How do I interpret RT-PCR results?

7th Jan 2021

Real-time reverse transcription polymerase chain reaction, or RT-PCR, is the standard method for diagnosing COVID-19. But the test is only as good as the sample put into it. Let’s explore how the timing of sample collection and type of specimen used affects RT-PCR results. 


How does the type of specimen used affect RT-PCR results?

Upper respiratory tract specimens

Most RT-PCR specimens are collected from the upper respiratory tract using nasopharyngeal swabs, although oropharyngeal swabs, and more recently, saliva, may also be used. The sensitivity and time course are thought to be similar between all three sample types. 

These samples are most accurate within the first week after symptoms appear, and viral loads decline over the next couple of weeks. Four weeks after symptom onset, the RT-PCR is more likely to be negative than positive.1

Probability of detecting SARS-CoV-2 using RT-PCR with a nasopharyngeal swab versus time since symptom onset. Graph.

Figure 1. Upper respiratory tract specimens provide the most accurate results for COVID-19 diagnosis using real-time reverse transcription polymerase chain reaction​ (RT-PCR) testing in the first week after symptom onset. By four weeks after symptom onset, RT-PCR is more likely to return a negative result (Adapted from Sethuraman, N, Jeremiah SS, and Ryo A. 2020).

One study using upper respiratory tract samples reported a false negative rate of 38% on the day symptoms appeared. It dropped to 20% on the third day after symptom onset but rose to 67% about two weeks later.2  Other studies have shown a similar trajectory, but with varying rates of false negatives. 

Probability of false negative test result on RT-PCR using a nasopharyngeal swab versus time since exposure to SARS-CoV-2. Graph.

Figure 2. When using real-time reverse transcription polymerase chain reaction (RT-PCR) to diagnose SARS-CoV-2 infection, the false-negative rates are 38% the day symptoms of COVID-19 appear. This rate drops to 20% on the third day after symptom onset and rises steadily to 67% just over two weeks after symptom onset (Adapted from Kucirka, LM, Lauer, SA, Laeyendecker, O, et al., 2020).

So, if you’re testing a patient within the first few days of symptom onset, the standard upper respiratory tract samples seem to be fairly reliable—although you’re still likely to miss a significant number of infected individuals. But often, by the time a patient presents to the hospital, they may have had symptoms for more than a week, leading to less reliability of these samples. 


Lower respiratory tract specimens

Specimens from the lower respiratory tract—including sputum and bronchoalveolar lavage samples—tend to be more accurate later in the disease process, especially for patients with pneumonia.

Viral loads in these specimens are high early in the disease, and they remain positive longer than upper respiratory tract samples, especially in patients with more severe disease.1 So, for a hospitalized patient with pneumonia, a sputum sample may be more reliable than a nasopharyngeal swab. 

Probability of detecting SARS-CoV-2 with RT-PCR using a nasopharyngeal swab or bronchoalveolar lavage or sputum versus time since symptom onset. Graph.

Figure 3. Lower respiratory tract specimens taken through bronchoalveolar lavage or sputum have higher initial viral loads that are retained for a longer period after COVID-19 symptom onset. So more accurate results are obtained through real-time reverse transcription polymerase chain reaction (RT-PCR), later in the disease process using these specimens compared to upper respiratory tract specimens (Adapted from Sethuraman, N, Jeremiah SS, and Ryo A. 2020).

But not all patients can produce a sputum sample, and bronchoalveolar lavage is an invasive, aerosol-generating procedure, which is why they are generally not the first tests of choice.  

Clickable call to action, "Start learning for free", with direct link to sign up for a free Medmastery trial account.


How can I be sure a negative test is a true negative result?

The decline in viral load isn’t always consistent, even in those with mild disease, and viral load can jump from day to day, as shown in a study of hospitalized patients.3

Viral load of SARS-CoV-2 versus days after COVID-19 symptom onset. Graph.

Figure 4. SARS-CoV-2 viral load in sputum and nasopharyngeal samples varies over time. On a given day, viral load may fall below a threshold, yielding a negative reverse transcription polymerase chain reaction (RT-PCR) test, while viral load may rise in the following days to yield a positive RT-PCR test result (Adapted from Wolfel, R, Corman, VM, Guggemos, W, et al., 2020).​

If we consider viral load over time, with both nasopharyngeal and sputum samples, then on any given day, a swab may produce a positive or negative result depending on the viral load. For example, there may be a day when a patient will test negative, followed by a positive result on a later day. This is why two negative tests, done at least 24 hours apart, are needed to declare a patient no longer infected, and might explain reports of patients who tested positive days or weeks even after two negative tests.  

