Hospital Quality · CMS-1849-F
The Seventh Condition
Sepsis readmissions enter the Medicare penalty program
For hospital quality and executive leaders, clinical documentation and health information management directors, and revenue cycle teams. Every figure below was verified against its primary source on August 20, 2026.
You discharged her on a Tuesday. The infection was controlled, the pressors were three days behind her, and her daughter drove her home with a folder of instructions and an appointment in three weeks. Nothing about the discharge was wrong. By every standard your hospital measures today, it was a good outcome.
Eleven days later she came back through the emergency department, and not with sepsis. She came back in heart failure, volume overloaded after the resuscitation that saved her. Nobody from your hospital had spoken to her since she left, and her appointment was ten days out.
She is not an outlier. She is the median. In the largest all-payer analysis of sepsis readmissions, median time from discharge to return was 11 days, so a visit set at two or three weeks lands after the middle of the distribution has come back. And because the measure Medicare just finalized counts readmission for any cause, her heart failure admission scores exactly as recurrent sepsis would.
That is the change. In the FY2027 Inpatient Prospective Payment System final rule (CMS-1849-F, July 31, 2026), the Centers for Medicare and Medicaid Services (CMS, the agency administering Medicare) finalized the Hospital 30-Day, All-Cause, Risk-Standardized Readmission Rate Following Sepsis Hospitalization into the Hospital Readmissions Reduction Program (HRRP, Medicare's penalty program for excess readmissions). Sepsis is the seventh HRRP condition.
The scoping decision that matters most
All-cause is the phrase to sit with. A patient who returns after a fall, in heart failure, or with a medication complication counts against the discharging hospital exactly as recurrent sepsis would. The measure prices the whole transition. Early-look reports with estimated payment adjustments arrive in FY2028 and FY2029; reductions begin FY2030.
01
What the measure actually counts
Specification Medicare fee-for-service and Medicare Advantage beneficiaries aged 65 and older, discharged alive with a principal discharge diagnosis of sepsis and 12 months of prior enrollment. The outcome is unplanned inpatient readmission for any cause within 30 days, scored by a hierarchical model of 161 risk variables. It is entirely claims-based, so it is cheap to run and sensitive to how hospitals code.
Exclusions Nine exclusions apply, among them in-hospital death, hospice discharge, transfer out, discharge against medical advice, and sepsis admissions already captured by the pneumonia readmission measure. That last one is large: CORE, the Yale New Haven center that built the measure, found roughly 24 percent of examined sepsis codes, 210,300 of 892,052, fell in the pneumonia cohort. The measure was not adopted into Hospital Inpatient Quality Reporting.
The Seventh Condition
August 26, 2026
02
Read the performance statistics honestly
Discrimination The model's c-statistic is 0.65 in both the development and validation datasets. That is modest, and comparable to CMS's hospital-wide readmission measure at 0.64 to 0.68. Readmission is simply not a highly predictable event from claims.
Reliability Across 3,053 facilities with at least 25 qualifying admissions, split-half reliability had a median of 0.682, a minimum of 0.205, and a maximum of 0.986. Sixty-nine percent of accountable entities met the 0.60 threshold, meaning close to a third did not. The American Medical Association (AMA, the national physician membership organization) opposed inclusion outright, citing that reliability floor, three simultaneous and untested specification changes, the absence of socioeconomic risk adjustment, and literature associating readmission penalties with increased mortality.
Distribution Risk-standardized rates run from 12.9 to 24.9 percent, but the middle half of hospitals sits between 17.3 and 18.9 percent. In a band that tight, small absolute movements shift percentile rank considerably.
| Owner | Action before the first early-look report | Basis |
|---|
| Quality and analytics | Model the cohort locally against the specification and its exclusions, and establish your observed rate now. The two-year claims lag means the FY2028 report will describe the patients you are treating today. | Measure specification |
| CDI and HIM | Sequence the underlying systemic infection first, then R65.20 or R65.21, then a discrete code for each documented acute organ dysfunction. R65.2- is never principal and is never reported alone. | ICD-10-CM Official Guidelines |
| CDI and HIM | Verify present on admission = Y on every pre-existing comorbidity, and capture frailty markers: malnutrition (E43), pressure injury (L89.-), oxygen dependence (Z99.81). | Causal-pathway risk model |
| CDI and HIM | Audit principal diagnosis selection on pneumonia-with-sepsis admissions. Sequencing decides which readmission measure grades the case. | 24 percent cohort overlap |
| Care transitions | Land follow-up contact inside 7 days. A visit at day 14 or 21 arrives after the median patient has already returned. | Claims-data analysis |
| Care transitions | Verify source control at discharge, and set surveillance for new infection at a different site. Infection accounts for 42.2 percent of 30-day readmissions, recurrent sepsis for 22.9 percent. | Claims-data analysis |
The incentive to watch
Because the measure is built entirely on claims, better documentation raises a hospital's expected readmission rate and therefore improves its risk-standardized result without a single patient outcome changing. That is legitimate: accurate coding is the point of risk adjustment. But a hospital that pursues only the coding half of this work will post a better score while its patients keep coming back. Both halves are required, and only one of them prevents a readmission.
