Every skilled nursing facility that bills Medicare Part A under the Patient Driven Payment Model now operates under a sharper microscope. CMS, the HHS Office of Inspector General, and Medicare Administrative Contractors have all signaled that PDPM coding patterns are a top integrity concern heading into fiscal year 2027. For a skilled nursing facility, the question is no longer whether your case-mix coding will be examined, but how well your documentation will hold up when it is.
Quick answer: PDPM case-mix creep is the steady drift of resident coding toward higher-paying classification groups without a matching rise in clinical complexity. CMS flagged it in the FY2027 proposed rule, and OIG has already recommended a $31.2 million recoupment from the first facility it audited. The defense is not a billing argument after the fact, it is contemporaneous MDS documentation that ties every coded condition to assessed clinical evidence and a skilled level of care.
What Does CMS Mean by Case-Mix Creep?
PDPM classifies each Part A resident into payment groups across five case-mix adjusted components: physical therapy, occupational therapy, speech-language pathology, nontherapy ancillary services, and nursing. Each component draws on data coded in the Minimum Data Set (MDS). Small coding choices, repeated across hundreds of assessments, move millions of dollars in reimbursement.
Case-mix creep describes what happens when the average case-mix index rises faster than the underlying acuity of the resident population would predict. In the FY2027 SNF PPS proposed rule, released April 2, 2026 and published in the Federal Register on April 7, CMS issued a Request for Information addressing exactly this concern. The agency observed that average case-mix indexes have climbed faster than changes in patient health status alone would explain, even as median per-diem costs declined.
CMS named specific clinical areas where coding rates have surged: malnutrition, depression, and swallowing disorders. These are not random examples. Each one can shift a resident into a higher nursing or nontherapy ancillary group, and each one depends heavily on assessment judgment rather than a single objective test. That combination is what makes them audit magnets.
The mechanics matter here. A case-mix index is a weighted measure of how complex a facility’s residents are, and PDPM ties that index directly to per-diem payment. When the index rises but documented patient acuity does not, the only remaining explanation is coding behavior. That is the gap CMS is now measuring, and it is the gap an auditor reconstructs claim by claim. The agency’s own framing is instructive: it compared the rate at which case-mix indexes have grown against the rate at which resident health status has changed, and found the two no longer move together. A payment system designed to track acuity is, in CMS’s view, increasingly tracking documentation practices instead.
The proposed rule carries a 2.4% payment update for FY2027, the net of a 3.2% market basket increase and a 0.8 percentage-point productivity adjustment, worth roughly $888 million in aggregate. The comment period on the proposed rule closed June 1, 2026, with a final rule expected by late summer and changes effective October 1, 2026. The payment bump is real, but it arrives alongside a clear message that CMS intends to police how facilities earn it. A larger pool of money and a heightened scrutiny of how that money is coded are arriving together, and that pairing is the defining feature of the FY2027 environment.
Why Does FY2027 Raise the Stakes for Every Skilled Nursing Facility?
The RFI is not an abstract policy exercise. It signals that CMS is building the evidentiary and methodological foundation to recalibrate PDPM and to support targeted enforcement. When a regulator publicly identifies the conditions it considers over-coded, every claim touching those conditions becomes easier to flag and harder to defend without strong records.
A Request for Information is also a structured way for CMS to collect data and stakeholder input before it writes binding rules. In practice, it tells facilities where the agency is headed and gives them a window to align documentation with the coming standard. Reading it as advance notice rather than as a comment exercise is the difference between preparing and reacting. Historically, the conditions a payment regulator singles out in an RFI become the conditions its contractors probe first, because the agency has already done the work of identifying where the data looks anomalous.
Enforcement is already underway. OIG opened an audit series, SRS-A-25-010, examining whether Medicare payments to skilled nursing facilities under PDPM complied with Medicare requirements. The OIG work plan confirms five projects remain active with completion expected in fiscal year 2026, so additional findings will surface throughout the scrutiny window. This is a coordinated posture rather than a single audit: the rulemaking signal from CMS and the enforcement activity from OIG point at the same coding behaviors at the same time.
The first completed audit set the tone. Reviewing a single facility, OIG found that 99 of 100 sampled claims failed to meet Medicare requirements, citing documentation that did not support medical necessity, incorrect rate-code assignment, and claims for patients who did not require a skilled level of care. The estimated overpayment came to $31.2 million. A finding of that size against one provider tells the entire sector how aggressively extrapolation can be applied when sampled claims are deemed noncompliant.
Extrapolation is the multiplier that turns a sample into a sector-level threat. An auditor reviews a limited set of claims, calculates an error rate, and projects that rate across the full universe of paid claims for the period. A handful of unsupported assessments in the sample can therefore drive a recoupment demand far larger than the dollars actually tied to those specific records. In the first PDPM audit, the directly identified overpayments in the sample were a fraction of the total, yet the projected figure reached eight figures. That arithmetic is why a coding weakness that looks minor at the claim level becomes existential at the facility level.
