Today, after 2 long years, antigen tests are finally available to Americans free of charge. (Actually, yesterday the system secretly went live. More than 900,000 got word and flooded the site in a few hours). Each household can order up to 4 tests on
"It’s possible to be contagious yet have a negative rapid test. Four of the 30 people in this study spread COVID19 between negative rapid tests." - How, if at all, would you change your recommendations around using rapid tests before you see elderly or at risk people? I think we'd hoped that a negative rapid test meant that at the very least you weren't (very) contagious and you could have some comfort from that before seeing at risk family/friends. This seems to contradict that and is so disheartening.
Is there any evidence to show that fully vaccinated and boosted (if relevant) are taking longer to test positive on rapid antigen tests even though they end up truly being covid positive? I know of 4 instances now among two different families where individuals continuously tested negative on serial testing but ultimately were positive on days 7 - 12 after exposure. Given how virulent omicron is reported to be and the replication of it in a human system being more rapid, is this just a fluke or do vaccines play a role in trying to shut the incoming virus down but ultimately succumb to its high replication rate? For the record, all illness in these individuals was mild and short-lived.
Given that PCR does not measure virus, and that the preprint reviewed here found that there isn’t a correlation between actual virus and PCR -shouldn’t we stop referring to PCR CT results as ‘viral load’?-it isn’t, it’s merely a proxy that may not hold up well, and the terminology used to date is misleading. Virologists have been very careful about this distinction for 2 years now-the rest of us need to catch up. Now that some are finally looking at actual virus it may turn out PCR testing has not been used appropriately and that we may need to narrow the situations in which it is used.
Why would antigen tests be getting darker late in infection? My son had mild symptoms last Wednesday and tested positive on the morning with all symptoms basically resolved. He bounced off the walls all week. On Sunday I started feeling sick and tested positive Monday. I tested him and saw a faint line. Tuesday would have been his back-to-school day so I tested him again and he still had a faint positive so I kept him home. My toddler also tested faintly positive for the first time on Tuesday. Since then, I've tested my son each morning and his lines are getting darker. Today, Thursday, 8 days post-onset, the line is as dark as it was on day 0. He remains asymptomatic. My daughter and I (I am vaxxed and boosted) have had coughs and runny noses, mild fevers and chills and aches. Could we be contaminating his sample just by being around? Or could our viral load be making him 'sick' again?
I would love some guidance. I’m on day 12 since first symptom; 11 days since positive PCR; 10 days since positive rapid test; 6 days since I last had symptoms!!! I’m still testing positive on rapid antigen test. Is this a personal anomaly or is there something about Omicron? I am still in isolation and will continue until I test negative, but I wonder how infectious I am and if I am, how common is it that people are following guidance to leave isolation after x number of days (be it 5 or 10 or 5 past last symptoms)…..And as a side note, my husband is still negative.
Does this data say anything about the relative likelihood of false positives among different brands of antigen tests? I have taken Quidel Quickvue tests twice, and both times got a positive result which was followed by a negative PCR test, and the second time, also followed by two negative Binax tests. The two Quidel tests that gave me false positives were from different boxes purchased months apart, one before Omicron and one last weekend. I've taken PCR and Binax tests at least a dozen times and never had a positive result, so it's hard not to conclude that the Quidel tests are untrustworthy. Because these false positives ruined my whole weekend, I started looking into the data for different brands, and these two studies do suggest that the PPV is much worse for Quidel than for Binax:
And I know that this isn't scientific at all, but there are way more online reviews complaining of false positives for the Quidel tests than for the Binax tests.
This update on testing (with antigen and PCR) is another very good addition to our understanding of how best to use these tests. If I may be so bold, I would make one suggestion, Dr Jetelina. The use of the term “viral load” in the context of PCR testing could be misunderstood. You come close to a clarification in the discussion of the fifth study reviewed. However, my preference would be to use the term “viral RNA load” in place of “viral load” in the context of PCR testing. As you know, neither antigen tests nor PCR tests detect infectious virion or measure infectious viral titers. Antigen and PCR tests measure PIECES of virus which are NOT infectious. Yes, there may be a relationship, but viral load (interpreted by many people as meaning infectious virus) is NOT the same thing as what PCR tests measure which is viral RNA load.
