13.1 Systematic Identification of Documentation Gaps
Key Takeaways
A review lead is not a diagnosis: medications, abnormal values, and treatment intensity require context and provider interpretation.
Specificity, etiology, validation, relationships, conflicts, and timing are useful review categories rather than an exhaustive official taxonomy.
In final MS-DRG v44, J18.9 and J69.0 are both secondary-diagnosis MCCs subject to exclusions; I50.31 is also an MCC. Specificity does not guarantee a grouping change.
Hypertension and CKD are presumed related unless the provider indicates nonrelationship; naming another CKD cause alone does not remove that presumption.
Use neutral patient-specific queries, preserve supporting and contrary evidence, and close unresolved requests under policy without inventing a diagnosis.
Identify the question before choosing a diagnosis
A documentation gap is an unresolved question about what happened to this patient, what a documented condition means, or whether the record supports it. It may affect communication, code assignment, quality data, or another use of the record. A costly treatment or a low working DRG is a review lead; neither establishes that documentation is deficient. A complete review can end with accurate symptom or unspecified coding when greater certainty is unavailable.
The CDI specialist connects observations with the provider's assessment and the applicable reporting rules. The specialist does not independently convert an abnormal value, medication, or radiology impression into an encounter diagnosis. Under FY 2027 ICD-10-CM I.A.19, code assignment follows the provider's diagnostic statement rather than the coder applying clinical criteria. That rule coexists with clinical validation: a meaningful inconsistency can require clarification of the diagnosis and its rationale.
A practical gap framework
These categories organize review; they are not an official, exhaustive four-category classification. Several can occur in one case.
| Question | What a real gap might look like | Appropriate next step |
|---|---|---|
| Is the diagnosis sufficiently described? | Heart failure is documented, but supported type or acuity is unclear | Ask for the clinically supported description without promising a severity tier |
| Is the cause established? | Anemia and gastrointestinal bleeding are documented, but their relationship is unclear | Ask about the anemia's nature or cause using the actual course and provider assessments |
| Is a documented diagnosis clinically supported? | A discharge diagnosis appears inconsistent with the available clinical course | Seek the provider's clinical rationale or clarification; do not independently delete it |
| Is a relationship needed? | A complication or relationship between two conditions is unclear | Check the classification's relationship rules before querying |
| Are the record's statements compatible? | One involved provider confirms a condition while another rules it out | Resolve the actual conflict through the attending under organizational policy |
| Is timing unclear? | A condition is established later, but its presence at the inpatient admission order is uncertain | Review the earlier evidence and clarify POA when necessary |
Specificity serves accuracy. A diagnosis need not become more severe to become more precise. A clinical validation query may confirm a condition, establish a different explanation, or clarify that it was ruled out. An unanswered query does not supply the missing diagnosis.
Specificity and grouping are different decisions
Suppose the provider documents pneumonia and the record includes a witnessed aspiration event, a swallowing assessment, and antibiotic treatment. Those observations can support a patient-specific question about the pneumonia's nature. They do not prove aspiration pneumonia: preserve their source, timing, and the clinician's interpretation instead of presenting aspiration as the required answer.
The FY 2027 Tabular List includes aspiration pneumonia under J69.0, whose title is pneumonitis due to inhalation of food and vomit. J18.9 describes pneumonia with an unspecified organism. In final MS-DRG version 44 Appendix C, both J18.9 and J69.0 are listed as MCCs when used as secondary diagnoses, subject to applicable exclusions and grouping logic. Thus, changing the former to the latter cannot be taught as an automatic non-CC-to-MCC upgrade. A principal-diagnosis change may affect the grouping pathway, but the complete case must be grouped to establish its result.
The same discipline applies to heart failure. I50.9 does not identify type or acuity. I50.31 identifies acute diastolic heart failure and is an MCC in version 44; it is not a CC. Even a correctly identified MCC does not guarantee a different DRG or payment. The code's role in the case, exclusions, procedures, and other diagnoses still matter.
Etiology requires interpretation
Consider a patient with documented anemia, gastrointestinal bleeding, serial hemoglobin results, and a transfusion. Organize the baseline and subsequent results, fluid administration, evidence of bleeding, and the provider's explanation. A hemoglobin decrease alone does not establish acute posthemorrhagic anemia. Transfusion supports the importance of the problem but does not independently identify its cause.
If clarification is warranted, ask the provider to characterize the anemia in the context of the documented bleeding and other relevant evidence. D62 describes acute posthemorrhagic anemia when that diagnosis is established. Do not prefill D62 or describe it as the patient's “true condition” before the provider answers. A clinically appropriate alternative, including inability to determine the cause, must remain possible.
Treatment reconciliation follows the same pattern. Check whether a medication was ordered, administered, held, or discontinued and why it was used. Norepinephrine's labeled indication includes severe acute hypotension in adults; its presence alone does not establish a particular shock diagnosis. Likewise, diuresis alone does not establish heart failure type or acuity. When the reason is already clear and the record is consistent, a treatment without a preferred diagnostic label may require no query.
