COVID-19 TalkBack follow-up. 

During the Talk Ten Tuesdays weekly Internet radio broadcast last week, we received many questions which we never got to answer. I am going to share some of those answers with you now.

Stacey asked, “can nurses charge for telehealth visits?”

There are two parts to this question. ‘can that service be provided as telemedicine now?’ and ‘Can the nurse provide the service?’

Telemedicine is serving many purposes right now. It is permitting folks to stay at home so they don’t expose others. Or get exposed themselves if they don’t have the coronavirus. It is allowing the providers to minimize their exposure. And it is being used in many different venues, including inpatient.

The Centers for Medicare & Medicaid Services (CMS) has put out a list of services which they are going to allow as telehealth visits, some of which are temporary additions for the COVID-19 pandemic. Here is the link to the approved codes: It is quite extensive. The modifier, -95, must be attached to designate it as a telemedicine service. This is a link to the Medicare Telemedicine HCP Fact Sheet:

Evaluation and management (E&M) services usually require a physician or other qualified healthcare professional, but there is an exception to this rule. 99211, the code for an established patient office or other outpatient visit, does not require the presence of a physician or other qualified health care professional. Usually the presenting problems are minimal and the visit takes five minutes or so. Let me emphasize that this is only for established patients to your practice.

99211 is on the list of Medicare-approved telemedicine services. Therefore, providers can bill for their nurses doing a 99211 by telemedicine the same as they could for in-person visits. Caution: this is only applicable for 99211. If the nurse spends 15 minutes or the patient or problem is complex, they cannot bill any higher E&M code.

If the patient is covered by a non-Medicare provider, I refer you back to the payer.

There were lots of questions about what to do if the COVID-19 test is negative.

Karen said her team is assigning Z20.828 for patients who test negative, per the AHA website.

Z20.828 is titled, “Contact with and (suspected) exposure to other viral communicable diseases.” If a patient has had known or suspected exposure to COVID-19, this code is applicable AS LONG AS THEY ARE NOT DEEMED TO HAVE COVID-19. If the provider receives results of a positive test, or, in their clinical judgment, they believe (and document) that the patient has COVID-19, then the code is U07.1. The fact that the patient has the disease makes the exposure code superfluous. Z20.828 may be used with the symptoms or manifestations that caused the presentation for medical care.

The next option is a rare occurrence, in my opinion. This is an asymptomatic patient who thinks they had exposure, but the exposure (and COVID-19) is ruled out. The code is Z03.818, Encounter for observation for suspected exposure to other biological agents ruled out.

  • Asymptomatic, because otherwise their diagnosis would be the symptoms.
  • If the exposure were still suspected, their code would be Z20.828 as above.
  • This code might be used for a patient being evaluated after exposure to another individual who turns out to not be a potential vector. If I came and shook your hand, and you were nervous and got checked out, Z03.818, Encounter for observation for suspected exposure to other biological agents ruled out, would be applicable when you discovered that I had COVID-19 in early March and am no longer infectious. You were not exposed.
  • Another scenario where Z03.818 could be used is if a patient presents to the hospital for some other reason, like myocardial infarction, without known exposure, during a pandemic (or future epidemic), and they are routinely checked for COVID-19 to determine whether they belong in a COVID-19 unit.

However, at the current time, with a false negative test rate of 30%, so many presymptomatic and asymptomatic infected individuals, and no ability to test for immunity, you can and should consider almost everyone in the general population a “suspected” exposure to COVID-19; therefore, Z20.828 is overwhelmingly going to be the correct code. The CDC says, “a negative result does not exclude the possibility of COVID-19.”

If a patient does not have any known or suspected exposure, they test negative, and they are asymptomatic (first of all, who could get a test done right now under those circumstances?!), this would be considered a screening, and Z11.59, Encounter for screening for other viral diseases, would be the appropriate code. Screening (for anything) is only done on asymptomatic patients. This will apply to PCR screening for acute infection and serologic testing for past infection. No symptoms, no exposure, routine testing, negative for acute COVID-19.

Screening is not happening much in the U.S. presently, but, in the future, when the testing is way more available and public health concerns are being addressed, it might happen more frequently. This code will be for the patient who is tested during a targeted testing of a population. Hypothetical: my town wants to know who has COVID-19 and tests all the inhabitants. I didn’t come to the doctor with a concern that I had it or was exposed to it; I was caught up in the epidemiological sweep. Think tuberculosis testing in the hospital every two years (Z11.7, Encounter for testing for latent tuberculosis infection) or a gynecologist getting a routine gonorrhea/chlamydia test on sexually active patients who present for their yearly PAP smear, Z11.3, Encounter for screening for infections with a predominantly sexual mode of transmission. A current scenario for screening could be a routine sweep through a nursing home facility without any suspected exposure just to see if there are any positive patients.

