15.3 Secondary Diagnosis Reporting Criteria
Key Takeaways
- UHDDS other diagnoses are additional clinically significant conditions that affect patient care by requiring clinical evaluation, therapeutic treatment, diagnostic procedures, extended length of stay, or increased nursing care and/or monitoring.
- Report conditions that coexist at admission, arise later, or affect treatment or stay length; exclude diagnoses from an earlier episode that have no bearing on the current hospital stay.
- A comorbidity coexists at admission; a complication arises during the stay—related to, but not the same as, present-on-admission assignment taught later in Domain VI.
- Do not report historical or remote conditions with no current evaluation, treatment, or relevance, and do not code abnormal findings unless the provider documents their clinical significance.
- Reportability is not limited to CC/MCC codes; a neither-severity diagnosis can still be required as a secondary diagnosis when it meets UHDDS additional-diagnosis criteria.
15.3 Secondary Diagnosis Reporting Criteria
Quick Answer: For reporting purposes, other (additional) diagnoses are additional clinically significant conditions that affect patient care by requiring clinical evaluation, therapeutic treatment, diagnostic procedures, extended length of hospital stay, or increased nursing care and/or monitoring. Comorbidities coexist at admission; complications arise during the stay. Do not report historical conditions from an earlier episode that have no bearing on the current hospital stay.
Section III of the ICD-10-CM Official Guidelines is the inpatient rule set for everything that is not principal diagnosis. CDI programs get into trouble here in both directions: under-reporting treated acute conditions that never made the discharge summary, and over-reporting remote history, copy-forward problem-list lines, and lab abnormalities no provider interpreted.
The UHDDS other-diagnosis definition
UHDDS item #11-b defines other diagnoses as all conditions that coexist at the time of admission, that develop subsequently, or that affect the treatment received and/or the length of stay. It then excludes diagnoses that relate to an earlier episode which have no bearing on the current hospital stay.
The Official Guidelines interpret that definition for reporting as additional clinically significant conditions that affect patient care in terms of requiring one or more of five services. Memorize the five; exam items quote them.
| UHDDS reporting criterion | What it looks like on an inpatient chart | Reportable example | Not, by itself, enough |
|---|---|---|---|
| Clinical evaluation | A clinician assesses the condition during the stay (consult, differential, documented assessment) | Nephrology consult and serial creatinines for acute kidney injury | A problem-list line nobody addressed |
| Therapeutic treatment | Medication, procedure, diet, device, or other therapy directed at the condition | Continuation or adjustment of insulin for diabetes; IV diuretics for decompensated heart failure | A remote diagnosis with no current therapy |
| Diagnostic procedures | Testing performed to evaluate the condition | Duplex ultrasound for suspected lower-extremity deep-vein thrombosis | An incidental past imaging report copied into the H&P |
| Extended length of stay | The condition kept the patient in the hospital longer | Ileus after surgery that delayed discharge | A stable comorbidity that never changed the discharge plan |
| Increased nursing care and/or monitoring | Extra vital-sign frequency, telemetry, intake/output, glucose checks, turning, isolation, or similar | Q1-hour neurologic checks after a new stroke; strict intake and output for hyponatremia | A history of seasonal allergies with no nursing implications |
Meeting one criterion is enough. A condition does not need to be a CC or MCC, does not need to change the MS-DRG, and does not need to appear in the first paragraph of the discharge summary. Conversely, sitting on the CC list does not make a diagnosis reportable if it was never evaluated or treated on this stay.
The entire record still matters. Treatment recorded only in the medication administration record, a consult that never reached the discharge summary, or nursing monitoring on a flowsheet can establish that a condition affected care. CDI’s concurrent review exists partly to get the provider to document the diagnosis those indicators support, because abnormal findings and non-provider labels are not a substitute for a provider diagnosis (see below).
Coexist, arise later, or affect the stay—then apply the five tests
Section III is broader than “problems present on admission.” A condition may:
- Coexist at admission (chronic kidney disease stage 3 that drives medication dosing all week).
