- No public price. Delivered within Epic licensing.
- Not disclosed
- Not disclosed
- —
- —
- US
Decision support
Epic Sepsis Model
by Epic Systems · US
Sepsis risk score calculated every 15 minutes inside the Epic EHR, with alert thresholds set by each hospital.
On this page
41/100
Editorial rating 3.0 / 5
- Regulatory & Compliance0/22
No FDA clearance listed
- Clinical Integration0/11.6
No EHR integrations listed
- Evidence Strength30/30
5 peer-reviewed papers
- Vendor & Market3/18
Seed-stage or unfunded
- Sentiment & Transparency1.3/2.5
Pricing on request, no public amounts
▸ Show all 11 dimensions▾ Hide dimension detail
- FDA clearance0/12
No FDA clearance listed
- HIPAA / SOC2 / BAA0/10
No public HIPAA/SOC2/BAA attestation
- EHR integrations (count)0/6
No EHR integrations listed
- Top-3 EHR coverage (Epic / Oracle / Athena)0/5
None of the top-3 EHRs covered
- Bidirectional write-back0/1
No bidirectional write-back documented
- Peer-reviewed papers21/21
5 peer-reviewed papers
- RCT / meta-analysis / systematic review9/9
1 RCT/Meta-Analysis/Systematic Review
- Funding & adoption signal3/12
Seed-stage or unfunded
- Years in market0/6
Founded year not recorded
- Pricing transparency1/3
Pricing on request, no public amounts
Sepsis risk score calculated every 15 minutes inside the Epic EHR, with alert thresholds set by each hospital.
No public price. Delivered within Epic licensing.
Evidence record
No FDA decision cited in our research as of November 2, 2026.
5 studies on PubMed: 3 other, 1 observational, 1 systematic review, published 2024 to 2026. Read the literature summary
- Score computed November 2, 2026, method 1.2.0
- Review written November 2, 2026
MedAI Verdict, Epic Sepsis Model record, https://medaiverdict.com/tools/epic-sepsis-model, version November 2, 2026
Bottom line
The Epic Sepsis Model is a score that Epic's EHR recalculates every 15 minutes for adult inpatients and emergency patients and that triggers a clinician alert above a hospital-set threshold. It has no FDA clearance. Epic's K260008 covers the Deterioration Index Version 2, a different model. Independent studies found the first version discriminated poorly and the second version did better, with a prospective 2026 study still reporting low positive predictive value and a high alert burden. It suits an Epic hospital with a team to tune thresholds and audit alerts. A hospital that wants a cleared sepsis product has other vendors to compare.
Who supplies it, and in what form?
Epic Systems Corporation delivers it inside its EHR license, and its pricing is not stated. Two versions exist. ESM v1 is a logistic regression. ESM v2 uses gradient-boosted trees and can be localized to a hospital (PMID 40729785). A 2021 paper describes it as implemented at hundreds of US hospitals.
Is it FDA cleared?
No. K260008 was decided on September 18, 2026 under product code QNL and covers the Deterioration Index Version 2. Sepsis products with FDA decisions include Prenosis Sepsis ImmunoScore (De Novo DEN230036, April 2, 2024), Bayesian Health's Sepsis Flagging Device (K250680, April 30, 2026) and Beckman Coulter's early sepsis indicator (K181599, March 18, 2019).
What did independent validations find?
None of these studies was run by Epic.
- Wong and colleagues, JAMA Internal Medicine 2021 (PMID 34152373), covered 38,455 hospitalizations at Michigan Medicine. Hospitalization-level AUROC was 0.63 (CI 0.62 to 0.64), and the model missed 67% of sepsis cases while alerting on 18% of hospitalizations. The authors cite poor discrimination and calibration.
- Currey and Tarabichi, American Journal of Emergency Medicine 2025 (PMID 40729785), studied 35,076 emergency encounters. AUROC was 0.77 for v1 and 0.90 for v2, falling to 0.70 and 0.85 before clinical recognition of sepsis.
- Wong and colleagues, JAMA Network Open 2026 (PMID 41758510), validated v2 prospectively at four US health systems. Encounter-level AUROC ranged from 0.82 to 0.92 and positive predictive value from 0.13 to 0.26. The authors describe "high institutional variability, low positive predictive value, and high alert burden".
The verdict
An Epic hospital already running the model has its best published numbers with v2, and they vary by institution. Prenosis and Bayesian Health hold FDA authorizations, and the Epic score comes with the Epic license instead.
Editorial review last generated November 2, 2026. Synthesized from clinician sentiment, peer-reviewed coverage, and our editorial silo picks. Refined by hand where vendor facts change.
What it costs
All tiers are custom-quoted. The vendor does not publish per-provider list pricing.
| Tier | Monthly | Annual | Notes |
|---|---|---|---|
| Enterprise | Not stated | Not stated | No public price. Delivered within Epic licensing. |
Source: vendor pricing page. Last checked November 2, 2026.
