AI Medical Intelligence
Kipu EMR AI Scribe Group Note Splitting
Thoughtfully curated clinical brief and documentation workflow for Kipu EMR AI Scribe Group Note Splitting on Merry AI.
Kipu EMR AI Scribe: Group Note-Splitting Without Cloned Charts
Merry AI · Thoughtfully curated clinical briefs.
TL;DR
One group session generates ten individually attested Kipu chart entries.
Cloned-note templates invite CARF and Joint Commission audit findings.
Merry AI separates each participant's contribution, affect, cravings, and CBT response.
A clinical director attests and injects notes in one workflow.
Loaded labor math favors documentation reform: 1.3% of one MA salary.
- Jump to Section:
- The Loaded Labor & Denominator Model
- Clinical Logic & Audit Defense
- Clinical Taxonomy: ICD-10 Standards
- The Note-Splitting Wedge
- Chrome Extension DOM Overlay
- Clinical Intelligence Layer
- Specialty Playbooks & Next Steps
- FAQ
Section 1: The Loaded Labor & Denominator Model
CLINICAL UPDATE 2026: Revised for new CMS CPT G2211 standards, SB 1120 compliance, and FHIR interoperability.
Behavioral health documentation carries a hidden denominator most vendors ignore. Below we model the fully loaded cost against a per-provider annual license. Merry AI anchors every claim to that denominator.
A single medical assistant costs roughly $48,000 fully loaded — a $35,000 base wage plus benefits and overhead. That figure is the honest denominator, not the salary line alone.
Against that denominator, Merry AI Pro at $648 per year represents 1.3% of one loaded labor line. The comparison is not marketing; it is arithmetic that survives a CFO review.
| Cost Line | Annual Figure | Denominator Share |
|---|---|---|
| Fully loaded MA labor | $48,000 | Baseline (100%) |
| Merry AI Pro license | $648 | 1.3% of labor |
| Recovered G2211 revenue | $15,600+ | Offsets license 24x |
| Provider time returned | 2.1+ hrs/day | JAMA benchmark |
The revenue capture line matters most: complexity coding recovers documented value clinicians already earned. G2211 rewards visit complexity that group facilitators routinely under-document.
Documentation is not overhead — it is unrecovered revenue.
Section 2: Clinical Logic & Audit Defense
A multi-site IOP therapist runs a 3-hour relapse-prevention group with 10 attendees in Kipu. Instead of cloning one template and editing names, Merry AI separates each participant's verbal contribution, affect, cravings and risk disclosures, response to CBT/MI interventions, and goal progress into 10 individualized notes.
The clinical director then reviews the split notes, attests to accuracy, and injects them into the correct Kipu chart entries in one workflow. This reduces cloned-note exposure during Joint Commission or CARF review — documentation integrity as the operating principle.
Cloned notes are the fastest path to a payer clawback in group settings. Identical language across ten charts is the pattern reviewers are trained to flag first.
Human-attested clinical metrics anchor defensibility — the reviewing clinician owns every attestation. Automation drafts; the licensed director signs.
| Audit Risk | Cloned-Note Exposure | Merry AI Split + Attestation |
|---|---|---|
| CARF individualization standard | Identical notes flagged | 10 distinct participant records |
| NCCI Modifier 25 scrutiny | Unsupported separate E/M | Documented distinct service |
| SB 1120 AI-use disclosure | Undocumented automation | Human review of record |
| Joint Commission chart review | Copy-paste pattern | Attested individualization |
Measurable clinical metrics travel into each note: affect ratings, craving severity, and goal-progress deltas. These are the details that make a record defensible under scrutiny.
Per the CMS NCCI edits guidance, paired-service edits require clear documentation of a distinct, separately identifiable service.
All handling conforms to HHS HIPAA Health Information Privacy Standards for protected health information across every split note.
Section 3: Clinical Taxonomy — ICD-10 Documentation Standards
Substance-use documentation lives or dies on precise diagnostic coding across split group notes. A shared code across ten participants is a documentation defect, not a shortcut.