Industry groups have pushed back, arguing that comparing pre-PDPM data to early PDPM data overstates the upcoding signal and that the transition adjustment already corrected for those effects. That debate matters for national policy, but it does not protect an individual facility in an audit. A reviewer evaluates your records against Medicare coverage and documentation requirements, not against a trade association’s macro critique. Facilities that depend on the CMS methodology being wrong are taking on risk they cannot control.
How Can a Facility Build a Defensible Coding File Before the Auditor Arrives?
A strong audit defense is constructed in the chart, in real time, long before any records request lands. The principle is consistency: the MDS must reflect what the interdisciplinary documentation actually shows, and the documentation must establish a daily skilled need. Where those two diverge, you have exposure regardless of intent.
Start with the conditions CMS named. For malnutrition, the record should show the clinical criteria used for the diagnosis, the physician’s involvement, the dietary assessment, and the active interventions, not merely a diagnosis carried forward from a prior setting. For depression, the PHQ assessment and interview responses should support the coded mood score. For swallowing disorders and the speech-language pathology component, the chart should document the assessed deficit, the diet texture orders, and the skilled intervention. Coding that outruns the supporting evidence is the textbook profile of case-mix creep.
Beyond the high-risk conditions, build internal controls that catch problems while they are still fixable:
- Concurrent MDS review. Audit a sample of assessments before submission, comparing each case-mix-driving item to the source documentation in the medical record.
- Skilled-need verification. Confirm that physician orders, nursing notes, and therapy logs together establish a daily skilled service, the threshold OIG repeatedly tests.
- Diagnosis substantiation. Require that any condition affecting payment trace to an active, physician-supported diagnosis with current clinical evidence, not an inherited problem list.
- Trend monitoring. Track your facility’s case-mix index over time and against benchmarks, because a rising index without a corresponding rise in documented acuity is precisely the pattern CMS and OIG screen for.
- Triple-check coordination. Ensure clinical, MDS, and billing staff reconcile the assessment, the documentation, and the HIPPS code before the claim goes out.
These controls also protect the financial reporting that flows downstream from coding. Because PDPM data feeds Medicare cost reporting, errors compound across compliance and reimbursement. Coordinating clinical documentation review with cost report preparation keeps a single coding weakness from surfacing in multiple places during an examination.
For operators who want an independent assessment of exposure, a coding-integrity review modeled on the OIG methodology is the most direct option. Pease Bell’s skilled nursing and long-term care advisory team can sample recent claims, test documentation against Medicare requirements, quantify potential repayment risk, and prioritize remediation before a contractor does it for you. Identifying a pattern internally costs far less than defending it under extrapolation.
What Should a Facility Do in the FY2027 Window?
Treat the period between now and the October 1, 2026 effective date as preparation time, not a pause. The final rule may refine the RFI into firmer policy, and audit activity is independent of the rulemaking calendar. Facilities that wait for certainty will be reacting to a records request instead of working from a clean file.
Document your coding rationale for the flagged conditions now, while the relevant residents are still in-house and the assessing clinicians are available. Reconstructing the basis for a malnutrition or depression code months later, after staff turnover, is the situation OIG findings consistently exploit. Contemporaneous evidence is the asset that decays fastest.
The timing is also a staffing question. The clinician who completed an assessment can explain the judgment behind it in a way that a successor reading the same chart cannot. Capturing that rationale while the team is intact converts individual clinical memory into a durable part of the record.
Finally, align leadership around the reality that case-mix optimization and case-mix creep look identical on a spreadsheet. The difference lives entirely in the chart. A skilled nursing facility that can show, claim by claim, that its coding reflects assessed clinical need is positioned to defend its reimbursement. One that cannot is exposed to the same extrapolation math that produced a $31.2 million recommended recoupment.
Frequently Asked Questions
What is PDPM case-mix creep?
Case-mix creep is the gradual movement of resident coding toward higher-reimbursing PDPM classification groups without a corresponding increase in actual clinical complexity. CMS measures it by comparing the rise in average case-mix index against changes in patient health status, and it flagged the pattern in the FY2027 proposed rule alongside declining median per-diem costs.
Which conditions is CMS scrutinizing most closely?
In the FY2027 SNF PPS proposed rule RFI, CMS specifically identified malnutrition, depression, and swallowing disorders as conditions with coding increases it wants to examine. Each can raise a resident’s nursing or nontherapy ancillary classification and each relies heavily on assessment judgment, which makes supporting documentation especially important.
How serious are the PDPM audits?
Serious. OIG’s audit series SRS-A-25-010 is reviewing whether SNF PDPM payments met Medicare requirements, and the first completed audit found 99 of 100 sampled claims out of compliance, producing an estimated overpayment of $31.2 million tied to insufficient documentation of medical necessity, incorrect code assignment, and claims for residents who did not need skilled care. Five projects remain active through fiscal year 2026.
How can a facility prepare before an audit?
Build the defense into the chart in real time. Run concurrent MDS reviews, verify that each case-mix-driving item traces to physician-supported diagnoses and skilled documentation, monitor your case-mix index for unexplained increases, and consider an independent coding-integrity review that tests your claims against the same standards OIG applies.
