17.1 CMS Hospital-Acquired Condition Payment Provision

17.1 CMS Hospital-Acquired Condition Payment Provision

Quick Answer: For Inpatient Prospective Payment System (IPPS) discharges, the Centers for Medicare & Medicaid Services (CMS) hospital-acquired condition (HAC) payment provision does not pay extra when a selected HAC is a secondary diagnosis and was not present on admission (POA). CMS groups the case as though that secondary diagnosis were not present, so the HAC cannot, by itself, buy a complication or comorbidity (CC) or major complication or comorbidity (MCC) Medicare Severity Diagnosis Related Group (MS-DRG) split. POA N (not POA) and U (insufficient documentation) generally block that extra payment; Y (POA) and W (clinically undetermined) do not. CMS currently lists 14 HAC categories. This claim-level rule is not the Hospital-Acquired Condition Reduction Program (HACRP).

This independent OpenExamPrep section helps inpatient clinical documentation integrity (CDI) specialists study how CMS treats selected HACs on an IPPS claim. It is not a CMS, Association of Clinical Documentation Integrity Specialists (ACDIS), or American Hospital Association (AHA) product, and it does not claim approval, partnership, or exact equivalence with those organizations’ materials. Domain VI already covered POA assignment in the prior chapter. This section’s job is the HAC payment provision: which conditions CMS selected, how POA changes grouping, and why a HAC on the claim is not the same event as a HACRP penalty on the hospital’s annual score.

What this provision is—and what it is not

Section 5001(c) of the Deficit Reduction Act of 2005 (DRA) directed CMS to identify conditions that are high cost or high volume (or both), that produce a higher-paying MS-DRG when reported as a secondary diagnosis, and that could reasonably have been prevented with evidence-based guidelines. For discharges on or after October 1, 2008, IPPS hospitals do not receive additional payment when one of those selected conditions was not POA. CMS’s own phrasing is the sentence you should be able to reconstruct on the exam: the case is paid as though the secondary diagnosis were not present.

Three boundaries keep this from swallowing the whole chart:

1. Secondary diagnosis only. The provision targets selected HAC codes used as secondary diagnoses. It is a grouping adjustment, not a license to omit a clinically valid diagnosis from the coded set.
2. Selected list only. Only diagnosis (and, where CMS requires them, associated procedure) combinations on the current fiscal-year ICD-10 HAC list count. “The patient had a complication” is not a HAC category.
3. CC/MCC mechanics. The payment hit appears when the HAC was the diagnosis that would have moved the case into a with CC or with MCC MS-DRG. If another secondary diagnosis that is a CC or MCC and is not in a selected HAC category remains on the claim, CMS has stated that payments are not adjusted solely because a HAC is also present. The HAC can still matter for quality reporting and for clinical care. It may not be what changed the check.

IPPS hospitals must still report a POA indicator on principal and secondary diagnoses. The diagnosis stays on the claim. The grouper is told, for selected HAC codes with a non-payable POA, not to let that code act as the CC/MCC that upgrades the MS-DRG.

HAC-POA payment is not HACRP

HACRP is a different CMS program (Social Security Act section 1886(p) / Affordable Care Act section 3008). CMS calculates a Total HAC Score from Patient Safety Indicator (PSI) 90 and selected National Healthcare Safety Network (NHSN) healthcare-associated infection measures. Hospitals in the worst-performing quartile receive a 1 percent reduction to IPPS payments for the program year. CMS’s HACRP pages state that HACRP is separate and distinct from the DRA HAC-POA provision.

Do not merge them on the exam:

If an item stem says “paid as though the secondary diagnosis were not present,” you are in HAC-POA. If it says “worst quartile” or “1 percent reduction to all IPPS discharges,” you are in HACRP. This section does not teach HACRP scoring, PSI-90 components, or NHSN measure definitions; those return in Chapter 21.

How POA values change HAC payment

CMS publishes the payment effect of each POA value for selected HAC codes. Pair this table with the POA definitions you already studied. Present on admission still means present at the time of the inpatient admission order. Conditions that arise in the emergency department, in observation, or during outpatient surgery before that order are POA = Y, even if the HAC “happened in the hospital building.”

