How a Regional Health System Shortened Sepsis Response Time
420-bed regional health system · Upper Midwest · 4 medical-surgical units, 2 ICUs · 7 min read
A 420-bed regional health system deployed Prescient's continuous sepsis early-warning scoring across its medical-surgical units to close the gap between the first physiological signal of deterioration and rapid-response activation. Because scoring ran continuously against live vitals and labs — rather than waiting for scheduled screening — the care team was alerted to rising risk earlier in the deterioration curve.
At a Glance
- Organization
- 420-bed regional health system
- Region
- Upper Midwest
- Scope
- 4 medical-surgical units, 2 ICUs
- EHR environment
- Single EHR, HL7 FHIR integration
- Pillar deployed
- Early Warning & Sepsis Prediction
- Time to go-live
- ~8 weeks from discovery
Representative example. This is an illustrative case study built from an anonymized customer archetype to show the shape of a Prescient engagement and its reporting. The scenario and all figures are representative examples, not published results from a named customer.
Results at a Glance
What Changed
Illustrative figures — representative of the reporting structure, not published results.
The Challenge
Where they started
Like most acute-care hospitals, this health system ran its sepsis bundle on periodic, nurse-driven screening. A patient's condition could shift meaningfully in the hours between checks, and the physiological signal of that shift often sat unexamined in the vitals and lab data until the next scheduled assessment.
By the time a patient formally met screening criteria, deterioration was frequently already well underway — and rapid-response activation, antibiotics, and fluids all started later than the clinical team wanted. Leadership also had no standardized, real-time way to see where deterioration risk was concentrating across units on any given shift.
- The health system's sepsis bundle depended on periodic nurse-driven screening, which meant a rising risk could sit unrecognized between checks.
- Rapid-response activation frequently happened only once a patient already met formal criteria — by which point the physiological signal had often been present in the data for hours.
- Quality leadership had no standardized, real-time way to see where deterioration risk was concentrated across units.
The Solution
Continuous scoring inside the existing workflow
Prescient connected to the health system's EHR over HL7 FHIR and began scoring every inpatient continuously — recalculating a Low-to-Critical risk band each time new vitals or lab data arrived, rather than at fixed screening intervals.
Crucially, nothing about the clinical team's day-to-day tooling changed. When a patient's risk band rose, the alert routed to the assigned nurse and rapid-response team inside the EHR they already used, with the specific contributing factors attached so the team could act on context rather than a bare number.
How data flowed in this deployment
Continuous vitals, lab trends, and notes stream in over HL7 FHIR.
Every new data point updates the patient's deterioration risk band.
Band changes route to the assigned nurse and rapid response team.
Response is documented; outcome feeds back into the model.
The Approach
How the Deployment Unfolded
- Weeks 1–3Discovery & FHIR connection
Scoped the EHR environment with the CMIO and IT team and established the HL7 FHIR integration for vitals, labs, and notes across the pilot units.
- Weeks 4–7Shadow mode validation
Risk scoring ran silently alongside existing workflows so clinical leadership could validate behavior against the system's own patient population before any alert reached a clinician.
- Week 8Go-live on med-surg units
Risk-band alerts activated inside the existing EHR workflow, routed to the assigned nurse and rapid-response team with contributing factors attached.
- OngoingThreshold tuning & review
Escalation thresholds were tuned with nursing leadership over the first weeks, and performance was reviewed against the system's own pre-deployment baseline.
The Results
Measured Against Their Own Baseline
Sepsis response — illustrative before/after
Illustrative example only, not a published result — measured against the system's own pre-deployment baseline.
Escalations acted on within the protocol window
Illustrative adoption curve as thresholds were tuned and teams built the workflow into their routine.
One patient's continuous risk score
Illustrative example: a patient's risk score crossing from Moderate into High before clinical intervention.
Key Takeaways
What Made It Work
Continuous scoring surfaced deterioration between scheduled screenings, not only at them.
Alerts routed inside the existing EHR workflow, so adoption didn't depend on a new system.
Contributing factors attached to every alert let teams act on context, not a bare score.
Results were measured against the system's own baseline, not an industry average.
The value wasn't a new dashboard — it was that the signal reached the right nurse inside the workflow they were already using, with the reason attached. That's what changed how quickly the team moved.
Representative quote illustrating the kind of feedback this engagement is designed to produce — not attributed to a named individual.
The Platform Behind This
Early Warning & Sepsis Prediction
See deterioration hours before it's charted.
Frequently Asked
Questions About This Scenario
More Case Studies