COVID-19 Daily Update, Public Health Compliance Data:
Over the last week we have been talking about the mystery of second wave resistance across the US. After about 25% of the population is infected and recovered, there is remarkable resistance to a resurgence of the virus, even after states "reopen".
This is mysterious because immunity thresholds should be around 66% if the R0 of SARS-CoV-2 is really around 3 as we believe.
One likely explanation for this is that personal behavior is providing the additional protection. In the past I have used Google mobility data to characterize the level of physical distancing. But this only captures how much we are moving around, and can't capture how careful people are being.
However, there is now some interesting polling data available that may capture some of these effects. Ryan Burge has provided excellent visualizations of the steps the American public is taking from the DFP COVID-19 Tracking Poll:
Approximately 75% of Americans are still avoiding crowded places. Over 75% say they have improved their personal hygiene (washed their hands more), 75% are avoiding personal contact, and mask wearing is at 94%.
Perhaps most interesting, 25% of Americans have intentionally avoided going to work because of the virus. If someone is willing to avoid going in to work, then they are also likely to be doing the other things, like avoiding crowds, personal contact with others, etc.
Note also that absolutely no sign of the "reopening" can be seen in the data. Reopening states had almost no impact on these behaviors. Note also that adherence is not evenly distributed across the population. For example certain political groups are more likely to adhere to COVID-19 health measures than others.
In sum, the US population's adherence consists of:
- 25% hard core physical distancers
- (who avoid work)
- an additional 50% medium physical distancers
- who go to work, but avoid crowds, wash hands, and avoid personal contact
- an additional 20% mask wearers:
- who do not do the above, but who are willing to wear masks
- And only 5% COVID-19 deniers
- who do nothing, and do not even wear masks.
Impact on Immunity Thresholds:
These results may be key to understanding the resistance to a second wave of infections.
If we assume that 25% of the population have been infected and are immune in a given hard hit location like New York, and then we assume that an additional 25% are staying home from work/working from home, we have already reached 50% of the population who are either immune or are hard core physically distancing. That alone would drop Rt from 3 down to 1.5.
If a high number of the 5% who aren't being careful at all, are among those who got sick and already had COVID-19 (and are thus immune, which seems likely), then that alone should be enough to reach Rt=1.
But if the efforts of the medium physical distancers and the mask wearers are added to the equation, then the immunity thresholds may be close to 15%-20%.
But ONLY when combined with physical distancing.
The mistake in expecting second waves of infections after reopening was in assuming that adherence to these health measures in personal behavior was going to change substantially when states opened (or closed). But the data does not support this conclusion.
Opening had little to no effect.
Certainly with restaurants open, the 25% who are not avoiding crowds and/or personal contact could be more likely to be infected, because environments are now available for them to put themselves at greater risk. But once 66% of the 25% who aren't being careful have been infected and are thus immune (that's only 17% of the total population), then a tipping point could be reached where the Virus's progress would naturally slow without further intervention.
But again, we would expect this protection to only hold when combined with the physical distancing efforts of the rest of the population.
Good analysis. Other contributing factors may be in play also. Two contenders:
ReplyDelete1. People with partial immunity due to a previous cold (coronavirus).
2. Population heterogeneity. Here's a new article (14 Aug 2020) that makes a case for herd immunity at ~40%. https://science.sciencemag.org/content/369/6505/846
There is a difference between immunity and social distancing that messes with the math.
ReplyDeleteConsider a hermetically sealed 25% of the population. What effect does this have on the transmission of the rest of the population?
Little. The sealed off population is almost irrelevant to what happens in the rest of the community.
The way I think about it is forest management. I need to thin the trees in every forest to have an effect. Removing all the forest in Alaska might remove a quarter of the trees, but the fires in California still burn.
So the herd immunity threshold would only fall 25% from 66% to 49.5% of the population in this extreme example.
The other thing that tends to mess up the math is that someone socially distancing by not going to work still may have a family at home. That family may have other members such as school age children to vector the virus in and out. Heck, nursing homes tell us being isolated from the world isn't enough to prevent Covid from coming in.
The modeling on this is not easy at all.
Bottom line is that social distancing is not the same as immunity and can't be used to recalculate the herd immunity threshold in this way. Social distancing helps, which is fantastic, but it's in no way immunity in practice or on paper.
The forest analogy is good. Consider using it in a different way. Imagine that 25% of the trees are much more flammable than others. Perhaps a different species. And imagine that these 25% of the trees are randomly distributed in the forest, and all of them have already burned. So the remaining 75% are not only thinned, but are also inherently less flammable. Perhaps you achieve herd immunity at 25%. The Covid analogy is that 25% of the population has far more contacts per day, or else riskier contacts. Cab drivers, meat packers, and Covid-deniers who don't distance, for instance. If the first wave mostly infects those people, then you have removed the most flammable trees from the forest. I think that is crudely what happens with "population heterogeneity" https://science.sciencemag.org/content/369/6505/846
DeleteThe interactions between immunity and physical distancing are definitely complicated.
DeleteAnother way to think about it is like moving balls on a billiard table. Each infected ball on average hits 3 others at R0. But then 25% of the balls just aren't there to hit (they are hiding in the pockets)... and another 25% are immune, and the rest of the balls are wearing masks... and now you have driven Rt down below 1, while only having around 25% immune.
ALL of these analogies over-simplify the true dynamics. But the point wasn't to do the full simulation, but to do a back of the envelope calculation to see if what is going on makes basic levels of sense.