Solving America’s Sepsis Crisis Using Sepsis Surveillance, Sepsis Alert, & Sepsis Detection
What you need to know about the recently released ‘Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock 2021’
By Tim Kuebelbeck, Chief Customer Officer at Ambient Clinical Analytics
At Ambient our mission is to help doctors, nurses, hospitals & health systems solve sepsis using innovative technology to drive bundle compliance and best practices while reducing alert fatigue and nurse and physician burnout. This whitepaper is a quick overview on the changes recommended by the Surviving Sepsis Campaign & the Society for Critical Care Medicine (SCCM) and our experience deploying technology designed to help hospitals achieve that mission.
According to newly released guidelines “The outcome of critically ill patients depends on timely application of critical care interventions in an appropriate environment. Outside the ICU, septic patients are typically seen in the emergency department (ED) and hospital wards. Delayed admissions of critically ill patients from (the) ED are associated with decreased sepsis bundle compliance and increased mortality, ventilator duration, and ICU and hospital length of stay.”¹
The above quote illustrates that in the pandemic world we find ourselves in, sepsis bundle compliance has become more relevant than ever, as has moving to the Hour-1 bundle. We have documented that there are six critical keys to solving sepsis, mostly focused on early sepsis detection, early intervention, and driving both the care and timely delivery of the bundle components using technology and automation. Those six keys are as follows.
In addition to the six key elements to solving America’s sepsis crisis, it is evident that providing a technology allowing for remote patient monitoring of a sepsis surveillance solution and if needed, tele-sepsis services, is critical. Shifting focus to automate sepsis best practices using innovative digital health technology to detect and treat sepsis, and to ensure that protocols don’t slip through the cracks, has become increasingly important. The COVID-19 pandemic has been devastating to our citizens, economy and incredibly challenging for our healthcare delivery system. This devastation is not limited to just providing care but has also stressed our physicians and nurses both emotionally and physically. With Covid, sepsis has been much more prevalent during the pandemic, making sepsis surveillance, sepsis detection, and a sepsis alert to drive sepsis bundle compliance more imperative than ever before.
About Sepsis
According to the World Health Organization, sepsis is a condition that affects over 30 million people globally every year and is responsible for some 6 million deaths. The Centers for Disease Control and Prevention report that 1.5 million Americans contract the disease every year and about a third of these patients die. Hospitals have access to the correct materials to treat sepsis, and with early detection and treatment delivered in the first critical 6 hours after diagnosis, these are, for the most part, preventable deaths. But the disease is also notoriously difficult to diagnose. The chronic under-diagnosing, delayed treatment and subsequent deaths are the motivation behind worldwide campaigns to improve sepsis recovery strategies.
In the United States hospital mortality for sepsis and septic shock are 14.9% & 34.2% respectively with over 250,000 deaths annually. Sepsis is the single most expensive condition treated in hospitals and accounts for over 13% of all hospital costs. Sepsis results in extended and extreme care, including long stays in the ICU, where patients need a high level of monitoring and attention. These requirements make it a particularly expensive condition to treat, costing the US $24 billion annually.
Surviving Sepsis Campaign Major Changes to Initial Resuscitation for Sepsis and Septic Shock
According to SCCM data strongly suggests that early treatment is associated with improved survival and that increased mortality is associated with any hourly delay in the management of sepsis. The biggest changes to initial resuscitation for sepsis and septic shock are twofold. Essentially changing what was the 3 hour bundle into a 1-hour bundle and additional guidance on which specific vasopressors to use, and in what order, for persistent hypotension. The logic to implement the first portion of this is very straightforward, in that we modify the 3-hour time window for the first 4 elements of the sepsis care bundle to a 1-hour window and modify Smart Reminders™ and Smart Escalations™ for those bundle elements accordingly. The second requires minor adjustments to additional logic on how to track and implement Smart Reminders™ and Smart Escalations™ for vasopressors. Since Sepsis DART™ already pulls and tracks all the vasopressors that are in the updated guidelines this is just a minor configuration setting. We have completed the required changes to Sepsis DART™ that accommodate the recent updates to the CMS tracking rules which is that any antibiotic on the antibiotic list "counts" for sepsis treatment.
Other Surviving Sepsis Campaign Changes to Initial Resuscitation for Sepsis and Septic Shock
The Hour-1 bundle should be viewed as a quality improvement opportunity moving toward an ideal state. For critically ill patients with sepsis or septic shock, time is of the essence. Although the starting time for the Hour-1 bundle is recognition of sepsis, both sepsis and septic shock should be viewed as medical emergencies requiring rapid diagnosis and immediate intervention.
The Hour-1 bundle encourages clinicians to act as quickly as possible to obtain blood cultures, administer broad spectrum antibiotics, start appropriate fluid resuscitation, measure lactate, and begin vasopressors if clinically indicated. Ideally these interventions would all begin in the first hour. Minimizing the time to treatment acknowledges the urgency that exists for patients with sepsis and septic shock.
