South Africa: Omicron about as severe as Delta
Omicron is barely milder when it comes to hospitalizations
We all have a burning question in mind: is Omicron milder than Delta?
Finally, we have enough South Africa hospitalization data available for the recent Omicron wave, with relatively large and growing hospitalization numbers allowing for apples-to-apples comparison.
I decided to look at South Africa’s hospitalizations and compare hospitalization rate recently (so with Omicron), with hospitalization rate around August (when South Africa had Delta).
I used two data sources: Worldometer for cases, and DATCOV hospital utilization report for hospitalizations. Worldometer has a handy “cumulative cases” graph, from which I can get cumulative cases for any two dates by hovering the mouse. Then I can get the cases that happened between dates, by subtracting those cumulative cases.
The hospital admissions report uses week numbers, so I had to get dates from week numbers. The admissions are in the DATCOV report:
First, I calculated Hospital Admissions for weeks 47-50: 1,460+4,011+7,659+8,348 = 21,478 patients admitted. Then I looked at cases that occurred within weeks 46-49, applying a one week lag between a “case” and a “hospitalization”. I got cases by hovering on appropriate dates of the worldometer cumulative report:
The number of cases for weeks 46-49 (Nov 15 to Dec 12) was 3,167,497-2,943,768 = 223,729 cases.
Dividing 21,478 hospitalizations by 223,729 cases, we get 9.60% hospitalization rate for weeks 47-50.
Let’s pick a set of weeks for the previous Delta wave: I decided to pick Week 32-35 as the Datcov data was available for these weeks. The number of hospital admissions for Weeks 32-35 is 9,523+9,836+8,618+6,901=34,878 hospital admissions.
The number of cases for weeks 31-34, also taken with a one week lag (Aug 02 to Aug 29) was 2,781,646-2,476,641 = 305,005 cases.
Dividing 34,878 admissions by 305,005 cases, we get 11.44% hospitalization rate for weeks 32-35.
What can we see? Omicron admissions rate in South Africa is estimated at 9.60%, and Delta admissions rate in South Africa, for August, is estimated at 11.44%.
In a very narrow sense, Omicron is milder because it has only 9.60% admissions rate, compared to 11.44% admissions rate for Delta in August.
But we can easily see that it is not much milder! Omicron has almost nothing to brag about, if it wanted to compare itself with Delta.
We now have to ask ourselves two questions:
Is South Africa now (Nov-Dec) different from August? The answer is yes, South Africa now has a much larger percentage of naturally immune, pure blood, unvaxxed citizens compared to August. So, we can expect this natural immunity to decrease Omicron’s hospitalization rates, as some recovered persons get reinfections, which are likely milder. In other words, Omicron would potentially be deadlier, if South Africa did not have so many already immune citizens.
Are Western countries like the UK, USA, or Denmark different from South Africa now? The answer is yes, the number of our naturally immune, unvaxxed citizens is much lower. At the same time, the number of vaccinated persons with questionable Covid specific immunity and potentially suppressed innate immunity, is much higher. We also have older population, different ethnic composition, and less physically active citizens.
We will soon consider death rates as well, as more data comes in. Deaths usually lag cases by 2-3 weeks and death information takes longer to collect.
The main objections here are:
It may be that mild cases of Omicron are undercounted. But it is also possible that mild cases of Delta in August were also undercounted.
It may be that Omicron patients are more scared and demand to be hospitalized over trivial complaints. I am not sure if this is the case, what do you think?
The above calculation is not a perfect way to figure out hospitalization rate. But it is a good attempt to compare “apples with apples”, that is, 4 weeks of Omicron with 4 weeks of Delta.
We have persons from South Africa here, I would love to hear your opinion, objections, etc. I also would love to hear your opinion if you reside anywhere else, too.
Thanks for reading Igor’s Newsletter! Subscribe for free to receive new posts and support my work.