Under the Patient Driven Payment Model, payment for a Medicare Part A skilled nursing stay is built from several components. Looking at a single total hides which components are moving, and why. Tracking each one separately gives clinical and business office leaders a clearer picture.
This post walks through the components at a high level and shows how to keep an eye on each. It is meant for operational understanding, not as coding guidance.
The components in plain terms
PDPM payment is built from case-mix groups across these categories:
- Physical therapy (PT)
- Occupational therapy (OT)
- Speech-language pathology (SLP)
- Nursing
- Non-therapy ancillary (NTA)
There is also a variable per diem adjustment that changes over the course of a stay for some components. Each resident's assessment data, primarily from the MDS 3.0, drives the classification in each category.
Why look at components separately
A total case-mix figure can stay flat while one component rises and another falls. Those movements may reflect real changes in your resident population, or they may reflect documentation and assessment practices. You cannot tell which without looking.
For example, if your SLP group distribution shifts over a quarter, the question is whether you admitted different residents, whether assessments are capturing information consistently, or whether something changed in how the data is entered.
A simple component dashboard
For each component, track:
- Distribution of case-mix groups over time, by month
- Average case-mix index for that component
- Differences across units or admission sources
- Comparison to your own history and, where appropriate, to sister buildings
Keep the view simple. A bar or stacked column per month is usually enough.
Questions to ask each month
For clinical leaders
- Do the groups reflect the residents we are actually admitting?
- Are there care areas where documentation might not be capturing what the team is doing?
For the MDS coordinator
- Are assessments completed on time and consistently?
- Are there items frequently left incomplete or coded differently by different people?
For administration
- Did our admission mix change, such as more post-surgical or more medically complex stays?
- Which hospital partners send which kinds of residents?
Connecting to length of stay
Because the per diem adjustment changes across a stay, length of stay and payment are linked. Tracking average length of stay by component group helps administrators see whether the building is serving residents well while also understanding how reimbursement may change as stays progress.
Guardrails
Analytics should support accuracy, not push it in a direction. A few reminders:
- Code what is true. The purpose of reviewing the data is to find places where documentation may not reflect care provided, not to chase higher groups.
- Involve clinicians. Interpretation of any pattern should include those closest to care.
- Consult your compliance resources. Anything that looks unusual deserves a careful, documented look.
- Respect the resident. Behind every group is a person with a plan of care.
Common patterns worth a closer look
- A component that shifts suddenly after a staffing change in the MDS department
- Large differences between similar units
- Frequent late or corrected assessments
- Wide variation between assessors
None of these is a conclusion. Each is a question.
Putting it into practice
A monthly PDPM review, thirty minutes with the MDS coordinator, therapy lead, DON and business office, can reveal many patterns. Bring one page, discuss one or two questions, and assign an owner for follow-up.
A hypothetical example
Imagine a hypothetical building that sees its nursing component distribution shift over two months. A review shows that a recent change in the MDS department's workflow led to some assessments being completed later than before, with a few items left unclear. The residents had not changed; the paperwork process had. Catching it in a monthly review let the team fix the workflow and clarify documentation expectations, with clinicians confirming what was actually true for each resident.
Where CarePulse fits
CarePulse can bring MDS and PointClickCare data together so each component is visible by month, unit and admission source. If you would like to see your own case-mix trends laid out clearly, we would be glad to show you in a demo.