Understanding PDPM: A Practical Guide for SNF Teams

The Patient-Driven Payment Model (PDPM) is the case-mix classification system that Medicare uses to determine per-diem payment for skilled nursing facility (SNF) stays under Medicare Part A. Introduced by CMS on October 1, 2019, PDPM replaced the older RUG-IV system and shifted the basis for payment away from the volume of therapy provided and toward the clinical characteristics and care needs of each resident.

For SNF administrators, nurses, MDS coordinators, and therapy leaders, understanding how PDPM classifies a stay is essential — both to deliver appropriate care and to ensure the facility is reimbursed accurately for the resources that care requires. This guide walks through the model at a practical level.

How PDPM structures a payment

Rather than a single rate, PDPM builds each resident's per-diem from several independently classified components. Each component reflects a different dimension of the resident's needs, and each is adjusted by a case-mix index derived largely from the Minimum Data Set (MDS) assessment.

• Physical Therapy (PT) — driven by the resident's primary diagnosis clinical category and functional status.
• Occupational Therapy (OT) — classified similarly to PT, with its own case-mix groups.
• Speech-Language Pathology (SLP) — reflects presence of acute neurologic conditions, swallowing disorders, cognitive impairment, and comorbidities.
• Nursing — captures the clinical and functional acuity of nursing care needs.
• Non-Therapy Ancillary (NTA) — a points-based component that reflects the cost of drugs, supplies, and services for medically complex residents.

The PT and OT components, and the NTA component, are also subject to a variable per-diem adjustment: payment rates change across the stay according to a set schedule, reflecting the typical pattern of resource use over time.

Why accurate MDS coding is the foundation

Because most PDPM components are calculated from MDS items, the accuracy of the assessment directly determines the payment classification. Small coding gaps — an omitted comorbidity, a missed swallowing or cognitive item, or an under-documented functional score — can move a resident into a lower-paying group even when the underlying care is more complex.

Common PDPM pitfalls

1. Missing NTA comorbidities — the NTA component depends on capturing every qualifying condition from the resident's history and current record.
2. Incomplete Section GG functional scoring — functional status feeds the PT, OT, and Nursing components, so incomplete or inconsistent coding has broad downstream effects.
3. Overlooking SLP-related conditions — swallowing disorders, mechanically altered diets, and cognitive impairment are frequently under-captured.
4. Late or inaccurate primary diagnosis mapping — the primary diagnosis must map to a PDPM clinical category; an unmapped or default code can distort the PT and OT classification.

How a modern EHR helps

Legacy SNF systems can capture MDS data, but they often leave the clinical reasoning to memory and manual review. A modern, AI-native EHR can surface likely gaps before an assessment is finalized — flagging comorbidities that appear in the chart but not in the MDS, cross-checking functional scores for internal consistency, and confirming the primary diagnosis maps to a valid PDPM category.

MedFlo is designed so that clinicians code what is true, and the software does the work of making sure nothing accurate is accidentally left out. That protects both compliance and appropriate reimbursement, without asking staff to become PDPM logic experts.