By July of 2020, 16.8 million people in the US had #COVID19 -- and didn’t know it. Read our pre-print @medrxivpreprint 
(1/13) https://www.medrxiv.org/content/10.1101/2021.01.27.21250570v1


We called out for volunteers that never had COVID-19 dx & got >400K. Our statisticians devised a code that was able to update daily to select donors that represented the US based off of age, sex, race, ethnicity, location, & send list to @NIAIDNews, @PittTweet, & @UABNews (2/13)
Mailed out @Neoteryx dried blood samplers to >11K folks & got >9K returned. Ran 6 ELISAs on each (full Spike & RBD IgG, IgM & IgA) & evaluated seropositivity by thresholds against pre-pandemic controls (check out @NatureComms paper on our assay) (3/13) https://www.nature.com/articles/s41467-020-20383-x
Utilizing both demographic AND behavioral data from @CDCgov, such as access to healthcare, medical conditions such as cardiovascular disease & diabetes, income, education level, children in house, etc. we weighted results, so our study was truly representative of the US. (4/13)
We found that 4.6% of those that never knew they had #COVID, were actually #seropositive, showing that they had been infected. (5/13)
Places with higher undiagnosed #seroprevalence were correlated with areas that had early severe outbreaks in the spring, whereas those subject to severe fall waves had very low undiagnosed seroprevalence by the summer of 2020. (6/13)
The largest burden of undiagnosed infections was in Black/African Americans (14.2%). Hispanic, younger (18-44 yrs), female, and those that lived in urban areas, all had undiagnosed seroprevalence estimates above the national average. (7/13)
Looking at other factors we saw that people who reported pre-existing health conditions such as #cardiovascular disease, #asthma, pulmonary disease, or #diabetes, had a lower undiagnosed seroprevalence (8/13)
This could be because of a variety of factors, including the tendency of COVID to present more severely in patients with #comorbidities, or, behavioral changes associated with avoiding exposure to COVID19 (9/13)
Furthermore, those who reported having flu/pneumonia #vaccinations also had lower undiagnosed seroprevalence rates, possibly due to increased adherence to #publichealth guidelines such as masking and social distancing. (10/13)
Though those with a reported COVID exposure (a lower # of participants) had higher undiagnosed seroprevalence, there was no correlation with reporting being sick, suggesting an important role for #asymptomatic transmission. (11/13)
So, are we a bit closer to vaccine-induced herd immunity? Will we see a massive uptick in “idiopathic” lung pathologies in the coming years? Time will tell. (12/13)
We thought there were 3 million cases by summer. There were actually 20 million. (13/13)