Author: Penny Jefferson, MSN, RN, CCDS, CCDS-O, CCS, CDIP, CRC, CHDA, CRCR, CPHQ, ACPA-C | August 11, 2026
Sepsis has never been a simple diagnosis in healthcare. Ask a bedside clinician, a clinical documentation integrity (CDI) specialist, a coding professional, a quality leader, and a payer to evaluate the same case, and you may not always get the same answer. Different clinical criteria, coding requirements, quality specifications, clinical validation expectations, and payer interpretations have made sepsis one of the most complex diagnoses we encounter.
Now there is another question hospitals should be asking: Are we even measuring the same sepsis patients?
That question becomes increasingly important considering two developments surrounding the Centers for Medicare & Medicaid Services (CMS) fiscal year (FY) 2027 Inpatient Prospective Payment System (IPPS) rulemaking.
In the FY 2027 IPPS final rule, CMS finalized the adoption of the Hospital 30-Day, All-Cause, Risk-Standardized Readmission Rate.
Following the sepsis hospitalization measure, with modifications, hospitals are expected to receive two years of confidential early-look reports during the FY 2028 and FY 2029 program years, including estimated Hospital Readmissions Reduction Program (HRRP) payment adjustments with the sepsis measure added.
The measure will enter HRRP payment-reduction calculations beginning with the FY 2030 program year [1]. In the proposed rule, CMS also sought public comment on the potential future use of the Adult Community-Onset Sepsis Standardized Mortality Ratio (SMR) measure [6]. CMS did not adopt that mortality measure in the FY 2027 final rule, so it remains a potential future measure rather than a current CMS reporting requirement [1,6].
These are not the same type of measure, and they do not rely on exactly the same data. However, when considered together, they raise an important issue for CDI, coding, quality, informatics, and clinical leaders: the same clinical condition may increasingly be identified and evaluated through different data pathways. As a result, hospitals must understand not only their sepsis outcomes, but also exactly who is being counted as a sepsis patient in the first place.
One Diagnosis, Different Data Pathways
The finalized sepsis readmission measure is claims-based. Partnership for Quality Measurement materials describe it as relying on claims data for identification of sepsis hospitalizations and risk adjustment [4]. In contrast, the developing Adult Community-Onset Sepsis SMR takes a substantially different approach.
The Centers for Disease Control and Prevention (CDC) National Healthcare Safety Network (NHSN) describes the SMR as a digital quality measure that uses Fast Healthcare Interoperability Resources (FHIR) to support algorithmic determinations from clinical data available in electronic health records [3]. The measure identifies Adult Sepsis Events using clinical indicators of presumed serious infection combined with evidence of organ dysfunction, including blood cultures, antimicrobial administration, laboratory abnormalities, vasopressor use, and mechanical ventilation [3].
This risk-adjustment methodology also draws heavily from EHR-based physiologic and laboratory data while incorporating selected claims-based variables such as comorbidities, present-on-admission codes for certain infection sites, and mechanical ventilation [3]. This is not simply a comparison of claims versus EHR data.
Rather, it represents a comparison between a claims-based outcome measure and a clinically richer, FHIR-enabled digital measurement approach that also incorporates selected claims information. One methodology primarily interprets what is reflected on the claim, while the other can access deeper structured clinical data within the electronic health record.
As a result, these approaches may not always identify exactly the same patients.
How Many Sepsis Populations Does One Hospital Have?
To understand the implications, it helps to follow how sepsis moves through a hospital today. A clinician recognizes and treats the condition. CDI reviews documentation and clinical evidence. Coding applies ICD-10-CM Official Guidelines for Coding and Reporting and assigns final codes. Quality teams then evaluate applicable cases based on defined measure specifications.
At the same time, electronic surveillance systems may identify patients through laboratory results, medication administration, cultures, organ dysfunction, and other structured clinical data. Later, a payer may evaluate whether the documented diagnosis was clinically supported, and if challenged, the denials staff and physician advisors may become involved.
Each of these processes is appropriate for its intended purpose, but each can produce a somewhat different sepsis population. A patient may be clinically treated for sepsis but not ultimately have sepsis represented on the final claim. Another may have sepsis coded but later face a clinical validation denial. A third may meet an electronic surveillance definition regardless of how the final diagnosis is coded.
This does not necessarily mean one department is right and another is wrong. Different methodologies may intentionally define populations differently. The challenge arises when an organization cannot clearly explain those differences.
When Claims Become Quality Data
During review of the sepsis readmission measure, stakeholders have raised concerns about the limitations of claims-based identification and risk adjustment. Comments submitted through the Partnership for Quality Measurement noted that variation in sepsis diagnosis, documentation, and coding practices could influence measured performance and that claims may not fully capture illness severity or clinical complexity [4].
For CDI and coding professionals, this is a critical point. Claims do not exist independently of the medical record. Provider documentation is translated into coded data, and coded data become the administrative foundation used for measurement. As a result, the integrity of the documentation-to-code pathway can influence not only reimbursement, but also how a patient is assigned to a quality measure population.