Using The EHR for Sepsis Surveillance
Unfortunately, health systems continue to deploy non-FDA cleared EHR electronic sepsis detection solutions that have proven to be ineffective rather than deploy FDA Class II cleared total sepsis solutions. With the current health system model using EHR sepsis detection only, driving a U.S. national average of 49% bundle compliance for sepsis², it’s no wonder that health systems lose enormous amounts of money on treating septic patients and have less than acceptable outcomes. This means less than 50% of patients with sepsis receive appropriate treatment and signifies a massive failure in the current deployment model. The current strategy needs to be fixed. Clearly the EHR isn’t the answer. In the U.S. we claim to have the greatest healthcare system in the world, and for sepsis patients we are failing in over half of that population when it comes to sepsis treatment.
Recently a Healthcare IT News article about a JAMA Internal Medicine published and peer reviewed study has proven that EHR’s like Epic do not deliver acceptable sepsis performance and are likely a major contributor to the America’s sepsis crisis. At least half of all U.S. hospitals are sitting well below 50% for sepsis CMS bundle compliance and the better ones are achieving 60% compliance. The reality is that 60% compliance is untenable and obviously in academic terms is a D. Yet many hospitals and health systems are very comfortable with treating 6 out of 10 septic patients appropriately and continue to accept poor algorithm performance and care delivery for sepsis treatment when it is a problem that can be easily solved using technology to drive better sepsis bundle compliance.
Sepsis detection solutions that are clinically vetted and FDA cleared facilitate sepsis surveillance, accurate and early sepsis detection and flexible sepsis alerts while driving delivery of the sepsis bundle within the appropriate time windows. Many IT departments still believe that the sepsis detection algorithm alone is a silver bullet for solving sepsis. Clearly, based on how the EHR vendor solutions have approached sepsis, they too believe a single algorithm is the answer. Unfortunately, this is not the case, and it continues to drive poor sepsis care across the U.S. and fueling America’s sepsis crisis. Hospitals and health systems find this crisis difficult to solve because of the large amounts of time, effort, and money previously spent with EHR vendors on a flawed approach causing many systems to have several failed attempts to fix sepsis. This flawed approach is leading to high mortality rates and poor outcomes. The common thinking is ‘if only our sepsis detector had higher specificity.’ What they are really saying is they alarm with way too many false positives. False positives are a huge issue with EHR provided sepsis detection and sepsis surveillance. False positives make the nurses and physicians lose faith in the sepsis alert which leads to disastrous results for septic patients.
Think of it this way, you can’t use a sepsis detection algorithm developed for the ICU in the Emergency Department (ED). If you did, the sensitivity would be too low, and you would not cast a wide enough net with your sepsis alert, resulting in missing many sepsis cases in your ED. On the other hand, if you tried using a sepsis surveillance algorithm designed for the ED in the ICU, your specificity would be too low and result in a very high false positive rate, eventually ending in doctors and nurses not trusting the sepsis alert. Which is why it is vital to have the ability to use multiple sepsis alert algorithms in your sepsis strategy. So why would you trust your EHR vendor to solve sepsis with a single algorithm?
The financial results of using EHR provided sepsis surveillance, sepsis detection, and sepsis alert solutions have been devastating to hospitals and health systems. In a study on hospital related costs of sepsis, the median of the mean hospital-wide cost of sepsis per patient was $32,421 (IQR $20,745–$40,835), and the median of the mean ICU cost of sepsis per patient was $27,461 (IQR $16,007–$31,251).³ According to the Agency for Healthcare and Research Quality (AHRQ), the average payment for a stay across all payers is $9,700.⁴ That is an average loss of $22,700 per sepsis patient treated. There are several sources and articles over the years that prove this. Specific to sepsis treatment, according to a recent article in HFMA, small hospitals experience net margin losses of ~$9.9M and larger hospitals experience a net margin loss of ~$33.9M on sepsis treatment⁵. All this data adds up to further proof that EHR provided sepsis alert, sepsis detection, and sepsis surveillance solutions are damaging our economy, failing our patients, our hospitals, and our health systems. The lack of impact EHR vendors are having on solving sepsis is costing lives, limbs, and long-term health issues in patients and contributing to the financial crisis our health systems face in today’s economic environment.
Additional proof that FDA certified sepsis alert solutions are desperately needed to solve sepsis and that EHR’s are ineffective can also be found at the state level. The inability of EHR vendors to address sepsis surveillance with existing tools and lack of expertise when it comes to FDA compliance has never been more glaring that it is today. The Massachusetts Sepsis Consortium released their findings on Advancing Sepsis Care in Emergency Medicine in 2019. In that report, healthcare systems listed their challenges providing high quality sepsis care. 29.5 percent of respondents reported that the facility EHR system was a barrier to successful implementation of standardized screening and treatment protocols for a variety of reasons⁶. Sepsis detection is a tiny sliver of the entire sepsis care solution. CMS data have shown that since implementation, organizations that follow all the steps have significantly lower mortality rates for patients diagnosed with severe sepsis and septic shock.⁷ Your sepsis surveillance solution must be able to handle a different sepsis alert and sepsis detection algorithms for different locations. Just as important, within a single hospital location, your sepsis surveillance solution must be able to handle different sepsis detection algorithms in different units.
