub-segmental, take it out. Was the DVT not treated? Take it out. You want to ensure that you're following every aspect of those definitions. You can actually take the supporting documentation that you find in your EMR and copy and paste it into a PI form and have your PI nurse verify it. It's giving it an extra set of eyes. This allows you to be absolutely certain that the hospital events being transferred to the NTDB and TQUIP are a true reflection based on their extensive definitions of your hospital's care. That way, when the next benchmarking report comes out, you know beyond a shadow of a doubt that all of the information in there is absolutely correct. So now, if you are a high outlier, you have answered the question, is it the data or is it our care? You have now validated all of the data going across for hospital events. So it is no longer a data question. It is now a care question. Now, by doing this every month or at least on a minimum quarterly, your hospital events are going to be valid. Now, doing it monthly is going to certainly decrease the anxiety come data submission time. The other part to remember and to keep a close eye on, what are the dates for the next benchmarking report? And you can find these and follow these because it depends on where your hospital is with concurrency because your benchmarking report is going to be the prior 12 months of complete data. So believe it or not, the data deadline today is the last chance to submit data that will impact your spring 2022 report. In other words, what you've reported by today is going to be the data in which your spring benchmarking report will be run. So after today's lecture and putting some of these data validation processes into place, reviewing these definitions, getting on board with your entire staff, all of your registrars, PI, nurse, et cetera, the next opportunity that you will have to make an impact on the fall benchmarking report and understanding what the date range of your patients will be for that report. So in other words, your last 12 consecutive months of complete data, no matter what that last 12 months will be, is what you're benchmarked on. So that is the window of opportunity. You don't have to go back and change data from 24 months ago. You can, and it will make things more accurate from a national research standpoint. But that's a lot of time and resources to be invested in not affecting your benchmarking report. So in other words, keeping your limited resources, because we're all low on those right now, focused on the window of opportunity that's going to affect your benchmarking report. Now, back to this, again, familiar slide, let's look at this third key factor. And that is the data analysis and report. So this is really your benchmarking reports. So here, you'll see that the benchmarking reports, again, are risk-adjusted outcome models that use that HLM adjusted outcomes model. Now, your odds ratio of 0 is average. If your odds ratio is greater than 1, that means you have a higher than average likelihood of a hospital event occurring. Likewise, if it's less than 1, you have a lower than average odds that a hospital event will occur. So now let's look at this graph, or this box plot, rather. You see the median. You'll see the bottom first decile, which is in green. Green is good. You have the 10th percentile and the 25th percentile. All of these are good. The 75th and the 90th percentile are above the average. And then the top 10 percentile that you see is in red. Red is not good. A top 10 percentile means that you are a high outlier. So let's look at an actual report. So here we see the green, that this sample hospital is a low outlier or a good performer for all patients, blunt multisystem trauma, and penetrating trauma. See the green diamonds? All of the odds ratios are less than 1. So that means they're at a lower risk of a hospital event occurring. They are all in the bottom first decile, or at least the lower 25th percentile. The confidence intervals are all below the median. So this sample hospital then is an average performer in all the other patient cohorts. The diamonds are in the black, which is OK. That's kind of average. And the odds ratio is 1 or slightly above with the diamonds in or on the 75th percentile. If the sample hospital was a high outlier, the diamond would be in red and in that top 10 percentile. And this is from ResearchGate. So let's look at one of our last case scenarios. This is case scenario number four. We have a 48-year-old white male, status post motorcycle crash. Initial workup reveals a C3 unstable fracture, bilateral rib fractures, pulmonary contusion, PE. He is intubated. He has an open bulk pelvic fracture. He has a renal lack with an initial creatinine of 0.8. Day eight, the patient is diagnosed or documented with AKI. His creatinine is 3.9. His urine output is down to 90 milliliters per day. His chest X-ray shows consolidation. The patient has a fever and an elevated white count. Bronchial washing reveals 3 plus strep pneumonia. And ceftraxone was started. What hospital events would you enter? All right. 