PDGM Payment Model Breakdown for Home Health Agencies

The Centers for Medicare and Medicaid Services (CMS) finalized a 3.0% temporary reduction and a 1.023% permanent adjustment to the 2026 home health payment rate under the Patient-Driven Groupings Model (PDGM), and recalibrated the case-mix weights using 2024 claims data. This PDGM payment model breakdown explains how five case-mix variables place each 30-day period into one of 432 payment groups, and which agency data entries control the result.

CMS issued the calendar year (CY) 2026 Home Health Prospective Payment System Final Rule on November 28, 2025. The rule estimates an aggregate 1.3% decrease in Medicare payments to home health agencies (HHAs) compared with CY 2025. It also recalibrated PDGM case-mix weights, Low Utilization Payment Adjustment (LUPA) thresholds, functional impairment levels, and comorbidity subgroups using CY 2024 claims data.

The rule applied a permanent adjustment of 1.023% and a temporary reduction of 3.0% to the 30-day payment rate. These followed permanent reductions of 3.925% in CY 2023, 2.890% in CY 2024, and 1.975% in CY 2025. Each reflects the gap CMS finds between assumed and actual behavior under PDGM.

Against this payment baseline, an error in any PDGM variable carries more consequence. Under PDGM, Medicare pays an HHA for each 30-day period of care, and each period falls into one of 432 case-mix groups. Each group has its own case-mix weight and its own LUPA threshold. This guide explains how each variable is set, where agencies lose payment under each one, and what QA controls prevent those losses.

Coding and Outcome and Assessment Information Set (OASIS) documentation both drive PDGM payment. Agencies should review the Complete Guide to Home Health Coding and OASIS Review in 2026 for the broader view. For the coding errors that most often reduce payment, refer to the 7 Home Health Coding Errors guide.

Key Takeaways

• PDGM sorts each 30-day period into one of 432 payment groups using five variables: admission source, timing, clinical group, functional impairment level, and comorbidity adjustment. Four variables come from Medicare claims. The functional impairment level comes from OASIS.
• Medicare pays the national standardized 30-day rate multiplied by the case-mix weight of the group, with the labor portion adjusted by the wage index. The CY 2026 rate is $2,038.22 for agencies that submit required quality data.
• A period with fewer visits than the LUPA threshold of its group is paid per visit, not at the full 30-day rate. CMS sets the threshold at the 10th percentile of visits for the group, with a minimum of 2 visits.
• CMS applied a permanent adjustment of 1.023% and a temporary reduction of 3.0% for CY 2026. These adjustments apply to every agency, so accuracy within an agency decides whether each period lands in the correct group.
• Each variable is set at a different handoff: intake, coding, clinical documentation, or scheduling. Each handoff needs its own QA checkpoint.
• PDGM optimization means that each variable reflects what the clinical record supports. It does not mean selecting a group that pays more than the record supports.

How PDGM Works in Home Health

The 30-Day Period of Payment

Medicare pays an HHA a national standardized 30-day period payment rate when a period of care meets a visit threshold. Periods below the threshold are paid a per-visit rate for each discipline that provided care.

The 30-day rate covers six disciplines and routine and nonroutine medical supplies (CMS Medicare Benefit Policy Manual, Chapter 7, Section 10). The disciplines are skilled nursing, home health aide services, physical therapy, speech-language pathology, occupational therapy, and medical social services. Durable medical equipment is paid separately under the fee schedule.

Recertification and plan of care review still occur every 60 days. Each 60-day certification can include two 30-day payment periods, and each period is assigned to its own payment group.

The Five Case-Mix Variables

PDGM assigns each 30-day period to one subgroup in each of five categories. The table shows what each variable measures, how many subgroups it has, and where the data originates.

Multiplying the subgroup counts gives 2 × 2 × 12 × 3 × 3, which equals 432 payment groups. The case-mix weight of each group reflects its predicted mean cost relative to the average across all groups.

CMS recalibrates the weights, functional impairment levels, comorbidity subgroups, and LUPA thresholds each year using recent claims data.