Sepsis Detection
There are multitudes of claims by companies to the efficacy of their sepsis detection in terms of sensitivity and specificity, but few have scientific, clinical, peer reviewed, and published studies to back it up. Ambient has several peer reviewed published studies with Mayo Clinic that scientifically prove the efficacy of our sepsis detection solutions. We have an end-to-end solution that is much more than just a sepsis patient tracker board, called Sepsis DART™. Sepsis DART™ was developed with Mayo Clinic, and it also includes an indicative sepsis alert that is configurable by each health system, and even configurable down to the unit level. Because different units have different sepsis detection requirements, we typically run a multitude of algorithms in each site we install at. We can start with the Mayo Clinic sepsis detection algorithm, but we also have other out of the box sepsis alert algorithms we have developed with other healthcare systems that we can utilize as starting points for you. Without the ability to run multiple sepsis alert algorithms, you will find yourself spending money, time, and effort on a sepsis alert solution without improving your sepsis bundle compliance or sepsis outcomes, leaving you in the same situation you were in the previous year.
Early Intervention (Sepsis Alert)
Above we have spent a bit of time talking about early sepsis detection with a sepsis alert, but it is also important to understand that with sepsis surveillance you must drive early intervention which is key to solving sepsis. With each hour that you don’t act, mortality for sepsis increases 7.6 percent. In fact, each hour of delay in antimicrobial administration over the ensuing 6 hours was associated with an average decrease in survival of 7.6%. Administration of an antimicrobial effective for isolated or suspected pathogens within the first hour of documented hypotension was associated with a survival rate of 79.9%. Making the Hour-1 bundle and a sepsis surveillance communication platform to drive early intervention a must for hospitals to implement.
Delivering Care and Quality Beyond the Bundle Elements
This brings us to delivering care and the sepsis bundle elements within the appropriate time windows. Beyond driving early intervention as described above, it is critical that your sepsis detection solution has sepsis patient tracker board functionality that provides automatic sepsis surveillance and at-a-glance awareness of patient status, enabling bedside and remote monitoring of large numbers of patients simultaneously. We also know that sepsis alert solutions must also employ Smart Reminders™ and Smart Escalations™ that directly notify the right care givers at the right time of a potential sepsis development in a patient and using those Smart Reminders™, continue to notify care givers until someone on the team takes the appropriate action. Unlike EHR’s, Sepsis DART™ does all of this and most importantly, graphically tracks the delivery of the sepsis care bundle (as appropriate for that patient) while providing both smart sepsis reminder notifications and escalations where elements of the bundle may be in jeopardy of not being delivered correctly or in a timely fashion.
CMS and Other Sepsis Reporting
Abstracting sepsis cases for CMS reporting is manually intensive and time consuming. Most hospital systems invest a substantial amount of nursing resource doing nothing but case abstraction and reporting. Having a database with 95% of the data you need in one place, and all with data points time referenced against each other, and a set of canned reports specifically designed to allow you to view and pull that information for CMS reporting, is not only important but often reduces very significant portions for a nurse analyst’s job. Typically 75% that most hospitals can cost justify a system with just the reduction in reporting time required for the analysts, never mind the lives saved, costs saved in LOS reductions and the quality-of-care improvements. In addition, the ability to generate ad hoc reports that can highlight things such as specific issues with bundle compliance, staff responsiveness, habitual alert silencing, and other associated process elements and patient health parameters can be a boon to driving overall quality improvement efforts across the care team.
Tele-Sepsis or Sepsis as a Service
This brings us to the newest innovation and maybe one of the most important for solving sepsis in every facility across the United States. Sepsis surveillance vendors need to provide Sepsis as a Service or Tele-Sepsis. With many hospital and health system organizations unable to fund the investment to create a centralized Clinical Control Tower environment or the inability to staff sepsis surveillance teams 24/7, it is important to pick a sepsis detection solution (like Sepsis DART™) that allows for Sepsis as a Service or Tele-Sepsis. Companies like Ambient can provide outsourced sepsis monitoring through their Clinical Control Tower technology. A services model for a tele-sepsis or sepsis as a service program should include the following:
Summary
It is evident that despite improvements in modern statistical techniques and machine learning approaches, few organizations have dramatically improved sepsis bundle compliance in the past decade. For those that have accomplished major compliance improvements, most have seen inconsistent performance on bundle adherence at best. The most effective way for hospitals to achieve long term and sustainable improvements in your hospital(s) are following the updated SCCM Hour-1 guidelines and using innovative sepsis surveillance technologies. Those sepsis surveillance solutions must utilize multiple algorithms for earlier and better sepsis detection and have automated communication that drives effective sepsis bundle compliance with your clinical staff. Sepsis bundle compliance induced more consistent and timely application of evidence-based care and reduced practice variability.
