This study adds some new information on the spread of Covid in schools.
Let's walk through this and see what we can learn!
Thread!
Let's walk through this and see what we can learn!
Thread!
https://twitter.com/DrDarria/status/1355185892240060416
The study follows the experience of two K-12 independent schools through the fall of 2020.
School A (Southern US): 2299 students & staff
School B (Mid-Atlantic): 1200 students & staff
The schools had similar plans re: opening, closing, mitigation
School A (Southern US): 2299 students & staff
School B (Mid-Atlantic): 1200 students & staff
The schools had similar plans re: opening, closing, mitigation
Here's where this study really stands out:
Universal Testing.
Regular testing of students & staff regardless of symptoms!
Testing was not as frequent as would be ideal, especially in the beginning. BUT this helps us glimpse how many infections are missed w/o this testing.
Universal Testing.
Regular testing of students & staff regardless of symptoms!
Testing was not as frequent as would be ideal, especially in the beginning. BUT this helps us glimpse how many infections are missed w/o this testing.
The study also included voluntarily reported tests out of school as cases & separately tracked these as well as those that were done as a result of possible exposure & during quarantine.
Positive cases were traced to try to determine source & exposures. No info on procedures.
Positive cases were traced to try to determine source & exposures. No info on procedures.
Prior to the school opening in the summer and following Thanksgiving vacation, everyone was tested.
This is helpful information (and a good safety protocol!) because it helps us see transmissions that are more conclusively linked to out-of-school exposures.
This is helpful information (and a good safety protocol!) because it helps us see transmissions that are more conclusively linked to out-of-school exposures.
Positive cases were tracked by how they were identified (eg, universal testing vs voluntary report), whether the person had symptoms, likely source of exposure, & student vs staff.
Using this data they could estimate the proportion of asymptomatic cases.
Using this data they could estimate the proportion of asymptomatic cases.
Clusters were defined as 2 or more linked cases in school.
Estimates of the rate of transmission, number of introductions, etc were calculated.
Estimates of the rate of transmission, number of introductions, etc were calculated.
Who gets included in those calculations?
+ Tests before school started or self-reported + test after the last round of testing, those tied to thanksgiving or out of school, more than a week after testing… 👇
Oh, and you had to have tracking data.
Remember that part.
+ Tests before school started or self-reported + test after the last round of testing, those tied to thanksgiving or out of school, more than a week after testing… 👇
Oh, and you had to have tracking data.
Remember that part.
RESULTS!
School A: 109 Covid+, apx 5% of students, faculty, & staff
54% of cases were identified through universal testing & another 20% based on contact tracing of these cases
27% were self reported
School A: 109 Covid+, apx 5% of students, faculty, & staff
54% of cases were identified through universal testing & another 20% based on contact tracing of these cases
27% were self reported
School B: 25 Covid+, apx 2% students, faculty, & staff
84% of cases were identified through universal testing & another 4% from contact tracing these cases
12% were self reported
84% of cases were identified through universal testing & another 4% from contact tracing these cases
12% were self reported
Both schools:
Not surprisingly, cases peaked after school breaks.
Good news: no cases were hospitalized
Interesting news: apx 50% of adults & 3% of lower school students had symptoms at testing.
Community incidence, but not test positivity, correlated with school infections.
Not surprisingly, cases peaked after school breaks.
Good news: no cases were hospitalized
Interesting news: apx 50% of adults & 3% of lower school students had symptoms at testing.
Community incidence, but not test positivity, correlated with school infections.
About 9% of cases introduced into schools resulted in transmission.
6 outbreaks in School A w a min of 3 secondary cases.
1 outbreak in School B w 1 secondary case.
No evidence of transmission between students & teachers in either school
6 outbreaks in School A w a min of 3 secondary cases.
1 outbreak in School B w 1 secondary case.
No evidence of transmission between students & teachers in either school
In School A: 72% of transmissions were associated with not following the school mask wearing rules.
