ONC HTI-2 USCDI version 5 data elements for ambulatory EHR documentation

ONC HTI-2 and USCDI Version 5 Data Elements: The 2026 Ambulatory EHR Documentation Playbook

Merry AI · Thoughtfully curated clinical briefs.

USCDI version 5 is voluntary, not a federal mandate. Ambulatory practices must meet the USCDI v3 baseline that HTI-1 required in certified EHRs by January 1, 2026. The HTI-2 proposed rule named USCDI v4, not v5, as the next certification baseline. USCDI v5 adds Name to Use, Pronouns, Interpreter Needed, Emergency Department Note, Operative Note, and a new Orders data class. Certified developers may adopt v5 early through the ONC Standards Version Advancement Process (SVAP). Your documentation workflow should capture these elements as structured, attestable data now, so a later federal baseline does not require rebuilding templates.

Core rule: Comply with v3, configure toward v4, document for v5.

USCDI v3 is the enforced 2026 baseline under HTI-1

The HTI-1 final rule, published in January 2024, set USCDI v3 as the required standard for ONC Health IT Certification. Certified developers had until January 1, 2026 to update their products. That makes USCDI v3 the operative floor for any ambulatory practice that uses Certified EHR Technology (CEHRT) for Medicare Promoting Interoperability or MIPS reporting.

The Office of the National Coordinator now operates as the Assistant Secretary for Technology Policy (ASTP/ONC), following its 2024 reorganization. It publishes a new USCDI version each July. Each version is a candidate for future certification rules. Publication alone does not make a version mandatory.

Practice Administrators should confirm three facts with their EHR vendor in writing: the USCDI version the certified product supports, the SVAP versions the vendor has voluntarily adopted, and the US Core FHIR Implementation Guide version behind the patient access and bulk data APIs.

HTI-2 proposed USCDI v4 as the next certification baseline

The HTI-2 proposed rule, released in July 2024, proposed USCDI v4 as the next required version, with a proposed compliance date of January 1, 2028. ASTP/ONC split HTI-2 into separate final actions. Later 2025 rulemaking (including HTI-4, finalized alongside CMS payment rules) focused on electronic prescribing, real-time prescription benefit, and electronic prior authorization. Federal deregulatory priorities have also prompted proposals to reduce certification burden.

The operational takeaway is simple: no final federal rule currently requires USCDI v5 in ambulatory certified EHRs. Practices that design documentation around v5 data elements are building ahead of the regulatory curve, not chasing a 2026 deadline.

Verify before you configure: track ASTP/ONC final rules, not proposed rule headlines.

USCDI v5 adds identity, language access, note types, and Orders

USCDI v5, published in July 2024, builds on v4. The v4 additions included Facility Information, Treatment Intervention Preference, Care Experience Preference, Alcohol Use, Substance Use, Physical Activity, Medication Adherence, and Average Blood Pressure. The v5 additions that matter most to ambulatory documentation fall into four groups.

• Patient demographics and identity: Name to Use and Pronouns support accurate patient identification and communication. Confirm the current federal enforcement posture on demographic elements before you build templates, because 2025 federal actions affected several demographic data policies.
• Language access: Interpreter Needed records whether the patient requires interpreter services. This directly affects visit time, care coordination, and documented barriers to care.
• Clinical note types: Emergency Department Note and Operative Note join the existing note types, including Consultation Note, Discharge Summary, History and Physical, Progress Note, and Procedure Note.
• Orders data class: Orders standardizes how ordered laboratory, imaging, clinical test, and procedure services are represented for exchange, and links each order to its clinical context.

Ambulatory practices receive the most immediate value from Interpreter Needed and the Orders data class. Both convert narrative workflow details into structured fields that payers, auditors, and receiving clinicians can query.

Structured SDOH capture supports both USCDI exchange and MDM scoring

Social Determinants of Health data elements (SDOH Assessment, SDOH Goals, SDOH Interventions, and SDOH Problems/Health Concerns) entered USCDI in version 2. Every later version, including v5, carries them forward. A structured SDOH problem uses ICD-10-CM Z codes, which give the data both interoperability value and billing value.

Under the 2021 AMA E/M guidelines, "diagnosis or treatment significantly limited by social determinants of health" is an example of Moderate risk in Medical Decision Making (MDM). The clinician must document how the social factor changed the plan. Listing the code alone does not meet this standard.

• Housing instability: Z59.819 (ICD-10-CM) captures housing instability, housed, unspecified. Pair it with a documented plan change, such as switching from an insulin regimen that requires refrigeration.
• Social environment problems: Z60.9 (ICD-10-CM) captures an unspecified problem related to the social environment. Use more specific Z60 codes when the encounter documentation supports them.