3%
Maximum payment reduction
HRRP is penalty-only. Sepsis enters the calculation from FY2030.
11 days
Median time to readmission
Interquartile range 5 to 19 days, on 2013 to 2014 all-payer data.
24%
Sepsis codes in the pneumonia cohort
210,300 of 892,052 examined. Sequencing decides the scorecard.
The Seventh Condition
August 26, 2026
03
Before you buy a transitional care program
The randomized record IMPACTS (691 patients, one health system) cut its 30-day composite, 28.7 versus 33.3 percent, adjusted odds ratio 0.80. Scaled to seven hospitals in ENCOMPASS (3,548 patients), the same program was null: 90-day death or readmission of 48.2 versus 48.0 percent, mortality lower (17.3 versus 20.5) and readmissions higher. A readmission-only measure will never credit a mortality benefit.
The observational best bet A first home health nursing visit within 2 days, one more that week, and a provider visit within 7 days were together associated with a 41 percent relative reduction in 30-day rehospitalization. Neither alone was significant, and only 28 percent of survivors sent to home health got the full pattern. I-TRANSFER, the prospective test, has published no outcomes.
Two kinds of work move the same number: what coders capture at the front door, and what clinicians do at the back.
Where MHMDAA Comes In
MHMDAA (Michael Hill, MD and Associates) is a quality, compliance, and revenue consulting and payor-provider dispute resolution firm. We work both halves because the measure does not: we audit sepsis documentation against the sequencing rules and the causal-pathway risk model, replicate the CMS cohort on your claims so you see your rate first, and pressure-test vendor proposals against the trial evidence. That work rests on more than 12,000 disputed-claim reviews, an internal figure.
| Phase | Work | Output |
|---|
| Baseline | Replicate the specification and its nine exclusions on your claims; calculate the observed rate and the pneumonia split | Your rate, an estimated percentile, and the boundary cases |
| Documentation integrity | Sequencing audit, present-on-admission verification, frailty capture, planned-readmission screening | A corrected coding pattern and a quantified risk-adjustment gap |
| Transition design | Map the 7-day contact window onto discharge workflow; build source-control and new-infection checks | A protocol aimed at the real failure modes |
| Readiness | Quarterly re-measurement ahead of the first confidential report | A documented trajectory, defensible to the board |
The economics are not marginal. On the all-payer data behind this measure, a 30-day sepsis readmission cost $16,852 on average and the annualized national burden exceeded $3.5 billion, before a decade of inflation and before the penalty. Across our engagements we project a 5:1 return on engagement investment, an internal projection validated against the client's own baseline, not an industry benchmark. The first confidential report lands in FY2028; the patients it describes are in your beds this week.
Michael Hill, MD & Associates
Quality, compliance, and revenue consulting and payor-provider dispute resolution · (877) 464-4556 · info@mhmdaa.com
Neutrality. This brief is educational commentary; it is not legal, coding, or clinical advice, and it is not a determination about any specific claim, payor, provider, or matter. MHMDAA is an independent quality, compliance, and revenue consulting and payor-provider dispute resolution firm; its principals provide dual-perspective analysis and independent expert services to payors and providers alike, and nothing in this publication constitutes advocacy for either side of any dispute or prejudges any question on which they may serve as independent experts. Authorship. This document was drafted with human authorship and may include AI-assisted formatting or summarization. All analysis, conclusions, and opinions are solely those of Dr. Michael Hill.
Sources
- CMS. FY2027 Inpatient Prospective Payment System and Long-Term Care Hospital Final Rule (CMS-1849-F), Fact Sheet. July 31, 2026.
- Partnership for Quality Measurement. Preliminary assessment, MUC2025-055, incl. AMA comment record.
- CMS / YNHHSC-CORE. Sepsis Readmission Measure Technical Expert Panel Summary Reports (TEP 1, 2, 3).
- Gadre SK, Shah M, Mireles-Cabodevila E, et al. Epidemiology and predictors of 30-day readmission in sepsis. Chest. 2019;155(3):483-490.
- Taylor SP, et al. ENCOMPASS. JAMA Intern Med. 2025;185(10):1238-1246. IMPACTS. Crit Care Med. 2022;50(3):469-479.
- Deb P, Murtaugh CM, Bowles KH, et al. Does early follow-up improve outcomes of sepsis survivors in home health care? Med Care. 2019;57(8):633-640.
- ICD-10-CM Official Guidelines for Coding and Reporting, FY2026, Section I.C.1.d.
