CMS divides predicted readmission rates by expected rates for an excess readmission ratio, compares it against the dual-eligible peer group median, weights the counted gaps by payment share, and caps the reduction at 3%. For FY2027, 2,334 hospitals received a reduction to their Medicare fee-for-service base operating DRG payments.
What the Hospital Readmissions Reduction Program does
The Hospital Readmissions Reduction Program cuts Medicare payments to hospitals that readmit more patients than CMS expects. Congress created it in Section 1886(q) of the Social Security Act, and the first reductions landed in FY2013.
The cut reaches a hospital through its rates rather than through a bill. CMS converts the reduction into a payment adjustment factor and applies it to every Medicare fee-for-service base operating DRG payment for the fiscal year. A factor of 0.97 is the 3% maximum. A factor of 1.0 means no reduction.
CMS scores six measures today: acute myocardial infarction, heart failure, pneumonia, COPD, elective total hip or knee arthroplasty, and coronary artery bypass graft surgery. Each counts unplanned readmissions within 30 days of discharge, risk-standardized for case mix. A seventh arrives on a published timetable.
How is the readmission penalty calculated?
Start with one measure at one hospital. CMS divides the hospital’s predicted readmission rate by its expected rate . Predicted reflects how that hospital performed with the patients it treated. Expected reflects how an average hospital would have performed with those same patients. The quotient is the excess readmission ratio, and a value above 1.0 means more readmissions than the case mix predicted.
No fixed threshold follows. CMS sorts every hospital into one of five similarly sized peer groups by the share of stays belonging to patients dually eligible for Medicare and Medicaid, then finds the median ratio inside each group. A measure contributes to your reduction when it carries at least 25 eligible discharges and its ratio sits above your peer group median. CMS weights each contributing gap by that condition’s share of your Medicare fee-for-service base operating DRG payments, sums the weighted gaps, and stops at 3%.
Two consequences deserve attention from anyone reading a Hospital-Specific Report. Your dual proportion determines the hospitals you are measured against. The median itself moves each year, so your reduction can shift while your own readmission rate holds steady.
What changed in FY2027?
CMS released the hospital-level figures on 7 October 2026. The reductions took effect on 1 October 2026.
| Figure | Number | What it means in practice |
|---|---|---|
| Hospitals evaluated | 2,912 | The base for every share below |
| Hospitals penalized | 2,334 | 80.2%, up from 78.2% of 2,945 a year earlier |
| Hospitals with no penalty | 578 | 19.8%, down from 641 hospitals (21.8%) in FY2026 |
| Average penalty among penalized hospitals | 0.44% | Unchanged from FY2026 |
| Hospitals facing 1% or more | 244 | 8.4% of those evaluated, up from 240 |
| Hospitals facing 2% or more | 34 | Includes the 10 at the cap |
| Hospitals at the 3% cap | 10 | Down from 15 in FY2026 |
| Performance period | Jul 2023 to Jun 2025 | Closed. FY2027 reflects care delivered two years ago |
Sources: Becker’s analysis of where readmission penalties hit hardest, Becker’s six things to know on the FY2027 penalties, Healthcare Dive’s coverage, and the CMS FY 2027 HRRP supplemental data file.
Read the year-over-year comparisons with care. FY2027 is the first year CMS calculated excess readmission ratios using Medicare Advantage data alongside fee-for-service, and the first year the performance period ran two years instead of three. Part of the movement belongs to the new method rather than to hospital performance.
Three patterns stand out.
CMS penalized more hospitals, and fewer at the top end. Ten hospitals hit the 3% cap, down from fifteen. The average reduction among penalized hospitals held at 0.44%. CMS caught more borderline performers rather than punishing outliers harder, which puts the risk on hospitals sitting just inside the line.
A fraction of a percent compounds. The reduction applies to base operating DRG payments across the full fiscal year. At meaningful Medicare inpatient volume, a cut below 1% still reaches every discharge for twelve months, and 244 hospitals now carry 1% or more.
Medicare Advantage data now feeds the score. The reduction still applies to fee-for-service payments alone. A hospital with a large MA population now has more of its care shaping a ratio that adjusts a smaller share of its revenue. Becker’s found the discharges behind the FY2027 scores rose 53% to 3.88 million even with the shorter window.
These figures describe FY2027. We update this article when CMS releases new HRRP data.
The sepsis measure arrives next, and the clock has started
In the FY2027 IPPS/LTCH PPS final rule published on 4 August 2026, CMS finalized the Hospital 30-Day, All-Cause, Risk-Standardized Readmission Rate Following Sepsis Hospitalization measure as the seventh HRRP measure. Commenters asked for more runway, and CMS added a year.
The phased schedule CMS finalized:
- FY2028 and FY2029: early-look reports. Confidential, with no public reporting and no payment effect. Hospitals see their sepsis readmission rate and an estimated payment adjustment that includes it.