What the literature says
5 peer-reviewed studies indexed on PubMed evaluate Epic Sepsis Model in clinical contexts. The most relevant are shown below, ranked by editorial relevance score combining title match, study design, recency, and journal tier.
- A Systematic Review and Meta-analysis of Externally Validated Epic Clinical Decision Support Tools. (opens in new tab)
- Patel H, Crusco S, Hansen D, et al.· J Gen Intern Med· 2026
- Clinical Decision Support (CDS) tools integrated with Electronic Health Records increasingly guide clinical practice. Epic Systems, storing more than 325 million patient records, offers various proprietary predictive models to healthcare systems. Despite widespread adoption, no systematic review has examined these tools' real-world performance compared to vendor-reported metrics. This study was prospectively registered on PROSPERO (CRD420251148571). We systematically searched PubMed, Scopus, and Embase (January 2018 to August 2025) for external validations of Epic's CDS tools. We pooled Area…
- Evaluation of Sepsis Prediction Models before Onset of Treatment. (opens in new tab)
- Kamran F, Tjandra D, Heiler A, et al.· NEJM AI· 2024
- Timely interventions, such as antibiotics and intravenous fluids, have been associated with reduced mortality in patients with sepsis. Artificial intelligence (AI) models that accurately predict risk of sepsis onset could speed the delivery of these interventions. Although sepsis models generally aim to predict its onset, clinicians might recognize and treat sepsis before the sepsis definition is met. Predictions occurring after sepsis is clinically recognized (i.e., after treatment begins) may be of limited utility. Researchers have not previously investigated the accuracy of sepsis risk pre…
- Multicenter Prospective Validation of an Updated Proprietary Sepsis Prediction Model. (opens in new tab)
- Wong A, Currey D, Schwinne M, et al.· JAMA Netw Open· 2026
- The Epic Sepsis Model version 2 (ESM v2) is a widely implemented proprietary sepsis prediction model, but no multicenter, external validation of its performance has been reported to guide adoption and use. To conduct a multicenter validation of the ESM v2 to compare performance against the original ESM v1, outline differences across heterogenous clinical sites, and compare model performance against clinician recognition of sepsis. This prognostic study included adult inpatient encounters at 4 large US health systems between August 31, 2023, and March 11, 2025. At each site, data were collecte…
- Patient Versus Prediction-Level Evaluation of a Dynamic Clinical Prediction Model of Sepsis. (opens in new tab)
- Tuttle M, Maas CCHM, An J, et al.· medRxiv· 2026
- The Epic Sepsis Model version 2 (ESMv2) is a prediction model embedded into the electronic medical record used to warn clinicians which hospitalized patients are at risk for sepsis. We conducted a retrospective cohort study of 31,951 hospitalizations of 25,760 patients to compare analyses conducted at the commonly used patient-level (where a maximum prediction prior to the onset of sepsis is used to measure performance) vs novel prediction-level (where each prediction is used to measure performance). Sepsis, defined by the Sepsis 3 criteria occurred during 1,049 hospitalizations (3.3%). Patie…
- Patient Versus Prediction-Level Evaluation of a Dynamic Clinical Prediction Model of Sepsis. (opens in new tab)
- Tuttle M, Maas CCHM, An J, et al.· Res Sq· 2026
- The Epic Sepsis Model version 2 (ESMv2) is a prediction model embedded into the electronic medical record used to warn clinicians which hospitalized patients are at risk for sepsis. We conducted a retrospective cohort study of 31,951 hospitalizations of 25,760 patients to compare analyses conducted at the commonly used patient-level (where a maximum prediction prior to the onset of sepsis is used to measure performance) vs novel prediction-level (where each prediction is used to measure performance). Sepsis, defined by the Sepsis 3 criteria occurred during 1,049 hospitalizations (3.3%). Patie…
Other decision support
See the full decision support rankingUpToDate Expert AI
by Wolters Kluwer
Generative AI version of UpToDate that answers only from UpToDate's expert-authored library. Included in UpToDate subscriptions in the US.
July 2026 figure from a third-party price comparison; the vendor page shows no amount. Wolters Kluwer states Expert AI is included in all UpToDate subscriptions in the US. Trainees $219 per year with proof.
OpenEvidence
by OpenEvidence
Free physician-only literature-grounded Q&A, 1M consults/day.
Free for NPI-verified physicians (ad-funded).
Doximity Ask
by Doximity
Free clinical AI assistant for verified US clinicians, built on Pathway's technology.
Free for verified US clinicians, funded by advertising, per third-party summaries. Doximity's pages returned 403 when read. Also part of enterprise contracts with health systems.DynaMed / Dyna AI
by EBSCO Health
Evidence-based clinical reference from EBSCO with Dyna AI, a generative Q&A add-on, and Epic integration.
Individual plan from the EBSCO store.