Inpatient Coding of Probable Diagnoses

A possible, probable, suspected, likely, questionable, or still to be ruled out condition can be coded if still documented as such at the time of discharge. Other similar terms used to describe possible conditions could include consistent with, compatible with, indicative of, suggestive of, and comparable with.

All the terms used above (and there could be others that would fit the definition) are terms used to describe conditions that the patient may have and that the patient is being treated or monitored for.

Per the OCG, coders can code these conditions as if they were confirmed as long as they are still documented as such at the time of discharge. Here are a couple of examples of this:

• Patient is admitted with SOB and has COPD. On admission, the patient is thought to have possible pneumonia. The patient is begun on IV antibiotics and responds to treatment. The patient is discharged with the diagnosis of COPD with acute exacerbation/acute bronchitis. There is no mention of pneumonia at the time of discharge or in the DS. In this case, the coder would need to query the MD to clarify if the possible diagnosis of pneumonia was ruled out. It may be that there is documentation in the record to elude to this and no query needed but if unclear it should be queried. The diagnosis of pneumonia would not be able to be reported unless the MD was queried and clarification of the diagnosis was given.
• Same patient as above BUT the DS does list the diagnosis of possible pneumonia as a final diagnosis. In this case, pneumonia is appropriate as a reportable diagnosis. It is a probable condition that is still being documented at discharge.

There are, however, exceptions to this rule. Here are some of the exceptions:

• Only confirmed cases of HIV/AIDS may be reported. If the diagnosis is listed with any of the terms used above and not confirmed, the disease should not be reported.
• Avian influenza, novel influenza, or other identified influenza should not be coded when they are documented as possible/probable or other terms. Only confirmed cases of Avian, novel and other identified influenza should be coded as such.
• Only confirmed cases of Zika virus can be coded.
• If there is a code for a suspected condition and the condition is listed as such, then this code would be reported instead of the code for the condition. An example would be for suspected adult and child abuse, neglect and other maltreatment. There are specific codes to report if this is suspected abuse.

Comparative/contrasting secondary diagnoses

These should be coded as if they were established in ICD-10-CM. The OCG for uncertain diagnoses should be followed. This was not the case in ICD-9 when instead of coding each of the comparative/contrasting SDX, the symptom was coded instead. Here are a couple of examples of this:

• Patient is admitted with pneumonia and during the admission complains of abdominal pain. Workup is done without confirmation of the cause of this pain. The physician documents in the record, at the time of discharge, abdominal pain was worked up with no clear cause. This was thought to be mild acute pancreatitis vs. alcoholic gastritis. In this case, both acute pancreatitis and alcoholic gastritis would be reported. No code would be reported for the symptom of abdominal pain.
• Patient is admitted with sepsis with no definitive cause. At the time of discharge, the MD documented that the sepsis was thought to be due to possible pneumonia vs. severe acute bronchitis. In this case, both pneumonia and acute bronchitis would be reported.

Comparative/contrasting principal diagnoses

This rule did not change with implementation of ICD-10-CM. If two or more contrasting/comparative diagnoses are documented at the time of discharge (and are considered as PDX) they are coded as if they were confirmed diagnoses. Sequencing would depend on the circumstances of admission. If treatment was considered equal, then either may be sequenced as the PDX.

Probable or possible malignancy

This may be the most difficult diagnosis that coders face. It is very difficult to assign a diagnosis of malignancy/cancer to a patient when this is only documented as a possible/probable or suspected condition. However, there is official coding advice that directs the coder to do just this. Even though most don’t like to do this it is what must be done. Here are a couple of examples (one PDX and one SDX):

• Patient was admitted with flank pain and workup was begun. The patient had CT scan that did show a mass in the left kidney. The patient wanted further workup as an outpatient. At the time of discharge, the physician documented the final diagnosis as kidney mass, most likely renal cell carcinoma. Workup was planned as an outpatient. In this case, the PDX should be reported as renal cell carcinoma.
• Patient was admitted with pneumonia. They had a history of lung malignancy that had been removed years ago. On this admission, patient had CT of the lung. This showed multiple lung nodules suspicious for metastasis. During the admission, the patient also complained of headache so a CT of the head was also performed. This showed multiple areas in the brain that were suspicious for metastasis. The patient refused further workup of these sites at this time. At the time of discharge, the physician documented the final diagnoses as pneumonia, history of lung cancer with possible metastasis to the lung and brain. Further workup will be discussed on follow up visit. In this case, both the lung metastasis and the brain metastasis would be reported as secondary diagnoses.

Borderline diagnoses

A diagnosis that is documented as “borderline” at the time of discharge is coded as a confirmed diagnosis, unless there is a specific index entry in ICD-10-CM for a borderline condition. Here are a couple of examples:

• If at the time of discharge there is a diagnosis of borderline diabetes, this is not coded as diabetes as there is a specific index entry for borderline diabetes.
• If at the time of discharge there is a diagnosis of borderline hypercholesterolemia, this would be coded as hypercholesterolemia since there is not a specific index entry for borderline hypercholesterolemia.

Impending/threatened conditions

These are NOT the same as a possible/suspected condition or uncertain diagnosis. If it occurred, code it. If the impending or threatened condition did not occur during the admission then the coder would need to reference the Alphabetic Index to see if there is a subentry term for “impending” or “threatened” for the condition and also reference the main term entries for “impending” or “threatened”. If they are listed, then assign the code that is given. If not, then the diagnosis that is listed as impending/threatened would NOT be reported.

Concern for/concerning for

AHA Coding Clinic First Quarter 2018 confirms that the term “concern for” should be interpreted as an uncertain diagnosis and coded using the guidelines for “uncertain diagnoses” in the inpatient setting. Please disregard previous publications/emails regarding the coding of “concern for” as it has been best practice to NOT code these in the past.

Authored by Kim Boy, RHIT, CDIP, CCS, CCS-P

References:
ICD-10-CM Official Guidelines for Coding and Reporting FY2019
Pages: 16, 18, 29, 53, 54, 83, 108, and 111
ICD-9-CM Coding Clinic, Third Quarter 2005 Page: 21
ICD-9-CM Coding Clinic, First Quarter 2006 Page: 4-5
ICD-9-CM Coding Clinic, First Quarter 2011 Page: 10
ICD-10-CM/PCS Coding Clinic, Second Quarter 2016 Page: 9
ICD-10-CM/PCS Coding Clinic, First Quarter 2018 Page 18-19

Atom Audit is an easy-to-use, intuitive coding review application that enables auditors to manage the entire audit life cycle efficiently and effectively. Features such as the customizable reporting dashboard and coder workflow give users the ability to turn audit insights into actionable next steps.

The information contained in this coding advice is valid at the time of posting. Viewers are encouraged to research subsequent official guidance in the areas associated with the topic as they can change rapidly.