What to choose for the principal diagnosis (Is there a choice?)

June 26, 2020 / By Audrey Howard, RHIA

In my opinion, one of the hardest aspects of coding is selecting the principal diagnosis. It is not simply assigning a code to a documented diagnosis. It takes critical thinking to ascertain “that condition established after study to be chiefly responsible for occasioning the admission of the patient to the hospital for care.” (Uniform Hospital Discharge Data Set definition of principal diagnosis). As often stated in AHA Coding Clinic, the sequencing of the principal diagnosis depends on the circumstances of the encounter. There is no shortcut to appropriately selecting the principal diagnosis. The coder must read through all reports and provider documentation to accurately identify that one condition which required the inpatient admission. But what happens when the patient is admitted with several conditions? There still can be only one principal diagnosis. 

The first thing I do when I review a record of a patient admitted with multiple diagnoses, which could potentially meet the principal diagnosis definition, is separate out the conditions and evaluate each one individually. I ask myself these questions for each condition:

  • Did this condition necessitate inpatient admission? Would the patient have been admitted to the inpatient setting for this condition? 
  • Did this condition meet admission criteria? Was the patient sick enough? Was the treatment significant enough?
  • Could this condition stand alone as the reason for admission?

If the answer to these questions is yes, then I can consider sequencing the condition as principal diagnosis. Let’s take a look at the following example:  A 65-year-old female presented through the Emergency Department with a chief complaint of weakness, fatigue and dyspnea/shortness of breath on exertion for the past week. The patient also complained of dysuria for the past month. The patient was alert, cooperative, and in no apparent distress without fever or chills. The chest was clear bilaterally with no rales, rhonchi or wheezing. No accessory muscle use or stridor. A COVID-19 test came back negative with no known exposure. The patient was noted to have 3+ bilateral pitting edema up to the knees. The following table shows a summary of the pertinent information:


Signs and symptoms

Lab and diagnostic tests


UTI, possible sepsis

Weakness, fatigue, dysuria, tachycardia (109), respiratory rate 26

Elevated WBC, elevated lactate, bacteria (multiple organisms in urine culture)

IV Zosyn

Deep venous thrombosis (DVT)

3+ bilateral pitting edema up to the knees

Day 2 ultrasound: Left lower venous: there is age indeterminate deep vein thrombosis visualized in the proximal femoral, mid femoral and distal femoral veins.

Heparin 5,000 units every 8 hours for 2 days, Eliquis (apixaban)

Pulmonary embolism (not documented by provider)

Dyspnea/shortness of breath on exertion, tachycardia (109), respiratory rate 26

Elevated D dimer (1445), CT PE “While no large central pulmonary embolus is seen, respiratory artifact limits evaluation for small peripheral emboli.”

Heparin 5,000 units every 8 hours for 2 days, Eliquis (apixaban)

The final diagnoses included UTI and DVT. Sepsis was ruled out. The provider never documented pulmonary embolism. Do we have a choice for principal diagnosis in this example? Based on the circumstances of admission and the existing documentation, the only option for principal diagnosis is UTI. A query related to pulmonary embolism would have been appropriate because the patient had signs and symptoms related to it on admission (dyspnea on exertion, tachycardia, elevated D Dimer) and it was treated (heparin, Eliquis) even though the CT PE only showed small peripheral emboli. What final corresponding diagnosis, if any, would the provider be willing to add to explain the symptoms? The DVT was not supported as principal diagnosis. Although the pitting edema was present on admission and the patient was started on heparin on admission, if I evaluate the DVT by itself, I cannot conclusively answer “yes” to all the questions I listed previously. After study, it does not appear to be the condition necessitating the inpatient status. The presenting signs and symptoms of weakness, fatigue, dysuria support UTI as the principal diagnosis. Although UTI does not always require inpatient care, it does appear to be the reason for admission in this case, especially since a diagnosis of sepsis was ruled out.

When reviewing the record for principal diagnosis selection, I take a step back and ask myself “What was the reason for the admission at that time?”  Another way of looking at this is to try and determine the intent of the admission. Sometimes, when the patient is transferred to the facility, the intent of the transfer will take sequencing priority over other conditions the patient brings with them. If the other conditions could have been treated appropriately at the first facility, then they would not be sequenced as the principal diagnosis at the second facility. The reason for the transfer would be sequenced as the principal diagnosis. 

For example, a patient was transferred from facility A to facility B due to worsening kidney function. The patient was originally admitted to facility A five days prior to transfer due to sepsis with an infected heel ulcer and was treated with IV antibiotics. According to the documentation, the patient received an incision and debridement at facility A for treatment of the heel ulcer. It was planned for the patient to receive further surgical debridement at facility A. However, due to the worsening kidney function, the surgery could not be completed at facility A at that time and the patient was transferred to facility B for care of the acute kidney failure. The principal diagnosis is the reason (intent) for transfer – the acute kidney failure. After the kidney function improved, the patient underwent surgery for care of the left heel. Neither the sepsis nor the heel ulcer should be sequenced as principal diagnosis at facility B because they did not necessitate the transfer for care.

Before finalizing the principal diagnosis selection, the coding professional also needs to apply any official coding guideline or convention that takes precedence and directs that one condition is sequenced as principal diagnosis over another condition. A good example of this is sepsis due to pneumonia. Although both conditions can necessitate inpatient admission, meet admission criteria and stand alone as the reason for admission, the sepsis coding guideline states that the systemic infection must be sequenced as principal diagnosis over the localized infection which does not allow for a choice between the two conditions;  therefore, sepsis is sequenced as principal diagnosis. 

There are definitely a lot of things to consider when selecting the all-important principal diagnosis. Happy sequencing!

Audrey Howard, RHIA, is a senior outsource services consultant with 3M Health Information Systems.

Learn best practices for documenting surgical complications.