U is a documentation failure, not a clinical maybe. If the record can support Y or N, a compliant query for POA is in order—without reimbursement language. W is the provider’s clinical inability to determine timing, not a coder’s shortcut when nobody looked. W still pays like Y under this provision; that is a high-yield inversion. Items that say “any non-Y value blocks payment” are wrong because W does not.

The diagnosis is not “deleted.” Coding still reports it when it meets Uniform Hospital Discharge Data Set (UHDDS) additional-diagnosis criteria. Quality programs may still see it. Only the MS-DRG CC/MCC upgrade from that selected HAC is what this provision withholds when POA is N or U.

The 14 CMS HAC categories

CMS lists 14 categories on the Hospital-Acquired Conditions page and publishes the ICD-10 diagnosis and procedure combinations each federal fiscal year (the ICD-10 HAC List, also in the MS-DRG Definitions Manual Appendix I). Through 19 September 2026, discharges in FY 2026 (effective 1 October 2025 through 30 September 2026) use the FY 2026 list. CMS has posted FY 2027 HAC files for discharges on or after 1 October 2026. Code files for the next year are a preparation packet, not a reason to invent unpublished category changes for a September 2026 discharge.

Do not memorize every ICD-10 string. Do memorize the category names, the procedure limits CMS built into several categories, and the CDI questions those limits create.

FY 2009 began with 10 categories. The FY 2013 IPPS final rule added four more (SSI after CIED, the orthopedic SSI category as refined, DVT/PE after hip or knee replacement, and iatrogenic pneumothorax with venous catheterization), producing the 14 still named on the CMS HAC page.

How to study the list without turning it into folklore

Retained object, air embolism, blood incompatibility. These are uncommon on a CDI census and almost never POA unless the patient arrived from another facility with the complication already present. If they occur during this admission, POA is N, and they cannot upgrade the MS-DRG. Query only for clinical validity and timing, not to “save the MCC.”

Stage III and IV pressure ulcers. This is the everyday HAC. Staging and POA are the two documentation failures that create U or a wrong Y/N. A Stage IV ulcer photographed in the emergency department before the inpatient order is Y. A Stage II on admission that is later documented as Stage IV needs a provider statement about whether the Stage III/IV was present at admission or evolved. Wound-care staging supports a query; it does not, by itself, establish the reportable diagnosis or the POA value. Unstageable and deep-tissue injuries are clinically important and may be query targets for staging. They are not automatically “the Stage III/IV HAC.” Check the current HAC list rather than promoting every pressure injury into category 4.

Falls and trauma. The HAC is the injury code CMS placed on the list (fracture, dislocation, intracranial injury, crushing injury, burn, other listed injury), not the word “fall.” A witnessed fall with no listed injury is not this category. A hip fracture from an inpatient fall is the classic N-POA HAC. A hip fracture that occasioned the admission is principal diagnosis and Y—this provision does not “take away” a principal fracture.

Poor glycemic control. CMS named DKA, hyperosmolar coma, hypoglycemic coma, and secondary diabetes with ketoacidosis or hyperosmolarity. In-hospital DKA in a patient admitted for something else can be an N-POA HAC if the code is on the list. Uncontrolled diabetes without one of those manifestations is not this category. Do not invent a HAC out of a glucose of 240.

CAUTI and vascular catheter-associated infection. Association with the device must be documented, not inferred from “Foley in place” plus a urinalysis. “UTI” and “bacteremia” are not automatically categories 7 and 8. These claims HACs sit near, but are not identical to, NHSN CAUTI and CLABSI used in HACRP. Same clinical story, different case-finding rules. This section stays on the claims list.

Procedure-limited SSIs. Mediastinitis after CABG. SSI after named bariatric operations. SSI after spine, neck, shoulder, or elbow procedures. SSI after CIED. Hip and knee replacement SSI is not category 11. Hip and knee replacement DVT/PE is category 13. Mixing those two orthopedic stories is a standard distractor: the 14-category list did not make every postoperative DVT or every orthopedic SSI a HAC.

DVT/PE after total knee or hip replacement. A PE after colectomy is not this HAC. A DVT after elective total hip arthroplasty, not POA, is. Prophylaxis documentation does not erase the diagnosis; it also does not create a HAC that CMS did not list.

