Long covid: probably worth avoiding—some considerations

This is not going to be kind, but it's true and necessary to state. I apologize in advance.

Had you asked me in advance, I would have said that Katja in particular is likely to buy into long covid even in a world where long covid is completely psychosomatic; I think you (Katja) are probably unusually prone to looking-for-reasons-to-"believe"-things-which-are-actually-psychosomatic, without symmetrically looking-for-reasons-to-"disbelieve".

On the object level: the "Long covid probably isn't psychosomatic" section of the post looks pretty compatible with that prior. That section basically says two things:

• Just because reports of long covid are basically uncorrelated with having had covid does not imply that long covid does not happen
• There is still evidence of higher-than-usual death rates among people who have had covid

If we take both of these as true, they point to a world where there are some real post-covid symptoms, but the large majority of reported long covid symptoms are still psychosomatic. That seems plausible, but for some reason it isn't propagated into the other sections of the post. For instance, the very first sections of this post are talking about anecdotes and survey studies (at least I think they're survey studies based on a quick glance, didn't look too close), and I do not see in any of those sections any warning along the lines of "BY THE WAY THE LARGE MAJORITY OF THIS IS PROBABLY PSYCHOSOMATIC". You're counting evidence which should have been screened off by the lack of correlation between self-reported long covid symptoms and actually having had covid.

I appreciate you describing brain fog as psychological torment.

I think most severe/life-disrupting cases of long COVID are indistinguishable from ME/CFS. (I think it would be correct to say long COVID "causes" ME/CFS.) As someone living with ME/CFS: yes, the cognitive dysfunction is torture. And pushing to do stuff anyway just makes things worse -- this is called post-exertional malaise. 

The comparison to sleep deprivation is good, but I'll note that the brain fog I regularly experience due to ME/CFS is much worse than any sleep deprivation I had ever experienced in my life beforehand.

The cognitive dysfunction itself is awful. And the way it crushes your dreams is heartbreaking. Say goodbye to your career and all your intellectual interests. 

(I mean, say goodbye to your interests in general. Your physical stamina gets wrecked too. I rarely have the physical stamina to cook simple meals without post-exertional malaise.)

The social isolation stemming from the mental fatigue is threefold:

1) I feel stupid when talking to my friends, because I kind of *am* stupid now. Especially in the context of real-time conversations. My memory and processing speed is not good.

2) When I try calling friends anyway, I am punished by post-exertional malaise afterwards. e.g. I spend the following day very uncomfortable and unable to function well enough to even distract myself. In this state, even watching simple TV can be uncomfortably exhausting.

3) My healthy friends just don't understand what it's like to be going through such a horrible illness. I feel boring. They don't know what to say. It works out to be very alienating and bad for mental health.

On top of all that, so many people don't believe this illness is "real" in some sense. They think it's just laziness, or aging, or psychological. (FWIW, my illness started at 23, and I saw multiple psychologists who concluded there was no plausible psychological basis for my illness.)

Imagine being psychologically tortured, socially isolated, and gaslit by a ton of people arguing that you aren't even sick. Potentially with no light at the end of the tunnel. It's extremely upsetting.

Lots of healthy people argue that lockdowns are bad for their mental health. I'd argue that ME/CFS is orders of magnitude worse -- even in my case, which is relatively mild. Up to you to decide if the risk is worth it in your case.

I've skimmed a lot of long COVID prevalence studies, and unfortunately, they often have huge sample biases and don't ask the right questions to track ME/CFS incidence. Overall, I agree with Scott's range of "a few tenths of a percent to a few percent" as the risk of really bad long COVID symptoms.

Thank you for this. I'd say prior to reading this I was around 70% that for someone recently vaccinated or boosted Omicron isn't really worth worrying about, and getting Omicron at some point might even be +EV, due to cross immunity effect [LW · GW]. I'd say now I'm around.. 45% on this?

My sense is that EAs in general have not been in the "strict lockdown" mode, trying to do some more careful tradeoffs to allow for things like EAGs. This struck me as.. reasonable-ish at the time, (even up through delta). But if Omicron has a similar long-covid story, this suggests that the upcoming EAGs may not be a good idea in person.

For the (K) "belief in having had long COVID" (psychosomatic) French study thing see also https://www.lesswrong.com/posts/LETXrTsGAqJrcik3T/french-long-covid-study-belief-vs-infection [LW · GW]

What does "VOI" mean?

Huge VOI in following behind the crowd

Frontpaging this despite the current policy to leave most Covid content on Personal Blog. Doing so because it is of interest and broad relevance to people's decisions and seems high-effort and high-quality, though note questioning of biased reasoning in the comments.

Q. Getting covid later is probably better than earlier.

As a counter consideration, vaccine effectiveness might wane quickly, and it's likely better to get COVID while better protected than while less protected. See, e.g. https://www.webmd.com/vaccines/covid-19-vaccine/news/20211227/covid-booster-protection-wanes-new-data

That being said, I'm still leaning towards avoiding COVID to avoid long COVID.