Admission Source and Timing

Admission source is institutional when the patient had an inpatient stay within 14 days before home health admission. The stay can be in an acute hospital, inpatient rehabilitation facility, skilled nursing facility, long-term care hospital, or inpatient psychiatric facility. All other admissions are community admissions.

Timing is early for the first 30-day period of care and late for all later periods. A gap of more than 60 days between periods restarts the sequence. Both variables come from claim history, not from the clinical assessment.

Clinical Group

The principal diagnosis reported on the claim assigns the period to one of 12 clinical groups. Five groups are musculoskeletal rehabilitation, neuro/stroke rehabilitation, wounds, behavioral health, and complex nursing interventions.

The other seven groups are subgroups of Medication Management, Teaching, and Assessment (MMTA). They are surgical aftercare, cardiac and circulatory, endocrine, gastrointestinal and genitourinary, infectious disease with neoplasms and blood-forming diseases, respiratory, and other.

Because the principal diagnosis sets the group, a change in the principal diagnosis can move the whole 30-day period into a different group with a different case-mix weight. The codes come from the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM). The PDGM Primary Diagnosis guide covers the selection logic.

Functional Impairment Level and Comorbidity Adjustment

The functional impairment level is the only variable drawn from OASIS. CMS assigns a low, medium, or high level based on responses to certain OASIS items. CMS recalibrates these levels each year together with the case-mix weights.

The comorbidity adjustment comes from secondary diagnoses reported on the claim. A period receives no adjustment, a low adjustment, or a high adjustment. CMS recalibrates the comorbidity subgroups each year, so the list of qualifying diagnoses can change.

How the Variables Combine

Two patients with the same principal diagnosis can fall into different payment groups. One patient is admitted five days after an acute hospital discharge, which makes the admission institutional. The other patient comes from a physician office, which makes the admission community.

Both periods are early if each is the first 30-day period, yet the two belong to different groups. If the patients also differ in functional level or secondary diagnoses, the groups differ again. The agency data entries, not the diagnosis alone, determine the final group.

PDGM Payment Model Explained: How Payment Is Calculated

Base Rate, Case-Mix Weight, and Wage Index

Payment for a 30-day period starts with the national standardized 30-day rate. CMS multiplies that rate by the case-mix weight of the group. CMS then adjusts the labor portion by the wage index for the location of the beneficiary.

The CY 2026 rate starts from the CY 2025 rate of $2,057.35. CMS applied a permanent behavior adjustment factor of 0.98977, a 2.4% payment update, two budget neutrality factors, and a temporary adjustment factor of 0.97000. The result is $2,038.22 for agencies that submit required quality data and $1,998.41 for agencies that do not (a 2 percentage point reduction from the payment update).

The temporary adjustment applies only to the year in which CMS finalizes it. CMS does not carry the factor into the starting rate for the following year.

Per-Visit Rates and LUPA

The LUPA threshold varies by payment group. CMS sets it at the 10th percentile of visits for each group, with a minimum of 2 visits. A period below the threshold is paid the wage-adjusted national per-visit amount for each visit instead of the full 30-day payment.

The CY 2026 per-visit amounts are $176.96 for skilled nursing and $193.42 for physical therapy. They are $194.74 for occupational therapy, $210.25 for speech-language pathology, $283.64 for medical social services, and $80.12 for home health aide services.

A LUPA that occurs as the only period, or as the first period in a sequence of adjacent periods, receives an add-on to the first visit. The CY 2026 add-on factors are 1.7200 for skilled nursing, 1.6225 for physical therapy, 1.6696 for speech-language pathology, and 1.7238 for occupational therapy.

Outlier Payments

Outlier payments add to the 30-day payment when the imputed cost of care exceeds a threshold. CMS calculates imputed cost by applying per-unit amounts, derived from the per-visit rates, to the visits on the claim.

The threshold for each group is the wage-adjusted period payment plus a fixed-dollar loss amount. The fixed-dollar loss amount is the same for all groups.

CMS pays a set share of the imputed cost above the threshold. The loss-sharing ratio has historically been 0.80, so Medicare pays 80% of the additional cost. The CY 2026 fixed-dollar loss ratio is 0.37, which keeps total outlier payments within 2.5% of estimated total payments.

