HealthSeptember 09, 2022

Evaluating which sepsis surveillance solution is right for your hospital

When it comes to sepsis surveillance, it's essential to know what you're looking for. Not all sepsis surveillance solutions are the same, and they can vary in cost, features, and usability. Before you embark on a search, identify your areas for improvement, your current challenges, and the priorities that your teams have in order to successfully adopt and use a solution to drive sepsis performance improvements.

Understanding how a sepsis surveillance solution works, key attributes, and how those can tie to your goals is essential.

Questions to ask when reviewing a sepsis surveillance solution

1. Does the surveillance solution support your hospital’s policies and care protocols?

Most hospitals rely on SEP-1, the National Quality Forum's standard recently re-endorsed.

2. Does it consistently search all hospital records for SEP-1 measures?

A real-time search of all hospital records enables early identification, which leads to early treatment. This is especially important given that each hour delay in sepsis treatment is associated with an additional 4% risk for death. Many systems look only retrospectively at sepsis performance when it’s too late to intervene on care.

3. Does the solution offer alerts that are both sensitive and specific?

Excessive alerts triggered by non-specific monitoring that picks up anything that might mimic early sepsis create alert fatigue, with all of its associated problems. As noted earlier, sepsis surveillance solutions now exist that have shown they can be both sensitive and specific.

Frequently, a hospital may rely on an EHR's generic sepsis tools and alerts to manage sepsis care. Unfortunately, the sepsis alert results are marginal at best.

4. Does the surveillance solution offer real-time monitoring of all prospective and established sepsis patients in a central view?

Separate from searching all hospital records, this more intensive monitoring is critical because real-time monitoring, including bundle compliance, empowers coordinators to get involved in active cases and change behavior as it's happening. The real-time tracking drives improved outcomes through rapid intervention in specific cases and its ability to prompt genuine teaching moments.

5. Does the solution provide detailed data for retrospective case review?

Being able to dig into how specific cases unfolded - when were sepsis criteria met, when was the care team alerted, and when were bundle steps administered - helps with identifying educational opportunities and can help abstractors understand past cases.

6. Does the solution drive accurate performance analytics that draws on all the data?

Having all of the data – including that unstructured data from clinical notes – allows sepsis coordinators to create trustworthy, easy-to-understand reports about hospital, unit, and individual performance, so all stakeholders understand where problems lie, why they exist, and, therefore, how to fix them. With the right solution, coordinators have all of the information they need at their fingertips to manage sepsis care throughout the hospital better and generate needed performance improvements.

7. Does the surveillance solution transparently draw on – and learn from – all relevant data?

When considering a surveillance solution, there are a few essential qualities in the algorithms that drive decision support recommendations.

As sepsis coordinators know well, SIRS (systemic inflammatory response syndrome) criteria drive many algorithms and alerts in current solutions. Without the use of artificial intelligence (AI) to draw on unstructured data, many current solutions typically have one or two critical flaws. Most suffer from inadequate specificity, thereby creating a slew of unnecessary alerts and corresponding alert fatigue, or they are slow to identify cases of sepsis, creating unacceptable lapses in initiating treatment. There is no doubt that AI will ultimately be necessary to create more timely and specific alerts that quickly and accurately distinguish between genuine sepsis and symptoms than merely mimic sepsis.

At the moment, natural language processing (NLP) is in the form of AI, which has the most robust track record in aiding early sepsis detection. Its power lies in its ability to analyze unstructured data, most commonly clinician notes, thereby considerably deepening an understanding of a patient's condition leading to accurate and timely sepsis detection.

Wolters Kluwer’s POC Advisor, which includes NLP, delivers highly sensitive and specific decision support to the point of care. More importantly, a combination of effective change management and that same electronic surveillance allowed one hospital to achieve a 53 percent reduction in sepsis mortality and a 31% percent reduction in readmissions due to sepsis — achieving the dual goals of saving lives and creating new levels of efficiency for the hospital, the effects of which ripple out positively in multiple ways – including a health system's bottom line.

Achieving accurate sepsis alerts goes beyond the algorithm

It is important to emphasize that part of the success of technology is tied to clinician trust. Because the genesis of clinical recommendations in POC Advisor is readily transparent to clinical decision-makers, they use and act on what appears on their screens. In contrast, some EHR sepsis solutions only offer black-box algorithms where the process is hidden from clinicians. The takeaway? When evaluating solutions for sepsis detection, demand transparency.

Driving sepsis performance goes beyond technology

Technology is one component needed to effectively address sepsis performance within your hospital. True and measurable sepsis improvement depends on three interdependent factors:

  • People: Ongoing training and feedback on sepsis care that makes clinicians aware of and holds them accountable for best practices.
  • Process: Creating efficient sepsis workflows and order sets throughout the hospital – not just the emergency department and intensive care unit – so dedicated clinicians have solid systemic support.
  • Technology: Identifying and implementing the right sepsis surveillance technology to support your people and your systems in the efforts to optimize sepsis care.
Dr. Itay Klaz is responsible for directing clinical efforts toward the development, implementation and support of Wolters Kluwer Sepsis Surveillance software solution.
Solutions
POC Advisor™
Scientifically proven to improve sepsis outcomes by ensuring every patient receives the right care at the right time, every time.

POC Advisor’s sophisticated , AI-powered analyses of patient data identifies sepsis early and accurately, and ensures clinicians respond with evidence-based care at the right time. All to save lives, reduce cost of care and improve CMS compliance.

Back To Top