[Crossposted from https://firstsigma.substack.com/p/omicron with a few small updates to the estimates from the last couple days]

My thoughts on how Omicron affects Covid risk and what we should do about it. In particular, I wanted to address some common questions I've seen along the lines of "since it's less severe shouldn't I worry less?" and "is it so transmissible that everyone's going to get it, and if so is there any point in trying to avoid it?"


tl;dr: it's worse than Delta and not as bad as during the 2020 surges (since that was before vaccines). Your risk of getting infected and hospitalized has likely gone up by over 5x since before the surge (that’s accounting for Omicron being less severe). I think we should take stronger protection measures during January in particular when the risk is highest and treatment most strained.

  • How much does Omicron increase the risk to me?
    • Your Covid infection risk for the same activity has already gone up by over 10x in most regions (mostly because Covid prevalence has gone up by that much, and also because of increased transmissibility) and your risk of severe illness has gone up by over 5x (see next bullet point). And the risk will continue to increase until the surge peaks.
  • How severe is Omicron?
    • It's estimated to be less severe than Delta (~50% less), but similar or more severe than Alpha (Delta was about 2x more severe than Alpha). Note that a lot of the apparent reduction in severity is because more of Omicron's cases are breakthrough infections, which are less severe.
    • And the overall chance of getting hospitalized with Omicron is still much higher because of the much higher rate of infections: if we estimate an Omicron infection as 50% less likely to cause a hospitalization, and Covid prevalence is 10x higher, then your hospitalization risk is 10x * 50% = 5x higher than before.
  • Will Omicron burn through the US and then be over quickly?
    • I think this is roughly 50% likely, I definitely wouldn't count on it. I think the main other possibility is a shape similar to Delta in the US - a quick peak, then a slow decline into an elevated plateau. I think we'll reach the peak quickly, the more important question is what the shape of the pandemic looks like after the peak.
  • Will Omicron infect most of the population? If yes, is it even worth it to try to avoid it?
    • I think it's worth taking low and moderate cost efforts to avoid it: vaccination, air filtration/ventilation, using highly effective masks (N95s), testing, moving interactions outdoors, working from home, etc (where feasible and appropriate). I don't think high-cost efforts like lockdowns are the answer right now (note that vaccines make a massive difference compared to 2020).
    • I'd expect Omicron to infect a large fraction of the population, but it's unlikely to infect "almost everyone". And note that Omicron may or may not protect you from reinfection by the next big variant.
    • Getting Omicron in January is worse than getting it in March, because in case you need treatment, January-February is when the healthcare system will be at its most strained. Also Paxlovid is becoming more available. I expect that in the US the peak in cases will be over by the end of January, but still high, and that the surge will be well over by the end of February in terms of both cases and hospitalizations. You should plan to take stronger protection measures than normal now, and return to more “normal” then.

So what should you do about it? I would say, it's worse than Delta and not as bad as during the 2020 surges (since that was before vaccines), so you should take protection measures accordingly. You should take stronger protection measures during January when the risk is highest and treatment most strained.

The best things for you personally to do to reduce your risk are:

  • Prefer outdoors interactions as opposed to indoors crowds
  • Use high quality air purifiers/filtration/ventilation
  • If you're still wearing a cloth or surgical mask, upgrade to an N95
  • Get boosted

See The best Covid protection measures you should take.

How much does Omicron increase the risk to me?

The Omicron surge has already increased your risk of infection for the same activity by over 10x in most US regions, comparing the beginning vs end of December. And the risk will continue to increase until the surge peaks. Most of the increase is simply due to the fact that a much larger percentage of the people around you will be infectious. A smaller but still noticeable part of the increase is due to the higher transmissibility and/or reduced vaccine efficacy.

See microcovid.org to estimate your Covid risk exposure and the effectiveness of different safety measures. (If cases in your area haven't peaked yet, you should probably multiply the risk estimate by about 2x to account for the fact that cases are rising extremely fast right now and test results may have a couple days lag - you can see what date the data is from under Details.)

How severe is Omicron?

