For sure, this wouldn’t be my choice of ways to greet the new year. But it is what it is.
We’re now two full weeks since the 12/17/2021 start of the Omicron wave in the U.S., and there’s still no letup in sight. The U.S. daily new COVID-19 case rate doubled last week.
Trees don’t grow to the sky. This period of extremely rapid new case growth has to end sometime. It’s just that the end is not yet in sight.
We’re now reaching the point where the sheer volume of cases overshadows the lower case severity of Omicron. Based on the most recent data, the total number of daily new COVID-19 hospitalizations is as high as it was at the peak of the Delta wave, although the number of cases in the ICU remains considerably lower.
Data source for this and other graphs of new case counts: Calculated from The New York Times. (2021). Coronavirus (Covid-19) Data in the United States. Retrieved 12/31/2021, from https://github.com/nytimes/covid-19-data.” The NY Times U.S. tracking page may be found at https://www.nytimes.com/interactive/2020/us/coronavirus-us-cases.html.
Lower case hospitalization rate is now cold comfort in the face of this astounding rate of growth.
My best guess is that Omicron in the U.S. has about one-third the case hospitalization rate that Delta had. It really is that much less virulent. That’s not as benign as the results from South Africa suggested, where the case hospitalization rate was about one-fifth. But it’s within the range of estimates from research done in Scotland and Great Britain. That one-third figure matches my most recent calculations from U.S. data on new cases and hospital admissions, assuming that what we’ve seen up to now has been a mix of Omicron and Delta.
If that’s true — there’s a fair bit of uncertainty there — then we would need three times as many Omicron cases, to generate the same level of hospitalizations we had under Delta.
I took considerable comfort in that. Because, seriously, who would have thought that we’d ever see that many new cases.
Here’s the problem. We’re going to hit that “three-times” new case rate the day after tomorrow. Look at the graph at the top of the page. Delta topped out at about 50 new cases / 100K / day. If the current trend keeps up, the U.S. will be at 156 new cases / 100K / day on Monday 1/3/2022.
The bottom line is that Omicron has a case hospitalization rate that’s only about twice that of typical seasonal flu in the U.S.. (That’s my calculation, from CDC data, a few posts back). But spreads vastly faster. Typical R-nought values for seasonal flu cluster in the 1.5 range. Estimated R-nought for Omicron is 10 to 15. And, although I would not have believed it two weeks ago, we’re now reaching the point where the sheer volume of cases has become the dominant factor in terms of overall morbidity burden from COVID-19.
Here’s the second problem: Admissions have now jumped in concert with new cases. Because of that, we’re already getting a new admissions rate that’s on a par with the peak of the Delta wave.
Unfortunately, the CDC COVID data tracker website is not updating over the holidays. I’ve gone back to processing the raw hospitalization data from U.S. DHHS. This file. And the picture as of yesterday looked like this. The seven-day moving average of new COVID-19 admissions topped 12,000, which was the level at the peak of the Delta wave.
Source: Calculated from U.S. DHHS Patient Impact and Hospital Capacity by State Timeseries, accessed 1/1/2022.
The sole bright spot in terms of hospitalizations is that a smaller fraction of the Omicron cases are ending up in the ICU, compared to Delta. And so, while new hospitalizations have more-or-less reached the same level as at the peak of the Delta wave, the share of ICU beds occupied by COVID-19 cases remains below that level (see chart below).
Source: Calculated from U.S. DHHS Patient Impact and Hospital Capacity by State Timeseries, accessed 1/1/2022.
Although ICU use lags hospital use a bit — it sometimes takes patients some time in the hospital go get sick enough to require an ICU level of care — by eye, I don’t think it’s the time lag that’s keeping that rate lower. The ICU rate per admission seems to have been lower throughout the Omicron era, which would be consistent with the generally milder nature of Omicron compared to Delta, and consistent with South African reporting of much lower rates of oxygen and ICU use among Omicron patients.
Technical note 1: Throughout the analysis above, I have stuck with convention and shown seven-day moving averages above. I’m not sure that’s warranted, given how smoothly the hospital-reported new admissions timeseries changes over time, and given how fast the situation is changing. Just as a point of reference, the raw number of the past couple of days has been 15,000 new admissions per day, or about 20% above the Delta peak.
Technical note 2: Just FYI, today’s data, used as moving averages, work out to a case hospitalization rate for COVID-19 of 3.1 percent, for the past week. That’s less than half the prior 6.5% rate for Delta. And I suspect that it isn’t lower than that because we’re still seeing a mix of Delta and Omicron cases. The case hospitalization rate should continue to decline somewhat as we move fully into the Omicron era.
Technical note three: One aspect of methods that nobody even questions is the use of a seven-day moving average. For new cases, we need at least that period of time to smooth out fluctuations in reporting. (E.g., more than half the states don’t report new cases on the weekend. But as Omicron has sped up the pace of change, that seven-day moving average increasing hides a true picture of the current situation. For sure, at this rate of growth, and a seven-day moving average, we’ll be a week past the peak before it even dawns on us that the worst is over.
