The CY 2026 Home Health Prospective Payment System Final Rule, published by the Centers for Medicare and Medicaid Services (CMS) on November 28, 2025, recalibrated PDGM case-mix weights and LUPA thresholds and projects an aggregate 1.3% decrease in Medicare payments to home health agencies. For agencies already operating on thin margins, the pressure that regulatory change places on coding and documentation accuracy is direct and measurable. Every Medicare payment a home health agency receives under PDGM begins with two decisions: how a clinician documents the OASIS and which ICD-10 diagnosis a coder assigns. When both are accurate and aligned, the episode payment is correct.

When either is wrong, the agency is placed in the wrong payment group and that revenue cannot be recovered after the claim clears. The CY 2026 Final Rule projects an aggregate 1.3% payment decrease, approximately $220 million across the industry. Agencies that allow coding and documentation errors to persist are losing revenue against a payment baseline CMS has already reduced.

This guide covers how home health OASIS coding and ICD-10 diagnosis coding work together under PDGM. It explains where errors originate, how QA should be structured, and what the in-house versus outsourced decision depends on in 2026.

Key Takeaways

Under PDGM, every 30-day episode payment is determined by the accuracy of two decisions — the ICD-10 primary diagnosis and the OASIS functional score. The CY 2026 Final Rule's case-mix recalibration has made errors in either more financially consequential than under any prior model.

• PDGM assigns each 30-day episode to one of 432 payment groups based on primary ICD-10 diagnosis, OASIS functional scoring, secondary diagnoses, and admission source. An error in any of these four variables changes the HIPPS code and the episode payment.
• The CY 2026 Final Rule recalibrated PDGM case-mix weights and LUPA thresholds using 2024 claims data. Agencies that have not reviewed coding patterns against the new weights are operating on assumptions that no longer reflect current payment rates.
• OASIS-E2 became the operative standard on April 1, 2026. Three items are now required at the Resumption of Care time point. Agencies whose EHR systems were not updated by that date began generating non-compliant submissions immediately.
• The FY 2026 ICD-10-CM code set includes thousands of code changes. EMR systems that have not been updated carry retired or replaced codes that generate clearinghouse rejections before a clinical review occurs.
• The most consistent coding errors across home health agencies are not individual mistakes from undertrained staff. They are systemic gaps in QA that allow wrong primary diagnoses, missing comorbidity codes, and inaccurate GG scores to reach the claim without correction.
• Reviewing OASIS and coding QA before billing, not after denial, is the most effective change agencies can make to protect reimbursement in 2026.

What This Guide Covers

• How PDGM Uses Coding and OASIS to Determine Payment
• ICD-10 Primary Diagnosis Selection Under PDGM
• OASIS-E Documentation and Its Coding Consequences
• Functional Scoring Under PDGM: The GG Items
• The 7 Most Common Home Health Coding Errors
• OASIS M-Item Scoring Errors and Audit Risk
• How to Structure Coding and OASIS QA in 2026
• In-House vs. Outsourced Coding: The 2026 Decision Framework
• Red Road's Coding and OASIS Review Process
• Frequently Asked Questions

How PDGM Uses Coding and OASIS to Determine Payment

PDGM replaced the prior visit-volume payment model with a patient-characteristic model. Each 30-day episode of care is assigned to one of 432 payment groups based on four variables. The primary ICD-10 diagnosis determines the clinical group. OASIS GG functional scores determine the functional impairment level. Secondary diagnoses determine the comorbidity adjustment. Admission source and timing determine whether the period is early or late, community or institutional. All four variables feed into the HIPPS code that drives Medicare payment.

An error in any of these four variables changes the HIPPS code and the payment group. When the payment group is wrong, the agency receives either more or less than it earned. Both outcomes carry audit exposure.

CMS published the CY 2026 HH PPS Final Rule fact sheet on November 28, 2025. The recalibrated case-mix weights and updated LUPA thresholds are now in effect. Agencies that have not reviewed coding patterns against those weights are operating on assumptions from a prior model.

ICD-10 Primary Diagnosis Selection Under PDGM

The primary ICD-10 diagnosis is the single most consequential coding decision under PDGM. It assigns the episode to one of 12 clinical groups, which determines the base case-mix weight and sets the payment foundation for the episode. An incorrect primary diagnosis does not simply reduce payment. It can move the episode into a different clinical group with an entirely different payment profile.

CMS maintains the accepted primary diagnosis list per clinical group. If the primary diagnosis submitted on a claim does not fall within the accepted codes for any clinical group, the claim is returned to the agency at the grouper level, before any clinical review.

