Thread by @Bob_Wachter

Covid (@UCSF) Chronicles, Day 665
I haven’t done a SF update for a week – it’s a good time to catch up. Any impact of holiday gatherings should now be baked in, we all know scores of people with Covid, & hospitals are getting overwhelmed. So is SF’s high vax rate helping? (1/20)

Let’s look first at data @UCSFHospitals before turning to SF. 10d ago, we saw a mini-plateau & I wondered if SF's “wall of immunity” would prevent a hospital surge. Answer: no such luck.
Today, we have 75 Covid patients at UCSF Medical Center, up from 22 one month ago. (2/20)

Issue of how many of them have “incidental Covid” (testing pos. but there for another reason) has become hot; @SaraMurrayMD, @RhiannonCroci & I wrote about this @washingtonpost. washingtonpost.com/outlook/2022/0… Using our method (combines diagnosis & use of antiviral Remdesivir)…(3/20)

...it looks like ~36% (27/75) are here for "incidental Covid." The chart (thanks to our awesome data science folks for these dashboards!) shows the breakdown by floor vs. ICU – the “for” vs. “with" Covid fractions about the same (2/3 for, 1/3 with) in both settings. (4/20)

For first time I’ve seen anywhere, they’ve done crosswalk between vax status & for/with. Outside the ICU, our 2/3-1/3 ratio holds for both vaxxed & unvaxxed pts. Note that there's a fair # (23) of fully vaxxed patients here FOR Covid (23); some are likely immunosuppressed. (5/20)

But look at ICU #'s on chart: 11 ICU patients; only 1 is fully vaxxed & here FOR Covid (I don’t know the case – perhaps a Delta patient from pre-Omicron days?). So fully vaxxed can get desperately ill, but it’s quite unusual. Whereas 6/11 ICUs are unvaxxed & here FOR Covid.(6/20)

A useful reminder from ER doc @Craig_A_Spencer @nytimes that even "incidental" Covid cases take more work & time, and some patients will deteriorate in hospital. Per Craig, it’s not quite March 2020 (vax plus “milder” helps), but it’s still awful. (7/20) nytimes.com/2022/01/10/opi…

My fave piece of @UCSF data: symptomatic/asymptomatic test + rates. Symptomatic: eye-popping 46%. Asymptomatic: even more astounding 12%, ergo ~1-in-8 UCSF patients w/ NO Covid symptoms are testing +. Reasonable to extrapolate this to the population; meaning that >1-in-10…(8/20)

... folk in SF are prob. infected. (FYI, this % @UCSF was 0.4% [1-in-250] 1 mth ago.
Remember, this means that if you go into an indoor space w/ 15 people, it’s almost certain that someone is breathing out live virus. Can’t think of better argument for N95 or equivalent.(9/20)

And SF is not an outlier in this regard. Here, virologist Trevor Bedford @trvrb estimates that, in fact, ~1% of Americans are being infected each day and 5-10% of people in U.S. are currently actively infected.

https://twitter.com/trvrb/status/1480610448563060738

These are mind-blowing figures. (10/20)

Bottom line: @UCSFhospitals has lots of Covid patients, at a time that hundreds of our docs & nurses are out with Covid. The stress level is very, very high.
And keep in mind that our Covid hospital numbers are low compared with places like NYC, Miami, & Houston.(11/20)

Let’s turn to the city of SF; it mostly mirrors @UCSF trends. Big omicron question for SF was if >80% full vax rate would protect from big surge. Answer is no, at least regarding cases: now averaging 1386 cases/day in SF, up from 84 a month ago – a 16-fold increase. Crazy.(12/20)

And remember, these are reported cases; all those cases confirmed by home tests (incl. my son’s) aren't included, likely thousands more daily. It's a wild amount of virus.
In late December, it looked like cases were surging but hospitalizations were not. As with UCSF,… (13/20)

…we haven't dodged the hospitalization bullet in SF.
Hospitalizations in city aren’t overwhelming, but they’ve certainly shot up: now 155. ICU patients, though, remain relatively low (29) – undoubtedly the impact of our >80% vaccine rate. (14/20)

Note that while cases in SF are more than 3x higher than last winter’s surge, hospitalizations are ~40% lower, & ICUs >50% lower. That’s the vaunted case-hospitalization-ICU dissociation – the combined effect of Omicron's "mildness" & vaccination. (15/20)

And it's not just SF: see this piece by @DLeonhardt in @nytimes, looking at vaxxed vs unvaxxed outcomes in NYC & Seattle, the latter resembling SF in vax rates & overall approach to Covid. Even with lots of breakthru cases, vax still matters hugely.(16/20) nytimes.com/2022/01/11/bri…

The lesson from these data? Even in highly vaxxed places, Omicron’s infectiousness & partial immune escape have led to caseloads that are so jaw-dropping and expectation-defying that @CDCgov is unable to model the next few weeks reliably (below). (17/20) cdc.gov/coronavirus/20…

But S. Africa & London curves, & possibly-plateauing NYC curve are reassuring. In U.S., we may be 1-2 weeks from Peak Omicron. Of course, the threat doesn’t end that day; risk stays high until we descend the mountain, which’ll take several weeks. (18/20) bloomberg.com/news/articles/…

Notwithstanding Omicron’s “milder” profile, Omi still causes lots of serious cases & hospitalizations, & the majority of the latter are still for Covid-related complications. By now it’s a familiar math equation: 50-70% milder x 5-10 times more cases = a lot of badness. (19/20)

But “mildness” & high vax rates DO lead to fewer ICU pts, which should translate into fewer deaths. Expect worse in less vaxxed towns.
Across U.S., there’s nothing I see in data that alters my projection that an awful January will lead to a good Feb. Till then, stay safe.(20/20)