At Ambient, we know that recognizing and treating sepsis effectively, requires implementing the six critical keys listed earlier in this whitepaper as they are critical to sepsis surveillance, improved clinical outcomes, and reduced mortality for affected septic patients. The treatment bundle for sepsis is well defined, and when delivered comprehensively within the defined time windows, provides remarkable improvements in not only the survivability of sepsis, but also in reductions of total care and length of hospitalization for patients who acquire sepsis. Achieving the six critical keys require using a solution outside of your existing EHR vendor.⁹
Center for Medicare and Medicaid Services (CMS) Sep-1 Reporting
The Center for Medicare and Medicaid Services (CMS), which collects data in order to push for treatment improvement and ensure transparency, has made the 2017 hospital sepsis (SEP-1) compliance rankings public as of July 25, 2018. Today, information regarding sepsis treatment compliance is readily available to the public on the CMS Hospital Compare website.
CMS’s focus on reporting is to drive improvement in sepsis recovery through measurement of, and to provide transparency to, hospitals’ compliance with sepsis treatment protocols. Since several reports have now been released about high numbers of preventable sepsis deaths, CMS (and other global healthcare quality bodies) have developed and reinforced new strategies for sepsis diagnosis and treatment, as well as improved visibility to sepsis treatment results.
Though sepsis treatment compliance has steadily improved over the past few years, more needs to be done to further decrease sepsis mortality.
Thus, there is now an increased need for innovative information technology solutions for better practitioner support, more accurate reporting, and most importantly improved patient care.
Impact on Hospitals
It can be tricky to manage sepsis effectively, and even a slight miscommunication between healthcare professionals (due to moving a patient or shift change) can cause major inefficiencies and even death. A January 2018 report from the Society of Critical Care Medicine based on 478 patients showed that most SEP-1 failures stemmed from a failure to measure lactate levels or a lack of fluids. Using solutions like Sepsis DART™ help healthcare providers eliminate those failures. For the past few years, health agencies around the world have been pushing for stricter monitoring and enforcement strategies, which include automated digital health tools like Ambient’s Sepsis DART™. However, problems continue with diagnosis, treatment, and reporting because of different definitions of sepsis, information overload and, in some instances, “alert fatigue.”
Sepsis DART™
Ambient Clinical Analytics has developed an FDA Class II cleared clinical decision support tool called Sepsis DART which assists hospitals with sepsis diagnosis, timely delivery and management of treatment, and reporting.
Sepsis DART™ (Detection And Response Tool) is designed to analyze patient data and identify potential sepsis conditions early, offering medical staff the right information for detection, and using smart sepsis alerts, support tracking of the treatment process. It moves with the patient through different hospital services and environments, integrates with any EMR system, and is customizable for various institutional purposes.
Sepsis DART™ is an FDA Class II approved solution that was clinically vetted with Mayo Clinic that has a strong return on investment. One of the reasons Ambient is the only FDA approved end-to-end sepsis solution on the market today, is the FDA requires that vendors have the capability to customize algorithms, which we do, and it allows us to tune algorithms to fit any health system.
The Sepsis DART™ alert system monitors and communicates regarding all aspects of sepsis treatment bundles to the right practitioners at the right time, maintaining information on septic patients even between care locations and shifting staff. This reduces errors and omissions, as the entire care team understands on a real-time basis what treatment elements have and have not been delivered, and how much time is left to successfully complete treatment. Published research from Mayo Clinic shows that AWARE Sepsis DART provides a high level of sensitivity and specificity as well as improved compliance with sepsis treatment delivery guidelines. And because Sepsis DART keeps all of the pertinent data for each case in a single repository, all centrally available and correlated to the “time zero” of the sepsis event, the effort required to abstract and report on sepsis cases is substantially reduced and that alone, not including improvements in outcomes and CMS compliance, often cost justifies any system costs. We Fix Sepsis!™
Citations:
By Tim Kuebelbeck, Chief Customer Officer at Ambient Clinical Analytics
At Ambient our mission is to help doctors, nurses, hospitals & health systems solve sepsis using innovative technology to drive bundle compliance and best practices while reducing alert fatigue and nurse and physician burnout. This whitepaper is a quick overview on the changes recommended by the Surviving Sepsis Campaign & the Society for Critical Care Medicine (SCCM) and our experience deploying technology designed to help hospitals achieve that mission.
According to newly released guidelines “The outcome of critically ill patients depends on timely application of critical care interventions in an appropriate environment. Outside the ICU, septic patients are typically seen in the emergency department (ED) and hospital wards. Delayed admissions of critically ill patients from (the) ED are associated with decreased sepsis bundle compliance and increased mortality, ventilator duration, and ICU and hospital length of stay.”¹
The above quote illustrates that in the pandemic world we find ourselves in, sepsis bundle compliance has become more relevant than ever, as has moving to the Hour-1 bundle. We have documented that there are six critical keys to solving sepsis, mostly focused on early sepsis detection, early intervention, and driving both the care and timely delivery of the bundle components using technology and automation. Those six keys are as follows.
- End to End Solution. It’s not just about the algorithm. It’s about a solution that that detects, tracks, and ensures complete delivery of the bundle elements.
- Early sepsis detection. Sepsis DART’s™ configurable algorithms are a must so that the detection approach matches local needs and supports deploying multiple and different algorithms in different units. Multiple algorithms are a must when it comes to an appropriate sepsis alert.