Two negative tests, done at least 24 hours apart, are needed to declare a patient free of SARS-CoV-2 infection.


But what about testing patients without symptoms?

One study reported a false negative rate of 100% on the day of infection, which was reduced to 66% on day four, presumably one day before symptoms typically appear.2

Probability of false negative rate for RT-PCR over time size exposure in COVID-19. Graph.

Figure 5. In asymptomatic patients with SARS-CoV-2 infection, there is a 100% false-negative rate with real-time reverse transcription polymerase chain reaction, or RT-PCR, on the day of infection and a 66% false-negative rate four days after infection (Adapted from Kucirka, LM, Lauer, SA, Laeyendecker, O, et al., 2020).​

Another study estimated that the peak of viral transmission occurs about one day before symptom onset, so viral loads are probably high enough to detect at that point.4

Graph of density of virus, or viral load, versus days after symptom onset.

Figure 6. Estimates suggest that SARS-CoV-2 viral load in asymptomatic patients is high enough to detect infection using real-time reverse transcription polymerase chain reaction (RT-PCR) one day prior to symptom onset (Adapted from He, X, Lau, EHY, Wu, P, et al., 2020).

It’s estimated that those who develop symptoms will likely start to test positive roughly around two days before symptom onset, but the problem is that we don’t know when that will be! 

Testing for SARS-CoV-2 too early, before the viral load has risen above the level of detection, can lead to a negative test.

So, it’s important to know that an asymptomatic person who had contact with a confirmed case or who was present at a superspreader event where others were also infected, might not yet test positive for the virus. They might be infected but we cannot see it yet on RT-PCR. That’s why this person still needs to take appropriate distancing or isolation measures to prevent transmission.


So, RT-PCR is positive, what comes next?

A positive RT-PCR test simply indicates that viral ribonucleic acid (RNA) is present in the sample, but it does not tell us whether the virus that is present is infectious, or even intact. When the virus is broken down by the immune system, it will leave behind some viral components, including RNA, which can be detected by RT-PCR. But, of course, this RNA alone is not infectious. 

In order to determine whether intact virus is present in the patient’s respiratory tract, researchers try to grow the patient’s sample in the lab to see if they can isolate the virus itself. If they are successful, this indicates that the patient is shedding intact virus that is capable of replicating.

Gloved hand with swab. Petri dish with virus. Illustration.

Figure 7. To determine if there is intact SARS-CoV-2 in a patient’s respiratory tract, researchers try to isolate the virus by growing the patient’s sample under laboratory conditions. If successful, it is clear this patient is shedding intact virus that can replicate. 

They can take this one step further, and culture the isolated virus with human cells. If the virus is able to infect the human cells, we can conclude that infectious virus is present.

Petri dishes with human cells and SARS-CoV-2 virus. Illustration.

Figure 8. If SARS-CoV-2 that has been isolated from a patient sample is cultured with and infects human cells, researchers can conclude that this patient has an infectious virus and can transmit it to others. 

Researchers have been able to isolate live virus from hospitalized patients in the first week after symptom onset, with sputum samples producing more live virus than nasopharyngeal swabs. Beyond eight days after symptom onset, researchers could no longer isolate viable virus from any sample type, suggesting that patients may not be infectious after the first week, even if RT-PCR results remain positive much longer.1

Graph comparing timeline for isolating SARS-CoV-2 in laboratory and positive RT-PCR results with nasopharyngeal and bronchoalveolar lavage or sputum specimens.

Figure 9. While real-time reverse transcription polymerase chain reaction (RT-PCR) results may still be positive, SARS-CoV-2 cannot be isolated beyond eight days after symptom onset regardless of sample type used (Adapted from Sethuraman, N, Jeremiah SS, and Ryo A. 2020).​

Another study showed a correlation between the patient's threshold cycle, or Ct, value and the ability to culture infectious virus from upper respiratory tract samples.5 A Cvalue under 40 is considered a positive test, the lower the C​value the higher the viral load. In their population, patients with a C​value of 34 or higher did not produce infectious virus. Some patients in this study tested positive but did not appear to be infectious. So there might be a Cvalue at which patients are no longer infectious even though they still shed the viral RNA.