The most consistent error pattern is selecting a diagnosis that reflects the patient's medical history rather than the primary reason for home health services at that episode. Agencies should ensure coders review the home health plan of care and physician orders before assigning the primary diagnosis. Coders who rely on the hospital discharge summary alone frequently assign a primary diagnosis that reflects the acute episode rather than the home health service focus, resulting in a clinical group mismatch that changes the episode payment.

ICD-10 specificity is a related and equally significant problem. The FY 2026 ICD-10-CM code set includes thousands of code changes. Codes valid in 2025 may be retired or replaced by more specific codes in 2026. EMR systems that have not been updated to the FY 2026 code set display retired codes as valid options. Coders using those systems generate claims that are rejected at the clearinghouse before a human reviewer sees them. Beyond invalid codes, unspecified codes carry lower case-mix weight than their specific counterparts and increase audit scrutiny. When the clinical record supports a specific code, using an unspecified one is an accuracy failure that reduces payment and creates documentation gaps.

Red Road Insight: The most common primary diagnosis error across the agencies we review is not a misunderstanding of PDGM clinical groups; it is a coder working from the hospital discharge summary rather than the home health plan of care. The discharge summary reflects the acute episode. The home health plan of care reflects the reason for home health services. When they differ, the coder must use the plan of care. Agencies that do not establish this as a written coding protocol generate the same clinical group mismatch repeatedly.

OASIS-E Documentation and Its Coding Consequences

OASIS documentation and ICD-10 coding must tell the same clinical story. When they diverge, agencies face both a payment accuracy problem and a compliance exposure. Reviewers compare OASIS responses and coding decisions as a pair, and a mismatch between the two is treated as a documentation flag, not a minor discrepancy.

OASIS-E added functional scoring depth, SDOH items, and standardized cognitive and mood assessments. OASIS-E1 in January 2025 made targeted refinements. OASIS-E2, effective April 1, 2026, added required items to the Resumption of Care time point and replaced several structural items. Agencies whose EHR systems were not updated by April 1 began generating non-compliant ROC submissions. A full version-by-version breakdown is in the OASIS-E Documentation Guide.

The GG functional scoring items and the M-items carry the most direct payment consequence. Both are covered in dedicated sections below. SDOH items covering housing stability, food access, and transportation are also required for all payers. The all-payer mandate became mandatory on July 1, 2025. Agencies that have not updated non-Medicare workflows are carrying compliance exposure across their full census.

Functional Scoring Under PDGM: The GG Items

The GG items, GG0130 (Self-Care) and GG0170 (Mobility), establish the functional impairment level that determines the PDGM payment group. Under the CY 2026 Final Rule, functional impairment levels and their case-mix weights were recalibrated using 2024 claims data. The scoring rules have not changed, but the financial consequences of inaccurate scoring are more precisely reflected in the updated model.

GG scoring requires documented observation of actual task performance. CMS requires clinicians to score based on what the patient demonstrates, not what the patient reports they can do. Complex patients with pain, fatigue, or cognitive impairment consistently report higher functional capability than they demonstrate when observed, and clinicians who rely on self-report systematically undercode impairment, placing patients in lower-weight PDGM groups.

This pattern reflects a workflow problem, not an individual training failure. When clinicians complete OASIS during time-pressured Start of Care visits without a structured observation protocol, self-report scoring becomes the default. Agencies that do not build structured observation into the SOC workflow will encounter systematic functional undercoding across the census.

The financial consequence is measurable. The reimbursement difference between adjacent PDGM functional impairment levels can exceed $500 per 30-day period. Across a 100-patient census, consistent undercoding of functional status by even one impairment level produces five-figure monthly revenue loss before any denied claims are counted.

Red Road Insight: Functional undercoding from self-report scoring is not a training problem that resolves after a refresher session. It is a workflow problem. When clinicians arrive at SOC visits with a full assessment load and no structured observation protocol, self-report becomes the default because it is faster. Agencies that do not build specific task observation into the SOC workflow, with documentation requirements that make the observation visible in the record, will see the same undercoding pattern across every coder who reviews those assessments.

The 7 Most Common Home Health Coding Errors

Across pre-billing reviews and post-payment audits, the same coding error patterns appear in home health agencies regardless of size or EMR platform. These are not isolated mistakes from individual coders. They are systemic gaps that persist when agencies run QA on a retrospective, sampled basis without connecting error findings to the specific PDGM variables they affect.