Thank you very much,
Dave Apgar, PharmD, CAPT(retired) US Public Health Service
"This study tells us that we need to be super careful when using rapid tests in the first few days of exposure or infection. To get the most from your rapid test, wait at least 48 hours after symptoms and 5 days after exposure before taking an antigen test. If you’re negative, test again 24 hours thereafter."
I don't think this is how most people are using rapid tests. From what I hear, people are testing as soon as they have any symptom. They are then acting on those results; negative, they feel free to carry on. And many people are testing as soon as they hear they have been exposed, and, again, with a negative result--the person feels free to act accordingly.
The recommendations to wait 48 hours or 5 days to test are just not part of their thinking, probably because recommendations appear to keep changing; federal and state and local officials provide conflicting recommendations; and employers put their own spin on those recommendations. AND people are tired and confused. I don't see most people being willing to follow the guidelines, if they even understand them or bother to try to figure them out.
What about the other important use for rapid tests: Not just how to manage isolation after a known exposure or symptoms, but as a safety tool for gatherings with people who don't expect they've been exposed at all?
You want to get together with a few people, indoors, for several hours. Perhaps you want to eat together, so you want a way to manage the risk of getting together without masks. None of you have a known exposure or symptoms. So, you each take an antigen test. If anyone tests positive, they don't attend, and you get together with people who all just tested negative within the past hour or two.
What do we know about how reliable that is? How do these studies inform how we can judge the risks of doing that?
I realize that overall prevalence of infection is an important factor, if you can estimate it. So another way of asking this question is: Given some estimate P% of the prevalence in the set of people you're drawing from, and given a number of people N, if each person tests negative (with throat + nasal swab method) and those N people get together very shortly after they all test, what are the odds that at least one of them is infectious over the next few hours?
I've had an epiphany regarding lateral-flow antigen testing over the last several months. For the organization I support, for our summer activities, we required a verifiable negative test (lab- or facility-based and documented) performed within 72 hours of arrival for entry and participation. A bit of research, and a requirement to rely on our participants' integrity allowed us to go to home-testing, with a picture of their negative result. Yes, we still had outbreaks, but we had a reasonable idea idea of our starting baseline. We're now considering a "home" test upon arrival observed by a designee with a repeat test 72 hours later. Also, we are employing these tests at our overnight events when someone presents with symptoms, as a rapid screen to see if we need, right now, to quarantine the cohort of participants they were with.
For another group I'm working with, where the social dynamics are a bit more fluid and not quite as well understood by Americans, I've been working with the local school district's plan, although I'm not sure I agree with it. I've been testing the school age kids who are identified as a classroom, or a close, contact at day 5 post-exposure and then 48-60 hours later with lateral-flow antigen kits (mostly, not exclusively, Abbott BinaxNow). Subjectively I'm pretty confident this approach has been an adequate screening approach. While the school doesn't recommend testing if asymptomatic, and doesn't require testing to return to school after a 5 day quarantine, I still employ it and we do NOT return a kid to school with a positive antigen test even after 5 days.
AS for cheek swabs being less effective than saliva or nasal swabs, recall that not all oral mucosa are created equal, and that omicron has been suggested to prefer a bronchial site rather than an upper-airway, or alveolar site (need to find that citation, sorry). With that in mind, I'd be surprised outperformed nasopharyngeal or posterior oropharyngeal samples.
"the number of viral particles does not equal the number of infectious particles" -- can someone define the difference between these two 'particles'? Fragments vs. intact virus?
"It’s possible to be contagious yet have a negative rapid test. Four of the 30 people in this study spread COVID19 between negative rapid tests." - How, if at all, would you change your recommendations around using rapid tests before you see elderly or at risk people? I think we'd hoped that a negative rapid test meant that at the very least you weren't (very) contagious and you could have some comfort from that before seeing at risk family/friends. This seems to contradict that and is so disheartening.
Is there any evidence to show that fully vaccinated and boosted (if relevant) are taking longer to test positive on rapid antigen tests even though they end up truly being covid positive? I know of 4 instances now among two different families where individuals continuously tested negative on serial testing but ultimately were positive on days 7 - 12 after exposure. Given how virulent omicron is reported to be and the replication of it in a human system being more rapid, is this just a fluke or do vaccines play a role in trying to shut the incoming virus down but ultimately succumb to its high replication rate? For the record, all illness in these individuals was mild and short-lived.
Might also want to swab the NAA swab on tonsils and then nose to increase yield. As they do in UK and at UCSF.
Given that PCR does not measure virus, and that the preprint reviewed here found that there isn’t a correlation between actual virus and PCR -shouldn’t we stop referring to PCR CT results as ‘viral load’?-it isn’t, it’s merely a proxy that may not hold up well, and the terminology used to date is misleading. Virologists have been very careful about this distinction for 2 years now-the rest of us need to catch up. Now that some are finally looking at actual virus it may turn out PCR testing has not been used appropriately and that we may need to narrow the situations in which it is used.
Why would antigen tests be getting darker late in infection? My son had mild symptoms last Wednesday and tested positive on the morning with all symptoms basically resolved. He bounced off the walls all week. On Sunday I started feeling sick and tested positive Monday. I tested him and saw a faint line. Tuesday would have been his back-to-school day so I tested him again and he still had a faint positive so I kept him home. My toddler also tested faintly positive for the first time on Tuesday. Since then, I've tested my son each morning and his lines are getting darker. Today, Thursday, 8 days post-onset, the line is as dark as it was on day 0. He remains asymptomatic. My daughter and I (I am vaxxed and boosted) have had coughs and runny noses, mild fevers and chills and aches. Could we be contaminating his sample just by being around? Or could our viral load be making him 'sick' again?
I would love some guidance. I’m on day 12 since first symptom; 11 days since positive PCR; 10 days since positive rapid test; 6 days since I last had symptoms!!! I’m still testing positive on rapid antigen test. Is this a personal anomaly or is there something about Omicron? I am still in isolation and will continue until I test negative, but I wonder how infectious I am and if I am, how common is it that people are following guidance to leave isolation after x number of days (be it 5 or 10 or 5 past last symptoms)…..And as a side note, my husband is still negative.
btw, the covidtests.gov page just contains the link to the special usps order form. if that helps anyone.
Does this data say anything about the relative likelihood of false positives among different brands of antigen tests? I have taken Quidel Quickvue tests twice, and both times got a positive result which was followed by a negative PCR test, and the second time, also followed by two negative Binax tests. The two Quidel tests that gave me false positives were from different boxes purchased months apart, one before Omicron and one last weekend. I've taken PCR and Binax tests at least a dozen times and never had a positive result, so it's hard not to conclude that the Quidel tests are untrustworthy. Because these false positives ruined my whole weekend, I started looking into the data for different brands, and these two studies do suggest that the PPV is much worse for Quidel than for Binax:
Binax: https://www.cdc.gov/mmwr/volumes/70/wr/mm7003e3.htm
Quidel: https://www.cdc.gov/mmwr/volumes/69/wr/mm695152a3.htm
And I know that this isn't scientific at all, but there are way more online reviews complaining of false positives for the Quidel tests than for the Binax tests.
This update on testing (with antigen and PCR) is another very good addition to our understanding of how best to use these tests. If I may be so bold, I would make one suggestion, Dr Jetelina. The use of the term “viral load” in the context of PCR testing could be misunderstood. You come close to a clarification in the discussion of the fifth study reviewed. However, my preference would be to use the term “viral RNA load” in place of “viral load” in the context of PCR testing. As you know, neither antigen tests nor PCR tests detect infectious virion or measure infectious viral titers. Antigen and PCR tests measure PIECES of virus which are NOT infectious. Yes, there may be a relationship, but viral load (interpreted by many people as meaning infectious virus) is NOT the same thing as what PCR tests measure which is viral RNA load.