- FY2030: payment adjustment. The applicable period runs 1 July 2026 to 30 June 2028.
That last date matters more than the first two. The performance window CMS will use for the first sepsis payment adjustment opened on 1 July 2026. Discharges from your hospital this week sit inside it.
The measure covers patients aged 65 and over discharged alive with a principal diagnosis of sepsis, including post-procedural sepsis, and counts any unplanned inpatient readmission within 30 days. CMS excludes patients who left against medical advice, patients discharged to hospice, and sepsis admissions already captured by the pneumonia readmission measure.
How coding connects to risk adjustment
CMS builds the expected rate from patient characteristics, and comorbidities arrive through coded claims. For the sepsis measure CMS named the data sources: risk adjustment variables come from inpatient, outpatient and physician Medicare administrative claims extending up to 12 months before the index hospitalization, plus secondary diagnoses documented as present on admission during the index stay. CMS keeps complications that arise during the index stay out of the model, treating them as a reflection of care rather than of baseline risk.
Two points follow for a coding leader. The risk picture for an inpatient measure draws on a full year of a patient’s record, including the clinic and physician encounters that preceded the admission. The present-on-admission indicator also carries weight, because it separates what the patient brought in from what happened during the stay.
Researchers disagree about how much coding moves these scores. A JAMA Internal Medicine research letter by Ibrahim and colleagues attributed 63.1% of the measured drop in risk-adjusted readmissions after HRRP to increases in coded severity, while stating that administrative data could not show whether those increases were clinically appropriate. A 2020 Medical Care study by Li and colleagues concluded that changes in patient coded severity did not affect readmission penalties.
Treat both as a compliance warning rather than a strategy.
What this does not mean
Coding should never change to move a metric. Every diagnosis on a claim needs clinician documentation behind it and has to satisfy the official reporting guidelines. A code the record does not support is a compliance exposure whatever it does to a ratio, and an auditor reading that chart later will reach the same conclusion.
The defensible goal is accurate capture. Documented conditions reach the claim. Undocumented conditions become a provider query.
Where AI medical coding fits
Risk adjustment reads the outpatient and physician record alongside the index admission, and those are the charts MediCodio AI codes . An auditor opening one of those charts sees which sentence supported each code and which rule allowed it. A chart goes to a human reviewer when the documentation runs thin, and where the record will not support a code, clinical documentation improvement raises the gap as a query.
What should hospital leaders do next?
- Pull your FY2027 Hospital-Specific Report and identify which measures drove your reduction, and whether any fell under the 25-discharge threshold. Your payment adjustment factor also sits in the CMS FY 2027 HRRP supplemental data file.
- Check your peer group. Your dual proportion sets the comparison, and that comparison moves on its own.
- Start watching sepsis now. The FY2030 performance window opened on 1 July 2026. Look at sepsis readmission rates and at how your coders are identifying sepsis before the first early-look report reaches you.
- Sample the charts behind the penalized measures and ask whether documented comorbidities reached the claim. Compare CC and MCC capture across facilities and across coders, and check present-on-admission assignment while you are in there.
- Look at where queries cluster. Repeated queries on one condition in one service line point at a documentation template rather than at a coder.
- Ask any coding vendor to show the evidence, not the code. If a reviewer cannot see the documentation passage and the rule behind an assignment, neither can your auditor. Ask which chart types the vendor codes, and get the answer before the pilot rather than after.
See it on the charts we code
Book a Demo and we will code a sample of your own charts, drawn from your outpatient, ED or professional-fee book. Your team audits the output, and you can open the documentation passage and the governing rule behind every assignment.
Related reading: how medical coding audit services surface inconsistency across facilities, and what RCM medical coding looks like across a multi-client book.
Sources
- CMS, Hospital Readmissions Reduction Program overview and statutory basis
- CMS QualityNet, HRRP measures and the excess readmission ratio
- CMS, HRRP payment reduction methodology and performance period
- CMS, FY 2027 IPPS/LTCH PPS final rule, published 4 August 2026, adoption and phase-in of the Sepsis Readmission measure
- CMS, FY 2027 HRRP supplemental data file, hospital-level payment adjustment factors
- Becker’s Hospital Review, where readmission penalties hit hardest
- Becker’s Hospital Review, CMS penalizes 2,334 hospitals for high readmissions: 6 things to know
- Healthcare Dive, Medicare penalizes more hospitals for readmissions
- Ibrahim AM and colleagues, Association of Coded Severity With Readmission Reduction After the Hospital Readmissions Reduction Program, JAMA Internal Medicine
- Li J and colleagues, Patient Coded Severity and Payment Penalties under the Hospital Readmissions Reduction Program, Medical Care, 2020