90 percent say B. And the answer is B. So there is a lot of information in this case scenario. The first thing you see is the PE. However, that PE was revealed on the initial workup. Therefore, he brought the PE with him to the hospital. We only want to capture the things that occurred in your hospital. We see on day eight, the doctor documents AKI. But we see that his creatinine is up to 3.9 from admission 0.8. So that is three times greater than the baseline. It also meets the criteria of decreased urine output because it's less than 100 milliliters of urine output per day. We see that the patient was intubated in the ED. Now, on hospital day eight, we see the consolidation, the bronchial washing, the elevated white count, the fever. So we see that he has strep pneumonia, which gives him antibiotics or started. So this is the documentation for the VAP. Now, the first thing as we close out that is so important is to review the National Trauma Data Standards Data Dictionary. Be absolutely completely familiar with all aspects of the data descriptions. You want to find the supporting data in the EMR and not always take things for granted at face value. You want to validate your hospital events to ensure accurate benchmarking reports. And when you review the data quality reports from TQIP, you want to follow up on any outliers that you may have. Now, when you look at this, you have a lot of resources available to you. As a TQIP participating center in the TQIP education, you have little mini recordings on the pre-existing conditions and the hospital events as well. If you still have questions after looking at the data description and the additional information, you can always go to these little recordings. They give you just a little bit more information, explanation, so forth in the little mini recording. So, for example, current smoker, when you look at that recording, it addresses vaping and smokeless tobacco, electronic cigarettes. That's not necessarily in print in the data dictionary, but they address it in that little recording. And each recording is anywhere from a minute to a couple of minutes. They're not very long, but it's a resource available to you as TQIP centers where you can find the additional direction if you need it. Looking in all of the aspects of the EMR, that is pivotal to making sure that you have what you need to support calling it what it is. Working with your PI nurses, your trauma program managers, even with your physician teams to make sure that the documentation is where it needs to be, that everything is there so that you are calling your pre-existing and hospital events accurately. The data quality report and the benchmarking reports, it is really, really important to work, registrars and program managers to work together on these. When you get your benchmarking reports and you see any outliers, to go to that drill down, pull out those patients, and then start to look at them. Were you a high outlier in PEs? What are your patients with PEs that were reported? Go to that drill down, look at them. Were they, now retrospectively because they're already coming out on the benchmarking report, were they true PEs? And if they are all true PEs, then the next step is looking at your care. But when you evaluate the PEs and they're not true PEs, then it may not be your care, it's your data. And that can be said for any hospital event. Just a list of the references from the articles and pictures and photos that I used for this presentation. I want to thank you for your attention. And I think we're going to open it up now for questions. Thank you, Michelle. Awesome presentation. And we do have several questions. So if I may, first question I have here is a patient is intubated in the operating room for a procedure and remains intubated for acidosis and the procedure is aborted. Is this patient or is this an unplanned intubation? Oh, that is a great question. When you look at the additional information for unplanned intubation, it talks about patients that were intubated for surgery. So the patient was intubated. Am I understanding that the question is the patient was intubated for surgery, but developed a respiratory acidosis while intubated and the procedure was canceled. Is that correct? That's my understanding. OK, so in the additional information for patients who were intubated in the field or the emergency department or those intubated for surgery, an unplanned intubation occurs if they require re-intubation greater than 24 hours after extubated. So if the patient was intubated for surgery, developed respiratory acidosis, surgery canceled, remained intubated, that is, in my opinion, not an unplanned intubation. Now, if they are extubated and have to be re-intubated because the respiratory acidosis returns, then that would be an unplanned intubation if it requires re-intubation greater than 24 hours. Great question, though. Awesome. Yeah, thank you very much. Another question here is it is suspected that the patient has sustained a pulmonary embolism but is too unstable to go to CT scan and therefore started on heparin for the treatment of PE. Do we consider that still a PE? That's another great question. I think for the PE, it is documented in the chart as a PE even though they don't have the scan. I think they're presumptively treating. They're presumptively treated. Under the additional information, it says consider the condition present if the patient has the scans and there's a high probability of a PE or a positive pulmonary angiogram or positive CT angiogram and then the last sentence says and or a diagnosis of PE is documented in the patient's medical record. So with that and or statement, I think that is the scenario that we're talking about here that when a patient has the rapid acute onsets of all the symptoms of a PE and they are too unstable to go to the scanner but they are diagnosed based on presentation and treated, then I would say yes, that is a PE. Great, thank you. Here's another one. Let's see. Unplanned intubation. So what if they're not initially intubated in the field, the ED or anywhere else but they need intubation and it wasn't anticipated? Okay, that's a great question. That one on the surface really falls under the unplanned intubation as a hospital event. The thing that the registrar needs to look for deeper is why were they intubated? You know, the intubation definition really talks so much about the respiratory distress. It talks about the acidosis and so forth. If this patient becomes you know combative for whatever reason and they're intubating them to keep them calm, that is I'm going to say not really the case because the unplanned intubation definition really revolves around respiratory distress. It revolves around being hypoxic, acidotic, and so if those are the reasons for this unplanned intubation in the