- Early sepsis intervention. Directly firing a sepsis alert for a patient requires a provider to step in, review case, & declare that the patient does indeed have sepsis. To drive early sepsis intervention, automated communication platforms are a requirement for any sepsis surveillance solution.
- Timely delivery of all required sepsis care elements. When the complete care bundle is delivered within defined time windows, remarkable improvements in the survivability of sepsis and in reductions of total care and length of stay are achieved. The only way to ensure that the sepsis bundle elements are not in jeopardy of delivery failure is using an automated communication platform.
- Automated Communication. Sepsis DART’s™ functionality provides automatic sepsis alerts, at-a-glance awareness of patient status, and smart reminders to ensure bundle elements are delivered successfully which enables bedside and remote patient monitoring of large numbers of patients simultaneously via Smart Reminders™ & Smart Escalations™.
- Certification. Certifications such as FDA, CE Mark, & ISO demonstrates a sepsis surveillance vendor has met a rigorous standard of excellence which is extremely important to health systems and enhance their competitive advantage and more importantly drives better care and saves lives. Patients pay the ultimate price when an organization’s solutions lack the required certifications.
In addition to the six key elements to solving America’s sepsis crisis, it is evident that providing a technology allowing for remote patient monitoring of a sepsis surveillance solution and if needed, tele-sepsis services, is critical. Shifting focus to automate sepsis best practices using innovative digital health technology to detect and treat sepsis, and to ensure that protocols don’t slip through the cracks, has become increasingly important. The COVID-19 pandemic has been devastating to our citizens, economy and incredibly challenging for our healthcare delivery system. This devastation is not limited to just providing care but has also stressed our physicians and nurses both emotionally and physically. With Covid, sepsis has been much more prevalent during the pandemic, making sepsis surveillance, sepsis detection, and a sepsis alert to drive sepsis bundle compliance more imperative than ever before.
About Sepsis
According to the World Health Organization, sepsis is a condition that affects over 30 million people globally every year and is responsible for some 6 million deaths. The Centers for Disease Control and Prevention report that 1.5 million Americans contract the disease every year and about a third of these patients die. Hospitals have access to the correct materials to treat sepsis, and with early detection and treatment delivered in the first critical 6 hours after diagnosis, these are, for the most part, preventable deaths. But the disease is also notoriously difficult to diagnose. The chronic under-diagnosing, delayed treatment and subsequent deaths are the motivation behind worldwide campaigns to improve sepsis recovery strategies.
In the United States hospital mortality for sepsis and septic shock are 14.9% & 34.2% respectively with over 250,000 deaths annually. Sepsis is the single most expensive condition treated in hospitals and accounts for over 13% of all hospital costs. Sepsis results in extended and extreme care, including long stays in the ICU, where patients need a high level of monitoring and attention. These requirements make it a particularly expensive condition to treat, costing the US $24 billion annually.
Surviving Sepsis Campaign Major Changes to Initial Resuscitation for Sepsis and Septic Shock
According to SCCM data strongly suggests that early treatment is associated with improved survival and that increased mortality is associated with any hourly delay in the management of sepsis. The biggest changes to initial resuscitation for sepsis and septic shock are twofold. Essentially changing what was the 3 hour bundle into a 1-hour bundle and additional guidance on which specific vasopressors to use, and in what order, for persistent hypotension. The logic to implement the first portion of this is very straightforward, in that we modify the 3-hour time window for the first 4 elements of the sepsis care bundle to a 1-hour window and modify Smart Reminders™ and Smart Escalations™ for those bundle elements accordingly. The second requires minor adjustments to additional logic on how to track and implement Smart Reminders™ and Smart Escalations™ for vasopressors. Since Sepsis DART™ already pulls and tracks all the vasopressors that are in the updated guidelines this is just a minor configuration setting. We have completed the required changes to Sepsis DART™ that accommodate the recent updates to the CMS tracking rules which is that any antibiotic on the antibiotic list "counts" for sepsis treatment.
Other Surviving Sepsis Campaign Changes to Initial Resuscitation for Sepsis and Septic Shock
The Hour-1 bundle should be viewed as a quality improvement opportunity moving toward an ideal state. For critically ill patients with sepsis or septic shock, time is of the essence. Although the starting time for the Hour-1 bundle is recognition of sepsis, both sepsis and septic shock should be viewed as medical emergencies requiring rapid diagnosis and immediate intervention.
The Hour-1 bundle encourages clinicians to act as quickly as possible to obtain blood cultures, administer broad spectrum antibiotics, start appropriate fluid resuscitation, measure lactate, and begin vasopressors if clinically indicated. Ideally these interventions would all begin in the first hour. Minimizing the time to treatment acknowledges the urgency that exists for patients with sepsis and septic shock.
