# What does testing positive for coronavirus actually mean? aka The numbers are almost meaningless **URL:** \ **Category:** The Quarantine Zone\ **Tags:** covid-quarantine\ **Created:** [April 9, 2020, 6:06am UTC](https://boards.straightdope.com/t/what-does-testing-positive-for-coronavirus-actually-mean-aka-the-numbers-are-almost-meaningless/851406 "2020-04-09T06:06:36Z")\ **Posts on this page:** 7\ **Page:** 1
**Author:** ![Nippless](https://avatars.discourse-cdn.com/v4/letter/n/df788c/32.png) [@Nippless](https://boards.straightdope.com/u/Nippless)\ **Post date:** [April 9, 2020, 6:06am UTC](https://boards.straightdope.com/t/what-does-testing-positive-for-coronavirus-actually-mean-aka-the-numbers-are-almost-meaningless/851406/1 "2020-04-09T06:06:36Z")
I’ve “heard” the test is testing for the presence of a coronavirus. ANY coronavirus. CVs are not so uncommon. You could’ve tested positive a year ago. It also explains why covid has such a wide range of symptoms, from asymptomatic, to mild flu, to on ventilator. We are not specifically testing sars-corvid-19. The only way to properly test is by antibodies, and this is not being done anywhere. Is this correct? ---
**Author:** ![UDS](https://avatars.discourse-cdn.com/v4/letter/u/ecae2f/32.png) [@UDS](https://boards.straightdope.com/u/UDS)\ **Post date:** [April 9, 2020, 7:17am UTC](https://boards.straightdope.com/t/what-does-testing-positive-for-coronavirus-actually-mean-aka-the-numbers-are-almost-meaningless/851406/2 "2020-04-09T07:17:36Z")
Well, it doesn’t _sound_ right. Where did you “hear” it? The SARS-CoV-2 genome was first sequenced on 12 January. The first viral testing protocol was published on 17 January, based on that sequence. I’d be astonished if the test wasn’t specific to the virus. If it’s just a general CV test, it would be all but useless, surely? Which would raise two questions; first, why would countries invest scarce medical resources in testing? Secondly, why would the countries that have been most effective in containing the pandemic all achieved this result with programmes that put very high levels of testing front and centre? If they were prioritising a useless strategy, you’d expect them to have lousier outcomes, not better outcomes. So, at first glance, what you have heard doesn’t pass the sniff test; it simply doesn’t stack up. That’s not to say that there aren’t answers to the questions I pose but, as of know, I’m pretty sceptical. ---
**Author:** ![DSeid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/dseid/32/20194_2.png) [@DSeid](https://boards.straightdope.com/u/DSeid)\ **Post date:** [April 9, 2020, 9:44am UTC](https://boards.straightdope.com/t/what-does-testing-positive-for-coronavirus-actually-mean-aka-the-numbers-are-almost-meaningless/851406/3 "2020-04-09T09:44:54Z")
Incorrect information, although of course false positive and false negative test results are an issue, as they are with all tests. The probes used are chosen with both for sensitivity and specificity to the SARS-CoV-2 as concerns. One of the issues with the CDC probe in its initial roll out was that they added a third probe to the two that were being used elsewhere to try to improve on the results and it failed miserably. Where what you heard may have originated is that the probe the CDC [tried to add](https://cen.acs.org/analytical-chemistry/diagnostics/Faulty-probes-blame-CDC-coronavirus/98/i9) IS nonspecific and lights up to other coronaviruses as well. Its intent was to be a “beacon” - > [@](#): > > Kelly Wroblewski, a microbiologist who leads infectious disease programs at the Association of Public Health Laboratories, says the reagents that didn’t work properly are nucleic acid probes designed to detect RNA sequences of all coronaviruses that infect humans, including milder strains that are endemic to the US population. These probes serve as a beacon, she says, that a coronavirus is present and that the sample should be tested further to determine whether that coronavirus is SARS-CoV-2 or another. It’s not clear at what step the probes failed or if they failed, for example, because of reaction conditions. Wroblewski also told Science that there were problems with signals appearing in a negative control experiment in the kit. > > The kit contains two other sets of probes that are specific to SARS-CoV-2 and are used to determine whether a patient is infected with the novel coronavirus. CDC officials have said that with a revised protocol, the older kits can be used with just those two sets of probes. New kits, which presumably contain a working third set of probes, are being made. CDC officials did not respond to requests to clarify the problems with the older kits. > > “It’s not necessary. It’s a nice to have, not a need to have,” Wroblewski says of the ability of the kits to detect other coronaviruses. “It doesn’t really impact the actual sensitivity of the assay when you are looking for novel coronavirus.” ---
**Author:** ![RivkahChaya](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/rivkahchaya/32/8919_2.png) [@RivkahChaya](https://boards.straightdope.com/u/RivkahChaya)\ **Post date:** [April 9, 2020, 10:47am UTC](https://boards.straightdope.com/t/what-does-testing-positive-for-coronavirus-actually-mean-aka-the-numbers-are-almost-meaningless/851406/4 "2020-04-09T10:47:48Z")
Thanks, DSeid. I was pretty sure the OP sounded wrong, because my first thought was “No one would ever test negative.” I’m glad to see a doctor with real information. You should start an “Ask the doctor” thread in “Quarantine,” for people with something they recently heard around, or read in a dubious source; instead of starting a new thread with the dubious info as the thread title, they could run it by you first. I know that’s a lot of work for you, but you end up debugging most of the “Hey, guess what I heard!” threads anyway. ---
**Author:** ![DSeid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/dseid/32/20194_2.png) [@DSeid](https://boards.straightdope.com/u/DSeid)\ **Post date:** [April 9, 2020, 3:19pm UTC](https://boards.straightdope.com/t/what-does-testing-positive-for-coronavirus-actually-mean-aka-the-numbers-are-almost-meaningless/851406/5 "2020-04-09T15:19:43Z")
I appreciate the kind thought but there are many people on this board debugging/fact-checking things in a crowdsourced way and physicians who are actually in the trenches (as a general pediatrician my line has been that if I’m in the trenches my trench is dug a mile back from the lines). I am qualified to offer the contributions I try to make, but not qualified for that role. ---