Each participant note carries its own attested diagnosis, never a group-cloned code. Individualization begins at the diagnosis line.
| ICD-10-CM Code | Description | Documentation Anchor |
|---|---|---|
| F10.20 | Alcohol dependence, uncomplicated | Craving disclosure, abstinence status |
| F11.20 | Opioid dependence, uncomplicated | MAT adherence, relapse-risk notes |
For alcohol-use coding, reference F10.20 Alcohol dependence (ICD-10-CM).
For opioid-use coding, reference F11.20 Opioid dependence, uncomplicated (ICD-10-CM).
For general taxonomy conventions, see ICD-10-CM coding standards (CMS).
Section 4: The Note-Splitting Wedge Competitors Left Blank
Every competing scribe treats a group as one recording that becomes one note. That is the structural miss, and it is where documentation integrity fails.
A 10-person IOP group is ten clinical encounters sharing one room, not one encounter with ten names. This is our Anchor Truth, and the entire workflow follows from it.
Freed lists "group session documentation" as a single bullet and stops. It never addresses individualization, attestation, or clawback risk.
| Capability | Generic AI Scribe | Merry AI |
|---|---|---|
| Group capture | One combined note | Ten separated notes |
| Per-participant affect/cravings | Not isolated | Individually parsed |
| Attestation model | Implicit | Director-reviewed |
| Kipu chart routing | Manual copy-paste | One-workflow injection |
The workflow wedge is individualization at scale with human sign-off — the exact gap payers audit. The Pro plan's multi-party group note-splitting, validated in the Path Recovery TN case study, is built for this.
Explore the mechanics via our Kipu Group Note-Splitting Integration.
Section 5: Chrome Extension DOM Overlay & EHR Field Injection
Kipu is a closed EHR without an open scribe API — so a browser-native approach is required. The Basic plan at $59/mo ($35/mo annual) delivers this overlay directly.
A Chrome extension DOM overlay reads and populates Kipu fields directly, needing zero IT setup. The clinician stays inside Kipu the entire time.
No integration ticket, no admin provisioning, and no system change is required to begin. The install is a browser extension, not a project.
| Requirement | API Integration | DOM Overlay Injection |
|---|---|---|
| IT involvement | Weeks of setup | None |
| Closed-EHR support | Blocked | Fully compatible |
| PHP/IOP group splitting | Unsupported | Native |
| Field-level injection | Vendor-dependent | Direct to Kipu fields |
Split PHP and IOP notes inject into each participant's correct chart entry, not a shared record. Routing accuracy is a documentation-integrity control, not a convenience.
Unlike beta EHR-push models, injection targets named DOM fields with clinician confirmation at each step. Nothing writes to a chart without a human step.
Section 6: Clinical Intelligence Layer — Closed-Pilot Orchestration
Documentation reform extends beyond the note into pre-visit, during-visit, and post-visit orchestration. The record is one artifact of a longer clinical arc.
The Practice Partner plan at $149/mo coordinates these phases across a closed-pilot rollout — five outpatient practices selected weekly for direct solutions engineering. It also carries the SB 1120 and NCCI audit shields.
| Phase | Orchestration Task | Clinical Output |
|---|---|---|
| Pre-visit | Prior-note surfacing | Continuity context |
| During-visit | Group turn separation | Ten draft notes |
| Post-visit | Attestation + injection | Charted Kipu entries |
Closed-pilot orchestration lets sites validate individualization accuracy before program-wide adoption. Multi-site directors get a controlled proof, not a leap of faith.
Review tiers via Merry AI Practice Partner Plans.
Section 7: Specialty Playbooks & Next Steps
Behavioral health is one node within a broader library of specialty documentation standards. The group-splitting logic here recurs across other multi-party settings.
Each playbook maps taxonomy, audit risk, and workflow to a specific clinical setting. The structure stays consistent; the standards change.
Browse adjacent guidance in our Specialty Clinical Playbook Library.
Ten voices, ten attested charts, one reviewed workflow.
Frequently Asked Questions
How does group note-splitting work in Kipu?
Merry AI separates each participant's contribution into individual notes for director attestation and chart injection.
Does note-splitting reduce audit risk?
Yes — individualized, attested notes reduce cloned-note exposure during CARF and Joint Commission review.
Is IT setup required for Kipu?
No — the Chrome extension DOM overlay operates browser-side with zero IT provisioning.
Which diagnoses are supported?
Common substance-use codes including F10.20 and F11.20 are documented per participant.