Case 01
Alexandra Hospital · NUHS
94%
AI intake accuracy
1,400
Beds in expansion pipeline
RFID tray scanning replaced manual intake charting — AI outperformed nurse estimation every single month.
The malnutrition coding gap
21.1%p
The gap
between estimated inpatient malnutrition risk and the rate at which it actually gets coded on the bill.
Coded diagnosis
8.9%
Of U.S. inpatients receive a malnutrition diagnosis on their claim.
Estimated clinical risk
≈ 30%
Of inpatients are flagged at malnutrition risk in clinical literature.
Why the gap persists
The diagnosis exists in the clinic.
It just never reaches the chart.
01
Subjective intake records
Recall-based charting introduces variability that undermines clinical confidence and audit defense.
02
No defensible evidence trail
Without objective proof, coding teams routinely downgrade or reject malnutrition claims.
03
Documentation never reaches the bill
Valid malnutrition cases sit unbilled because the supporting evidence was never captured.
Annual opportunity
$4.8M
recoverable for a representative 400-bed hospital — sitting unbilled because the evidence was never captured.
Source: Journal of Hospital Medicine, 2024 · Nutrition in Clinical Practice, 2021. Illustrative financial model based on a 400-bed hospital with $200M Medicare revenue.
Clinical workflow
From tray to claim,
in five fully traceable steps
Nutri-Guard sits in the existing meal-service workflow and writes structured evidence directly into the EHR. No new device. No extra clicks for the RD.