Using The EHR for Sepsis Surveillance
Unfortunately, health systems continue to deploy non-FDA cleared EHR electronic sepsis detection solutions that have proven to be ineffective rather than deploy FDA Class II cleared total sepsis solutions. With the current health system model using EHR sepsis detection only, driving a U.S. national average of 49% bundle compliance for sepsis², it’s no wonder that health systems lose enormous amounts of money on treating septic patients and have less than acceptable outcomes. This means less than 50% of patients with sepsis receive appropriate treatment and signifies a massive failure in the current deployment model. The current strategy needs to be fixed. Clearly the EHR isn’t the answer. In the U.S. we claim to have the greatest healthcare system in the world, and for sepsis patients we are failing in over half of that population when it comes to sepsis treatment.
Recently a Healthcare IT News article about a JAMA Internal Medicine published and peer reviewed study has proven that EHR’s like Epic do not deliver acceptable sepsis performance and are likely a major contributor to the America’s sepsis crisis. At least half of all U.S. hospitals are sitting well below 50% for sepsis CMS bundle compliance and the better ones are achieving 60% compliance. The reality is that 60% compliance is untenable and obviously in academic terms is a D. Yet many hospitals and health systems are very comfortable with treating 6 out of 10 septic patients appropriately and continue to accept poor algorithm performance and care delivery for sepsis treatment when it is a problem that can be easily solved using technology to drive better sepsis bundle compliance.
Sepsis detection solutions that are clinically vetted and FDA cleared facilitate sepsis surveillance, accurate and early sepsis detection and flexible sepsis alerts while driving delivery of the sepsis bundle within the appropriate time windows. Many IT departments still believe that the sepsis detection algorithm alone is a silver bullet for solving sepsis. Clearly, based on how the EHR vendor solutions have approached sepsis, they too believe a single algorithm is the answer. Unfortunately, this is not the case, and it continues to drive poor sepsis care across the U.S. and fueling America’s sepsis crisis. Hospitals and health systems find this crisis difficult to solve because of the large amounts of time, effort, and money previously spent with EHR vendors on a flawed approach causing many systems to have several failed attempts to fix sepsis. This flawed approach is leading to high mortality rates and poor outcomes. The common thinking is ‘if only our sepsis detector had higher specificity.’ What they are really saying is they alarm with way too many false positives. False positives are a huge issue with EHR provided sepsis detection and sepsis surveillance. False positives make the nurses and physicians lose faith in the sepsis alert which leads to disastrous results for septic patients.
Think of it this way, you can’t use a sepsis detection algorithm developed for the ICU in the Emergency Department (ED). If you did, the sensitivity would be too low, and you would not cast a wide enough net with your sepsis alert, resulting in missing many sepsis cases in your ED. On the other hand, if you tried using a sepsis surveillance algorithm designed for the ED in the ICU, your specificity would be too low and result in a very high false positive rate, eventually ending in doctors and nurses not trusting the sepsis alert. Which is why it is vital to have the ability to use multiple sepsis alert algorithms in your sepsis strategy. So why would you trust your EHR vendor to solve sepsis with a single algorithm?
The financial results of using EHR provided sepsis surveillance, sepsis detection, and sepsis alert solutions have been devastating to hospitals and health systems. In a study on hospital related costs of sepsis, the median of the mean hospital-wide cost of sepsis per patient was $32,421 (IQR $20,745–$40,835), and the median of the mean ICU cost of sepsis per patient was $27,461 (IQR $16,007–$31,251).³ According to the Agency for Healthcare and Research Quality (AHRQ), the average payment for a stay across all payers is $9,700.⁴ That is an average loss of $22,700 per sepsis patient treated. There are several sources and articles over the years that prove this. Specific to sepsis treatment, according to a recent article in HFMA, small hospitals experience net margin losses of ~$9.9M and larger hospitals experience a net margin loss of ~$33.9M on sepsis treatment⁵. All this data adds up to further proof that EHR provided sepsis alert, sepsis detection, and sepsis surveillance solutions are damaging our economy, failing our patients, our hospitals, and our health systems. The lack of impact EHR vendors are having on solving sepsis is costing lives, limbs, and long-term health issues in patients and contributing to the financial crisis our health systems face in today’s economic environment.
Additional proof that FDA certified sepsis alert solutions are desperately needed to solve sepsis and that EHR’s are ineffective can also be found at the state level. The inability of EHR vendors to address sepsis surveillance with existing tools and lack of expertise when it comes to FDA compliance has never been more glaring that it is today. The Massachusetts Sepsis Consortium released their findings on Advancing Sepsis Care in Emergency Medicine in 2019. In that report, healthcare systems listed their challenges providing high quality sepsis care. 29.5 percent of respondents reported that the facility EHR system was a barrier to successful implementation of standardized screening and treatment protocols for a variety of reasons⁶. Sepsis detection is a tiny sliver of the entire sepsis care solution. CMS data have shown that since implementation, organizations that follow all the steps have significantly lower mortality rates for patients diagnosed with severe sepsis and septic shock.⁷ Your sepsis surveillance solution must be able to handle a different sepsis alert and sepsis detection algorithms for different locations. Just as important, within a single hospital location, your sepsis surveillance solution must be able to handle different sepsis detection algorithms in different units.
