Evidence-based design

Research

The peer-reviewed evidence behind Zora's design decisions and the growing body of research into AI-assisted clinical documentation.

Research cited on this page is used to inform product direction and does not imply clinical validation of the current platform.

The evidence is clear.
The systems haven't
caught up.

Zora is built on a growing body of peer-reviewed research into clinician burden, AI-assisted documentation, and Australian primary care. Every design decision traces back to published evidence or direct clinical input. We do not build features because they are technically possible. We build them because the research says they matter.

16'
documentation time saved per 8 hours of patient care (Rotenstein et al., JAMA 2026)
0.5
additional weekly visits delivered per clinician with AI scribe adoption (Rotenstein et al., JAMA 2026)
8,581
clinicians studied across 5 academic health systems in the largest multisite AI scribe study to date
Key research

The studies that
shaped Zora.

JAMA 2026

Changes in Clinician Time Expenditure and Visit Quantity With Adoption of Artificial Intelligence–Powered Scribes

Rotenstein LS et al. · JAMA. 2026;335(16):1408–1417

A multisite study across five US academic health systems found that AI scribe adoption was associated with meaningful reductions in documentation time and modest increases in weekly visit volume. Benefits were greatest for primary care clinicians and those who used AI scribes in more than half of their consultations. The study provides the strongest multisite evidence to date for AI scribe efficacy.

Read Paper
AJGP 2025

Is AI A-OK?: Medicolegal considerations for general practitioners using AI scribes

Bradfield OM, Mahar PD · Aust J Gen Pract. 2025;54(5):304–310

Published by the Royal Australian College of General Practitioners, this article sets out the medicolegal framework Australian GPs must follow when adopting AI scribes — including patient consent requirements, data storage obligations under Australian privacy law, and record accuracy standards. Zora is designed from the ground up to meet each of these requirements.

Read Paper

Every feature traces
back to evidence.

📋

Pre-consult surveys

Patients who provide structured intake information before a consultation allow clinicians to spend less time on history-taking and more time on clinical reasoning.

💬

SMS not app

Removing app download friction increases patient completion rates for intake surveys. An SMS link that opens in the mobile browser eliminates the biggest barrier to engagement.

🚦

Triage colour coding

Clinician-approved red flag logic reduces the risk of missed urgent presentations in high-volume practices. Red, Yellow, and Blue give every patient a visible priority before the consult begins.

🎙️

AI scribe

Real-time transcription removes the documentation burden that accounts for the majority of after-hours EHR time. The clinician reviews and approves — the AI never acts alone.

Two-step referral proof-read

Mandatory dual confirmation before sending reduces referral errors. The send button stays locked until both proof-read checkboxes are ticked by the clinician.

⏱️

Medicare timer

Passive timing reduces underclaiming, which is common in busy practices. The timer runs automatically through every module, with milestones at 6, 20, and 40 minutes.

Results from the field.

Mileva is currently onboarding pilot clinics across general practice and psychiatry in Melbourne and Sydney. Quantitative outcomes including consult time, triage accuracy, and GP satisfaction will be published here as data becomes available.

Subscribe to our research updates

Read on Substack →