Bayesian Health's continuous AI sepsis monitor wins Medicare add-on payment

Overview

Updated Oct 6

Since October 1, 2026, hospitals using Bayesian Health's continuous AI sepsis monitor can bill Medicare for an extra payment of up to $61.84 per eligible discharge. The Centers for Medicare and Medicaid Services (CMS) approved the add-on payment in its FY 2027 Inpatient Prospective Payment System Final Rule.

The rule covers an estimated 739 Medicare Severity-Diagnosis Related Groups (MS-DRGs)—roughly 97% of all DRGs in the inpatient system. The approval makes Bayesian's device the first continuous sepsis monitoring technology with both FDA clearance and a dedicated Medicare reimbursement pathway. NTAP provides a temporary add-on payment, on top of the standard MS-DRG payment, for cases where a qualifying technology was used and costs exceed what the bundled payment covers.

Why it matters

Sepsis kills 270,000 Americans in hospitals each year; hospitals now have a financial pathway to adopt AI that cuts mortality 18%.

Questions about this story

Free account needed to ask — your question is kept and asked for you right after sign-up. Answers are public.

No questions yet — be the first to ask.

Key Indicators

$61.84

Maximum NTAP add-on per eligible discharge

Hospitals using Bayesian's monitor can bill up to this amount above the standard MS-DRG payment.

739

Eligible MS-DRGs

Roughly 97% of all DRGs in the inpatient prospective payment system, spanning adult Medicare fee-for-service admissions.

18.2%

Mortality reduction with clinician engagement

From a prospective study published in Nature Medicine spanning 764,707 patient encounters across five hospitals.

1.85 hours

Median faster antibiotic delivery

Patients whose clinicians engaged with Bayesian alerts received antibiotics a median of 1.85 hours sooner.

3 years

Duration of NTAP pathway

The Medicare add-on payment remains in place for up to three years.

270,000

Annual US hospital sepsis deaths

Sepsis takes 270,000 lives in US hospitals every year and drives over $50 billion in annual hospital costs.

Voices

Curated perspectives — historical figures and your fellow readers.

Ever wondered what historical figures would say about today's headlines?

People Involved

Suchi Saria

Founder and CEO, Bayesian Health; lead researcher at Johns Hopkins

Leading commercialization of the FDA-cleared sepsis monitor

Organizations Involved

Bayesian Health

Clinical AI company

Commercializing the first FDA-cleared continuous AI sepsis monitor

A clinical AI company commercializing real-time patient monitoring technology developed at Johns Hopkins University.

Centers for Medicare & Medicaid Services

Federal agency

Approved NTAP for Bayesian's sepsis monitor in FY 2027 Final Rule

The federal agency that administers Medicare, Medicaid, and the Children's Health Insurance Program.

U.S. Food and Drug Administration

Federal Regulatory Agency

Granted 510(k) clearance to Bayesian's sepsis flagging device in April 2026

The federal agency responsible for protecting public health by regulating drugs, medical devices, and biologics.

Timeline

2017 October 2026

6 events Latest: October 1st, 2026 · 3 weeks ago

Tap a bar to jump to that date

October 2026
NTAP payments take effect
Latest Implementation

Hospitals using Bayesian's monitor can begin billing Medicare for add-on payments on eligible adult inpatient discharges.
October 1st, 2026
September 2026
CMS approves NTAP for Bayesian's sepsis monitor
Regulatory

Medicare approves New Technology Add-on Payment under FY 2027 IPPS Final Rule, enabling up to $61.84 per eligible discharge starting October 1, 2026.
September 2nd, 2026
May 2026
Bayesian announces FDA clearance publicly
Announcement

Company formally announces the 510(k) clearance, noting NTAP decision expected in early August.
May 12th, 2026
April 2026
FDA grants first-ever 510(k) clearance for continuous AI sepsis monitor
Regulatory

Bayesian Health receives FDA clearance for its sepsis flagging device, the first continuous AI sepsis monitor ever cleared.
April 30th, 2026
January 2022
Nature Medicine publishes landmark sepsis AI study
Research

Prospective study across 764,707 patient encounters at five hospitals shows 18.2% mortality reduction when clinicians engage with alerts.
January 12th, 2022
2017
Saria begins sepsis research after personal loss
Origin

Suchi Saria starts developing the Targeted Real-Time Early Warning System at Johns Hopkins after her nephew dies of sepsis.
2017

Scenarios

NTAP Accelerates Hospital Adoption of Continuous AI Sepsis Monitoring

Likely Resolves by Oct 1, 2027

Discussed by: Industry analysts and the company—Bayesian's press materials frame NTAP as the missing financial piece for broad adoption.