Omicron's intrinsic severity is estimated to be lower than Delta, but higher than other previous variants like Alpha (Zeynep Tufekci). So that's moderately good news, but probably not as good as you thought based on the early reports about low average severity in South Africa. This is because of the difference between intrinsic severity and contextual severity. The right question to be asking when thinking about your personal risk is: for people with a given vaccination status, how severe is an Omicron infection compared to earlier variants? This is a tricky question because (a) Omicron is better at infecting vaccinated people, and (b) a lot more people have gotten vaccinated in the meantime. Both of these mean more of Omicron's cases are breakthrough cases, which are much less severe - so on a population level, the average case is much less severe. But for you as an individual, you should think of Omicron as significantly more likely to infect you, and if you are infected, a bit less severe than Delta but still more severe than other previous variants - based on early estimates, it appears about 25%-50% less severe than Delta in terms of hospitalizations per infection, and Delta was about 2x as severe as Alpha. More recent data seems to suggest even lower severity, so I'll estimate 35%-75% less severe, which aligns with Zvi's probability updates.

It’s important to remember that although Omicron’s rate of hospitalizations per infection is lower, that lower risk is blown away by the massively higher chance of you getting infected by Omicron due to the massively higher Covid prevalence around you. With your risk of infection increasing by over 10x in most regions, you can estimate your risk of getting infected and then having severe illness requiring hospitalization to be e.g. 10x * 50% = 5x higher than before Omicron.

Vaccine effectiveness against infection appears to have dropped a lot - early studies on Pfizer showed original 2 doses had effectiveness 35% (95% CI 20-50%), while with a booster effectiveness came back up to 73% (95% CI 58-83%). But vaccine effectiveness against severe hospitalization appears to be holding strong (Zvi, Vaccine Effectiveness section).

Since young healthy vaccinated people are already at very low risk for severe disease, we'll probably be ok in terms of hospitalizations and deaths - worse than the flu, but by less than 2-3x, I'd guess. And still much worse than Delta during the peak. For young vaccinated people, the potential risk of long covid is the main concern. Long covid risk is highly correlated with severity of illness, so basically this just means you should still act based on minimizing your risk of severe disease, but the risk to you is much higher than it looks on the surface (like if you just look at death rate). See my covid risk analysis article for more details.

Separately from the personal risk is the societal strain on hospitals and other infrastructure. Large numbers of people getting Covid infections, even mild ones, means staffing shortages, including medical staff and other essential workers. A small percentage of hospitalizations on a very large number of infected people still means many more hospitalizations than before Omicron. These two combined means a lot of strains on the healthcare system. At least it's not looking like a catastrophic crisis. See Zvi for recent updates on this topic.

Will Omicron burn through the US and then be over quickly?

One common prediction I've seen is along these lines: “Omicron will blow through the US by 3/1/2022, leading to herd immunity and something like the ‘end’ of the COVID-19 pandemic”, as discussed on Bet with Zvi about Omicron - I think this is a good high-level summary of the thinking for and against this prediction. Zvi predicts this at a 70% chance, Holden at 50%. I am also putting this scenario at 50%. Metaculus currently has this at 47% (note Metaculus is asking the reverse question). 

I do think there will be a steep, quick climb to a peak. The bigger question in my mind is: what will the shape of the pandemic look like after the peak? Will it go down as quickly as it went up, or remain elevated for months?

One way to try to answer this is to look at the shape of the Delta surge in different countries, and then think about what may be different with Omicron. I think the key differences are that it's spreading even faster than Delta, but on the other hand a lot (maybe most) of that faster spread is due to its high vaccine escape, which means that it also has more population it would need to burn through to be done. I don't see a very clear reason why we should expect the post-peak trendline to look terribly different from Delta, but I'm not an epidemiologist.

Delta burned through some countries quickly, but not others - why?

As Delta was starting to hit the US, I made a low-confidence prediction that Delta would peak quickly, burn through the non-immune population, and then fall back down quickly. This was based on the shape of the India and UK Delta surges as of the beginning of August, which did indeed seem to be falling from the peak quickly; and some napkin math estimating the total percentage of the US and UK populations with (some degree of) immunity through either vaccination or past infection and the rate at which we'd reach herd immunity. (Note Delta certainly wasn't the first time people talked about this, there'd been discussion like this already on earlier surges, see e.g. this thread from Zeynep back in April 2021 about some of the many confounders and unknowns involved, which are part of the reason it was a low-confidence prediction.)