- Develop subsequently (hospital-onset pneumonia; acute blood-loss anemia after surgery).
- Affect treatment or length of stay even if it was not the reason for admission (a new oxygen requirement that delays discharge).
Each still has to pass the five-criterion filter. “Developed subsequently” is not a license to code every nursing comment. “Coexist at admission” is not a license to dump the entire past medical history onto the claim.
Historical conditions with no current relevance
This is the exclusion half of UHDDS #11-b, and it is one of the highest-yield CDI rules in Domain VI.
Do not report diagnoses that relate to an earlier episode and have no bearing on the current hospital stay. Remote childhood appendectomy with a well-healed scar and no abdominal issue this admission is history, not a secondary diagnosis. A tonsillectomy twenty years ago does not belong on an IPPS claim for community-acquired pneumonia. Resolved gestational diabetes after a pregnancy that ended a decade ago does not become a diabetes code on a current orthopedic stay unless diabetes is a current, evaluated condition.
History codes (for example, personal-history Z codes) are a different tool. They may be reported as additional diagnoses when the past condition impacts current care or influences treatment—a history of total hysterectomy that explains why a pelvic diagnosis is impossible, a history of anaphylaxis to penicillin that changes antibiotic choice, a history of malignant neoplasm under current surveillance that is being evaluated. The test is current relevance, not “the H&P copied the list from last year’s note.”
Copy-forward electronic problem lists are the usual failure mode. Yesterday’s resolved acute kidney injury (AKI) still sitting as an active line can look like a current MCC. If no one evaluated or treated AKI this stay, it is not a UHDDS other diagnosis. If they did evaluate it, the attending needs to say whether it is still valid. That is a clinical-validation / query problem (Domain III), not a reason to auto-code the fossil.
Comorbidity versus complication
CDI language uses these terms constantly; they are not interchangeable, and they are not synonyms for “CC.”
| Term | Timing relative to the inpatient admission | Typical documentation story | Related later rule |
|---|---|---|---|
| Comorbidity | Coexists at the time of admission (present when inpatient care starts) | Longstanding COPD, CKD, or heart failure that still required evaluation or treatment this stay | Often present on admission (POA) = Y when the inpatient order is written; POA values are taught in the next Domain VI chapter |
| Complication | Arises during the hospital stay (not present at admission) | Postoperative wound infection; hospital-onset AKI after nephrotoxins; new respiratory failure after aspiration on day 4 | Often POA = N; some of these conditions also intersect CMS hospital-acquired condition payment rules, which are a later section |
A comorbidity can be a CC, an MCC, or neither. A complication can be a CC, an MCC, or neither. CC/MCC is a CMS severity class. Comorbidity/complication is a timing-and-relationship class. Mixing them produces the false statement “every CC is a complication.” It does not. Essential hypertension continued from home is a comorbidity, not a complication, and in many years it is not a CC at all—yet it is still reportable if it was evaluated or treated.
For this sitting, learn the timing distinction cleanly. Do not try to assign the four POA values (Y, N, U, W) here; that is its own guideline cluster. Do not collapse this distinction into the HAC payment provision or the HAC Reduction Program—those are different CMS constructs taught later. The only crossover to remember now is conceptual: a complication that arises after admission is the kind of condition POA and HAC rules will later scrutinize, but reportability under Section III still starts with the five UHDDS criteria.
Abnormal findings are not diagnoses
Laboratory, imaging, pathology, and other diagnostic abnormal findings are not coded and reported unless the provider indicates their clinical significance. A sodium of 128 mmol/L is a number. Hyponatremia is a diagnosis. CDI may query when indicators are strong and no provider interpretation exists. CDI may not instruct coding to mint hyponatremia from the lab file alone. The same rule protects the claim from “hyperglycemia” that was a single postoperative glucose, never assessed, never treated, and never named.
