CDC DRAMATICALLY REDUCES LAST WEEK’S OMICRON VARIANT PREVALENCE PROJECTIONS DOWN FROM 73% TO 22%; SAYS IT’S NOW LIKELY AROUND 59%
By Andrew McKay
Last week, many news outlets made a big deal out of the released CDC data saying that Omicron was 73 percent of all new cases in the United States. At the time, I raised three questions about this projection:
- It seemed likely to be an over-estimate that represented a high concentration of genetic sequencing being done in those areas that were seeing Omicron and surges rather than a true measure of the whole country new case share.
- I also noted that the 73% number was based on modeling / projections, not hard data.
- I also reported that the 73% number was actually an average within a range between 34% and 95% (because of the modeling), which is a pretty large variation (see Chart 1).
Well, CDC has now revised the model dramatically down from 73% to 22% for last week, and even so that’s from a range as low as 15% and as high as 31%. This week’s number is presented (based on modeling) as an average of 59% within a range between 41% and 74%. CDC has also (to their credit) now added some notable asterisk explanations of the data that indicate it is heavily based on modeling. The explanation had been in the graphic, but the two big asterisks at the top of the chart are new since last week.
So, what does all of this mean? Here’s how I interpret it. Omicron is certainly still a fast-transmitting variant, but it is not the radical nightmare transmission rate the 73% would have indicated. All available evidence points to waves that rise rapidly, peak, and recede quickly compared to Delta with a lot of uncertainty about the virulence (hospitalizations and deaths), although we expect outcomes to be worse for unvaccinated people than for vaccinated people.