This is why the traditional boundary between “coding data” and “quality data” is becoming increasingly difficult to maintain. Some quality measures rely on coded administrative data, while others increasingly use discrete clinical data from the EHR. Hospitals must now understand both.
Why Readmissions Change the Sepsis Conversation
The Hospital Readmissions Reduction Program (HRRP) is a Medicare value-based purchasing program that reduces payments to hospitals with excess readmissions [2]. CMS currently includes six condition- or procedure-specific 30-day risk-standardized unplanned readmission measures, including acute myocardial infarction, COPD, heart failure, pneumonia, coronary artery bypass graft surgery, and elective hip and knee arthroplasty [2].
In the FY 2027 IPPS final rule, CMS finalized the sepsis readmission measure with a phased implementation. Hospitals will receive confidential early-look reports in FY 2028 and FY 2029, and the measure will be incorporated into HRRP payment-reduction calculations beginning with FY 2030 [1]. HRRP broadly defines readmissions, capturing unplanned returns within 30 days of discharge regardless of whether the readmission occurs at the same or a different applicable acute-care hospital and whether the principal diagnosis matches the index condition [2].
This is particularly significant for sepsis. A patient may survive the acute infection but return days later with heart failure, renal dysfunction, recurrent infection, medication-related complications, or functional decline. The question therefore shifts from whether sepsis was documented and coded correctly during the index admission to whether the original record accurately reflects the clinical complexity of the patient whose outcomes are now being measured.
Medicare Advantage Adds Another Dimension
Another important change is the inclusion of Medicare Advantage data in HRRP measures. In the FY 2026 IPPS final rule, CMS finalized modifications to include Medicare Advantage data alongside Medicare fee-for-service data beginning with the FY 2027 program year and shortened the performance period from three years to two years [5]. However, CMS did not finalize the inclusion of Medicare Advantage data in aggregate payment calculations for excess readmissions [5].
Even so, the broader implication is clear: beginning with the FY 2027 program year, Medicare Advantage beneficiaries will be included alongside Medicare fee-for-service beneficiaries in the six existing HRRP measure populations [5]. For organizations where Medicare Advantage, CDI, coding, quality, risk adjustment, and denials operate in separate silos, this reinforces the need to understand how these data streams intersect. The medical record does not recognize departmental boundaries, and a single patient’s data can influence multiple downstream systems.
Mortality Brings Us Closer to the Clinical Record
The developing Adult Community-Onset Sepsis SMR highlights a different approach to measurement. The CDC describes it as an annual risk-adjusted standardized mortality ratio for adult inpatients with community-onset sepsis, where the numerator includes in-hospital deaths or hospice discharges and the denominator reflects predicted outcomes [3].
Unlike claims-based approaches, this measure incorporates detailed physiologic and laboratory data in risk adjustment, including blood pressure, lactate, creatinine, platelets, white blood cell count, bilirubin, sodium, albumin, vasopressor use, and hypothermia, all derived from EHR/FHIR data [3]. Selected claims-based variables and NHSN survey data are also incorporated into the methodology [3]. This represents a shift toward evaluating outcomes using more granular clinical information rather than relying solely on administrative claims.
For CDI professionals, this evolution is significant. For decades, the focus has been on translating clinical documentation into coded data. Now, digital measurement allows quality systems to directly evaluate portions of the underlying clinical record. While claims and documentation remain essential, structured clinical data increasingly plays a direct role in how patients are measured and compared.
Clinical Validation Becomes More Important, Not Less
As sepsis measurement expands across reimbursement, readmissions, mortality, risk adjustment, and payer review, clinical validation becomes even more critical. The appropriate response is not simply to increase sepsis capture, but to ensure that documentation accurately reflects clinical reality.
Documentation integrity must work in both directions. Clinically supported conditions that are missing should be identified, but unsupported or inconsistent documentation must also be addressed. The goal is not maximum diagnosis capture, but an accurate, clinically supported record that can withstand multiple forms of review and measurement.
This distinction becomes increasingly important as data is consumed not only by clinicians, but also by claims systems, quality programs, risk adjustment models, external auditors, and automated digital measurement tools.
Stop Starting with the Query Rate
For CDI programs, preparation should not begin with increasing query volume. It should begin with understanding the data. Ask CDI, coding, quality, and analytics teams how many sepsis patients the organization had last year. If the answers differ, that variation itself is meaningful and should be explored.
The next step is to understand why differences exist. They may stem from clinical definitions, coding guidelines, measure specifications, data extraction methods, timing, exclusions, clinical validation processes, or payer methodologies. Following a sample of patients across these systems can reveal where and why divergence occurs.
Was the patient clinically septic? Was sepsis documented? Was it coded? Was it captured in surveillance systems? Was it included in quality measures? Was it queried or denied? Was the patient readmitted within 30 days? These questions often reveal gaps that traditional CDI dashboards do not capture.
