Thread by @jburnmurdoch

NEW: today’s update from Gauteng, now on a log scale to better show current trajectories.

Steepness of lines shows how much faster the growth in cases and positivity is now vs past waves, and hospital admissions are now steepening too as the acceleration in cases feeds through.

Here’s same thing back on linear scale.

The eagle-eyed among you may spot something here: hospital admissions were tracking the Beta wave and first wave in yesterday’s chart, but today they’re slightly steeper. How is that possible, since it still shows data for the same dates?

This is why:

South Africa’s weekly hospital admission counts are back-filled every day as new patient data comes in, so the figure for the week ending Nov 28 has been rising in recent days, from 580 in the data published on the 29th, to 788 in today’s report.

In other words we need to be cautious in interpreting admissions data for two reasons:
1) It lags cases by ~ a week, so admissions line now is reflecting cases line a week ago, not today
2) It’ll be revised upwards, steepening even before including admissions in last 5 days

And that’s only the quantitative reasons.

We also need to be aware that Omicron wave in Gauteng currently skews very young, as @tomtom_m shows here, which would push case-hospitalisation rate down [but only while cases remain concentrated among young]:

https://twitter.com/tomtom_m/status/1466121405322260482

And of course the other possibility, the one we’re hoping for, is that immunity could be reducing the share of cases that end up in hospital, much as we saw in Europe recently where vaccination greatly weakens the link between cases and severe disease

https://twitter.com/jburnmurdoch/status/1458748635609550848

But a clear warning on that from data published earlier today: researchers in South Africa find that the risk of re-infection is much higher with Omicron than with past variants, suggesting some evasion of immunity from prior infection (no data on vax)

https://twitter.com/SACEMAdirector/status/1466459311673159685

Final points:
1) Vax-induced immunity works differently to prior-infection, and there’s reason to believe that with a heavily mutated variant, vax may fare better due to vax-induced antibodies’ ability to bind to more parts of the virus than infxn-induced Abs. BUT don’t know yet.

2) Some people are making the mistake of comparing deaths *today* to cases and concluding low severity. Current wave of cases is so steep it’s climbed as high in 20 days as Delta got in 42. If you compare deaths over a short period to a much longer one, of course they’re lower.

Please let me know if you have any questions, feedback, things I’ve missed etc, and thank you to everyone who has pointed out new data that I’ve been incorporating into these threads.

Please also follow experts on the ground in SA including @tomtom_m @SACEMAdirector @Tuliodna