No HHA may receive more than 10% of its total home health payment as outlier payments.

Partial Payment Adjustment

A partial payment adjustment applies when an intervening event ends the original period early. The two events are a beneficiary-elected transfer, and a discharge and return that warrants a new OASIS assessment, certification, and plan of care.

CMS prorates the original period by the span of days from the first billable service through the last billable service under the original plan of care.

Red Road Insight: Two types of data set the 30-day payment. Claim-based variables are set when the billing record is built. The functional impairment level is set when the OASIS is completed. A coding review that checks only the claim does not test the OASIS-driven variable, and an OASIS review alone does not test the other four.

PDGM Optimization: Where Agencies Lose Payment Under Each Variable

Each PDGM variable is set at a different handoff in the agency workflow, and each carries its own failure pattern. The patterns below follow from how CMS defines the variables. The escalation level shows how fast an agency should act.

Admission Source and Timing Entries

Admission source and timing come from claim history. The intake record must hold the discharge date and setting of any stay in the 14 days before admission. The billing system must also track the sequence of adjacent periods.

An agency that skips this verification risks submitting a claim in a group that does not match claim history. A billing system that misapplies the 60-day gap rule can label a period early or late in error. A coding audit that starts with the principal diagnosis tests neither.

Escalation Level: Immediate: Agencies should confirm admission source from discharge documents and confirm the period sequence from claim history before the billing record is created.

Principal Diagnosis and Clinical Group

The principal diagnosis assigns the clinical group, and the clinical group describes the primary reason for home health care. A principal diagnosis that reflects the hospital discharge condition, not the reason for home health care, can place the period in the wrong group. A diagnosis coded with less detail than the record supports can produce the same result.

Escalation Level: Immediate: Coders must query the clinician before submission when the principal diagnosis does not match the primary reason for skilled care documented at start of care.

Functional Impairment Level

The clinician's OASIS responses set the functional impairment level. A response that conflicts with the narrative note can place the period in a level that the observed function does not support. A level lower than the record supports can reduce the case-mix weight.

A level higher than the record supports exposes the claim to review by a Medicare Administrative Contractor (MAC). CMS recalibrates the functional impairment levels each year, so the same responses can produce a different level in the next year. The OASIS Functional Scoring Under PDGM guide explains how the items are scored.

Escalation Level: Monthly: Agencies should review the distribution of low, medium, and high levels by clinician and branch each month. They should also compare OASIS responses with the narrative note for the same visit date.

Comorbidity Adjustment

The comorbidity adjustment depends on secondary diagnoses reported on the claim. A secondary diagnosis that appears in the physician record but not on the claim does not count toward the adjustment. Agencies must check the capture workflow against the subgroup list for the current year.

Escalation Level: Monthly: Agencies should track the share of periods with no, low, and high adjustments each month, and compare it with prior months and with the clinical group mix.

LUPA Thresholds and Visit Counts

A period below the LUPA threshold of its group is paid per visit. The threshold is specific to the group and changes with annual recalibration. A visit count that avoided a LUPA one year may not avoid it the next.

Visit frequency follows the plan of care and the clinical need of the patient. The threshold check is a monitoring step. Agencies must never add visits to reach a threshold when the visits are not medically necessary.

Escalation Level: Immediate: Clinical managers should review any period where visits fall below the group threshold while days remain, and confirm that visit orders reflect documented patient needs.

Red Road Insight: Each variable fails at a different handoff. Intake sets admission source and timing, coding sets clinical group and comorbidity, clinical documentation sets functional level, and scheduling drives LUPA. An end-of-period review finds errors after the period has closed. Checks placed where each variable is set find the error while the agency can still correct it.

Financial and Compliance Consequences of PDGM Behavior Adjustments

The law requires CMS to assume behavior changes under the 30-day payment unit and PDGM, and to compare them with actual behavior each year. CMS finalized three behavior assumptions in the CY 2019 rule: clinical group coding, comorbidity coding, and the LUPA threshold.

When actual behavior differs from assumed behavior in aggregate, CMS applies permanent or temporary adjustments to the 30-day payment amount. For CY 2026, CMS applied a permanent adjustment of 1.023% and a temporary adjustment of 3.0%.