There are implications for getting the codes right and avoiding using the Z11.59 screening code too liberally. Only certain conditions can be put through under claims for COVID-19 treatment as per the National Uniform Billing Committee with a “DR” condition code ( The only conditions included are:

  • 29, Other coronavirus as the cause of diseases classified elsewhere
  • 1, COVID-19 for services from April 1 on
  • 828, Contact with and (suspected) exposure to other viral communicable diseases
  • 818, Encounter for observation for suspected exposure to other biological agents ruled out

Note Z11.59 is NOT on the list.

If the patient came in with a cough (or some other symptom/s), no known or suspected exposure, and the test is negative, the code would be R05, Cough (or the other symptom/s). Z03.818 is not used in conjunction with symptoms because the Z03 category has a Types1 Excludes note to code to signs or symptoms.

There is a final code which no one else really talks about, Z71.1, Person with feared health complaint in whom no diagnosis is made. This is for an asymptomatic patient who self-refers, whom the clinician does not deem to be at risk of COVID-19 (e.g., not during the pandemic), and the healthcare provider does not believe any investigation is warranted. This is not useful at this juncture in time but might be warranted in the future.

Jessica wants to know if you hold a claim for the results of the pending test, does the provider need to add an addendum to include the results post-discharge?

The ICD-10-CM Official Coding and Reporting Guidelines, April 1, 2020 through September 30, 2020 ( declare that “documentation of a positive COVID-19 test result, or a presumptive positive COVID-19 test result” is grounds for coding a diagnosis of COVID-19 (U07.1). Does this mean that the provider needs to physically document that the test result was positive?

Normally, coders can’t assume an interpretation of a test result. If the test comes back positive post-discharge and you can access it, you can capture U07.1 without provider query. It is important to get  good epidemiological data and to have COVID-19-related medical care paid for. It is like documenting, “probable malignancy,” and the pathology returns with renal cell carcinoma. You are permitted to capture the malignant neoplasm without query.

If the test returns negative, the provider just used symptoms, like “upper respiratory tract infection,” as their diagnosis and didn’t suggest in their documentation that they suspected COVID-19, then I would not ask for an addendum to establish that COVID-19 had been ruled out.

The quandary arises when the provider suspects COVID-19 and uses an uncertain diagnosis, but the final results of the test are negative. There is a pretty high rate of false negatives, and some tests take days to be resulted. That patient may still have COVID-19. This patient may require a query to the clinician to settle the question.  

The AHA is recommending that, if testing is pending and the provider does not have enough evidence to make a clinical diagnosis, inpatient admissions and outpatient encounters should be held back for coding and billing until the results return. The hope is that this will give more robust and accurate data on the prevalence of the virus. They even go so far as to suggest that institutions develop facility-specific coding guidelines stipulating this bill-hold procedure.

Evangeline wanted clarification as to whether sepsis should always be sequenced first.

Real sepsis (that is, life-threatening organ dysfunction due to a dysregulated host response to infection as opposed to fishing-for-dollars or medicine-by-checkbox sepsis) that is present on admission is the principal diagnosis (PDx). The underlying infection is a secondary diagnosis as are the manifestations and comorbidities.

POA-Y sepsis where you question clinical validity requires a clinical validation query. If the clinician supports it, then sepsis is PDx. If they remove it, then sepsis is no longer a valid diagnosis and U07.1 will be the PDx.

POA-N sepsis needs to be addressed. If it occurs within ~24 hours of admission and the clinical indicators might suggest that the patient was ill and the sepsis really was POA-Y, then you need to query for POA status. If it really is not POA, then the reason, after study, that occasioned the admission wasn’t the sepsis, so sepsis should be a secondary diagnosis.

A very interesting question posed by Patricia out of Henry Ford regarded COVID-19 infection, pneumonia, and HIV+ status. Her encoder takes it to HIV infection, and she wondered if this was correct.

First, if your encoder takes you someplace that seems fishy, you must go back and do it the old-fashioned way (check your Codebook!). Don’t blindly code or sequence diagnoses if the results don’t seem correct.

My advice is to consider linkage and figure out what bought the bed. If the provider documents, “Pneumocystis pneumonia due to AIDS,” and “COVID-19,” then HIV infection is the PDx. If the provider documents viral pneumonia due to COVID-19 and lists HIV+ as a secondary diagnosis, then U07.1, J12.89, Other viral pneumonia, and Z21, Asymptomatic human immunodeficiency virus infection status may be the correct sequencing. The viral load may be a helpful piece of data to determine if a query is required to establish B20, Human immunodeficiency virus disease instead of Z21 (in a newly diagnosed patient). If the provider doesn’t give linkage and you can’t tell from the encounter story, you may need to query to determine PDx.

Piper wanted to know what the percentage of false positives for COVID-19 is.

