COVID-19 as a primary diagnosis, comorbid or ancillary condition

Clinical Documentation Integrity (CDI) is more important than ever in the era of COVID-19.

I have heard grumblings on TV and in social media that the diagnosis of COVID-19 in some patients who are counted as COVID-19 patients overwhelming the country’s hospital systems is not really legitimate.

The implication is that unless the condition is the primary diagnosis, it should not count. Oh good? Is it a phantom PCR test, fake personal protective equipment, fake ventilation needs in an isolation room of a COVID-19 ward? Doesn’t rotating rooms require the same amount of terminal cleaning time? Shouldn’t precautions be taken to prevent these patients from transmitting the virus to other patients?

If this is a secondary diagnosis, is it really incidental (eg, not affecting the patient at all, as if they were asymptomatic), or is it the secondary cause of the admission? How can we tell?!

Clinical Documentation Integrity (CDI) is more important than ever in the era of COVID-19. It is essential that we ensure that suppliers document accurately and that coders retrieve the correct codes in the correct order.

There are actually five buckets of COVID-19-related patients in our hospitals now. There are admitted patients:

  1. With acute or symptomatic persistent COVID-19 infections with life-threatening manifestations like COVID-19 pneumonia or acute respiratory distress syndrome (ARDS). MRDx is U07.1, COVID-19 plus manifestation, Present on admission-Yes (POA-Y);
  2. For an underlying condition that has been exacerbated or caused by contracting COVID-19. Consider a severe exacerbation of chronic obstructive pulmonary disease (COPD) or heart failure worsened by COVID-19 hypoxemia. U07.1 is a secondary diagnosis and a major comorbid condition or complication (MCC);
  3. For a fully unrelated condition, which also has POA-Y COVID-19 infection. An example would be a motor vehicle accident patient with a fractured femur whose admission PCR is positive. These cases are really fortuitous;
  4. For a condition that COVID-19 is believed to be responsible for, but which U07.1 has resolved. For example, a patient who has a pulmonary embolism, kidney failure, or organizing pneumonia, but no longer has active, acute COVID-19. These patients have a secondary diagnosis of U09.9, post-COVID-19 status, unspecified. They don’t have U07.1; and
  5. With another condition that contracts COVID-19 as a nosocomial infection. They also get U07.1 as a code, but that’s a secondary condition and POA-N.

Coders should remember that uncertain diagnoses of COVID-19 are not coded as U07.1. Everyone should remember that coders are allowed to assign U07.1 if a positive COVID-19 test result has been obtained, even if no one has documented it.

Are the statistics correct? I see quoted 40% of COVID “accessories”. Is the sequencing accurate? Are the POA indicators correct? Is it really incidental?

The most important relevant official directive is IC 1. g. 1) (b) Sequencing of codes (of COVID-19). The first part of the first sentence says, “when COVID-19 meets the definition of primary diagnosis”. This invokes Section II, which deals with MRDx selection, as long as the tabular listing and alphabetical listing do not contradict the guidelines. People are misinterpreting the guidelines and putting undue emphasis on the second sentence, which continues, “U07.1, COVID-19 should be sequenced first.”

When does a condition meet the definition of PDx? PDx is the “condition that is established, after study, to be primarily responsible for the admission of the patient to hospital for care”. He is present at admission. It usually consumes the lion’s share of resources. The only bucket above that meets these criteria is bucket #1. Only in active infections is U07.1 PDx.

The guidelines openly advise following sepsis guidelines if COVID-19 has progressed to sepsis. If the patient is obstetrical or neonatal, these guidelines prevail. If there is a lung transplant recipient with COVID-19 pneumonia, the T86 code indicating lung transplant infection would be PDx.

There are two small wrinkles in relation to this problem. The first is the 20% increase in the Medicare Severity DRG, which is applied in the event of a positive COVID-19 laboratory test. Some people may think that incidental COVID-19 does not qualify for the upward adjustment. There are precautions and actions taken solely on the basis of SARS-CoV-2 infection that warrant this adjustment. I can’t imagine a situation right now, during the Omicron/Delta surge, that would negate COVID-19 meeting legitimate secondary diagnostic criteria.

The second issue is the Health Resources & Services Administration (HRSA) sequencing instruction for COVID-19 program eligibility for the uninsured. Uninsured patients are covered under the American Rescue Plan Act of 2021 through the Provider Relief Fund. HRSA guidelines for submitting claims are that U07.1 should be listed as the primary diagnosis. Paraphrased, HRSA acknowledges that using COVID-19 as the primary diagnosis is contrary to official coding guidelines; however, HRSA’s COVID-19 Uninsured Program is not a health plan, so it is not subject to HIPAA requirements.

Does it give the impression that the health facility is playing with the system? The government sets the rules; hospitals are just trying to follow them to get paid appropriately.

Nosocomial COVID-19 could be determined by analyzing the POA indicator. This could be an important quality metric to determine and monitor.

How could we disentangle incidental diagnostics from additional preparation? Without a chart review, I would recommend compiling a list of diagnoses that could likely be exacerbated or initiated by the presence of COVID-19 infection. They are likely to be cardiac or pulmonary in etiology, but not exclusively. If one of these (eg COPD exacerbation, acute on chronic heart failure) is PDx and COVID-19 is a secondary diagnosis, the virus is a comorbidity and not an accident. Accidental COVID-19 is also likely to be asymptomatic or minimally symptomatic, as opposed to true manifestations.

This will be miss some cases. For example, let’s say an elderly patient who has COVID-19 and has been weak becomes dehydrated and falls, fracturing his hip. Is her hip fracture sequenced PDx to match the primary procedure, or is her PDx COVID-19 due to him being the admitter?

Does it really matter? The public probably won’t be savvy enough to understand the nuanced difference between comorbid and incidental if they can’t understand that an mRNA vaccine doesn’t alter a person’s DNA. Hospitals are overflowing with patients with COVID-19, patients with something else caused by COVID-19, patients with something else who happen to be COVID-19 positive, and a few patients who don’t have COVID -19 , now or again.

An absurdly high positivity rate indicates that there is some crazy transmissibility variant ravaging our population. Accessory, schmincidental; isn’t that bad enough already? !

Note on programming:

Listen to Dr. Erica Remer today as she co-hosts Talk Ten Tuesdays with Chuck Buck at 10 a.m. ET.