Audit, Provenance, and Release Gates
How Regenemm approaches traceability, source evidence, human review, agentic audit, patient-facing release, and external disclosure controls.
Audit, Provenance, and Release Gates
In healthcare, it is not enough to generate an output.
A clinical document, patient summary, referral, result review, education sheet, billing support note, or medicolegal chronology must be traceable. It must be reviewable. It must be clear where the information came from, who reviewed it, and when it was released.
Regenemm Healthcare is designed around that principle.
Why Provenance Matters
Clinical workflows depend on trust.
A healthcare team needs to know:
- what information was used;
- where it came from;
- whether it was patient-provided, clinician-entered, imported, retrieved, or generated;
- whether AI assisted;
- whether a human reviewed it;
- whether it was released to a patient or external party;
- whether a later version changed the clinical meaning.
Regenemm is designed to preserve this chain of context.
Hub-Governed Audit
Regenemm Voice acts as the Hub and clinical control plane.
It governs:
- clinical state;
- source evidence;
- audit events;
- human review;
- release gates;
- patient-facing outputs;
- external disclosures;
- agentic workflow records;
- data permanence.
Spokes provide workflow surfaces. Durable audit records and material workflow events return to the Hub.
What Regenemm Audits
Depending on the workflow and deployment, Regenemm is designed to audit material events such as:
- clinical data access;
- document creation;
- document editing;
- document release;
- patient sharing;
- external disclosure;
- agentic assistant actions;
- tool use;
- evidence retrieval;
- Edge Connector ingestion;
- interoperability events;
- consent checks;
- policy decisions;
- human review;
- escalation;
- release gate approval or denial.
Agentic Audit
When agentic assistants support a workflow, the audit record should identify:
- which agent acted;
- what task was assigned;
- what data class was accessed;
- what tool was used;
- what evidence was referenced;
- what output was generated;
- whether a policy check occurred;
- whether human review was required;
- whether the output was approved, revised, rejected, or escalated.
Agentic assistants do not replace clinical accountability. They operate inside governed workflow boundaries.
Provenance for Clinical Outputs
Regenemm is designed so material outputs can be linked back to source context.
This may include:
- source consultation;
- source transcript;
- imported report;
- pathology or radiology result;
- patient-provided record;
- retrieved literature;
- Knowledge Base reference;
- prior Hub state;
- Edge Connector message;
- human reviewer;
- release decision.
Provenance helps clinicians trust, verify, correct, and explain outputs.
Release Gates
A release gate is a required checkpoint before information is used, shared, or sent onward.
Release gates may apply before:
- patient-facing summaries;
- patient education;
- external correspondence;
- referrals;
- medicolegal documents;
- billing support records;
- FHIR or EMR-bound outputs;
- MHR-related outputs;
- care-team escalation messages;
- high-risk clinical workflow changes.
Release gates may require clinician review, patient-safe language review, consent check, purpose-of-use check, role check, source evidence check, privacy review, medicolegal review, or audit packet creation.
Patient-Facing Release
Patient-facing material requires particular care.
Before content is released to a patient through Regenemm Link or another authorised pathway, Regenemm's intended model supports:
- clinician review where clinically material;
- patient-safe language;
- consent and sharing controls;
- source traceability;
- versioning;
- audit of release.
External Release
External release may include communication with GPs, hospitals, insurers, legal teams, allied health providers, EMRs, MHR pathways, or other authorised systems and organisations.
External release should be governed by role, purpose, consent, clinical review, and audit.
Summary
Regenemm is designed around a simple principle:
If clinical information is read, changed, generated, reviewed, released, or escalated,
the system should know what happened, why it happened, who or what acted,
and what evidence supported it.
Evidence is preserved.
Clinical state is governed.
AI assistance is auditable.
Humans review.
Release is controlled.
Audit persists.