The seven most consistent errors are:

• Primary diagnosis reflects the admitting condition rather than the reason for home health services. This is the most common PDGM clinical group mismatch. Origin: coder protocol, requires a written rule requiring review of the plan of care before primary diagnosis assignment.
• Secondary diagnoses present in the clinical record are not coded to the specificity required for comorbidity adjustment.
• Unspecified ICD-10 codes are assigned where physician documentation supports a more specific code.
• Outdated ICD-10 code sets in the EMR display codes that were retired or replaced in FY 2026 as still valid. Origin: EHR configuration, requires IT verification of FY 2026 code set update.
• Symptom codes are used as primary diagnoses when the underlying condition is documented and codeable.
• Coding decisions are made from the hospital discharge summary alone, without reviewing the home health plan of care.
• There is no reconciliation between coding decisions and OASIS responses before the claim is submitted. Origin: QA process gap, requires a pre-billing sign-off step that checks both functions against the same HIPPS code.

Clinical examples and correction guidance for each error type are in the 7 Most Common Home Health Coding Errors guide.

OASIS M-Item Scoring Errors and Audit Risk

The M-items cover wound status, skin conditions, and specific health conditions. They are among the most frequently reviewed items in Medicare documentation audits because they carry both payment and quality implications.

The most common M-item failure is a mismatch between the coded OASIS response and the narrative clinical note. A nurse documents in free text that a wound is present, actively treated, and measured at a specific stage. The corresponding M-item is coded as absent or at a lower severity. The coded OASIS response is what drives payment and public quality reporting. The narrative note is what a Medicare auditor reads. When they conflict, the agency has no defensible position on either.

The second consistent pattern is M-item responses left blank when the clinical condition is ambiguous. CMS guidance on when to use the dash response versus a scored response is specific. Clinicians who default to blank or dash when a condition is present but uncertain generate incomplete OASIS submissions. In audit, these appear as underdocumented care.

Red Road Insight: The narrative-code mismatch on M-items is the most defensible error type in audit, and the most avoidable. The clinical evidence is in the note. The OASIS response contradicts it. MAC reviewers flag this not because the care was undocumented but because the coded response and the narrative tell different stories. A pre-billing review that checks M-item responses against the corresponding narrative note catches this before submission. Without that check, the mismatch reaches the payer uncorrected.

M-item scoring accuracy, common response errors, and narrative alignment guidance are in the Common OASIS M-Item Scoring Errors guide.

How to Structure Coding and OASIS QA in 2026

Most home health agencies maintain some form of coding and OASIS QA. The structural failure is not the absence of review processes. It is review processes that operate retrospectively, on sampled charts, without a decision-based framework. Retrospective sampled QA identifies errors after claims have been submitted and does not prevent those errors from affecting payment.

Effective QA in 2026 operates at three points in the episode timeline.

Start of Care: The Highest-Leverage Point

Start of Care is where the HIPPS code is established, and errors caught at this stage are corrected before any payment consequence occurs. Agencies should use Start of Care QA to confirm that the primary diagnosis reflects the reason for home health services, secondary diagnoses are coded to comorbidity-qualifying specificity, and GG functional scores are supported by observed performance documented in the clinical notes. This is where most revenue protection occurs.

Recertification: Continuity and Clinical Story

Recertification QA confirms continuity with the Start of Care clinical record. Diagnoses must reflect current clinical status, not the admission diagnosis from weeks prior. Functional scores must reflect documented current status, not values copied from the prior period. Episodes where functional scores shift significantly without visit note support are a consistent ADR trigger.

Pre-Billing: The Final Gate

Pre-billing review confirms the claim matches the clinical record before submission. Agencies should verify that OASIS responses and ICD-10 codes are aligned, the HIPPS code reflects the correct PDGM group, and all required OASIS items are present. For ROC assessments completed after April 1, 2026, providers must also confirm that OASIS-E2 required items are included before the claim is submitted.

QA Cadence: Daily, Weekly, Monthly

A detailed QA checklist and pre-billing review framework is in the Pre-Billing Review Checklist guide.

In-House vs. Outsourced Coding: The 2026 Decision Framework

The decision between internal and external coding is a capacity and accuracy question, not primarily a cost question. In-house coding is sustainable when staffing is stable, training is current with annual ICD-10 and PDGM updates, and denial rates stay below 10%. When any of those conditions break down, the cost of inaccuracy exceeds the cost of addressing it through external support.