By mid-August, I saw that the UK was not following India's trajectory, and instead trailing into a plateau that was still half or so the height of the peak - which then continued for months with periodic ups and downs. The US then followed a similar trend, but with a slower spike up and a slower drift down. And that wasn't just because India's peak was higher or something like that - in terms of confirmed cases per capita, India's peak was not much higher than the US's plateau (and yes, confirmed cases is not a great way to compare countries, but the shapes are still clearly different). 

Looking back now at the shape of the Delta surge in different countries (e.g. on this chart), I see two typical shapes:

  • In some countries, like India, Germany, France, and Canada: Delta peaks in about 2 months, then drops back down over the next 2-3 months to much lower than the peak and not much higher than before Delta.
  • In other countries, like the US, UK, Brazil, Russia: Delta peaks in about 2 months, then over the next 2-3 months gradually trails off into a plateau that's still 1/3 or 1/2 as high as the peak, and far higher than pre-Delta, maybe continuing on a slow decline but not dropping anywhere near where it was before the surge.

These happen in different countries and different times, and I don't really have an explanation of why yet. E.g. it's not just low vs high vaccination rates. And it also doesn't seem to be explained by differences in test rates. Another common proposal I've heard is about differences in the social graphs in terms of rates and types of social interactions, but I'm not aware of a good attempt to actually quantify that. Anyone know of a satisfactory explanation?

Will Omicron infect most of the population? If yes, is it even worth it to try to avoid it?

I think there's several questions to ask here:

  • How much of the population will Omicron infect before another variant displaces it?
  • How much short-term and long-term immunity does an Omicron infection provide to infections by Omicron, and to the next big variant?
  • And how does that compare to the immunity provided by vaccines to the next big variant?

If the vast majority of people are likely to get Omicron exactly once before the next big variant arrives, then Covid precautions would be of little value aside from flattening the curve considerations (hospital and ICU capacity, etc). Which is certainly still important - your risk of severe outcomes is much worse if you get infected in January than in March, after the peak when treatment will be much less strained.

This assumes that Omicron provides reasonably strong protection from Omicron reinfection, which I think is likely, and that this protection remains strong at least long enough to protect you until Omicron is displaced by the next variant, which I think is also likely. If not, then you might expect to get Omicron multiple times, so you should still try to minimize the number of times you get it.

But let's unpack the question of "How much of the population will Omicron infect before another variant displaces it?" In our experience so far, it seems more likely than not that the next big variant will arrive in the next half year to year. For Delta, the vaccinated population was getting infected with Covid at a rate of about 20% per year (based on CDC with a 4x adjustment for confirmed vs actual case count). Omicron case rates are far higher during the current surge, currently maybe 4% of the population per week, which is crazy high. But we don't know what the post-peak shape of cases will be (as discussed in the previous section). Also, the bulk of the surge has always been among the unvaccinated population. Overall it seems like there's a good chance that you will get Omicron, but also a good chance that you'll never get Omicron before it is displaced by another variant. I'll very roughly predict that a median of 40% of the US population will get infected by Omicron.

The next consideration is: how much immunity does one variant grant from infection from a future variant? And how does that compare to vaccine efficacy? Omicron infection does seem to grant strong immunity to Delta based on early studies, so that's good, and seems to be an indicator that Omicron is likely to grant decent cross-immunity against new future variants as well. But our vaccines have also proven great at protecting against infection by new variants, and even more against severe illness. It seems to me that immunity from vaccines and previous infection are going to follow generally similar trends, most of the time. In other words, I wouldn't bet on Omicron granting me hugely more immunity from the next big variant than my vaccine does, and therefore I'd rather not get Omicron if I have the choice. 

That said, I don't think we will (or should) be going back to early pandemic lockdown-style life. Lockdowns are a very costly, short-term intervention, and one important thing to remember is that despite the variants, vaccination (especially with a booster) massively reduces your risk compared to back then. Masks and testing are important too, especially during a surge. But for the long-term, I put most of my stock in lasting countermeasures, namely vaccines and air filtration/ventilation. I think that's the only way we as a society are going to effectively handle this and future pandemics.

