According to the official coding guidelines, ICD-10 code U07.1 for COVID-19 may be reported under the following circumstances:
- As documented by the provider;
- Documentation of a positive COVID-19 test result; or
- A presumptive positive COVID-19 test result.
“Presumptive” was the description given for cases that local testing initially indicates are positive, but have not been validated with additional testing by the Centers for Disease Control and Prevention (CDC). We no longer need to do two tests, as the CDC now accepts local testing and no longer recognizes “presumptive” reporting as a distinct category.
Uncertain cases of COVID-19 are not to be reported; however, if the provider verifies that they believe the patient has the condition, nothing else is needed to report ICD-10 code U07.1. Notice the first circumstance in which COVID-19 is reported: “as documented by the provider.”
The official coding guidelines go on to say that “confirmation” does not require
documentation of the type of test performed; the provider’s documentation that the individual has COVID-19 is sufficient.”
We seem to have a problem here. If we give in to a reckless disregard for compliance, we could fall victim to fraudulent billing of COVID-19 when the diagnosis is uncertain. “Clinically confirmed COVID-19” can be documented despite negative, pending, or no test results being available; however, the provider should include supporting clinical indicators of concern to justify the diagnosis.
Unfortunately, the Centers for Medicare & Medicaid Services (CMS) did not give us ICD-10 code U07.2, COVID-19, SARS-CoV2 (virus) not identified. That code would have solved much of our reporting conundrum (once we figured out a standardized definition for U07.2, that is). However, we are left in a quandary. What do we do if we have an overwhelming preponderance of clinical evidence to suggest that a patient has COVID-19, in the absence of an available test? Perhaps even more perplexing: what do we do if we have an overwhelming preponderance of clinical evidence to suggest that a patient has COVID-19 in the presence of a suspected false-negative PCR test (up to 30 percent of cases)?
How do we proceed while maintaining compliance with not only our personal ethics, but also the spirit of the guidelines we have received from CMS?
Surely, we are not going to go with the “assume everyone has COVID-19” approach for the inpatient population!
My answer: tread very carefully. I do not believe that CMS (nor anyone, for that matter) wants to see a mountain of fraudulent activity, such as billing for COVID-19 for regular cold and flu or pneumonia patients who likely do not have COVID-19. (For reference, see here: https://www.cms.gov/files/document/se20015.pdf regarding why some may find extra motivation to bill a U07.1 from a financial standpoint.) Spoiler alert: you get a 20-percent bonus on your relative weight. Do not do it for the money, do it only when it is the right thing to document clinically.
If your physician has a strong and valid clinical opinion that a patient has COVID-19, there ought to be a way that we can document that. There should be a rationale, on paper, that is defensible and reflects the best information and the best standards we have to date, in order to portray an accurate clinical presentation, while also justifying why we believe that a patient has COVID-19, even in the absence of a COVID-19 test (and in some cases, in spite of it).
This means that you need to make sure that the record supports every possible shred of evidence that an expert diagnostic work-up can overturn.
Reference this recent coding guidance:
Question: Please provide guidance on correct coding when the provider has confirmed the documented COVID-19 after the test results come back negative. How should this be coded?
Answer: If the provider still documents and confirms COVID-19 even though the test results are negative, or if the provider documented disagreement with the test results, assign code U07.1, COVID-19. As stated in the Official Guidelines for Coding and Reporting for COVID-19, “code only a confirmed diagnosis of the 2019 novel coronavirus disease (COVID-19) as documented by the provider . . . the provider’s documentation that the individual has COVID-19 is sufficient.”
Danger ahead: if you are going to do this, make sure you can support it clinically, which leads us to the discussion below;
If you are going to query for a case of COVID-19 with absent or conflicting test results:
The patient should present ill enough to be admitted to the hospital with hypoxia and/or acute respiratory distress syndrome (ARDS), along with evidence of pneumonia and at least some (three or more) of the following:
- Known or suspected contacts with COVID-19 infected patients;
- Signs or symptoms: cough (86 percent of cases) fever/chills (30.7 percent), and nausea (24 percent), RR > 24 (17.3 percent), or supplemental oxygen required at presentation (27.8 percent);
- Hypoxemia on pulse oximetry beyond what would be clinically expected. Patients often present with less-than-expected distress and work of breathing and compliant of lung distress, yet a very concerning level of hypoxemia in the early phases (P/F ration <300 +/- RR > 30 with room air oxygen saturations less than 94 percent), progressing to respiratory failure (12.2 percent, requiring mechanical ventilation at 88.1 percent mortality) or ARDS (P/F Ration is often less than 200 in ARDS), which will correspond to ever-increasing oxygen demands;
- Brain: while viral entry into brain/CNS not proven definitively, seizures, encephalitis, anosmia may be all related to direct involvement;
- Nose: anosmia may be related to direct viral involvement of nasal nerve endings (or olfactory bulb in brain);
- Blood: pulmonary embolism, elevated D Dimers, elevated Ferritin levels, thrombocytopenia, lymphopenia, microvascular thrombosis, C-reactive protein elevations;
- Heart: elevated troponin and BNP, MI, heart block, arrhythmia, systolic dysfunction, cardiogenic shock, myocarditis (20 percent of severe);
- Liver: LFTs (frequently elevated);
- Intestines: diarrhea (not uncommon);
- Kidneys: renal dysfunction very common in critical disease. Virus detected in proximal tubules (CRRT between 3.2 percent up to 15 percent of the time);
- Endothelium: may be the factor that ties all the above together as endothelial cells lining blood vessels become present in all organs (rare);
- Skin: different lesions including urticaria, petechiae and others documented (rare);
- Positive COVID-19 test: 70 percent of all COVID-19 (as suggested by the limited studies we have). False positives are rare, so once this once this criterion is met, the patient has it has the condition. This could change in the future if chronic carriers become known.
