The Centers for Disease Control and Prevention (CDC) report that at least one in three inpatient deaths is related to sepsis. The condition is a leading cause of hospital deaths and readmissions in the U.S., with a total 2019 health system economic impact of $1.5 billion. Many health systems experience high sepsis mortality rates and their costly ramifications, but the good news is that most sepsis-related deaths are preventable.
This sepsis prevention opportunity means that health systems can cut high costs associated with sepsis deaths, and most importantly, improve outcomes for patients with sepsis. With a sophisticated data platform that can aggregate sepsis data and immediately surface relevant insights, improvement teams understand their sepsis populations and which interventions best support sepsis detection, treatment, and surveillance.
Health systems already have a great desire to combat sepsis, and many have numerous sepsis prevention and treatment interventions in place but that’s not enough to overcome a sometimes difficult-to-detect disease. Data infrastructure, such as the Health Catalyst Sepsis Analytics Accelerator, combined with commitment and education, is critical for health systems to prevent and treat sepsis. Access to granular-level sepsis data is the best way health systems can understand the causes of sepsis (which are often preventable in the in-patient setting) and how to remedy them before it’s too late.
For example, comprehensive analytic insight might reveal a delay in clinicians administering antibiotics for patients with sepsis. The sepsis data tools’ drill-down capabilities would unearth a reason—such as a cumbersome standard procedure of ordering two blood cultures before administering antibiotics—that delays medication for patients in desperate need of treatment. Based on this insight, clinical improvement teams might change the blood culture order process or educate providers to administer antibiotics after one culture instead of two, allowing clinicians to give potentially life-saving antibiotics sooner. These small changes can lead to drastic improvements in early sepsis treatment and reduced sepsis rates.
By regularly turning to data, organizations can tackle sepsis prevention and treatment. Specifically, health systems can apply five data-informed strategies to reduce sepsis mortality rates:
Health systems can use their data infrastructure (e.g., the Health Catalyst Data Operating System (DOS™)) to aggregate their population’s sepsis data. Improvement teams can use this comprehensive sepsis data to define pre-sepsis thresholds and then create custom EHR alerts that notify the provider of early sepsis warning signs and the recommended course of action.
For example, the EHR could alert a provider about a patient’s potential for infection, low blood pressure and rapid heart rate, and prompt the facility’s best practice, such as administering antibiotics immediately. Using past sepsis data to surface insights and recommended next steps at the point of decision making helps clinicians detect sepsis earlier in the process when patients have the best chance of recovery.
Survival for patients with sepsis depends on the care team’s interventions. To ensure effective interventions, health systems should follow sepsis guidelines, such as the three-hour sepsis bundle, that prompts immediate action. The three-hour bundle includes four specific actions care teams should take within three hours of suspecting a patient has sepsis:
It is important that health systems have access to accurate sepsis recognition and treatment data, such as in the Sepsis Analytics Accelerator dashboard (Figure 1) to understand performance and identify areas to improve their three-hour bundle compliance. For example, sepsis data might reveal a delay in measuring a patient’s lactate levels within the three-hour timeframe. This insight would signal a need for a change in lactate collection processes or an opportunity to educate clinicians about measuring lactate levels emergently for optimal patient outcomes.
Sometimes delays in care for other conditions (e.g., AMI, stroke, traumatic injury) can lead to delays in sepsis care. Using data to understand the causes of a delay in the care process can reveal opportunities to improve triage practices in the emergency department (ED). For example, if a health system’s ED averages a 90-minute wait and a 40-minute wait for lactate collection, patients who enter the ED with sepsis wait over two hours for care. This time frame leaves less than one hour for care teams to complete the four recommended actions in the three-hour bundle, which proves deadly for some patients. With this insight from the ED data, the improvement team can investigate ED triage processes and possibly obtain blood cultures or lactate testing during triage instead of after admission.
To diagnose sepsis more accurately, organizations can leverage the CDC’s Sepsis Surveillance Toolkit. The toolkit includes instructions for creating a patient cohort based on physiological data (e.g., vital signs) that the CDC considers true indicators of sepsis. Using physiological data helps care teams accurately prevent and diagnose sepsis the first time. This approach differs from the traditional administrative coding identification approach based on a clinician’s notes (e.g., written orders, progress notes, physical exam, etc.), which leaves more room for error.
To effectively use the physiological sepsis identification method, organizations need the reliable data infrastructure mentioned in strategy #1 (DOS) to rapidly collect and organize data from multiple sources. With comprehensive physiological patient data, organizations are now ready to accurately detect and treat sepsis.
Although health systems should prioritize sepsis prevention, they cannot ignore the reported 40 percent readmission rate among patients who have recovered from sepsis. To reduce readmission rates, care teams can implement smoother post-acute care transitions with patients, their families, or long-term care facilities and discuss possible prevention tactics.
As part of the transition process, care teams can identify post-discharge follow-up appointments to ensure the patient is leaving the hospital with adequate support. The care team can also provide education to the patient and family about reoccurring infection and the best course of action if the patient relapses. For patients who are at higher risk for readmission (e.g., patients with comorbidities), clinicians can schedule regular follow-up calls to regularly monitor the patient’s progress or involve home health agencies to detect setbacks as early as possible.
Along with high sepsis mortality rates, readmissions related to sepsis are also more costly than readmissions related to heart disease, including heart failure and chronic obstructive pulmonary disease. As organizations continue to face diminishing profit margins, they cannot afford to face sepsis without analytic insight from sepsis data and supporting data infrastructure. Using the five strategies above allows clinicians to decrease sepsis mortality rates by detecting the condition early, immediately applying the best intervention based on the patient’s individual status and reducing readmissions for patients who have recovered from sepsis.
Would you like to learn more about this topic? Here are some articles we suggest:
Would you like to use or share these concepts? Download this presentation highlighting the key main points.