1/ Maybe it’s my #epitwitter bias, but I get frustrated by the outsized attention these hospital-based viral load comparison studies (kids vs adults) get. And interpreted as children as a group highly infectious... https://twitter.com/nicolamlow/status/1296550523798257665

2/ the epidemiologic evidence of actual transmission in real-life settings (household studies, popn-representative seroprevalence studies, overall experiences of daycares and elementary schools globally) has indicated than young children lower risk than adults re: spread...

3/ the viral load studies seem much less relevant evidence...
* viral load is not live virus nor infectiousnes. For instance, viral load often stays high for a while in #COVID19 patients who are no longer infectious. Viral load is necessary but not sufficient for infectiousness

4/ *the samples viral load studies often select & highly susceptible to bias. For instance, relatively few children are hospitalized w #SARSCoV2 infection. Re: viral load of hospitalized kids, u r choosing tiny proportion of kids with worst #SARSCoV2. Maybe useful clinically...

5/ But questions of school and community transmission are fundamentally different questions bc the people who drive #SARSCoV2 transmission are not the sickest people, not people in the hospital. Usually an adult w relatively mild symptoms at a family party, bar, in workplace...

6/ ..non-regulated social environments (or wrt some workplaces, where employers restrict safety procedures that interfere w usual workflow). This is part of the reason why sleep away camps & universities dangerous...

7/ @UNC didn’t close bc the classroom environments were unmanageably high-transmission. It was all the other elements of residential college life (sleeping, eating, and socializing together).

8/ #SARSCoV2 is tricky in other ways too. Its “dispersion factor” is high. A relatively small proportion of cases seed outbreaks (probably during a narrow window of high infectiousness in period before they’d be in a hospital)... https://www.smithsonianmag.com/science-nature/how-superspreading-events-cause-flare-ups-covid-19-180975588/

9/ this is fundamental, key information for predicting which settings will be riskiest. It’s info you can’t get from a viral load study. It’s info you have to get from epi studies that actually examine transmission in realistic settings

10/ So if i want to understand transmission and I see the bulk of epi evidence pointing one way and I see a bunch of viral load studies (not live virus) of select poorly characterized folks from hospital settings pointing another way, I’m going with the epi studies

11/ not to say that the epi studies of #SARSCoV2 are perfect. They are deeply limited, especially in US, where our testing is still symptom & exposure driven — and slow - which means we mostly miss the ability to track asymptomatic and presymptomatic spread btw kids

12/ it’s a huge problem. But triangulating among the global studies and the observed experiences of US childcare centers vs other settings offers important insights.

13/ There are groups who are fielding great school-based epidemiologic studies as we speak. In what is fundamentally an epidemiologic question about disease transmission, these will be the important sources of evidence

14/ note: the particular study linked in tweet 1 did not restrict to hospitalized, also drew from urgent care clinics & most of the “children” (up to age 22 yrs) were not highly symptomatic. I remain skeptical if interpretations in lay press but did want to clarify this