Iatrogenic pneumothorax with venous catheterization. Central-line placement with a documented iatrogenic pneumothorax is the intended pairing. Pneumothorax after lung biopsy, thoracentesis, or barotrauma is a different documentation problem. Do not drop every iatrogenic pneumothorax into category 14.

Worked encounters

ED-present ulcer. A patient arrives with a sacral wound. The emergency physician and wound nurse describe full-thickness ulceration to fascia before the inpatient order. The attending later writes “Stage IV pressure ulcer.” POA is Y. If the code is an MCC, it may still split the MS-DRG. This is not a non-POA HAC payment event. CDI’s remaining job is staging accuracy and clinical support, not “hiding” a community ulcer.

Hospital-acquired Stage IV, other MCC present. On hospital day 10 a heel ulcer is documented as Stage IV; admission skin exam was intact. POA is N. CMS will not let that HAC code buy the MCC split. The same claim also has acute respiratory failure as a validated MCC that is not a selected HAC. The MS-DRG can still be with MCC from respiratory failure. Staff who say “the HAC wiped the whole MCC DRG” are describing the provision incorrectly.

Hospital-acquired Stage IV, no other CC/MCC. Same ulcer, no other CC or MCC. Grouping now behaves as though the Stage IV were absent, so the case does not pick up an MCC from that ulcer. The ulcer is still coded. Quality review may still see it. The check does not.

Insufficient POA. Wound care writes “Stage III, likely hospital-acquired.” The attending never addresses timing. Coders cannot infer N from a consultant’s guess. POA U pays like N for the HAC provision. A nonleading POA query is the repair, not a silent U forever.

Wrong orthopedic HAC. Postoperative DVT after lumbar fusion is a clinically important secondary diagnosis. It is not category 13 (that category is DVT/PE after hip or knee replacement). SSI in the lumbar incision can be category 11 (spine). Read the category limits before you tell a surgeon “this will be a HAC no matter what.”

Venous catheter versus lung procedure. Iatrogenic pneumothorax after subclavian central-line attempt matches category 14 when the codes pair on the CMS list and the event is not POA. Iatrogenic pneumothorax after transthoracic needle biopsy does not become category 14 by analogy.

What CDI actually does on a HAC chart

• Confirm the diagnosis is clinically supported and meets additional-diagnosis reporting rules. Clinical validation queries belong here when the HAC label is unsupported.
• Confirm POA against the inpatient order time, including ED and observation time that precedes that order.
• Confirm the category fit, especially procedure-limited SSIs, hip/knee DVT/PE, and pneumothorax with venous catheterization.
• Leave reimbursement, MS-DRG, and quality-score language out of the query. The 2026 ACDIS/AHIMA query brief still forbids leading the provider with payment or quality outcomes.
• Do not tell coding to drop a valid HAC so the DRG “looks better.” The provision already withholds the CC/MCC upgrade when POA is N or U. Omitting a reportable diagnosis is a compliance problem, not a HAC workaround.
• Know where the current list lives: CMS ICD-10 HAC List ZIP files and Appendix I of the MS-DRG Definitions Manual for that fiscal year. Encoder folklore is not the list.

Traps this section is built to catch

• Treating HAC-POA and HACRP as one program or applying the 1 percent hospital cut to a single claim’s HAC.
• Teaching that any non-Y POA blocks HAC payment—W is payable like Y.
• Using U as a quiet default instead of querying when the record can support Y or N.
• Calling every inpatient UTI, every DVT, every SSI, or every pneumothorax a CMS HAC.
• Putting hip/knee SSI in the DVT/PE category, or hip/knee DVT in the spine/neck/shoulder/elbow SSI category.
• Treating ED-acquired conditions as N because “it happened in the hospital.”
• Believing the diagnosis is removed from the dataset; only the CC/MCC grouping effect of that selected secondary HAC is withheld.
• Inventing extra HAC categories from Joint Commission never-events or from NHSN measure lists.

The payment sentence is short on purpose: selected secondary HAC, not POA (N or U), paid as though that diagnosis were not present—unless another non-HAC CC or MCC still supports the split. HACRP is a different program.