Thread by @hjillianh

New preprint 🧵🧵🧵: bit.ly/3zH0IYr
Led by @ SaahithiRao @MattKellyMD and @Duke_Childrens @DukeCHDI colleagues Wen Zhao, Jason Lang, Laine Thomas, and Ben Goldstein
#asthma #pediatrics #SARSCoV2

Remember how in #DukeBRAVEKids we found that kids with asthma seemed to be at lower risk of SARS-CoV-2 infection?
bit.ly/3x4F7ah
We decided to try to run that down using a population-level, retrospective EHR-based analysis from @DukeHealth

We identified every kid in the system (6,515) who met our asthma definition started (bit.ly/2V8DDPj), lived in Duke’s main catchment area, and had at least one encounter in the two years before the pandemic

We then looked at the characteristics of kids with and without asthma, and kids with and without asthma who were tested for SARS-CoV-2 infections

3,594 (7.3%) kids had SARS-CoV-2 tests during the study period (March-October 2020).
Kids were more likely to be tested if they:
1) had asthma
2) identified as Hispanic
3) lived in neighborhoods with a lower average socioeconomic status (neighborhood deprivation index)

Kids who had more prior healthcare interactions also were more likely to be test (they probably regularly get care at Duke)
Non-Hispanic Black kids were less likely to be tested than kids of other races/ethnicities

Here’s where it gets tricky – we wanted to see if children with asthma were less likely to test positive for SARS-CoV-2 than kids without asthma, but it’s comparing apples and oranges. We already know kids with asthma are more likely to get tested in the first place

We needed to match the kids with asthma to a group of kids who had all of the same characteristics that we were able to measure, minus asthma. We generated propensity scores for the probability of having asthma and matched each kid with asthma to a kid without asthma

If you have any real questions about this, I’m going to have to direct you to @SaahithiRao who is an actual statistician, unlike me, who’s main statistical skill is that I remember how to do an ANOVA by hand

In the matched asthma/no asthma cohort, 13% of kids tested positive for SARS-CoV-2:
9.8% of kids with asthma
16% of kids without asthma
Only one child was hospitalized for COVID-19 (who did not have asthma)

Asthma was associated with a DECREASED RISK (!) of SARS-CoV-2 infection [risk ratio (RR): 0.67, 95% confidence interval (CI): 0.49–0.92]

But why?
Here are a couple of ideas we looked into...

Kids with asthma are frequently prescribed inhaled corticosteroids to help control their asthma, and there’ve been a few studies suggesting that these drugs may reduce expression of the SARS-CoV-2 receptors, ACE2 and TMPRSS2

We looked at whether kids with asthma how had inhaled corticosteroid prescriptions were less likely to test positive for SARS-CoV-2 and the risk was even a little lower (RR: 0.60, 95% CI: 0.38–0.94)

Kids with asthma also frequently have atopic disease (sometimes related to their asthma, sometimes other allergies). Atopy is usually driven by a specific type of immune response (Th2), which could potentially alter the effects of SARS-CoV-2.

Kids with atopy also had an even slightly lower risk of testing positive for SARS-CoV-2 (RR: 0.59, 95% CI: 0.39–0.88)!

Our original observation that kids with asthma were less likely to test positive for SARS-CoV-2, even with an infected close contact, may not have been a fluke

There are a couple of potential mechanisms of protection suggested by the even greater reduced risk of testing positive in kids with inhaled corticosteroid prescriptions or atopy

We can’t know for sure that kids with asthma and their families weren’t behaving differently during the pandemic, and it’s likely that if your kid has a disease that affects their lungs, you’re trying to avoid COVID exposures

We can’t account for what wasn’t observed – this study is based on the assumption that if you didn’t get tested for SARS-CoV-2, you never had it. We all know that’s not true, particularly in kids, many of whom will have only mild or no symptoms. #knownunknowns

As cases ramp back up and most kids are still unvaccinated, we really need to think about how to make sure kids expand testing, particularly for populations of kids who have traditionally been underserved.

In good news, kids with asthma at least don’t appear to be more at risk for SARS-CoV-2. It’s also likely very important to keep up with asthma medications and asthma action plans to make sure kids stay healthy.

Many thanks to my brilliant colleagues @saahithirao @MattKellyMD and those not on Twitter – I learned more about study design and statistics from them in the past 6 months than I’ve learned in my life.

Special shoutout to @SaahithiRao who handled a very complicated analysis brilliantly (particularly when it turned out to be much trickier than we originally thought!