Sepsis Detection
There are multitudes of claims by companies to the efficacy of their sepsis detection in terms of sensitivity and specificity, but few have scientific, clinical, peer reviewed, and published studies to back it up. Ambient has several peer reviewed published studies with Mayo Clinic that scientifically prove the efficacy of our sepsis detection solutions. We have an end-to-end solution that is much more than just a sepsis patient tracker board, called Sepsis DART™. Sepsis DART™ was developed with Mayo Clinic, and it also includes an indicative sepsis alert that is configurable by each health system, and even configurable down to the unit level. Because different units have different sepsis detection requirements, we typically run a multitude of algorithms in each site we install at. We can start with the Mayo Clinic sepsis detection algorithm, but we also have other out of the box sepsis alert algorithms we have developed with other healthcare systems that we can utilize as starting points for you. Without the ability to run multiple sepsis alert algorithms, you will find yourself spending money, time, and effort on a sepsis alert solution without improving your sepsis bundle compliance or sepsis outcomes, leaving you in the same situation you were in the previous year.
Early Intervention (Sepsis Alert)
Above we have spent a bit of time talking about early sepsis detection with a sepsis alert, but it is also important to understand that with sepsis surveillance you must drive early intervention which is key to solving sepsis. With each hour that you don’t act, mortality for sepsis increases 7.6 percent. In fact, each hour of delay in antimicrobial administration over the ensuing 6 hours was associated with an average decrease in survival of 7.6%. Administration of an antimicrobial effective for isolated or suspected pathogens within the first hour of documented hypotension was associated with a survival rate of 79.9%. Making the Hour-1 bundle and a sepsis surveillance communication platform to drive early intervention a must for hospitals to implement.
Delivering Care and Quality Beyond the Bundle Elements
This brings us to delivering care and the sepsis bundle elements within the appropriate time windows. Beyond driving early intervention as described above, it is critical that your sepsis detection solution has sepsis patient tracker board functionality that provides automatic sepsis surveillance and at-a-glance awareness of patient status, enabling bedside and remote monitoring of large numbers of patients simultaneously. We also know that sepsis alert solutions must also employ Smart Reminders™ and Smart Escalations™ that directly notify the right care givers at the right time of a potential sepsis development in a patient and using those Smart Reminders™, continue to notify care givers until someone on the team takes the appropriate action. Unlike EHR’s, Sepsis DART™ does all of this and most importantly, graphically tracks the delivery of the sepsis care bundle (as appropriate for that patient) while providing both smart sepsis reminder notifications and escalations where elements of the bundle may be in jeopardy of not being delivered correctly or in a timely fashion.
CMS and Other Sepsis Reporting
Abstracting sepsis cases for CMS reporting is manually intensive and time consuming. Most hospital systems invest a substantial amount of nursing resource doing nothing but case abstraction and reporting. Having a database with 95% of the data you need in one place, and all with data points time referenced against each other, and a set of canned reports specifically designed to allow you to view and pull that information for CMS reporting, is not only important but often reduces very significant portions for a nurse analyst’s job. Typically 75% that most hospitals can cost justify a system with just the reduction in reporting time required for the analysts, never mind the lives saved, costs saved in LOS reductions and the quality-of-care improvements. In addition, the ability to generate ad hoc reports that can highlight things such as specific issues with bundle compliance, staff responsiveness, habitual alert silencing, and other associated process elements and patient health parameters can be a boon to driving overall quality improvement efforts across the care team.
Tele-Sepsis or Sepsis as a Service
This brings us to the newest innovation and maybe one of the most important for solving sepsis in every facility across the United States. Sepsis surveillance vendors need to provide Sepsis as a Service or Tele-Sepsis. With many hospital and health system organizations unable to fund the investment to create a centralized Clinical Control Tower environment or the inability to staff sepsis surveillance teams 24/7, it is important to pick a sepsis detection solution (like Sepsis DART™) that allows for Sepsis as a Service or Tele-Sepsis. Companies like Ambient can provide outsourced sepsis monitoring through their Clinical Control Tower technology. A services model for a tele-sepsis or sepsis as a service program should include the following:
- Monitoring the sepsis surveillance solution using a clinical command center technology such as Sepsis DART’s Clinical Control tower
- Working with the bed-side clinical teams in identifying and treating sepsis patients after a sepsis alert fires
- Provide training for sepsis response teams
- Managing the smart sepsis alert and escalation process to ensure treatment is in compliance with CMS bundle targets
- Lead sepsis related quality improvement projects for the hospital
- Develop process standards that meet best medical practices and regulatory requirements for supporting sepsis detection treatment and corresponding training modules
- Set-up a staffing program to include clinicians that have experience in treating sepsis patients
Summary
It is evident that despite improvements in modern statistical techniques and machine learning approaches, few organizations have dramatically improved sepsis bundle compliance in the past decade. For those that have accomplished major compliance improvements, most have seen inconsistent performance on bundle adherence at best. The most effective way for hospitals to achieve long term and sustainable improvements in your hospital(s) are following the updated SCCM Hour-1 guidelines and using innovative sepsis surveillance technologies. Those sepsis surveillance solutions must utilize multiple algorithms for earlier and better sepsis detection and have automated communication that drives effective sepsis bundle compliance with your clinical staff. Sepsis bundle compliance induced more consistent and timely application of evidence-based care and reduced practice variability.
