I've heard a lot about how the Delta variant is much more contagious and how vaccination both reduces transmission and dramatically reduces deaths, but I wanted to get a better sense of it by looking directly at the data. As with last time, I pulled the data from JHU's CSSE ( code). First, here's the pandemic so far:

Cases (blue) and Deaths (orange), 7-day moving average, February 2020 to present. Deaths are scaled 100x relative to cases, and then for each state both are scaled to fit.

You can see how Northeastern states (NY, NJ, etc) had massively disproportionate deaths in their first wave, though the statistics can't tell us how much was undertesting vs overwhelming the medical system vs older people being infected. You can also see that deaths generally move along with cases, with a small lag.

Looking at the rightmost edge of each chart, you can see the effect first of vaccination bringing cases down, and then Delta and bringing them back up. Let's look at just the most recent 45 days, this upswing, and sort the states by the vaccination status (CDC, "Percent of Total Pop Fully Vaccinated by State of Residence"):

Cases per 1,000 population, 7-day moving average, last 45 days.

Since this is now in cases per 1,000 people, we can compare values between states and not just curve shapes. Vaccination does seem to be helping, since states with a larger portion of the population vaccinated are seeing less increase. Still, it's roughly exponential growth everywhere, including in high-vaccination states, as we can see by scaling the charts:

Cases, 7-day moving average, last 45 days, scaled to fit.

Now, let's overlay deaths, again scaled 100x relative to cases:

Cases (blue) and Deaths (orange), 7-day moving average, last 45 days. Deaths are scaled 100x relative to cases, and then for each state both are scaled to fit.

Based on past waves, you would expect deaths to be rising along with cases. While we do see this in some states (AR), in most states deaths are rising much more slowly than cases, or even holding steady. This is really good to see! The pattern is stronger in states with higher vaccination rates, which makes sense, since we expect the vaccines to do more to prevent death than infection because (a) more vulnerable people are disproportionately vaccinated and (b) that's what we saw in studies.

Overall, it looks to me like the large public health push toward vaccination is justified.

Comment via: facebook

New to LessWrong?

New Comment