Geriatrics
Everyday medical support built on trust, quality checkups, and personal attention to your overall wellness.
TL;DR: PDPM reimbursement lives or dies on Section GG functional scoring captured within the 3-day admission window. Weekend admissions routinely default to "supervision" coding, silently collapsing PT/OT case-mix groups and costing facilities up to $320/day. Scribing.io captures OT/PT eval audio, extracts self-care and mobility verbal cues, auto-codes GG items (06/01 logic), crosswalks to the I0020B primary clinical category, flags discrepancies before the assessment window closes, and generates an auditable PDPM logic tape ready for ADR defense. CMS gives you the mappings; Scribing.io operationalizes them at the bedside.
PDPM Logic Fundamentals
Scribing.io Clinical Logic Walkthrough
The Auditable PDPM Logic Tape
ICD-10 Documentation Standards
Manual vs. Scribing.io Workflow
Pricing and Compliance Governance
PDPM Logic Fundamentals: Section GG as the Reimbursement Engine
CLINICAL UPDATE 2026: Revised for new CMS CPT G2211 standards, SB 1120 compliance, and FHIR interoperability.
Under the Patient-Driven Payment Model, a Skilled Nursing Facility's per-diem rate is no longer volume-driven—it is documentation-driven. The classification hinges on five case-mix components (PT, OT, SLP, Nursing, and NTA). Two of the highest-value components are calculated directly from Section GG functional markers.
The Anchor Truth for every Clinical Operations Director is simple: SNF documentation must capture Section GG functional markers to satisfy PDPM. Those markers must originate from what the clinician actually observed and said during the eval. Scribing.io auto-extracts the "Self-Care and Mobility" verbal cues and maps them to the correct GG items to justify the facility's reimbursement tier.
The failure point is almost never clinical—it is transcriptional. A therapist verbally documents "two-person assist," but the note reaching the MDS coordinator reads "supervision." That single-word drift separates a high-acuity CMG from a denied claim.
CMS publishes the PDPM GROUPER logic and ICD-10 mappings, but publishes nothing that captures the spoken clinical reality before it decays into an under-coded note. That capture layer is the gap. See how functional capture varies by discipline in our Clinical Specialties Directory.
Clinical Logic: The Late-Friday Post-Stroke Admission
This is the scenario that defines the ROI case for Medical AI Scribing in SNFs. Walk through it exactly as it unfolds on the floor.
The setup is deceptively routine: A 79-year-old post-stroke resident is admitted late Friday afternoon. The OT and PT evals happen, but weekend MDS coverage is thin. Weekend notes default to "supervision" for transfers and dressing.
That single default silently drops the resident out of the high PT/OT case-mix group. The cost is $320/day for 7 days—$2,240 before anyone notices, and it exposes the claim to an ADR denial.
What Scribing.io does, step by step, is convert the spoken eval into defensible coded evidence:
Step | Scribing.io Action | Clinical Artifact Produced |
|---|---|---|
1. Capture | Records and transcribes the OT eval audio at bedside on Friday | Timestamped transcript |
2. Extract | Isolates functional cues: "requires two-person assist for sit-to-stand," "unable to button shirt" | Structured self-care and mobility markers |
3. Map | Maps sit-to-stand transfer to GG0170E; upper-body dressing to GG0130F | GG item assignment |
4. Code | Applies coded values 01 (Dependent) and 02 (Substantial/max assist) per captured cues | Auto-coded GG scores |
5. Flag | Detects the "supervision" default conflicts with captured audio; raises discrepancy alert before the 3-day window closes | Discrepancy alert |
6. Push | Pushes corrected GG scores plus I0020B primary clinical category to MDS | Corrected MDS entry and PDPM logic tape |
The result restores accurate reimbursement: The corrected GG0170E and GG0130F scores return the resident to the accurate PT/OT case-mix group. The discrepancy is resolved inside the assessment window.
The facility then carries an auditable logic tape into any future ADR—preventing the denial and restoring the full reimbursement tier. Run your own facility's numbers with the AI Medical Scribe ROI Calculator.