CMS finalized a smaller permanent adjustment than it proposed. Commenters noted that behavior change after CY 2022 could stem from factors unrelated to PDGM, including the OASIS-E assessment, expanded home health value-based purchasing, and growth in Medicare Advantage enrollment.

The temporary adjustment recoups part of a retrospective overpayment amount that CMS calculated at $4.76 billion for CY 2020 through CY 2022 (CMS-1828-F, Section III.C).

These adjustments apply to the national standardized 30-day payment, so they affect every agency. Accuracy within an agency decides whether each period lands in the group that the clinical record supports.

In this context, PDGM optimization has a defined meaning. Each of the five variables reflects what the clinical record supports, which prevents underpayment from missed or misclassified data. It does not mean selecting a group that pays more than the record supports.

Documentation that does not support a coded variable exposes the claim to MAC review. A review that finds an overpayment results in repayment.

Red Road Insight: A workflow built around a fixed expectation of the weight or LUPA threshold of a group is exposed to each annual recalibration. Checks that reference the current CMS tables and apply to every period stay valid when the numbers change.

Building a QA Process for PDGM Accuracy

Effective QA for PDGM operates before the claim is submitted, at the handoff where each variable is set. Retrospective review finds errors only after the 30-day period has closed.

Pre-Submission Checklist

• Confirm admission source from discharge documents for any stay in the 14 days before admission, and confirm the period sequence in the billing system.
• Confirm the principal diagnosis reflects the primary reason for skilled care documented at start of care, and that the clinician and coder agree on that reason.
• Compare each OASIS functional response with the narrative note for the same visit date before the assessment is locked.
• Reconcile secondary diagnoses in the physician record with the comorbidity subgroups for the current year.
• Compare visits delivered with the LUPA threshold of the group while days remain in the period.
• Confirm that the reference tables in use match the current CMS tables, including case-mix weights, thresholds, per-visit rates, and subgroup lists.

QA Cadence

Escalation Level: Immediate: Agencies must resolve any period flagged by the daily checks before the claim is billed.

Escalation Level: Monthly: Agency leaders should review the group distribution, the LUPA share, and the comorbidity capture rate together each month. A shift in one often explains a shift in another.

How External PDGM Review Supports Payment Accuracy

For agencies managing high admission volumes across many clinicians and branches, review capacity can limit how many periods receive a full check on all five variables. Intake, coding, clinical documentation, and scheduling each set a different variable.

An external review function acts as a concurrent checkpoint before submission. It compares the principal diagnosis, the secondary diagnoses, and the OASIS functional responses against one another. It also checks the claim-based variables against intake records.

The function reports findings by coder, clinician, and clinical group. That reporting shows which handoff produces a recurring error and supports targeted training.

The Bottom Line

PDGM pays each 30-day period according to five variables, and each variable is a data entry that an agency sets. Against a payment baseline that CMS estimates fell by 1.3% in CY 2026, an error in any variable affects the payment for the whole period.

Agencies that place a QA checkpoint at each handoff, and refresh their reference tables with each annual rule, keep each period in the group that the clinical record supports. That is the practical meaning of PDGM optimization.

How Red Road Supports PDGM Accuracy

Agencies that manage PDGM variables across intake, coding, clinical documentation, and scheduling often need a review layer that connects the handoffs. Red Road provides coding and OASIS review for home health agencies through certified coders and clinical reviewers.

Reviewers check the principal diagnosis, the secondary diagnoses, and the OASIS functional items against one another before the claim is submitted. Red Road reports findings by coder and by clinical group, which shows which handoff produces a recurring error.

This support operates as an extension of the agency team and follows the CMS tables in effect for the period under review.

Explore how Red Road's coding and OASIS review services support PDGM payment accuracy.

Regulatory Sources

• CMS-1828-F: CY 2026 Home Health Prospective Payment System Final Rule fact sheet
• CMS Home Health Prospective Payment System overview: case-mix groups and outlier payments
• CMS Medicare Benefit Policy Manual, Chapter 7, Section 10: payment adjustments, LUPA, and outliers
• CMS Transmittal 13488: CY 2026 home health payment rate update