See The best Covid protection measures you should take for how to best protect yourself from Covid.



New Comment
9 comments, sorted by Click to highlight new comments since: Today at 9:37 PM

upgrade to an N95


Elastomeric respirators are better than N95s in almost every way, although they do tend to somewhat muffle speech.

I mostly agree with you and go into that in detail on https://firstsigma.github.io/masks but I think the "average" reader (especially one who is still wearing cloth/surgical) is better served by a recommendation to upgrade to N95 first. I find that P100s are great for many situations, but they do have their downsides.

I see almost no significant reason for anyone (with a few exceptions such as people with speech issues) to use disposable N95s instead of elastomeric respirators. As I pointed out in the comment I linked to, N95s generally provide a poor seal. Another problem is that the straps N95s use are more difficult to put on and take off. The straps also lack adjustability and a head harness. Compared to elastomerics, differences in weight and bulkiness (while wearing) aren't that significant. Muffled speech is noticeable with elastomerics but can be compensated for by speaking louder or using an electronic voice amplifier.

Here are some technical notes about elastomerics: 1) Elastomerics should not be referred to as "P100s," because that just refers to a filter standard and most elastomerics can use different types of filters including N95. 2) The 3M 2291 filter is better than the 3M 2091 due to its lower pressure drop (that means it takes less effort to inhale). 3) Another alternative (for people with breathing difficulties, for instance) is to use an N95 filter which probably has an even lower pressure drop. 4) The N95 cartridge filters (3M 603 adapter, 3M 501 retainer, and 3M 5N11/5N71 filter pads) don't weigh significantly more than the P100 pancakes. 5) I've seen some anecdotal reports that the 7500 series is more comfortable than the 6500 series due to the (supposedly) slightly more pliable silicone. 6) You reported on counterfeiting on Amazon, yet still linked to Amazon without offering verified vendors for the filters.

I agree with most of your points, especially about poor seal for typical N95s, but in my personal experience, I am often wearing a mask intermittently, not wearing it for the whole day, and in that context I typically prefer to use a N95 which I can fold and put in my pocket - that convenience is very important for me. I definitely do recommend an elastomeric P100 for long-duration wear where the convenience factor is not an issue. And I found it basically impossible to make myself audible in a loud crowded environment, although it's fine in quieter environments.

I have not done a lot of detailed research into different P100 filters, my cursory reading suggested that they were generally "good enough" for my purposes (N95->P100 being a large difference, while different P100 filters having only a relatively small difference, as far as I know), but I appreciate your info and suggestions.

The counterfeiting issues I've heard of mostly apply to N95s/KN95s/etc. I had not previously heard of it in the context of elastomeric respirators and their filters, although it wouldn't surprise me. My experience with a couple other vendors I tried has been unpleasant (issues with long delays and no stock etc), so I ended up just going with Amazon. Do you have recommended vendors?

I'm not sure about the level of counterfeiting, but there does seem to be at least some of it going on.


Zoro.com (owned by Grainger) is a reliable and reputable vendor.

I've made several updates to my guide to incorporate your input. I'll be trying the 7500 and 2291s next. Thanks again!

Thanks! I didn't realize that Zoro was owned by Grainger, that makes me feel much better about them. I also had a decent experience buying from them, it was still much slower than Amazon but not too bad. I'll add a recommended link to them.

And to add to my last comment, just to explain why I wrote things the way I did, I think the more important thing in terms of recommendations to a general audience is that in my personal experience, most people I've talked to have indeed upgraded to an N95, but extremely few people I've talked to have actually ended up wearing a P100 elastomeric. They have a ton of advantages, but most people simply don't want to use them. So in my mask article I explain the advantages, but in my general advice article I feel like it's more useful to simply say upgrade to a N95 or P100.

They upgraded to N95s because that's what the media and public health experts have talked about. You've started to right this wrong, but you could go a little further.

Wearing a respirator such as an N95 is mostly about risk reduction rather than convenience. While there's no reason not to mention convenience, any recommendations should be ordered mostly by risk reduction.

New to LessWrong?