If your quality team has been reviewing the July 2026 Care Compare preview files, there’s a good chance they’ve run into things that don’t add up. If they haven’t looked yet, this is your warning to look now. CMS’ official preview window closes at 11:59 p.m. PT on June 20th (CMS, 2026a).
The reason this matters at the executive level: the patient-level discharge files CMS provides through the HQR system are the only tool your team has to verify that your hospital-level scores on mortality, readmissions, and EDAC are accurately telling your quality story. Right now, those files have five things wrong with them.
What We’re Seeing
1. Denominators Don’t Add Up
One of the first checks your team performs on patient-level discharge files is confirming that the number of patients included in the measure, the denominator, matches the hospital-level score. This tells you exactly which patients and outcomes contributed to your score.
Most years these match. This year, they frequently don’t.
Across measures and cohorts, we’re seeing discharge file denominators running both too high and too low compared to the hospital-level scores. That inconsistency is a signal of multiple underlying data problems in and of itself. One likely contributor: errors in the inclusion/exclusion column of the discharge file, which this year shows far fewer exclusion categories than we typically see.
2. Medicare Advantage Patients Are Missing From Key Files
This is the first year Medicare Advantage (MA) patients are included in claims-based quality scores (CMS, 2025). For the 30-day readmission scores by patient cohort and stroke mortality, discharge-level denominators appear too low. The most likely explanation is that MA patients are simply absent from those files.
The evidence points in this direction. MA patients are present in some files, including hospital-wide readmissions and hospital-wide mortality. And in the cohort readmissions discharge file, there’s a column specifically for “Medicare Advantage (Yes/No)” — yet every single row shows “No” (CMS, 2026d). This points to a systematic omission.
3. A Full Year of COPD Mortality Data Is Missing
For COPD mortality, the problem is different and more specific. The mortality denominators reconcile correctly for AMI, CABG, HF, and PN cohorts. They don’t for COPD.
All cohort 30-day mortality measures except stroke should cover a three-year period: July 2022 through June 2025 (CMS, 2026c). For COPD, we’re only seeing two years of data covering July 2023 through June 2025. The entire July 2022 through June 2023 period appears to be missing from the COPD file, while patients from that same period exist in every other cohort. An entire year of COPD patients appears to have been dropped.
4. No Unplanned Readmissions, For Any Patient
Your hospital had unplanned 30-day readmissions. Every hospital does. But the patient-level discharge file doesn’t appear to agree.
The “Unplanned Readmission within 30 days (Yes/No)” column shows “No” for every single record, including rows where a readmission date is clearly populated (CMS, 2026d). This is an internal contradiction in the file itself.
Without an accurate indicator in that column, your team cannot validate the numerator. This means you cannot confirm which cases CMS is counting against you. For a measure that directly affects your ratings, that’s not an acceptable gap.
5. EDAC Calculations Are Wrong. Even After CMS Issued a Fix.
On the afternoon of June 11th, CMS distributed a notification through the various mailing lists stating that EDAC data issues had been identified and resolved. They asked hospitals to download updated reports. Looking at the freshly downloaded reports, we still found the same problems.
In addition to denominator mismatches, we’re seeing two specific errors in the “Days from Index Discharge to First Event” column (CMS, 2026b). First, EDAC looks at events occurring within a 30-day window following discharge, yet we’re routinely seeing values greater than 30 days. Second, even setting aside that timing issue, the calculated values don’t match the underlying dates. There is an accompanying file listing all events and their dates, and the actual difference between the index discharge and first event doesn’t align with what’s reported in the column. If that calculation is wrong, the question becomes – what else may be miscalculated downstream?
What Your Team Should Check Right Now
Here’s exactly what your team should look for in your hospital-specific reports:
Mortality
- Do your COPD and Stroke cohort patient counts match the denominators in your PDF report?
- In the COPD cohort, are patients present for discharges between July 2022 – June 2023?
- In the Stroke cohort, are Medicare Advantage patients present in the discharge file?
Readmissions
- Do any of your six readmission cohort patient counts match the overall measure denominators?
- Note that the Hip & Knee Replacement cohort may show too many cases in the discharge file relative to the overall measure denominator.
- Are Medicare Advantage patients listed in your readmissions discharge file from CMS?
- Look at the “Unplanned Readmission” column. Does it show “No” for every record — even rows where a readmission date is populated?
EDAC
- Does your Index Stay Summary patient count match the overall measure denominator for each of the three EDAC cohorts?
- Are there records where “Days from Index Discharge to First Event” exceeds 30 days?
- Do the calculated “days between” values match the dates listed in the two patient-level files?
What to Do Before June 20th
Step 1: Check your files today. Even a 30-minute spot-check by a data analyst will tell you whether you’re seeing the same patterns we are.
Step 2: Submit an inquiry through the QualityNet Q&A Tool. This is the official channel for preview period questions and corrections. CMS requires all requests be submitted no later than 11:59 p.m. PT on June 20, 2026:
https://cmsqualitysupport.servicenowservices.com/qnet_qa?id=ask_a_question
Step 3: In your inquiry, request two things specifically:
- Confirmation of whether the patient-level file issues described above are known and whether corrected files are planned
- An extension of the preview period to allow adequate time for hospitals to re-validate once corrected files are available
The Broader Point
We are not asserting that any hospital’s published scores are wrong. We are highlighting these issues because many hospitals cannot determine whether their scores are right. That is the entire purpose of the preview process.
If you are seeing these issues in your files, you are not alone. The more quality leaders who submit inquiries to CMS, the stronger the signal to CMS that this is a systemic problem requiring a systemic response. Individual hospitals raising the same concern independently is how the process gets taken seriously.
The integrity of public quality reporting depends on hospitals having a genuine opportunity to validate their data before it goes public. Right now, that opportunity is at risk. Your voice, through the official channels, is the most direct way to protect it.
CareKate Analytics is actively supporting hospitals through the review process. If you’re seeing these issues in your files or want a second set of eyes before the window closes, reach out directly at tom@carekate.com.
References:
Centers for Medicare & Medicaid Services. (2025, August 4). Medicare program; hospital inpatient prospective payment systems for acute care hospitals (IPPS) and the long term care hospital prospective payment system and policy changes and fiscal year (FY) 2026 rates; changes to the FY 2025 IPPS rates due to court decision; requirements for quality programs; and other policy changes; health data, technology, and interoperability: Electronic prescribing, real-time prescription benefit and electronic prior authorization (CMS-1833-F). U.S. Department of Health and Human Services. https://www.govinfo.gov/content/pkg/FR-2025-08-04/pdf/2025-14681.pdf
Centers for Medicare & Medicaid Services. (2026a, May 22). CMS releases July 2026 public reporting hospital and inpatient psychiatric facility data for preview. QualityNet. https://qualitynet.cms.gov/news/69fe0ddecc117044f745ab3a
Centers for Medicare & Medicaid Services. (2026b). Excess Days in Acute Care measures hospital-specific report user guide: 2026 public and confidential reporting. QualityNet. https://qualitynet.cms.gov/inpatient/measures/edac/reports
Centers for Medicare & Medicaid Services. (2026c). Mortality measures hospital-specific report user guide: 2026 public and confidential reporting. QualityNet. https://qualitynet.cms.gov/inpatient/measures/mortality/reports
Centers for Medicare & Medicaid Services. (2026d). Readmission measures hospital-specific report user guide: 2026 public and confidential reporting. QualityNet. https://qualitynet.cms.gov/inpatient/measures/readmission/reports
Leave a Reply