Reportable is not “severity,” and severity is not “reportable”
Domain I taught CC/MCC recognition. This section is the gate before that recognition matters:
- Does a provider document the condition (or a compliant query response that reaches the record)?
- Does it meet UHDDS other-diagnosis criteria (one of the five)?
- Is it current to this stay rather than an irrelevant historical episode?
- Then ask whether CMS currently classes it as MCC, CC, or neither, and whether an exclusion list pairing applies.
Skipping to step 4 is how programs code decorative MCCs. Stopping at step 4’s “neither” and deleting a treated diabetes diagnosis is how programs under-report resource use that utilization and quality teams still need on the claim.
Worked encounter: report AKI, omit the relic
Mr. Patel is admitted for community-acquired pneumonia (principal diagnosis after study). The H&P past medical history lists remote appendectomy in 2004 and seasonal allergic rhinitis. Neither abdominal nor allergy issue is evaluated, treated, tested, or monitored. On day 2 his creatinine rises; the team orders serial laboratories, holds lisinopril, gives fluids, and obtains a nephrology consult. The attending documents acute kidney injury.
- AKI required clinical evaluation, diagnostic procedures, treatment, and increased monitoring → report as a secondary diagnosis.
- Remote appendectomy has no bearing on this stay → do not report as a current additional diagnosis.
- Seasonal allergic rhinitis with no current relevance → do not report.
Pneumonia remains PD. AKI does not “steal” PD just because it might be a CC or MCC. Appendectomy does not become a secondary diagnosis just because it is true.
Worked encounter: comorbidity versus complication
Ms. Brooks is admitted for acute on chronic systolic heart failure. Type 2 diabetes mellitus is present on the home medication list, glucose is monitored, and insulin is continued—comorbidity, reportable. On hospital day 3 she develops a catheter-associated urinary tract infection that is cultured, treated with antibiotics, and documented by the attending—complication, reportable if it meets the five tests (it does). Both can appear as secondary diagnoses. Only the infection is a hospital-course complication. POA assignment and any HAC payment consequence are later questions; Section III already says both conditions affected care.
Worked encounter: history that does matter
Mr. Nguyen is admitted for cellulitis. He has a documented severe penicillin allergy with prior anaphylaxis. The team avoids beta-lactams and documents the allergy as the reason. That history influences treatment on this stay and is reportable as an additional diagnosis (allergy/history coding as appropriate). The same chart’s “history of wisdom-tooth extraction, 1998” still has no bearing and stays off the claim.
How CDI should apply Section III
- Hunt for treated or monitored conditions missing from the provider’s diagnosis list; those are query candidates when indicators exist.
- Hunt for problem-list fossils and uninterpreted labs heading toward the billed codes; those are “do not report” or “query for clinical significance” candidates.
- Teach providers that continuing a home drug, ordering a test, or increasing monitoring is exactly the UHDDS evidence HIM needs—if they also name the condition.
- Never tell a provider “we cannot code this because it is not a CC.” Reportability and severity class are different gates.
- Never tell a provider “please add this remote history so SOI looks higher.” Irrelevant history is excluded by UHDDS on purpose.
Traps: coding every past medical history line; coding only CCs and MCCs; treating comorbidity and complication as synonyms for CC; coding labs without provider significance; leaving a treated acute condition off the claim because it was not in the discharge summary’s first paragraph. Secondary diagnosis reporting is a current-stay clinical-significance test with five published on-ramps and a hard off-ramp for history that no longer matters.
For inpatient reporting, the Official Guidelines interpret UHDDS other diagnoses as additional clinically significant conditions that affect patient care by requiring:
Which statement correctly applies the UHDDS exclusion for historical conditions?
In inpatient CDI usage consistent with UHDDS timing, which distinction is correct?
A patient is admitted for community-acquired pneumonia. The H&P lists remote appendectomy 20 years ago with no current abdominal issue. During the stay, acute kidney injury is documented and treated with serial creatinines, fluids, and a nephrology consult. Which reporting choice follows UHDDS additional-diagnosis rules?