Follow the Patient Beyond the Claim
The finalized sepsis readmission measure also extends the CDI perspective beyond the inpatient stay. While CDI traditionally concludes at discharge and final coding, HRRP follows patients for 30 days post-discharge [2]. CMS’s two confidential early-look years give hospitals an opportunity to examine which patients return, why they return, and whether the index-admission record accurately reflects the complexity of their condition before the measure begins affecting HRRP payment-reduction calculations in FY 2030 [1].
This does not mean CDI should own readmission outcomes. Rather, it means CDI should understand how documentation contributes to the data used in those evaluations. The patient journey does not end at discharge, and neither does the impact of the record.
Are We Measuring the Same Patient?
The CMS decision to finalize the sepsis readmission measure makes this question more than theoretical. Hospitals now have two confidential early-look years—FY 2028 and FY 2029—before the measure begins affecting HRRP payment-reduction calculations in FY 2030 [1]. The broader insight emerges when the claims-based measure is viewed alongside the developing Adult Community-Onset Sepsis SMR, which CMS identified in the proposed rule as a potential future measure [6]. Both aim to measure outcomes related to sepsis, but they do so through different lenses—claims-based versus clinically enriched digital data.
This creates an opportunity for CDI, coding, quality, informatics, and analytics leaders not to force uniformity across systems, but to understand why differences exist. The goal is not to change documentation to improve scores, but to ensure that regardless of the system interpreting the data, the patient’s story remains accurate, clinically supported, and explainable.
For years, we have asked whether sepsis was documented correctly. The next question is more complex: When we say, “our sepsis population,” are we all talking about the same patients?
Before FY 2030 payment calculations begin, hospitals should use the early-look period to understand these differences and be able to answer that question with confidence.
References
- Centers for Medicare & Medicaid Services. FY 2027 Hospital Inpatient Prospective Payment System (IPPS) and Long-Term Care Hospital Prospective Payment System (LTCH PPS) Final Rule — CMS-1849-F. July 31, 2026. CMS finalized adoption of the Hospital 30-Day, All-Cause, Risk-Standardized Readmission Rate Following Sepsis Hospitalization measure, with confidential early-look reports during the FY 2028 and FY 2029 program years and use in HRRP payment-reduction calculations beginning with FY 2030. https://www.cms.gov/newsroom/fact-sheets/fy-2027-hospital-inpatient-prospective-payment-system-long-term-care-hospital-prospective-payment
- Centers for Medicare & Medicaid Services. Hospital Readmissions Reduction Program (HRRP). Current CMS program guidance describing the existing HRRP measures, 30-day unplanned readmission methodology, excess readmission ratios, and payment-adjustment framework. https://www.cms.gov/medicare/quality/value-based-programs/hospital-readmissions
- Centers for Disease Control and Prevention, National Healthcare Safety Network. Adult Community-Onset (CO) Sepsis Standardized Mortality Ratio (SMR). Describes the NHSN digital quality measure, Adult Sepsis Event identification, FHIR-enabled reporting, outcome calculation, and risk-adjustment variables derived from EHR, claims, and NHSN Annual Survey data. https://www.cdc.gov/nhsn/fhirportal/dqm/co-smr.html
- Partnership for Quality Measurement. Hospital 30-Day, All-Cause, Risk-Standardized Readmission Rate Following Sepsis Hospitalization (MUC2025-055). Measure specifications and stakeholder comments addressing claims-based sepsis identification and risk adjustment, documentation and coding variability, clinical specificity, and other methodological considerations. https://p4qm.org/prmr-measures/muc2025-055
- Centers for Medicare & Medicaid Services. FY 2026 Hospital Inpatient Prospective Payment System (IPPS) and Long-Term Care Hospital Prospective Payment System (LTCH PPS) Final Rule — CMS-1833-F. July 31, 2025. CMS finalized the addition of Medicare Advantage data to the six existing HRRP readmission measures and shortened the applicable performance period from three years to two years beginning with the FY 2027 program year. CMS did not finalize the proposed inclusion of Medicare Advantage data in calculations of aggregate payments for excess readmissions. https://www.cms.gov/newsroom/fact-sheets/fy-2026-hospital-inpatient-prospective-payment-system-ipps-long-term-care-hospital-prospective-0
- Centers for Medicare & Medicaid Services. FY 2027 Hospital Inpatient Prospective Payment System (IPPS) and Long-Term Care Hospital Prospective Payment System (LTCH PPS) Proposed Rule — CMS-1849-P. April 10, 2026. CMS requested comment on the potential future use of the Adult Community-Onset Sepsis Standardized Mortality Ratio and described its development using EHR/FHIR and claims data. https://www.cms.gov/newsroom/fact-sheets/fy-2027-hospital-inpatient-prospective-payment-system-ipps-long-term-care-hospital-prospective
This article was originally published on RACmonitor.