HOWEVER:
The source of *the majority of infections* could not be identified.
HOWEVER:
The source of *the majority of infections* could not be identified.
Authors Key Findings:
- Kids can get & spread Covid
- But they didn't seem to lead to large chains of infection
- With mitigation measures, transmission was lower than has been seen for other respiratory viruses when mitigation measures were not in place.
*Mitigation is key*
- Kids can get & spread Covid
- But they didn't seem to lead to large chains of infection
- With mitigation measures, transmission was lower than has been seen for other respiratory viruses when mitigation measures were not in place.
*Mitigation is key*
Author Key Findings Cont:
- Community incidence is a better predictor of cases in schools than community test+ %
- Transmission appears to have occurred mostly in clusters
- Clear evidence of out-of-school transmission during breaks & lower incidence w continuous school.
- Community incidence is a better predictor of cases in schools than community test+ %
- Transmission appears to have occurred mostly in clusters
- Clear evidence of out-of-school transmission during breaks & lower incidence w continuous school.
Author Key Findings Cont 2:
Testing was key not only for research but for preventing further spread, identifying risky behaviors, and monitoring adherence & performance of mitigation strategies.
It also provides clear evidence of safety for families & teachers.
Testing was key not only for research but for preventing further spread, identifying risky behaviors, and monitoring adherence & performance of mitigation strategies.
It also provides clear evidence of safety for families & teachers.
Temperature screening was not helpful (no surprise there)
Outbreak investigations in schools with assessment of secondary household transmission are also needed.
Pooled rapid saliva testing allowed >freq of testing
Limitation: Only 2 schools with MASSIVE RESOURCE INVESTMENT
Outbreak investigations in schools with assessment of secondary household transmission are also needed.
Pooled rapid saliva testing allowed >freq of testing
Limitation: Only 2 schools with MASSIVE RESOURCE INVESTMENT
CONCLUSION:
With massive investments in mitigation, including regular universal testing & contact tracing, with significant behavioral change, schools can open in person reasonably safely.
Important:
THIS DOES NOT MEAN ALL SCHOOLS CAN OPEN SAFELY.
With massive investments in mitigation, including regular universal testing & contact tracing, with significant behavioral change, schools can open in person reasonably safely.
Important:
THIS DOES NOT MEAN ALL SCHOOLS CAN OPEN SAFELY.
Now onto my thoughts:
This is some of the best information we have on school safety with mitigation measures because of the universal testing and contact tracing. 🥳
And as with any study it's not without flaws.
Let's evaluate those & see how our interpretation changes.
This is some of the best information we have on school safety with mitigation measures because of the universal testing and contact tracing. 🥳
And as with any study it's not without flaws.
Let's evaluate those & see how our interpretation changes.
Thought 1:
We know that testing wasn't occurring daily or even weekly, at least at the beginning.
Given that most cases were identified by this testing some almost certainly fell through the cracks.
Hard to know how many & how this might change results.
We know that testing wasn't occurring daily or even weekly, at least at the beginning.
Given that most cases were identified by this testing some almost certainly fell through the cracks.
Hard to know how many & how this might change results.
Thought 2:
Some of the + tests were excluded.
This makes perfect sense for the cases identified before schooling started, bc they were kept home.
But numbers for others would be helpful to better understand the context.
Some of the + tests were excluded.
This makes perfect sense for the cases identified before schooling started, bc they were kept home.
But numbers for others would be helpful to better understand the context.
Thought 2.5:
The methods say those without tracing information were excluded but it's not clear what this means.
From the whole study?
From analyses of sources / school transmission?
Does this include mean no tracing done or no source found?
The methods say those without tracing information were excluded but it's not clear what this means.
From the whole study?
From analyses of sources / school transmission?
Does this include mean no tracing done or no source found?
https://twitter.com/Epi_D_Nique/status/1355236481275748357?s=20
Thought 3:
Even with the big improvement in methodology, we are still seeing multiple biases that go in the direction of *underestimating* cases.