Per the National Institutes of Health (NIH): “false-negative test results can occur. In people with a high likelihood of infection based on exposure, history, and/or clinical presentation, a single negative test does not completely exclude SARS-CoV2 infection, and testing should be repeated;”
- Presence of the characteristic failure to respond to antibiotics and/or steady worsening of pulmonary/oxygenation status and hemodynamic decompensation despite supportive treatment;
- Focus of treatment may be shifted to limiting effects of cytokine storm rather eliminating infection;
- For bacterial secondary infection add: + PCT elevation + sputum cultures and rapid onset worsening of infiltrates; and
- + Elevated IL-6 levels
Note: we are not recommending any attempt to capture a presumption of COVID-19 in patients with mild symptoms who are not hospitalized and are not requiring high levels of resource utilization to stabilize them or make them comfortable. For the outpatient population, we recommend that you go strictly by OP guidelines. Follow only what symptoms are known. If the COVID-19 test results in a negative and you suspect that it is a false negative (unfortunately), those just may end up going unreported due to compliance and coding guidelines. We recommend a retest when possible, if the provider has a strong belief that the test may be wrong. Keep in mind that there may always be exceptions; this is just a general discussion.
I would like to add that we don’t know why these things happen with COVID-19, but we are at least moving towards a taxonomy to explain what things happen to COVID-19 patients. We can use those “what’s” as clinical indicators to justify a query. We should also be diligent to document them as criteria to justify a diagnosis, when appropriate to do so.
While I don’t have any research to back this up, from a query standpoint, I would recommend trying to find at least three or more of the items listed. Perhaps in the future, we can get some studies to show exactly how many of these items correspond to a reasonable statistical likelihood that COVID-19 is present. Perhaps by then we will have more access to faster/better testing. Until then, we are all just doing the best we can.
Perhaps even more importantly, we strongly recommend that the doctor list these criteria as their “evidence of” a presumptive diagnosis to make it clear this is an evidence-based decision and not a wild shot at collecting cash for care. Coding Clinic has previously stated that “evidence of” is not an uncertain diagnosis, but rather a confirmed one (2014).
Another rule of thumb: not every patient who presents with symptoms has the disease. Not every patient who has the disease will present with exactly three or more symptoms from the list above. The provider’s professional judgement will now, as always, supersede all other considerations. That is not to say that we should not be performing clinical validation of nonsensical claims, but right now, the clinical documentation improvement (CDI) and coding world (along with the rest of the world) is still playing catch-up, and probably not doing much “clinical validation of COVID-19,” especially not under the current circumstances.
Anyway, let’s not forget that for coding purposes, we are still forced to operate under the “Guideline 19,” which states that the coding of a diagnosis is not based on clinical criteria, but rather on the provider’s diagnostic statement. In this write-up we are mostly addressing query, not coding practice.
Keep in mind that the opinions and suggestions here are not “peer-reviewed medical science.” This is merely an attempt to ensure that there is enough evidence to appropriately support complaint query practice, and to support the reporting of a COVID-19 diagnosis to CMS in a justifiable and defensible way.
The goal here is simply to put some methodology out there to help prevent over-reporting COVID-19, while also maintaining the spirit of compliance and allowing the doctor to exercise his or her very necessary critical judgment and medical authority to treat an illness, save a life, and/or reduce or prevent long-term morbidity.
Let’s conclude with one last piece of recent coding advice.
“Question: please provide guidance on correct coding when the provider has documented COVID-19 as a definitive diagnosis, before the test results are available, and the test results come back negative.
Answer: coding professionals should query the provider if the provider documented COVID-19 before the test results were back and the test results come back negative. Providers should be given the opportunity to reconsider the diagnosis based on the new information.”
Commentary: we recommend a retest if you suspect a false negative. Sometimes, a retest may confirm the diagnosis for you. If you get two negative tests back to back, then in some cases, the provider may opt to rule out the diagnosis (make sure this is documented clearly as well, or the provider will almost definitely receive a query).
One final word: do not “go rogue” and proceed to put these practices into place without first working with your own physician advisors, infectious disease specialists, and coding and compliance staffs. A team approach is required here.
*At the time of publication, many of the occurrence rates for the presenting signs and symptoms were updated at the last minute from here: