COVID: quarantine, along with testing strategies

 a recent mathematical modeling study suggested tweaking the current quarantine recommendations in the hopes of decreasing quarantine times by adding testing strategies, especially important with our current huge surge and more indoor exposure during the upcoming holidays and prolonged winter,  (see covid quarantine testing strategies medrxiv2020 in dropbox, or doi.org/10.1101/2020.10.27.20211631) 

 

Details: 

-- the recommendation in general is to self-quarantine for 14 days after an exposure to someone who has Covid-19, per the CDC 

-- the goal of this current mathematical modeling study is to blend the time of self-quarantine along with SARS-CoV-2 testing, to achieve optimal risk reduction for post quarantine transmission (PQT) of the virus 

-- the current modeling assumes a one-day delay for testing results 

-- they also assume that PCR specificity is 100%

-- these models were based on clinical studies over time, finding for example: 

    -- false positive PCR results become less likely 5 to 6 days after contact (that is the time when true positives are most likely)

    -- testing too early raises a higher risk of a false negative result 

 

    -- For the mathematical modeling, they relied heavily on testing protocols used in the setting of offshore oil and gas platforms, where there is a guaranteed isolation during periods of quarantine (a captive audience). 

    -- of 4040 SARS-CoV-2 PCR tests, a negative exit test 96 hours after the start of quarantine for an offshore oil rig population identified all known asymptomatic cases that had previously tested negative at entry. they project that this scenario would have prevented an expected 9 offshore transmission events if there had been no exit testing.  no offshore worker having negative tests at entry and exit from quarantine was later diagnosed with a Covid-19 infection during their offshore work 

    -- comparing a 3-day quarantine and testing only on entry: 

        -- extending the current quarantine duration to 7 days and adding testing on exit (96 hours after the 1^st test) reduced the probability of PQT by 98% 

        -- doing the same for 5 days reduced PQT by 93%

-- the goal of these strategies is to reduce the effective reproduction number R₀ (the average number of secondary infections caused by a primary case) to <1, using what is known about the infectivity of Covid-infected people, the temporal diagnostic sensitivity of PCR test, and an incubation period of 8.29 days 

 

Results: 

-- if an individual self-quarantines immediately at symptom onset without testing, the probability of PQT is <0.25 if people quarantine for 3 days, and this falls to <0.05 for quarantines of 8 days or longer  

-- they found the following were efficient approaches: 

    -- a 13-day quarantine with Covid testing on entry 

    -- a 7-day quarantine with testing on exit (i.e., the test done one day prior to the end of quarantine, given the 24-hour turnaround for test results) 

    -- a 7-day quarantine with testing on both entry and exit from quarantine

-- all of these provide at least as much benefit as a 14-day quarantine without testing, though testing at both entry and exit decreased the probability of PQT the most 

   -- for example, a quarantine as short as 3 days with the test of both entry and exit had a 64.5% reduction in the probability of PQT relative to no testing, vs a 22.8% decrease with testing only at entry and a 61.4% decrease for testing only at exit

-- single testing on exit was more effective in reducing the probability of transmission than testing upon entry, for all of the testing strategies 

-- they calculated that the largest reduction in the probability of PQT when conducting a single test is: 

    -- for quarantine less than 7 days, test on the last day 

    -- for a 7-day quarantine, test on day 6 

    -- for quarantines between 8 and 13 days, test on day 5 

    -- for quarantines of 14 days or longer, test on day 6 

-- they also assessed the effect on their model by disease prevalence in the community, finding that with a disease prevalence of 1%, a cohort size of 40, and a 5-day quarantine: 

    -- entry testing only: probability of PQT is 0.06 

    -- entry and exit testing: PQT probability .005 

    -- and, as the disease prevalence increases in the community, this would justify increased testing frequency, and that a series of tests would decrease the likelihood of false negatives

--their graph of PQT probability vs duration of quarantine, with the effect of different testing strategies (this graph below probably optimizes the different strategies, since those identified through contact tracing will have more exact time of exposure to the virus and less ability to have spread the virus prior to symptoms).  they had a similar graph for those just entering quarantine (ie not by contact tracing), though somewhat less impressive

The probability of post-quarantine transmission for no testing and three testing strategies and durations of quarantine of 1–14 days, with an incubation period of 8.29 days, 30.8% asymptomatic infections and perfect self-isolation of symptomatic infections. Curves for the probability of post-quarantine transmission show the probability of one or more post-quarantine infections when an infected individual enters quarantine through contact tracing within the incubation period without testing (red), and with testing upon entry to quarantine (orange), on exit from quarantine (blue), and on both entry to and exit from quarantine (purple). With a delay in sample collection to results, we assumed that testing on exit occurred the day before the end of quarantine

 

Commentary: 

-- given the huge economic, social, political, personal, emotional, psychological, etc. impact of prolonged quarantining because of the Covid-19 pandemic, it certainly makes sense to determine the minimum time needed for quarantine. Of course, every strategy is not perfect, and the goal is to reduce risk to an acceptable level (which will be different for different people). 

-- this modeling study, based on their assumptions and data, does suggest that a strategy integrating testing, and especially testing prior to stopping quarantine, would effectively allow for curtailed quarantine for many people

-- And, as a reminder, we know that over half of Covid-19 transmission cases occur from people who are presymptomatic or asymptomatic, so screening for symptoms is inadequate to prevent the continuation of an outbreak 

Limitiations:

--this model is based on PCR, and results would differ for other tests. and shifting to saliva tests might allow for increased numbers of tests to be done, which would affect the above model

--PCR can be positive for a lot longer than they modeled (up to 3 months has been reported, in a person who recovered from clinical Covid-19 infection), and may reflect dead virus or viral remnants (ie, PCR positivity is not necessarily specific for live and transmissible virus). though a negative test at exit would certainly be reassuring, and this would shorten the quarantine for a lot of people

    --in particular, they assume that PCR specificity is 100%. the issue is specificity for what? it seems true that the specificity for detection of SARS-CoV-2 is >95%. But the specificity for viable transmissible virus is not clear, and undoubtedly less (eg, see covid pcr specificity lancet2020 in dropbox or doi.org/10.1016/S2213-2600(20)30453-7)

so, PCR testing, added to quarantine, can effectively shorten the time of quarantine (and all of its attendant issues, such as getting back to important family and work responsibilities, and avoiding the rather profound psych issues related to prolonged quarantine). 

--indeed, a welcome change would be cutting the quarantine in half (only 7 days), if the SARS-CoV-2 were negative at the start and end of the quarantine.

--but it would really be useful to know if positive test results at the end of the quarantine really meant transmissible virus (and the best test we have for viral infectivity, as far as i know, would be viral culture, an in vitro test). and it would be great to do viral cultures on asymptomatic and symptomatic people, at different times after exposure to SARS-CoV-2. the best group of subjects probably would be those identified through contact tracing, where there is a discrete exposure at t₀. this data would really help make the modeling of the amount of quarantine time needed more accurate and applicable

geoff

 

to get access to all of the blogs:

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list