Are we quarantining millions of people unnecessarily?
Article by @apoorva_nyc on pitfalls of #COVID19 PCR for quarantine / public heatlth action / policy.
*Disclaimer* the article is based on my discussions with her.
I want to clarify here:
1/17
nytimes.com/2020/08/29/hea…
Article by @apoorva_nyc on pitfalls of #COVID19 PCR for quarantine / public heatlth action / policy.
*Disclaimer* the article is based on my discussions with her.
I want to clarify here:
1/17
nytimes.com/2020/08/29/hea…
The basic gist is that PCR is an exceedingly powerful useful tool to detect infections. So powerful that it can remain positive for much longer than people are infectious
More in this thread:
2/17
More in this thread:
2/17
So PCR is great for diagnosing sick ppl. But if testing non-sick people, many get their first +ve test weeks after being contagious.
I wrote about it in this thread here:
But There is a simple solution...
3/17
I wrote about it in this thread here:
But There is a simple solution...
3/17
We can use all the information obtained from the #COVID19 PCR test - including the amount of virus detected in a sample.
Unfortunately, we convert a numeric value to just say positive... without distinguishing 10 viruses from 10 billion viruses!
4/17
Unfortunately, we convert a numeric value to just say positive... without distinguishing 10 viruses from 10 billion viruses!
4/17
Failure to use all of this rich quantitative data is a big loss for public health and medicine.
In May we published about how interpreting the Ct value (~viral load in the sample) could greatly benefit clinical decision making and public health
academic.oup.com/cid/advance-ar…
5 / 17
In May we published about how interpreting the Ct value (~viral load in the sample) could greatly benefit clinical decision making and public health
academic.oup.com/cid/advance-ar…
5 / 17
If you're not familiar with the way we use PCR to quantify viral load, it is with something called the Ct value. I describe it (in a simplified way) in this thread:
But now I want to make some clarifications to the article.
6/17
But now I want to make some clarifications to the article.
6/17
I propose that we use the quantitative values to direct public health efforts and to make decisions about whether we need to quarantine or contact trace people or isolate them in the hospital, for instance.
7/17
7/17
If someone is asymptomatic and has low viral load (ex. Ct >33), then likely they have already recovered from infection and are no longer transmitting virus.
These ppl probably do not need to be quarantined... but... (I can hear people saying 'but' now)
8/17
These ppl probably do not need to be quarantined... but... (I can hear people saying 'but' now)
8/17
What if the viral load is low because it's a new infection and they are about to become infectious??
Agreed, this is important!
For public health, if no symptoms and very low viral load, then test again 15-24 hrs after the first sample collection...
9/17
Agreed, this is important!
For public health, if no symptoms and very low viral load, then test again 15-24 hrs after the first sample collection...
9/17
If viral load is low because person is at the start of infection, then in 24hrs the viral load will be much higher.
Note, The time when this is happening is a small window of time, relative to the positive duration, so this is relatively unlikely w low frequency testing
10/17
Note, The time when this is happening is a small window of time, relative to the positive duration, so this is relatively unlikely w low frequency testing
10/17
Most of the time, a low viral load will remain low the next day, so if someone has a Ct value >33 two days in a row - they probably do not need quarantining and if the exposure is unknown, then likely not fruitful to contact trace either.
11/17
11/17
But unfortunately, we don't use these (Ct) numbers today for medicine or public health.
We just quarantine everyone with a first +ve result for 10 days, falsely assuming we happened to get a swab at the beginning of their infection - this is perhaps irresponsible.
12/17
We just quarantine everyone with a first +ve result for 10 days, falsely assuming we happened to get a swab at the beginning of their infection - this is perhaps irresponsible.
12/17
Instead it turns out that most people who are found positive via asymptomatic testing are long past their infectious stage - perhaps over 70%!
So perhaps 70% of asymptomatics who get quarantined don't need to be!
13/17
So perhaps 70% of asymptomatics who get quarantined don't need to be!
13/17
If we want a chance at finding ppl b4 they infect others, we need very frequent testing - one reason I and others have been calling for new approaches to testing, using cheap, rapid tests for infectiousness.
theatlantic.com/health/archive…
14/17
theatlantic.com/health/archive…
14/17
So, I hope that this @nytimes article by @apoorva_nyc (top of this whole thread) is seen as a call to action to start using all of the information (the quantitative values) available from the test to help make important decisions about how to act on a positive result...
15/17
15/17
Interpreting the PCR viral load value can help optimize contact tracing efforts to follow leads most likely to be fruitful, will prevent unnecessary quarantining of millions after their their infectious period and will help save huge resources in hospital settings
16/17
16/17
Finally, to be clear, I do NOT want to change the PCR threshold to call someone with low viral load negative. This was unclear in the article. I simply want to use all the information in the most informative way to guide downstream actions.
17/17
17/17