An ambient AI scribe should extract the clinician's spoken reasoning ("she can't keep insulin cold at the shelter, so we're starting a weekly GLP-1") into a structured assessment. The output should name the Z code and state the treatment limitation in a single, auditable sentence.

USCDI v5 elements map directly to MDM thresholds and audit defenses

The table below links each data element to its clinical indicator, the MDM element it supports, and the documentation that defends it under payer review.

Audit principle: a structured field without clinical reasoning will not survive review.

CPT G2211 complexity capture recovers $15,600+ per clinician annually

CMS activated HCPCS add-on code G2211 on January 1, 2024 for office and outpatient E/M visits. It applies when the clinician is the continuing focal point for all needed services, or provides ongoing care for a single serious or complex condition. Beginning in CY 2025, CMS also allows G2211 when the base E/M is reported with modifier 25 on the same day as an annual wellness visit, a vaccine administration, or a Medicare Part B preventive service.

At a national payment of roughly $16 per claim, a clinician who captures G2211 on about 975 eligible Medicare visits per year recovers $15,600+ in annual revenue that would otherwise be missed. Most practices lose this revenue through documentation gaps, not because the visits fail to qualify.

The documentation fix requires the note to state the longitudinal relationship explicitly. USCDI-structured Problems, Care Team Member, and Progress Note data provide the evidence trail. Review the full attestation logic in our CPT G2211 Complexity Capture Architecture.

A $648 annual scribe costs 1.3% of a $48,000 MA

A fully loaded medical assistant costs about $48,000 per year once you include wages, payroll taxes, benefits, and training. Merry AI Pro costs $54 per month on annual billing, or $648 per year. That equals 1.3% of the MA labor denominator.

CMOs should frame the decision as reallocation, not replacement. MAs move from transcription to rooming, care gap closure, and SDOH screening. Those tasks produce the USCDI-structured data the scribe then turns into defensible MDM.

Browser overlay scribes write into certified EHR fields without API contracts

An AI scribe is not itself Certified Health IT. The certified EHR remains the system of record that stores and exchanges USCDI data through US Core FHIR APIs. The scribe's job is to place clinically accurate content into the correct EHR fields, so certified exchange carries complete data downstream.

Browser overlay architecture places structured output into existing EHR note sections without a vendor integration project, an API fee, or a multi-month interface queue. Compare deployment models in our analysis of EHR Browser Overlay Architecture.

Architecture rule: the EHR certifies the data; the scribe ensures completeness.

State AI laws add obligations on top of federal USCDI requirements

Federal USCDI rules govern data standards. State law governs how AI touches clinical decisions and patient communications. California SB 1120 (the Physicians Make Decisions Act) requires that a licensed physician or qualified health care professional make medical necessity determinations in utilization review, not an automated tool alone. California AB 3030 requires disclaimers on generative AI patient communications about clinical information that a licensed provider has not reviewed.

Multistate groups need a single compliance map that pairs each state's AI disclosure, consent, and review requirements with the documentation workflow. Use the State AI Scribe Rules & Statutory Compliance Hub to confirm requirements before rollout in each jurisdiction.

A 90-day rollout aligns documentation with USCDI v5 readiness

The following sequence fits a typical 5 to 25 clinician ambulatory group.

1. Days 1 to 15: Obtain written confirmation of your EHR's certified USCDI version, SVAP adoptions, and US Core version. Audit 50 recent notes for missing Interpreter Needed, SDOH, and order-reason fields.
2. Days 16 to 45: Deploy ambient scribing with templates that output SDOH Z codes, the treatment limitation statement, order reasons, and G2211 longitudinal attestation language.
3. Days 46 to 75: Run a coding audit comparing MDM levels and G2211 capture before and after deployment. Document the variance and the clinical reasons behind it.
4. Days 76 to 90: Finalize state AI disclosure workflows. Train clinicians to attest every AI-generated note. Set a quarterly review of ASTP/ONC rulemaking.

Frequently asked questions about HTI-2 and USCDI v5

USCDI v5 is not required for 2026 certification

No final federal rule mandates USCDI v5. HTI-1 requires USCDI v3 as of January 1, 2026. HTI-2 proposed v4 as the next baseline. Developers may adopt v5 voluntarily through SVAP.

Ambulatory practices should still document toward v5 now

Structured capture of Interpreter Needed, order reasons, and SDOH supports MDM scoring, G2211 attestation, and audit defense today. It also prevents template rework when a later federal baseline takes effect.

SDOH Z codes raise MDM risk only with documented impact

Codes such as Z59.819 and Z60.9 support Moderate risk only when the note explains how the social factor significantly limited diagnosis or treatment.

AI scribes support certified EHRs but do not replace certification

The certified EHR stores and exchanges USCDI data. The AI scribe makes sure clinician reasoning reaches the correct structured fields, with clinician attestation on every note.

Bottom line: v3 is enforced, v4 is proposed, v5 is your documentation target.