In addition, HALF of cases didn't have an identified source - why is in school apparently ruled out then?
Even with the big improvement in methodology, we are still seeing multiple biases that go in the direction of *underestimating* cases.
In addition, HALF of cases didn't have an identified source - why is in school apparently ruled out then?
https://twitter.com/Epi_D_Nique/status/1355242157737717766?s=20
Thought 4:
As @OcaptMyObvious notes, the very fact of testing changes the outcome!
While this shows an important value of testing (it works!) it also means that the results of this study CANNOT be considered to represent what happens in other schools.
As @OcaptMyObvious notes, the very fact of testing changes the outcome!
While this shows an important value of testing (it works!) it also means that the results of this study CANNOT be considered to represent what happens in other schools.
https://twitter.com/OcaptMyObvious/status/1355246138664824832?s=20
To sum up my thoughts for now:
Although infections are likely underestimated, the data show that with very strong mitigation (see paper for details), universal testing & tracing, and adaptation of mitigation plans in response to testing schools can achieve low transmission.
Although infections are likely underestimated, the data show that with very strong mitigation (see paper for details), universal testing & tracing, and adaptation of mitigation plans in response to testing schools can achieve low transmission.
BUT
Do our public schools have the resources to implement this? If not now, when?
Can public schools get community buy in from teachers, staff, students, & families to adhere to these strict strategies? And if not, how should schools respond?
Do our public schools have the resources to implement this? If not now, when?
Can public schools get community buy in from teachers, staff, students, & families to adhere to these strict strategies? And if not, how should schools respond?
Finally, the universal testing in this study suggests that MOST infections in schools where there is not universal testing are not being caught!*
And even this is an underestimate (see above).
Prior studies w/o testing likely caught a very small fraction of actual cases.
And even this is an underestimate (see above).
Prior studies w/o testing likely caught a very small fraction of actual cases.
*Caveat here: At least 50% of the cases in adults were symptomatic at testing so might have been identified without testing.
But also, why were they coming to school w symptoms??
For lower grades, it's likely very few cases are identified without testing.
But also, why were they coming to school w symptoms??
For lower grades, it's likely very few cases are identified without testing.
Okay, seriously now…
FIN
FIN
JK.
Since the data collection only went through December this probably does not include the B117 variant which is likely more transmissible than what was spreading during the study.
Again, this points to the importance of frequent universal testing!
Since the data collection only went through December this probably does not include the B117 variant which is likely more transmissible than what was spreading during the study.
Again, this points to the importance of frequent universal testing!
https://twitter.com/ccleighton/status/1355252356317913089?s=20
This is really important context for how this study fits into the question of public school openings:
Mitigation measures at many schools are very poor and there is almost certainly transmission happening in these schools.
amp.usatoday.com/amp/4257689001…
Mitigation measures at many schools are very poor and there is almost certainly transmission happening in these schools.
amp.usatoday.com/amp/4257689001…
The rosy picture painted by the best resourced schools is not reflective of the picture across the US.
This is genuinely frightening.👇
We need to be clear on the difference between school safety under best practices vs safety under usual practices & those in state guidelines.
This is genuinely frightening.👇
We need to be clear on the difference between school safety under best practices vs safety under usual practices & those in state guidelines.
Yet another point is how contact tracing determines who a likely "close contact" is and what that might mean for the detection of in-school transmission.
Thankfully, the brilliant @Theresa_Chapple has a thread for that!
Thankfully, the brilliant @Theresa_Chapple has a thread for that!
https://twitter.com/Theresa_Chapple/status/1355193117348130818?s=20
And going back to this idea of what is happening in the best resourced & motivated places vs everywhere else, sometimes the criteria used in contact tracing is designed to underestimate infections (intentionally or not).
Btw, lunch is prob riskiest time.
Btw, lunch is prob riskiest time.
https://twitter.com/jistBhappy/status/1355202920355139590?s=20
• • •
Missing some Tweet in this thread? You can try to
force a refresh
