Read on Twitter
Of the threads I’ve read exploring the causes of falling rates of #SARSCoV2 infections (and hospitalizations and deaths), I like this one best 1/ https://twitter.com/nataliexdean/status/1362100680492601350
2/ Even if Rt transmission rates spike back up starting today, falling mortality rates are already baked into the data bc of lag time between infection and death. We should get below 1,500 deaths/day in early March (the red line is key & roughly follows the black line) https://twitter.com/_stah/status/1362717910083452932
3/ 1,500 #COVID19 deaths/day is *still* as high as the awful summer surge & average of >40,000 deaths a month. Like losing a small city every month. That’s nothing to celebrate. But still the downward trend is notable...
4/ #epitwitter knows that dissecting cause-and-effect is complex. Not only are most health phenomena multifactorial, but there are different right answers to a question depending on how you ask the question
5/ For instance, I used to work in obesity epi (a field where many armchair experts think they know all the answers). One could ask, Why do Americans have higher BMI than X country? Or why do they have average lower BMI than Y country? ...
6/ Or why did BMI rise so quickly between the specific years of 1985 and 1995? Or why am I skinnier than my sister? All of these are more complex questions than they seem, and they all have *different* answers bc they are all making different “causal contrasts”
7/ We know in #epitwitter that the answer to a causal question is usually different depending on exactly how you frame the question
8/ For instance, there are people on Twitter who seem very upset that #epitwitter is not talking more about seasonality as a cause of declines in infection rates. Of course seasonality matters. We predicted that winter would be awful and it has been...
9/ But the question that’s most intriguing to me is why did rates start falling so suddenly precisely in early January. As @nataliexdean’s thread points out, that’s not generally how seasonality works https://twitter.com/nataliexdean/status/1362100680492601350
10/ My leading theory had focused heavily on the fact that early January was on the back end of the staggering surge of infections all over the country during the traditional holiday season (Midwest hit peak earlier so were already on backside of huge surge during the holidays)
11/ The pattern all over US has been that massive slams of infection in a city or region are often followed by dramatic decreases as people and orgs adjust behavior to lower transmission rates. But often these surges happened at different times in different regions...
12/ ...so a sharp decline in one region doesn’t look that dramatic on a national map because other regions could be increasing in transmission or holding steady even as one area experiencing dramatic drops in transmission and cases
13/ But the national holidays are different (& yes, impact more extreme bc in winter & characterized by long indoor gatherings w people outside household, lots of mixing of networks). So every region was seeing a surge around same time & everyone on downward slope at same time
14/ But recent data journalism by Vox also hit me with another sickening but familiar way of understanding the dramatic change in behavior (I think) early in 2021: a *lot* more White people died of #Covid19 during this recent surge... https://twitter.com/juliaoftoronto/status/1361704937412038659
15/ bc the graph above is just proportions and not raw numbers, it doesn’t convey the huge increase in numbers of hospitalizations and deaths among White people in the late fall/early winter surge.
16/ This image from the @voxdotcom story ( @juliaoftoronto @zhoyoyo) shows how the parts of the country’s hardest hit in the recent surge were much Whiter parts of the country.
17/ In the spring & summer & early fall, I’d often see White people downplaying the outbreak by asking online, “but do you actually know anyone who’s gotten sick from it?” My answer (and answer of so many other Black and Brown folks) was, “YES!! We do!”
18/ But they weren’t asking about our loved ones, who were clearly dying in awful numbers. This @CDCMMWR graph of excess mortality from Jan 2020 through early October 2020 always gets me. Before November, #COVID19 really was a pretty non-White disease https://www.cdc.gov/mmwr/volumes/69/wr/mm6942e2.htm
19/ The Oct-December 2020 surge changed that. Maybe part of the huge drop is as simple as people who didn’t care before (when it was overwhelmingly non-White people dying) care more now that it’s come closer to home for them. https://twitter.com/juliaoftoronto/status/1361704942076121091
20/ Honestly this idea is so depressing. I’m an equity researcher. Part of the power of this work is that focusing on the most marginalized gives us an early warning system. It gives us a chance to say, “This is awful. Let’s stop this before if affects even more people”
21/ We don’t want to eliminate disparities by having other people suffer as much as we have. We want our lives to matter so much that our suffering inspires the action to prevent anyone else from feeling this kind of heartbreak.
22/ This image in the @voxdotcom story hit me the hardest. Yes, #COVID19 primarily kills older people, but many young people have died too. I know the tremendous ache I feel from all the Black people (esp men) in their 30s, 40s, 50s, 60s we’ve lost... https://twitter.com/juliaoftoronto/status/1361704932966031368
23/ the numbers among the Hispanic group (graphic above) appear even worse. I am so sorry for your losses. I mourn with you. This is all so awful.