When In-House Coding Is Sustainable

Internal coding teams have direct access to clinical staff and immediate knowledge of agency-specific workflows. For agencies with stable staffing, structured QA, and denial rates below 10%, in-house coding is operationally viable. Agencies must treat coder training as a continuous function rather than an annual event, as annual ICD-10 changes, PDGM recalibration, OASIS version updates, and MAC bulletin changes require real-time distribution to coding staff to maintain accuracy.

When External Support Solves a Real Problem

Outsourcing coding and OASIS review addresses specific failure conditions. Denial rates above 10% indicate a systematic accuracy problem that internal QA has not resolved. Staff turnover in coding or clinical positions creates coverage gaps that cannot be filled quickly without accuracy declining. Census growth or new acquisitions generate volume that in-house teams cannot absorb without error rates rising.

External coding support gives agencies access to coders trained in home health ICD-10 and OASIS, kept current on PDGM recalibration and MAC bulletins, and checked daily for accuracy. Building that specialization internally is costly and fragile when staff leave.

How Red Road Supports Home Health Coding and OASIS Review

Home health agencies that perform consistently under PDGM use layered review processes that connect OASIS documentation to coding decisions before any claim is submitted. Red Road's home health coding services operate as an embedded clinical back-office function, with direct access to the clinical team, no intermediary account management layer, and daily QA accountability on every coder.

Multi-Layer Review

Every chart moves through a structured sequence. Coder review confirms ICD-10 specificity, primary and secondary diagnosis selection, and alignment with physician documentation and the plan of care. Clinical validation by Registered Nurses confirms OASIS responses, GG functional scoring, M-item accuracy, and comorbidity capture. Using Red Road's CHAP-verified Skilled Nursing Clinical Audit Tool, agencies can effectively identify and communicate opportunities for improvement in clinical care, documentation quality, and compliance with CHAP Standards.

Compliance sampling compares records against CMS guidelines and MAC-specific bulletins, with corrective actions tracked to closure. Final reconciliation confirms clean submission with documented rationale.

PDGM Alignment Checks

Every review includes OASIS-to-code crosswalks confirming that functional scoring and diagnosis selection support the same HIPPS code. Red Road checks comorbidity coding against the CMS qualifying pairs list. The review flags LUPA risk where visit frequency planning puts an episode below the HIPPS-specific threshold. MAC bulletins are tracked so local coverage requirements are reflected in real-time coding decisions.

Turnaround and EHR Integration

Standard turnaround is 24 to 48 hours, with a same-day pathway for priority cases. Red Road operates within the EHR systems agencies already use, including Kinnser (WellSky), Axxess, Homecare Homebase (HCHB), and KanTime. Red Road tracks and reports accuracy, on-time completion, and query response rates weekly and monthly, with trend analysis identifying root causes rather than individual errors.

Onboarding Structure

Onboarding runs across six phases: scope definition by service line, payer mix, and PDGM profile; HIPAA controls and BAA execution; playbook creation for coding rules, escalation paths, and turnaround targets; a parallel review period of one to two weeks to calibrate decisions; go-live with daily trackers and a single point of contact; and ongoing governance through weekly huddles, monthly reviews, and quarterly audits with action logs.

The Bottom Line

Accurate home health OASIS coding is not a compliance function that runs parallel to agency operations. It is an operational function that determines whether those operations are financially sustainable under a payment model that rewards documentation precision.

The CY 2026 Final Rule recalibrated case-mix weights and updated LUPA thresholds. Coding inaccuracy now costs more per episode than it did under prior weights. Agencies that allow errors in functional scoring, primary diagnosis selection, or comorbidity capture to persist are losing revenue against a payment baseline CMS has already reduced.

Explore how Red Road's home health coding services support reimbursement accuracy and audit readiness.

Regulatory Sources

• CMS CY 2026 Home Health PPS Final Rule Fact Sheet, November 28, 2025 (cms.gov)
• CMS Patient-Driven Groupings Model (PDGM) — Clinical Groupings and Payment Structure (cms.gov)
• FY 2026 ICD-10-CM Official Guidelines for Coding and Reporting (cms.gov)
• CMS OASIS-E2 Guidance Manual, effective April 1, 2026 (cms.gov)
• CMS LUPA Threshold Reference by HIPPS Code, CY 2026 (cms.gov)

Compliance Disclaimer

This content reflects CMS guidance and regulatory standards as of the publication date. Agencies should verify current requirements against the most recent CMS transmittals, Final Rules, and MAC bulletins. Consult your legal and compliance advisors before making coding or operational decisions based on this content.