**Author:** ![Stranger\_On\_A\_Train](https://avatars.discourse-cdn.com/v4/letter/s/13edae/32.png) [@Stranger\_On\_A\_Train](https://boards.straightdope.com/u/Stranger_On_A_Train)\ **Post date:** [April 9, 2020, 3:34pm UTC](https://boards.straightdope.com/t/what-does-testing-positive-for-coronavirus-actually-mean-aka-the-numbers-are-almost-meaningless/851406/6 "2020-04-09T15:34:44Z")
To expand upon that, the SARS-CoV-2 virus which is responsible for COVID-19 disease is part of a group of [_Betacoronaviruses_](https://www.cdc.gov/coronavirus/types.html). Within the genera there are four different lineages of which SARC-CoV-2 (subgenus _Sarbecovirus_) of which the only known viruses that infect humans are the SARS-CoV(-1) (responsible for the 2003 SARS outbreak). The antigen tests, which us real time reverse transcription polymerase chain reaction (RT-PCT) look for RNA segments and nucleoproteins (proteins associated with the genome) known to be specific to this lineage, so it should not test positive in response to other coronaviruses that do not contain these specific fragments. Different countries have developed different test protocols looking for different gene segments so it is unsurprising that they may have varying reliability, and as noted, the CDC test was designed look for more general targets (three different segments in the ‘N’ gene of the virus) presumably under the thesis that there might be different variants of the gene and they wanted to be able to cast as wide a net as possible. There are a variety of reasons why both false positives and false negatives could occur in testing, including contamination of the samples, poor quality in the reagents used, poor sample technique (not getting a sample deep enough in the respiratory tract), and variations in viral load during the infectious phase (there is evidence that the virus can initially colonize in the upper respiratory tract, then seemingly disappear only to reappear in the lungs, so a symptomatic patient may still appear to be negative for infection), in addition to errors in the replication process. CT scans are used to verify characteristic signs of the COVID-19 illness (damage in the lung tissue), and since treatment for severe conditions for such signs is essentially to support the patient with supplemental oxygen and positive pressure inflation of the lungs regardless of cause, a positive test is mostly a diagnostic and tracking tool rather than one that informs care, e.g. there is not a need to make repeated tests to gauge the amount of virus in the patient’s system for treatment. The antigen test cannot test for someone who has had the infection and eliminated the virus, so a serology test which looks for the protein-specific immunoglobulins. Since there is no validated antibody test yet, the reliability of tests are unknown, although there are over thirty different efforts around the world to develop accurate and rapid tests so it is likely someone will develop a highly reliable antibody test which can be rapidly deployed, hopefully a pin-prick test that could be done by non-medical authorities or in the home. The antibody test is not useful for early detection because it requires the body to develop a significant immune response (though there is some evidence that patients may start producing antibodies very early in the symptomatic phase) but is vital for quantifying how much the virus has spread through the population and particularly in how many asymptomatic but now immune people there are so that both the assurance of a degree of herd immunity can be assessed and the likelihood of the then-endemic virus popping back up in another outbreak. Since most viral epidemics are followed by another outbreak ~6 months after the peak of the first, that is vital to limiting further infections among at-risk people. The problem with the current testing isn’t the accuracy of the testing, per se–with enough testing we could estimate the incidence of both false positives and false negatives, and adjust the overall predictions accordingly–but that there simply isn’t enough testing anywhere to get high confidence in sample estimates. This is almost always a problem in early epidemics where the roll-out of testing necessarily follows the spread of the epidemic but it has been particularly problematic in this case both because of how rapidly the disease is transmitted and the apparent degree of asymptomatic carriers in the population. Stranger ---
**Author:** ![psychobunny](https://avatars.discourse-cdn.com/v4/letter/p/54ee81/32.png) [@psychobunny](https://boards.straightdope.com/u/psychobunny)\ **Post date:** [April 9, 2020, 4:19pm UTC](https://boards.straightdope.com/t/what-does-testing-positive-for-coronavirus-actually-mean-aka-the-numbers-are-almost-meaningless/851406/7 "2020-04-09T16:19:00Z")
Just to clarify: The current test (nasal or oral swab) tests for the actual presence of the virus. It is designed to be specific for this corona virus. What the initial (failed) tests tried to do was to initially screen for any corona virus and if it was positive to then screen further for this specific one. Since it tests for the actual presence of the virus, it will not come up positive unless someone is currently infected. However, they had control problems with the initial more broad-based screening where even the controls were sometimes coming up positive. The current testing eliminates that step and goes straight to testing for the SARS-COV-2. If you have had the disease and gotten over it, this test will be negative which is why they are using two negative tests as criteria for resolution. Antibody tests show whether you have developed an immunity to the disease. They are currently being developed. Generally it can take 2-6 weeks to develop antibodies but this can overlap with the acute infection (there are early and late antibodies which vary depending on the course of the disease). Once we have the antibody tests, we can test known patients to tell at what point the antibodies develop in the course of the disease. Depending on these results, we should be able to have a blood test that can show whether you have been infected at any point and by using the correct set of early and late antibodies, should be able to tell if you are fully recovered and presumable immune, at least temporarily. Depending on how early the antibodies develop, you may be able to diagnose an acute infection with an antibody test but the current test is still the gold standard.