hero-valley.png
A path leading from a small hospital on a hill down to a warm lit house at dusk
The hospital is coming home.
We're the platform that helps you run it.
[name] helps community hospitals launch and operate hospital-at-home programs in an AI-native era — finding the right patients, running the daily operation, and sharing the infrastructure no small hospital should build alone.
- Built on the research that is defining the field
- Medicare reimbursement guaranteed through 2030
- Bringing experience from Harvard Med, Mass General Brigham, and the NIH
The gap
Hospital-at-home works. Running it is the hard part.
The evidence stopped being the obstacle years ago. What stops a community hospital now is arithmetic: a program costs roughly the same to run at three patients as at fifteen, and the census is the one thing nobody hands you. Screening is manual, so eligible patients are discharged to a bed upstairs before anyone notices. Staffing is sized for a census you don't have yet. The fixed costs land on day one while the volume takes a year. It is the same wall whether you have never enrolled a patient or you have been running at five a day for eighteen months.
That's the problem we exist to remove.
empty-beds.png
A calm hospital corridor with a window showing a neighborhood of homes beyond
~400
hospitals approved
2 in 3
have never enrolled a patient
Just 8
patients a day to break even
The platform
One AI-native command center for care at home.
Four jobs, one system, one login. The AI does the reading, the ranking, and the routing. Your clinicians make every call.
Community-acquired pneumonia
Meets criteria: CURB-65 of 1, room air sat 95%, lives with spouse at home, within service radius.
Cellulitis, lower extremity
Meets criteria: afebrile 18h, IV antibiotics only, no ICU history. Awaiting social work note.
CHF exacerbation
Borderline: diuresing well, but lives alone — needs caregiver confirmation.
COPD exacerbation
Meets criteria: off BiPAP 24h, ambulatory, home O₂ already in place.
Under the hood
Three models, one loop.
Hospital-at-home is a data problem wearing a clinical hat. Screening, triage, and enrollment are all prediction, and all three are done today by people reading charts by hand. That's the wedge.
Eligibility
Reads ED, observation, and inpatient feeds continuously — diagnosis, acuity, vitals trend, distance, support at home — and ranks candidates with the evidence for each one exposed. Not a rules engine somebody on your staff has to maintain forever.
Triage
Scores every incoming signal on deterioration risk rather than threshold breach. Each dismissal is a label, so false-alarm volume falls week over week instead of staying flat for the life of the program.
Enrollment
Predicts which families accept and which decline, and adapts how the program gets explained at the bedside. Built on discrete-choice research into what actually moves a patient from no to yes.
Every accept, decline, dismissal, and outcome is a training signal. Your models tune to your hospital. The network's models improve for every member — federated, with patient data never leaving your tenant.
How the network compounds thisFind every eligible patient
AI reads your ED and inpatient data continuously and surfaces candidates with the evidence behind each one.
Run the day safely
AI triage ranks every signal by what could actually go wrong — and learns from the alerts your nurses dismiss.
Grow the census
Closed-loop referrals, and enrollment built on what actually persuades a family to say yes.
Prove the value
The eight numbers your CFO actually wants, current as of this morning.
Opportunity Index
How many patients did your hospital send to a bed that could have healed at home?
We build that estimate from public Medicare data. The preview below runs on example hospitals while we finish the national dataset — try one, then request your own hospital's numbers.
SAMPLE — illustrative estimate for a fictional hospital
Cedar Hollow Regional Medical Center
Cedar Hollow, OH · 148 beds
Estimated eligible admissions
612
per year
Estimated missed contribution margin
$1,940,000
per year
Modeled census ramp
Average daily census by month, first 18 months
Break-even at 8/day — reached month 7
Wherever you are
Three kinds of hospital call us.
Only one of them is starting from scratch. The platform is built for all three.
Not yet approved
You've read the evidence and you want the program. We write the waiver application, design the workflows around how your hospital actually operates, and get you to a first patient in about ninety days.
Approved, never enrolled
You hold the waiver and it has been sitting. Two out of three approved hospitals are here. We start from the approval you already have and go straight to workflow, screening, and first patients.
Live, but under break-even
You're running four or five a day and the arithmetic still doesn't work. This is what the platform is best at — surfacing the eligible patients you're missing today and taking the operational load off a team that is already stretched.
network-map.png
A stylized regional map of small hospitals connected by soft glowing paths to homes between them
The Network
Too small to do it alone? That's the point of the network.
- One negotiated contract for monitoring, DME, imaging, couriers, and home pharmacy — instead of vendor minimums sized for a program ten times yours.
- Benchmarks from every member, so you know whether your escalation rate is good or just normal.
- A platform that gets measurably better with every hospital that joins.
Bring your patients home.
Or start with the Opportunity Index preview — it takes 30 seconds. Try the preview
doorstep.png
A clinician greeted warmly at a front door in golden-hour light