At Ambient, we know that recognizing and treating sepsis effectively, requires implementing the six critical keys listed earlier in this whitepaper as they are critical to sepsis surveillance, improved clinical outcomes, and reduced mortality for affected septic patients. The treatment bundle for sepsis is well defined, and when delivered comprehensively within the defined time windows, provides remarkable improvements in not only the survivability of sepsis, but also in reductions of total care and length of hospitalization for patients who acquire sepsis. Achieving the six critical keys require using a solution outside of your existing EHR vendor.⁹
Center for Medicare and Medicaid Services (CMS) Sep-1 Reporting
The Center for Medicare and Medicaid Services (CMS), which collects data in order to push for treatment improvement and ensure transparency, has made the 2017 hospital sepsis (SEP-1) compliance rankings public as of July 25, 2018. Today, information regarding sepsis treatment compliance is readily available to the public on the CMS Hospital Compare website.
CMS’s focus on reporting is to drive improvement in sepsis recovery through measurement of, and to provide transparency to, hospitals’ compliance with sepsis treatment protocols. Since several reports have now been released about high numbers of preventable sepsis deaths, CMS (and other global healthcare quality bodies) have developed and reinforced new strategies for sepsis diagnosis and treatment, as well as improved visibility to sepsis treatment results.
Though sepsis treatment compliance has steadily improved over the past few years, more needs to be done to further decrease sepsis mortality.
Thus, there is now an increased need for innovative information technology solutions for better practitioner support, more accurate reporting, and most importantly improved patient care.
Impact on Hospitals
It can be tricky to manage sepsis effectively, and even a slight miscommunication between healthcare professionals (due to moving a patient or shift change) can cause major inefficiencies and even death. A January 2018 report from the Society of Critical Care Medicine based on 478 patients showed that most SEP-1 failures stemmed from a failure to measure lactate levels or a lack of fluids. Using solutions like Sepsis DART™ help healthcare providers eliminate those failures. For the past few years, health agencies around the world have been pushing for stricter monitoring and enforcement strategies, which include automated digital health tools like Ambient’s Sepsis DART™. However, problems continue with diagnosis, treatment, and reporting because of different definitions of sepsis, information overload and, in some instances, “alert fatigue.”
Sepsis DART™
Ambient Clinical Analytics has developed an FDA Class II cleared clinical decision support tool called Sepsis DART which assists hospitals with sepsis diagnosis, timely delivery and management of treatment, and reporting.
Sepsis DART™ (Detection And Response Tool) is designed to analyze patient data and identify potential sepsis conditions early, offering medical staff the right information for detection, and using smart sepsis alerts, support tracking of the treatment process. It moves with the patient through different hospital services and environments, integrates with any EMR system, and is customizable for various institutional purposes.
Sepsis DART™ is an FDA Class II approved solution that was clinically vetted with Mayo Clinic that has a strong return on investment. One of the reasons Ambient is the only FDA approved end-to-end sepsis solution on the market today, is the FDA requires that vendors have the capability to customize algorithms, which we do, and it allows us to tune algorithms to fit any health system.
The Sepsis DART™ alert system monitors and communicates regarding all aspects of sepsis treatment bundles to the right practitioners at the right time, maintaining information on septic patients even between care locations and shifting staff. This reduces errors and omissions, as the entire care team understands on a real-time basis what treatment elements have and have not been delivered, and how much time is left to successfully complete treatment. Published research from Mayo Clinic shows that AWARE Sepsis DART provides a high level of sensitivity and specificity as well as improved compliance with sepsis treatment delivery guidelines. And because Sepsis DART keeps all of the pertinent data for each case in a single repository, all centrally available and correlated to the “time zero” of the sepsis event, the effort required to abstract and report on sepsis cases is substantially reduced and that alone, not including improvements in outcomes and CMS compliance, often cost justifies any system costs. We Fix Sepsis!™
Citations:
- Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock 2021, SCCM, Oct 4, 2021
- Castellucci, M., Just 49% of hospitals follow CMS’ sepsis treatment protocols, Modern Healthcare, July 27, 2018
- Arefian H, et al Hospital-related cost of sepsis: A systematic review. J Infect. 2017 Feb;74(2):107-117. PMID: 27884733.
- AHRQ, Center for Delivery, Organization, and Markets, Healthcare Cost and Utilization Project, Nationwide Inpatient Sample, 2010
- Kulick, D., et al., Sepsis poses a cost-containment challenge in the face of the COVID-19 pandemic, HFMA, July 20, 2020
- Massachusetts Sepsis Consortium, Advancing Sepsis Care in Emergency Medicine, April 2019
- HHS.gov, Largest Study of Sepsis Cases among Medicare Beneficiaries Finds Significant Burden, HHS.gov, February 14, 2020
- Kumar A, et al Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med. 2006 Jun;34(6):1589-96. PMID: 16625125.
- Jozwiak M, Monnet X, Teboul JL., Implementing sepsis bundles, NCBI, September 2016