With Medicare reimbursement now available, hospitals weigh the $61.84 add-on against the cost of deploying the platform. The reimbursement helps carry integration costs—EHR workflow, staff training, alert tuning—that previously fell entirely on the hospital. If adoption grows as the cost barrier drops, Bayesian could expand its footprint from existing customers like Cleveland Clinic, Johns Hopkins, and MemorialCare to a broader set of mid-size health systems. The company's partnership with Mayo Clinic in palliative care suggests the platform may extend beyond sepsis.

NTAP Approval Opens Door for Competing Continuous Sepsis AI Monitors

Possible Resolves by End of 2028

Discussed by: Healthcare technology analysts noting CMS's willingness to grant NTAP for continuous AI monitoring devices

CMS's decision establishes a precedent that continuous AI monitoring devices can qualify for NTAP. Competing companies developing similar pre-suspicion sepsis detection tools may pursue their own FDA clearance and NTAP applications. The three-year NTAP window for Bayesian's device gives competitors a timeline: they would need their own approvals within that window to capture hospital budgets before Bayesian becomes entrenched.

NTAP Fails to Move Adoption Needle, Sepsis AI Stays Niche

Unlikely Resolves by Oct 1, 2027

Discussed by: Healthcare IT analysts skeptical that reimbursement alone changes adoption dynamics

The $61.84 add-on covers only a fraction of the total cost of deploying continuous AI monitoring. Hospitals with tight IT budgets and strained clinical staff may still decline adoption, particularly if their existing sepsis protocols already achieve acceptable outcomes. The add-on also only applies to Medicare fee-for-service patients, leaving commercial, Medicaid, and uninsured admissions outside the reimbursement pathway. If adoption stays concentrated among academic medical centers with existing AI infrastructure, the NTAP's practical impact would be modest.

Historical Context

3 moments from history that rhyme with this story — and how they unfolded.

April 2018

FDA's first AI/ML-enabled medical device clearance (2018)

The FDA cleared IDx-DR, an AI system that detects diabetic retinopathy, marking the first authorization of an AI/ML-enabled medical device for marketing in the United States. The system allowed primary care providers to diagnose a condition traditionally requiring specialist referral.

Then

Opened the floodgates for hundreds of AI-based medical devices to seek FDA clearance.

Now

Established a regulatory pathway for clinical AI, but reimbursement lagged—devices were cleared but hospitals had no dedicated payment mechanism.

Why this matters now

The gap between FDA clearance and Medicare reimbursement that IDx-DR faced is exactly the gap NTAP now closes for Bayesian's sepsis monitor.

2001–2015

CMS's early NTAP approvals for novel technologies (2001–2015)

CMS used the NTAP program, created by the Medicare, Medicaid, and SCHIP Benefits Improvement and Protection Act of 2000, to fund early adoption of technologies like drug-eluting stents and ventricular assist devices. These technologies faced the same barrier Bayesian now faces: standard DRG payments based on historical costs did not account for the new technology's expense.

Then

Hospitals gained a financial bridge to adopt costlier but more effective technologies during their first years on the market.

Now

Established NTAP as the standard mechanism for Medicare to support novel inpatient technologies, typically providing 2-3 years of add-on payments.

Why this matters now

NTAP's track record shows it can shift adoption economics for new medical technology—the mechanism Bayesian now benefits from.

2015–present

Sepsis bundled payment initiatives (2015–present)

CMS and private payers experimented with sepsis-specific payment models, including the Sepsis Quality Measures and various bundled payment programs. These initiatives focused on rewarding faster treatment protocols and better outcomes but did not create a dedicated pathway for AI-based early detection tools.

Then

Improved sepsis documentation and treatment timeliness at participating hospitals, but adoption of predictive AI tools lagged due to lack of reimbursement.

Now

Demonstrated that payment reform can drive sepsis care improvement, but left a gap for pre-suspicion AI monitoring.

Why this matters now

NTAP for Bayesian's device fills the specific gap these earlier initiatives left: reimbursement for pre-suspicion detection rather than post-diagnosis treatment.