Thank you very much,
Dave Apgar, PharmD, CAPT(retired) US Public Health Service
Another remarkable and informative post, thank you Dr Jetelina.
"This study tells us that we need to be super careful when using rapid tests in the first few days of exposure or infection. To get the most from your rapid test, wait at least 48 hours after symptoms and 5 days after exposure before taking an antigen test. If you’re negative, test again 24 hours thereafter."
I don't think this is how most people are using rapid tests. From what I hear, people are testing as soon as they have any symptom. They are then acting on those results; negative, they feel free to carry on. And many people are testing as soon as they hear they have been exposed, and, again, with a negative result--the person feels free to act accordingly.
The recommendations to wait 48 hours or 5 days to test are just not part of their thinking, probably because recommendations appear to keep changing; federal and state and local officials provide conflicting recommendations; and employers put their own spin on those recommendations. AND people are tired and confused. I don't see most people being willing to follow the guidelines, if they even understand them or bother to try to figure them out.
So helpful, and so clearly written and explained. Thank you thank you...
What about the other important use for rapid tests: Not just how to manage isolation after a known exposure or symptoms, but as a safety tool for gatherings with people who don't expect they've been exposed at all?
You want to get together with a few people, indoors, for several hours. Perhaps you want to eat together, so you want a way to manage the risk of getting together without masks. None of you have a known exposure or symptoms. So, you each take an antigen test. If anyone tests positive, they don't attend, and you get together with people who all just tested negative within the past hour or two.
What do we know about how reliable that is? How do these studies inform how we can judge the risks of doing that?
I realize that overall prevalence of infection is an important factor, if you can estimate it. So another way of asking this question is: Given some estimate P% of the prevalence in the set of people you're drawing from, and given a number of people N, if each person tests negative (with throat + nasal swab method) and those N people get together very shortly after they all test, what are the odds that at least one of them is infectious over the next few hours?
I've had an epiphany regarding lateral-flow antigen testing over the last several months. For the organization I support, for our summer activities, we required a verifiable negative test (lab- or facility-based and documented) performed within 72 hours of arrival for entry and participation. A bit of research, and a requirement to rely on our participants' integrity allowed us to go to home-testing, with a picture of their negative result. Yes, we still had outbreaks, but we had a reasonable idea idea of our starting baseline. We're now considering a "home" test upon arrival observed by a designee with a repeat test 72 hours later. Also, we are employing these tests at our overnight events when someone presents with symptoms, as a rapid screen to see if we need, right now, to quarantine the cohort of participants they were with.
For another group I'm working with, where the social dynamics are a bit more fluid and not quite as well understood by Americans, I've been working with the local school district's plan, although I'm not sure I agree with it. I've been testing the school age kids who are identified as a classroom, or a close, contact at day 5 post-exposure and then 48-60 hours later with lateral-flow antigen kits (mostly, not exclusively, Abbott BinaxNow). Subjectively I'm pretty confident this approach has been an adequate screening approach. While the school doesn't recommend testing if asymptomatic, and doesn't require testing to return to school after a 5 day quarantine, I still employ it and we do NOT return a kid to school with a positive antigen test even after 5 days.
AS for cheek swabs being less effective than saliva or nasal swabs, recall that not all oral mucosa are created equal, and that omicron has been suggested to prefer a bronchial site rather than an upper-airway, or alveolar site (need to find that citation, sorry). With that in mind, I'd be surprised outperformed nasopharyngeal or posterior oropharyngeal samples.
"the number of viral particles does not equal the number of infectious particles" -- can someone define the difference between these two 'particles'? Fragments vs. intact virus?
Excellent reviews. Thanks for all of your hard work.