Beyond CMS Mappings: The Auditable PDPM Logic Tape
Every SNF has access to the CMS PDPM GROUPER logic and the FY 2026 ICD-10 mappings. What no fact sheet, ZIP file, or GROUPER download provides is the operational bridge between the spoken encounter and the coded MDS record—captured in real time and defensible under audit.
The Original Insight, grounded in the Anchor Truth of Section GG capture, is that Clinical-Grade Scribing does not stop at extraction. It performs a chain of governed actions:
Real-time 06 to 01 auto-coding: Functional performance cues convert to GG-compliant coded values as the eval is spoken, not reconstructed days later from memory.
3-day admission-window enforcement: The system tracks the assessment reference period and forces discrepancy resolution before the window closes—the exact failure point in the late-Friday scenario.
MDS 3.0 v1.18.11 export: Corrected scores export in the current item-set version rather than requiring manual re-entry.
I0020B crosswalking: Transcripts map to the primary ICD-10 clinical category that drives the PDPM clinical category assignment.
FHIR R4 interoperability payloads: GG observations post as structured resources, aligning with 2026 CMS interoperability standards.
Auditable PDPM logic tape for ADRs: Every coding decision traces back to the source phrase in the audio transcript.
What CMS resources leave unaddressed is that the mappings assume the input data is already correct. They govern the math, not the capture. A GROUPER cannot detect that "supervision" should have been "two-person assist."
Ambient Clinical Intelligence closes that gap by anchoring every code to an auditable spoken source. Explore how this exports into your record system via the EHR Integration Library.
Technical Reference: ICD-10 Documentation Standards
The I0020B primary diagnosis anchors the resident to a PDPM clinical category. Two of the most common and most frequently miscoded post-acute diagnoses are documented below.
ICD-10 Code | Description | Documentation Note | Reference |
|---|---|---|---|
I69.30 | Unspecified sequelae of cerebral infarction | Common post-stroke primary diagnosis; sequelae specificity affects clinical category. Pair with captured GG functional deficits. | |
Z47.1 | Aftercare following joint replacement surgery | Drives Major Joint Replacement clinical category; requires supporting orthopedic history documentation. |
In the case study above, an I69.30 primary diagnosis paired with under-coded GG scores is the exact combination that triggers ADR scrutiny. The diagnosis signals high acuity while the functional coding contradicts it.
The Scribing.io I0020B crosswalk keeps the clinical category and the functional evidence aligned, preventing the contradiction that examiners search for first.
Manual Weekend MDS vs. Scribing.io Workflow
The operational difference becomes concrete when the weekend workflow is placed side by side against the ambient capture workflow.
Checkpoint | Manual Weekend Workflow | Scribing.io Workflow |
|---|---|---|
Functional data source | Reconstructed from memory or abbreviated notes | Verbatim eval audio transcript |
GG coding timing | Deferred to Monday, often past window | Coded live during the eval |
Discrepancy detection | Manual, if noticed at all | Automated alert before 3-day close |
ADR defensibility | Narrative recall, weak audit trail | Phrase-linked logic tape |
Reimbursement exposure | Up to $320/day silent loss | Case-mix tier preserved |
The single decisive variable is timing against the assessment reference period. Manual workflows fight the clock; Ambient Clinical Intelligence enforces it.
Pricing and Compliance Governance
Governance under SB 1120 requires that any algorithmic contribution to clinical documentation remain under licensed clinician review. Scribing.io routes every auto-coded GG value to the MDS coordinator for attestation before MDS submission.
The financial model scales with census and eval volume rather than flat seat counts. Review the tiers and audit-defense inclusions on Scribing.io Pricing and Plans.
For Clinical Operations Directors weighing a facility-wide rollout, the recommended path is a single-unit pilot measuring recovered case-mix days against the manual baseline. The AI Medical Scribe ROI Calculator models that recovery directly against your PDPM per-diem.