We don’t have this data. We are not doing enough testing to be able to determine this. What I can tell you is this:

  • The false positive rate of COVID-19 testing is believed to be relatively low, especially compared to the false negative rate. The CDC 2019-nCoV Real-Time RT-PCR Diagnostic Panel was designed to minimize the likelihood of a false positive.
  • Right now, the prevalence is extremely high. That means that lots of people have the virus (whether they are demonstrating symptoms).
  • A false positive may be less problematic than a false negative. If you are told you are positive, you will likely self-quarantine. The issue is that after you think you are no longer contagious, you might relax, get exposed for real, and come down with COVID-19 then. At this time, you would have a true positive test result. This may lead people to think that you got re-infected when it is believed that COVID-19 infection leads to immunity (duration to be determined).

Beth asked what to use if a pregnant patient comes in without symptoms – Z11.59 or an O code?

  • If the pregnant patient has COVID-19, she would have an O code as PDx, O98.5- (viral disease) with or without O99.5- (respiratory disease – if symptomatic). U07.1 would be a secondary diagnosis. This would apply even if she were asymptomatic because what to do with the patient and what intensity and frequency of follow-up would most assuredly take into consideration the fact that she is pregnant. She still has the virus and it is complicating the peripartum condition.

O98.5-, Other viral diseases complicating pregnancy, childbirth, or the puerperium, would only be appropriate if the disease were felt to be present. It would not be appropriate if the disease were ruled out by the provider.

  • If she had exposure to COVID-19 and were asymptomatic and test-negative, her code would be Z20.828, Contact with and (suspected) exposure to other viral communicable diseases. I can’t find any coding advice but believe her pregnancy would be captured by applying the Z3A- code and encounter for supervision of normal (Z34-) or high-risk (O09.-) pregnancy, or other applicable O code (like anemia, obesity, or diabetes in pregnancy).

[I considered, but don’t think Z36.89, Encounter for other specified antenatal screening, would be applicable because the other tests included in Z36.- seem very maternofetal oriented. Similarly, I do not think Z03.79, Encounter for other suspected maternal and fetal conditions ruled out would apply, for the same reasons.]

  • Asymptomatic, no known or suspected exposure, not COVID-19 positive, routine testing of a pregnant patient (in the future) could be captured with Z11.59, Encounter for screening for other viral diseases, with an additional Z3A- and encounter for supervision of pregnancy code, or other applicable O code as noted above.

Finally, Annette asked, “If tests are so unreliable and everyone needs two, why are we testing? We don’t we just treat?”

The volume of testing in the United States is wholly inadequate at the present time. We have no idea of the magnitude of the denominator for death and incidence rates. We have no clue of the scope of asymptomatic infection. We have no numbers on recovered patients.

The positives seem to be reliable. It’s the negatives that are potentially suspect. It can be that the specimen was collected incorrectly or that it was done at the wrong time (e.g., too early or after virus was cleared from nasopharynx). The patients who are being tested are the ones admitted and being treated. These are probably the wrong patient population to be testing. Unless there is a concern that there is a different infection in play, these patients should be assumed to be infected and treated as such. The other folks are the ones where early and ample testing could make an impact.

We no longer need to do two tests; the CDC is accepting the testing of local and regional centers (“Presumptive” wasn’t verbiage for an uncertain diagnosis. It meant the local/regional/state test was positive without CDC confirmation.). If the result is negative, there is the option of repeating it and doing a confirmatory test. In fact, the CDC “no longer (recognizes) the designation of ‘presumptive positive,’” as of March 14, 2020.

The biggest issue is that there is no proven treatment. There are experimental treatments, and we provide supportive care for COVID-19. However, it would be good to determine if COVID-19 was ruled out and some other infection were the culprit. Perhaps there would be an effective treatment for the alternative etiology.

Thank you all for the great questions! This topic is obviously so crucial and all-consuming that we, at MedLearn and ICDUniversity, are feverishly working to put together a webinar to make this all crystal clear for you. I believe it is scheduled for April 29 at 1:30 pm.

Until then, stay safe, sane, and at home.

Programming Note: Listen to Dr. Erica Remer every Tuesday on Talk Ten Tuesdays, 10-10:30 a.m. EST.


Erica E. Remer, MD, CCDS

Erica Remer, MD, FACEP, CCDS, has a unique perspective as a practicing emergency physician for 25 years, with extensive coding, CDI, and ICD-10 expertise. As physician advisor for University Hospitals Health System in Cleveland, Ohio for four years, she trained 2,700 providers in ICD-10, closed hundreds of queries, fought numerous DRG clinical determination and medical necessity denials, and educated CDI specialists and healthcare providers with engaging, case-based presentations. She transitioned to independent consulting in July 2016. Dr. Remer is a member of the ICD10monitor editorial board and is the co-host on the popular Talk Ten Tuesdays weekly, live Internet radio broadcasts.

You May Also Like

HCCs: The Role of CDI and Risk Scores

HCCs: The Role of CDI and Risk Scores

Predicting coding patterns using the HCC risk scores can be a valuable endeavor. EDITOR’S NOTE: Longtime RACmonitor contributing correspondent Frank Cohen, a senior healthcare analyst,

Read More

Leave a Reply

Your Name(Required)
Your Email(Required)