About ADAPT

ADAPT is a browser-based clinical decision support tool that executes a site's own locally approved enteral feeding protocol. Given an infant's weight and gestational age, ADAPT generates a protocol-concordant feeding plan in under 30 seconds.

Why it exists

Standardized feeding regimens are associated with reduced NEC incidence (Jasani 2017). The remaining challenge is reliably delivering an adopted protocol under real-world conditions: time pressure, shift changes, and manual arithmetic at the bedside. ADAPT addresses this implementation gap by building reliability into how an existing, locally approved protocol is executed.

How the pieces fit together

ADAPT is a suite of four tools sharing one protocol format. The Protocol Builder lets a site encode its locally approved feeding protocol as a structured configuration. The Feeding Calculator consumes that configuration to generate per-admission feeding plans. The Guideline Doc Composer renders the same configuration as a clinical guideline document for committee review. The Growth Tracker monitors longitudinal growth against Fenton 2025 curves. The four tools live at the ADAPT launcher.

Scope

ADAPT runs entirely in the browser. Site protocols are authored locally and remain under each site's control. Every recommendation is reviewed and approved by a licensed provider before order entry. ADAPT is a clinical reference aid. It is not a substitute for clinical judgment, not an order-entry system, and not an EMR integration. Browser-based; no installation, accounts, or licensing fees.

Privacy and disclaimer

ADAPT runs entirely in your browser. Clinical values entered into the calculator are not saved or transmitted. Patient identifiers (name, MRN, date of birth) are not inputs and are never collected. The site stores a small amount of UI preference and the protocol template a site has selected as its default; never patient data. No cookies, no analytics, no third-party tracking.

ADAPT is a clinical reference aid, not a substitute for professional judgment. Output should be independently verified by a licensed clinician before use in patient care. Provided without warranties of any kind, express or implied.

Status

Validation

100 test cases, 100% computational accuracy against protocol-derived recommendations.

Software class

Non-device clinical decision support. Provider reviews and approves every recommendation before order entry.

Pre-pilot signal

At a single NICU, observational unit-level surveillance was associated with a 42% reduction in NEC incidence and a 72% reduction in discharges with feeding tubes. These are single-site observational results without formal pre-period methodology; a multi-site pilot is being developed to test whether the signal reproduces.