ICU, then yes, unplanned intubation. But if it's for some other kind of, for no better term, oddball reason that's not respiratory driven, then I'm going to say no because this definition is really looking for the patients that get into respiratory distress. Great, thank you. Let's see, here's another. What if a DVT is diagnosed on ultrasound but the team does not think it is a true DVT and do not treat it and do not document it? For context, they give an example where they had a upper extremity or ulnar vein DVT under that situation. That's great and I think the person that asked the question kind of answered it themselves. They kind of followed that same algorithm of, you know, yes it needs to be diagnosed in the chart but it also needs to be documented that it was there and it needs to be treated. So they documented that they had a scan but the doctor doesn't believe it, the doctor doesn't document it in the chart, and the doctor doesn't treat it. So all three of those elements, you know, it's not a DVT. Great. Now here, even if it was documented in the chart and they do not treat it, it's not a DVT. So remember that was something that was brought out this year by the college pretty heavily is the treatment, the treatment of those DVTs. That's great and a follow-up to that is do you have a preferred reference or whatever to identify what are deep veins? You can Google it, you know, everything can be found on Google. That slide that I used was a great one for the lower extremities. I'm sure you can find one for upper extremities. Just like with Google, verify if you're, if you find a great picture printed out, put it on your desk, have it verified by your PI nurse or your, your team that yes, these are considered deep veins. And we all kind of get so focused on the word DVT and it's like, oh, it's a DVT. And then you start reading the report and it's really not a deep vein, but we get sucked into, it's there, it's documented. And then you kind of ask the question, like, was this treated? And they're like, it's not a DVT, it's not a deep vein. And you kind of go, oh, but yeah, find a great reference picture, print it out or have it, you know, bookmarked on your desktop, verifying with the team that those are the right vessels to be looking for upper, lower, and even in the pelvic region. Great. Here's one. If you have a trach patient that is going on and off the ventilator and a positive lavage comes back, does this patient count since they've been, not been on the ventilator for over two days? That one's a little tricky. I guess the thing would be, is that patient getting any type of ventilator support, such as a respite at night? So in other words, they, you know, they have a trach, uh, maybe they're just getting, you know, uh, you know, kind of like the face tent oxygen around the trach during the day, and they're doing really, really well, but at night they kind of go back on the ventilator, uh, to have a little bit of rest. Um, if that is the case, that's still considered a vet day because the definition is for ventilator days, any portion of a day is considered a day. So if you look at that definition for vet days and carry that over to that, then that is considered still on the ventilator, even though they're on respite. Now, if they've been off of any ventilator support for greater than two days, that's going to be a little bit trickier. That's going to be one to really, um, um, hmm. If they've been completely off of any ventilator support for greater than two days, I'm going to have to go, that's one to email, uh, the TQIP group about and let them weigh in on, because I don't want to say that one's, that one's pretty tricky, but if they're having any type of respiratory ventilator support at any portion of the day, that's still considered being on the vet. So the answer there would be, yes, it's a vet if it meets all the other criteria. If they're completely off of any ventilator support, um, um, I'm going to kind of lean towards no, but I'm thinking that would be kind of an unusual situation that they've gone all of the days on a vent, all the days with a trach, they've been completely weaned, and then they develop a vat would be, or an infection would be maybe a little bit less likely, but sorry, Tim, I don't have a definitive answer for you on that one. No, good. I like these thoughtful, provoking questions here. Um, one other here is for the severe sepsis event, does the patient have to meet CMS criteria, two SERS plus end organ dysfunction? And the, it says the NTDS data dictionary does not really specify what constitutes organ dysfunction. Right. So for the severe sepsis, you know, they are looking, well, for the severe sepsis, let's take out the, the organ dysfunction. Um, are they hypertensive? Are they hypo perfused? Um, is it affecting one or more organs? So liver, kidneys, heart, um, those would be looking at, I'm going to say yes, that that would be severe sepsis, even though the multiple or the, the, um, organ dysfunction criteria isn't, uh, explicitly spelled out. Um, that would be one that I would bump to your physician group at your hospital on a case by case basis. So for example, um, these are going to be your uber sick patients in the ICU. They're going to be on all kinds of pressures. Um, they're going to be on fluids. Um, a lot of things are going to be going on. They're going to have positive cultures somewhere. Um, because once you get a positive blood culture, um, that little infection, that little bug is in the bloodstream. So remember what we were talking about with the, the, the flow of traffic through all the vessels. So that little bug is going to go all through the body. So heart, lungs, liver, um, kidneys, brain, so forth. So when those little bugs travel everywhere and start to create havoc in the organs, your physicians are going to document the severe sepsis. They're going to document, you know, organ dysfunction. And if it's a case that you think is headed that