---
title: 'Michigan Medicine / Epic Sepsis Model v1: Real-time proprietary sepsis alerts'
slug: epic-sepsis-model-negative
stable_id: 998bcea29b46d14e
company: Michigan Medicine / Epic Sepsis Model v1
function_code: clinical_ops
pattern_codes:
  - threshold_as_control
  - continuous_decisioning
  - creator_to_judge
evidence_strength: verified
publication_tier: showcase
freshness: current
reviewed_at: '2026-08-23'
updated_at: '2026-08-22'
source_quality_summary: 1 peer reviewed; publication outcomes are verified.
caveat_summary: One health system; v1 only, not later v2.
collections:
  - regulated-autonomy
  - negative-results
bundle_version: 1.0.0
bundle_fingerprint: sha256:c23c6cc2b88153d008ea8fda928f632ce0011fc2d4c5036672a16e5d895bab93
canonical_url: https://brianletort.ai/transformations/epic-sepsis-model-negative
---

# Michigan Medicine / Epic Sepsis Model v1: Real-time proprietary sepsis alerts

A deployed alert can add workload while missing most cases.

Function: Clinical operations. Patterns: Threshold as the human control; Continuous decisioning; Creator to judge. Evidence: verified.

Freshness: current. Reviewed: 2026-08-23. Updated: 2026-08-22.


Source quality: 1 peer reviewed; publication outcomes are verified.

## Before

1. **Hospital clinicians** — Monitor admitted patients and recognize sepsis through contemporary clinical practice. (control: Timely antibiotics are the study proxy for clinical recognition; no historical manual queue is reconstructed.)
2. **Clinical team** — Diagnoses and treats patients using ordinary EHR information. (control: Human diagnosis and treatment authority.)

## After

1. **Epic Sepsis Model v1** — Calculates a proprietary risk score every 15 minutes and generates alerts at the selected threshold. (control: Hospital-selected threshold within the recommended range; model missed 67% and alerted on 18% of stays in the validation.)
2. **Clinician** — Reviews or disregards the alert and retains all treatment decisions. (control: Ordinary surveillance remains the failure path for missed cases; alert fatigue is an explicit control concern.)

## Decision rights

Clinicians own treatment; hospitals choose thresholds.

## Exception path

Clinical judgment can override alerts; missed cases rely on ordinary surveillance.

## Outcomes

- **Missed sepsis and alert burden** (verified): Contemporary practice → Missed 67% of sepsis while alerting on 18% of hospitalizations; AUROC .63. One health system; v1 only, not later v2.

## Executive lesson

Local validation is a precondition, not a refinement.

## Anti-pattern

Do not generalize v1 results to v2.

## Questions for leaders

- Where is the operating threshold set and who can override it?
- What measured result would trigger rollback or retraining?
- Which residual decisions must remain human-owned?

## Sources

- [External Validation of a Widely Implemented Proprietary Sepsis Model](https://pmc.ncbi.nlm.nih.gov/articles/PMC8218233/) — JAMA Internal Medicine