Percentage of positive SARS-CoV-2 culture versus threshold cycle. Graph.

Figure 10. Patients with a threshold cycle (Ct) of 34 or higher did not produce a positive SARS-CoV-2 culture and did not appear to be infectious (Adapted from La Scola, B, Le Bideau, M, Andreani, J, et al., 2020).

These findings are partly why the CDC allows healthcare workers to return to work ten days after symptom onset as long as they have been symptom-free for at least 72 hours, even if RT-PCR results are still positive. 


What do I do if my patient has tested negative?

While a positive SARS-CoV-2 RT-PCR test is a pretty reliable indicator that your patient is infected, you can’t rule out the disease based on a single negative test. 

Man with positive test. Man with negative test. Illustration.

Figure 11. A positive SARS-CoV-2 real-time reverse transcription polymerase chain reaction (RT-PCR) test is a reliable indicator of infection. However, it is not possible to rule out SARS-CoV-2 infection with a single negative test. 

It’s important to interpret any negative test in the clinical context. Look at the patient’s clinical presentation and ask yourself some questions: 

  • Is this clinical presentation typical of COVID-19?
  • What is their risk? 
  • Did they have close contact with an infected person? 
  • Do they work at a high-risk occupation? 
  • Have they been practicing physical distancing and wearing a mask in public? 

When in doubt, retest, possibly using a different specimen type. Consider ordering other tests, such as computer tomography (CT) or serology testing to provide cross-validation. And be sure to adhere to strict infection control protocols and isolate the patient until you have more confidence in the result. 

When you are unsure if a negative SARS-CoV-2 test is reliable, make sure to do the following: 1.Redo the test using a different specimen type. 2. Cross-validate using CT or serology testing. 3.Implement infection control protocols and isolate the patient.


That’s it for now. If you want to improve your understanding of key concepts in medicine, and improve your clinical skills, make sure to register for a free trial account, which will give you access to free videos and downloads. We’ll help you make the right decisions for yourself and your patients.


  1. Sethuraman, N, Jeremiah, SS, and Ryo, A. 2020. Interpreting diagnostic tests for SARS-CoV-2. JAMA. 323: 2249–2251. PMID: 32374370
  2. Kucirka, LM, Lauer, SA, Laeyendecker, O, et al. 2020. Variation in false-negative rate of reverse transcriptase polymerase chain reaction-based SARS-CoV-2 tests by time since exposure. Ann Intern Med. 173: 262–267. PMID: 32422057  
  3. Wölfel, R, Corman, VM, Guggemos, W, et al. 2020. Virological assessment of hospitalized patients with COVID-2019. Nature. 581: 465–469. PMID: 32235945
  4. He, X, Lau, EHY, Wu, P, et al. 2020. Temporal dynamics in viral shedding and transmissibility of COVID-19. Nat Med. 26: 672–675. PMID: 32296168
  5. La Scola, B, Le Bideau, M, Andreani, J, et al. 2020. Viral RNA load as determined by cell culture as a management tool for discharge of SARS-CoV-2 patients from infectious disease wards. Eur J Clin Microbiol Infect Dis39: 1059–1061. PMID: 32342252

Recommended reading

  • Fang, Y, Zhang, H, Xie, J, et al. 2020. Sensitivity of chest CT for COVID-19: comparison to RT-PCR. Radiology. 296: E115–E117. PMID: 32073353
  • Wang, W, Xu, Y, Gao, R, et al. 2020. Detection of SARS-CoV-2 in different types of clinical specimens. JAMA323: 1843–1844. PMID: 32159775
  • West, CP, Montori, VM, and Sampathkumar, P. 2020. COVID-19 testing: the threat of false-negative results. Mayo Clin Proc. 95: 1127–1129. PMID: 32376102
  • Zheng, S, Fan, J, Yu, F, et al. 2020. Viral load dynamics and disease severity in patients infected with SARS-CoV-2 in Zhejiang Province, China, January-March 2020: retrospective cohort study. BMJ. PMID: 32317267