way, um, by documentation, that is definitely one to, to pull in your physician group to verify is this severe sepsis. And you may even have to kind of copy and paste the NTDS definition in that email to them or show them in person, but you're getting their physician way in yes or no on the, on the severe sepsis. And I, I have another interesting one here regarding unplanned, uh, operations. Uh, it's kind of a two part, um, two situations. So if the patient goes straight to the operating room from the ED is packed, goes back to the ICU and then returns to the OR for further bleeding within an hour, is this an unplanned OR? And secondly, the second situation patient comes in with a brain bleed. Um, they try conservative measures for a few hours and then the bleed worsens and the patient's taken the OR. Is this an unplanned OR? Okay. Let's start with that first one. So the patient goes from ED to OR, um, they're packed, they go up to the ICU and within that short period of time, they have developed additional bleeding or severe bleeding and they're taken back to the OR. That one is definitely an unplanned, uh, return to the operating room because they went to the OR for bleeding. They had some procedures done. They were packed. They went to the ICU anticipating that they were going to be okay for a period of time until they could get them back to the OR, unpack them, close them. That's kind of that anticipation. But in that very short timeframe, bleeding occurred, which required them to go back to the OR. So that one I'm going to say in my opinion is, is definitely, um, an unplanned visit to the operating room because they're going back, um, to a related, um, procedure to what was initially done, you know, um, with the second example, I'm going to say that one is a no, that's not an unplanned because they know there's a bleeding going on in the brain. They are optimistic that it will stabilize and they don't have to go to the OR, but there is still that likelihood that the definitive treatment for that condition is surgical intervention. And so when the non-operative doesn't, um, uh, work, then they are going on and doing the initial procedure that they thought they might need to do. So I know that kind of is a kind of around the tree kind of thing, but the same would be, um, accurate about a splenic lac. They know that there's a splenic lac. They know there's bleeding in the spleen. They're going to try to non-operatively manage it. They're going to monitor serial crits. And all of a sudden one crit takes a, you know, a nosedive and they rush the patient to the operating room for a splenectomy. That splenectomy was likely from the beginning, if non-operative did not, um, work. Whereas with the first example, we've gone to the OR, we've surgically tried to fix something. We thought we fixed it. They're up in the ICU and what we thought we fixed breaks loose again or something worsens and we have to take them back to the OR, you know, for a, a very similar procedure. So that's where that unplanned, um, comes in because it talks about the definition is unplanned operative procedure or patients return to the operating room after initial operative management of a related previous procedure. So her second or his second example question, they've not had an initial procedure yet. Great point. And then perhaps for the last one, I'll combine a couple here, um, related to unplanned intubations, um, a couple of different situations also, uh, patient that's intubated for acute mental status change and inability to clear secretions, uh, as well as, um, just documented for airway protection. Um, they don't include, uh, any other, uh, explanation, but in those two circumstances, would they one or both be unplanned? So for those kind of going back to an earlier question, um, mental status changes, protection of airway, none of those things have to do with, um, respiratory, um, distress. So again, the definition is patients require placement of an endotracheal tube and mechanical or assisted ventilation manifested by severe respiratory distress, hypoxia, hypercardia, or respiratory acidosis. So dementia, mental status changes, protection of airways, secretion of, uh, clearing of secretions really, those are, are things that the patient is getting intubated for, but aren't necessarily respiratory distress related in which they needed to be ventilated. So for secretion management, you don't want them to aspirate. Okay. For mental status changes, they're too sleepy to, um, you know, clear their secretions or to take deep breaths or to stay well oxygenated. Those aren't respiratory distress. So I'm going to say in those examples that that is not unplanned intubation. Great. Michelle, thank you very much. A lot of really good thought provoking questions here. Uh, presentation was awesome. Um, any closing, uh, closing remarks or I just, um, we have a lot of resources available to us, um, within the electronic medical record, um, within infectious disease within our PI coordinators, um, using those resources, understanding, um, where they get their information from and any little special tidbits, like the bugs and drugs that they can shed on being able to capture data more accurately and more efficiently, you know, tap into them, but always remember that those additional, um, resources may not be looking at the same, um, definitions that you are, but it's a place to start. It's a place to kind of, Hey, I didn't know this bugs and drugs report, um, was out there. Now this is a great resource that I can use and interpret it my way based on the national trauma data standard definitions. Great. Well, thank you very much. Thank you everyone. Thank you, everyone for joining us. I hope you have a wonderful rest of your day. Great.

video

trauma registries

data abstraction

trauma care

TQIP

hospital events

documentation

National Trauma Data Standards

data validation

benchmarking

data analysis

classification

quality improvement