Validation without an automatic rejection rule
When a diagnosis appears unsupported, first check whether the apparent mismatch reflects missing context, earlier treatment, an unavailable baseline, or genuine contradictory evidence. Include information that argues against the diagnosis as well as information supporting it. Normal results after treatment do not necessarily describe the untreated presentation.
For example, a patient has a discharge diagnosis of severe sepsis, but the reviewer cannot identify documentation explaining the infection-associated acute organ dysfunction. The useful question concerns the provider's clinical assessment and the relationship of any organ dysfunction to sepsis. It is not a demand to remove the diagnosis because cultures are negative or one lactate result is normal. Current SSC guidance treats sepsis as a clinical diagnosis that cannot be established or excluded by one biomarker or test. The sepsis lesson distinguishes that clinical assessment from quality abstraction and coding.
The 2026 ACDIS/AHIMA query brief supports clarification of clinically inconsistent or unsupported diagnoses. Cite the actual observations, not reviewer-created diagnoses masquerading as evidence. A value should retain its units, time, and source; do not relabel an elevated lactate as “lactic acidosis” unless that interpretation is documented. Avoid financial or quality-score language in the query.
Check presumed relationships before asking
FY 2027 I.C.9.a.2 directs hypertensive CKD coding when hypertension and CKD are both present, unless the provider indicates that the CKD is not related to hypertension. A separate statement linking them is unnecessary. Merely naming another cause of CKD is not the guideline's exception: multiple causes can coexist.
For documented hypertension and stage 4 CKD, with no applicable heart involvement or statement of nonrelationship, I12.9 and N18.4 identify the hypertensive CKD and stage. If the provider additionally documents diabetic CKD, do not assume that this automatically removes the hypertension relationship. Follow all applicable Index, Tabular, and combination-code instructions. Hypertension with both qualifying heart disease and CKD may require I13 rather than separate I11 and I12 coding.
Other presumed relationships must be checked individually. The ICD-10-CM “with” convention applies where the classification links the conditions; it does not authorize linking every condition occurring in the same patient. When clarification of stage, a genuine conflict, or an unclear relationship is needed, query that unresolved issue instead of requesting a relationship the classification already presumes.
Work a gap to closure
- Define the encounter and question. Identify the setting, responsible provider, and the exact missing or inconsistent element.
- Build a short evidence timeline. Include relevant current findings, baseline information, actual interventions, responses, and contrary evidence. Use prior records in a current clinical scenario under the 2026 brief; do not import historical diagnoses without a current trigger.
- Check whether a query is needed. Review existing compatible provider documentation, classification presumptions, legitimate unspecified coding, and applicable documentation exceptions.
- Ask neutrally. Use the appropriate involved provider and a patient-specific format. A multiple-choice query requires supported choices and an open alternative, not a fixed set of severe diagnoses or a universal number of options.
- Reconcile the answer. Ensure the response meets permanent-record requirements, resolve any resulting conflict, and apply reporting, POA, and grouping rules separately. Record closure under policy when no answer is obtained.
An effective review explains why clarification mattered and what evidence supported it. It does not measure success by agreement with a proposed diagnosis or by an increase in reimbursement.
Sources: FY 2027 ICD-10-CM guidelines, I.A.15, I.A.19, I.B.14 and I.C.9; 2026 compliant query practice brief, V–X; FY 2027 final MS-DRG v44 manual, Appendix C; FY 2027 Tabular List and Index; 2026 SSC adult guidance; norepinephrine prescribing information.
A provider documents pneumonia. The record also contains a witnessed aspiration event and a swallowing assessment. A reviewer proposes an aspiration-pneumonia query because J18.9 is supposedly non-CC and J69.0 an MCC. What is the best evaluation under final MS-DRG v44?
The event itself establishes J69.0, so a provider response is unnecessary.
J69.0 always becomes principal and guarantees an MCC DRG when a query is answered.
Clarification may be appropriate on clinical grounds, but both codes are listed as secondary-diagnosis MCCs subject to exclusions; the proposed automatic tier upgrade is incorrect.
Because the two codes have the same listed severity, clarification of the pneumonia type can never serve documentation integrity.
The provider documents hypertension and stage 4 CKD attributed to diabetes. The record contains no statement that CKD is unrelated to hypertension. Which approach follows FY 2027 I.C.9.a.2?
Remove the hypertension relationship because any other documented CKD cause overrides it.
Require a separate provider sentence linking hypertension and CKD before considering a combination code.
Use I13 for every hypertension–CKD pair, even without qualifying heart disease.
Retain the presumed hypertension–CKD relationship unless the provider indicates nonrelationship, while following applicable diabetic CKD and other classification instructions.
Anemia, gastrointestinal bleeding, a falling hemoglobin, and a transfusion are documented, but the anemia cause is unclear. What should a clinically justified CDI query seek?
The provider’s characterization of the anemia and its cause, using the actual course and a neutral format that permits other conclusions.
Confirmation of D62 because the transfusion independently establishes acute posthemorrhagic anemia.
A more severe anemia diagnosis selected by the reviewer from the hemoglobin decrease alone.
Removal of the anemia because no diagnosis can be reported when its cause remains unspecified.
Sections you finish are checked off in the contents.