I think we should look further into this study, which seems somewhat reassuring, but I have reservations about it:

Kuodi et al., January 2022, "Association between vaccination status and reported incidence of post-acute COVID-19 symptoms in Israel: a cross-sectional study of patients tested between March 2020 and November 2021" (pdf, not yet peer-reviewed)

Their previous version understated the results, since it didn't include the uninfected. The new version does, and says (cutting out some text before and after, and emphasis added by me):

Methods: We invited individuals who were PCR tested for SARS-CoV-2 infection at participating hospitals between March 2020-November 2021 to fill an online questionnaire that included baseline demographics, details of their acute episode and information about symptoms they were currently experiencing. Using binomial regression, we compared vaccinated individuals with those unvaccinated and those uninfected in terms of self-reported symptoms post-acute infection.

Results: We included 951 infected and 2437 uninfected individuals. Of the infected, 637(67%) were vaccinated. The most commonly reported symptoms were; fatigue (22%), headache (20%), weakness (13%), and persistent muscle pain (10%). After adjusting for follow-up time and baseline symptoms, those who received two doses less likely than unvaccinated individuals to report any of these symptoms by 64%, 54%, 57%, and 68% respectively, (Risk ratios 0.36, 0.46, 0.43, 0.32, p<0.04 in the listed sequence). Those who received two doses were no more likely to report any of these symptoms than individuals reporting no previous SARS-CoV-2 infection.

Conclusions: Vaccination with at least two doses of COVID-19 vaccine was associated with a substantial decrease in reporting the most common post-acute COVID-19 symptoms, bringing it back to baseline. Our results suggest that, in addition to reducing the risk of acute illness, COVID-19 vaccination may have a protective effect against long COVID.

One thing that's pretty suspicious here and could cast doubt on these results is that the uninfected were more likely than the infected doubly vaccinated to have some of these long COVID symptoms, including the most common one, fatigue (and that seems to be the only one for which there's a statistically significant difference, driven by those aged 36+).

On what they did:

A first series of binomial regression models were fitted to the data for the ten most commonly reported post-COVID-19 symptoms according to vaccination status. We adjusted for the difference in follow-up time and proportion of asymptomatic patients at the time of diagnosis between the groups. We then compared vaccinated and infected individuals to never infected individuals in terms of reported symptoms, also using binomial regression models. We used two distinct steps since it was not possible to adjust for follow-up time and symptoms at baseline among uninfected patients.

To take the anticipated age differences into account, the analysis was age-stratified and differences in the length of time from the beginning of symptoms to responding to the survey were adjusted for in the model. Vaccination status was recorded as either one dose or two doses. At the time of data collection, very few individuals had received a third dose and those who did were recorded as two doses.

It's not clear to me whether or not they adjusted for the demographic characteristics they collected (summarized in table 1).

Some possible explanations and potential weaknesses of the study:

1. The sample sizes are not very large, so it could be noisy. They only had 59, 135 and 100 doubly vaccinated + infected individuals in the 19-35, 36-60 and >60 age groups, respectively. The number of uninfected was much larger, though, 2437 together. As far as I can tell, they didn't make any adjustments for multiple comparison tests, and they did >160 tests, for the different symptoms and age groups.

2. I don't know if they made any adjustments (other than age stratification) for the comparisons to the Uninfected group, and this could leave bias. If they did make any such adjustments, they might not have made enough. They didn't do any matching.

There could be group differences not accounted for that would bias it this way, e.g. people who got vaccinated and infected could have been more health-conscious on average than those who were uninfected (with any vaccination status). We might think the uninfected are demographically between the unvaccinated+infected and vaccinated+infected. Compared to the unvaccinated+infected, the doubly vaccinated+infected were more likely to have tertiary education (62.9% vs 52.1%), more likely to have hypertension (13.9% vs 5.1%) and more likely to have diabetes (8.2% vs 3.8%). Hypertension and diabetes should point to the opposite bias, though, if unadjusted for. (I think it's unlikely they adjusted for anything they should not have adjusted for in a way that increased bias, although such adjustments can increase variance.)

3. Some of those considered uninfected could actually have been infected, and ended up with long COVID.  They used "uninfected in terms of self-reported symptoms post-acute infection", but it's possible some asymptomatic infected individuals get long COVID, and some of these asymptomatic infected individuals were probably not vaccinated. Their positive cases were based on "self-reported positive PCR results".

4.The vaccines could protect against long COVID-like symptoms from other causes. Maybe through the placebo effect?

Overall, given how large the gap is between uninfected and doubly vaccinated+infected, and the reasonable assumption that vaccination + infection should actually be worse than never getting infected, that gap could easily hide a risk of long COVID of around 1% (at least for fatigue).

If half of my friends are getting covid right now, there seems to be massive value in waiting another three months before reassessing precautions, so I can see how it goes for them. While noisy, I expect to get a better sense of the distribution of outcomes among people in my reference class—and especially whether there is a macroscopic chance of really dire outcomes—from this (I think regrettable) experiment than I get from cobbling together different studies about vaguely defined symptoms among ambiguous or not-quite-relevant demographics, with various guesstimated adjustments.

I recommend registering in advance how you will evaluate the new information you get over the next several months. In particular, it would be valuable to write down your credence on various outcomes, conditional on covid being something that is worth paying a substantial cost to avoid.

My guess is that you and I will disagree on what we're likely to see in low vs high risk worlds. For example, I think that A and B in your analysis are almost guaranteed, whether long covid risk is large or small. Also, C, F, G, H, and L are entirely consistent with low risk, and only modestly more likely in worlds where long covid is a major concern.

EDIT: I forgot (or didn't know) that loss of sense of smell was less likely with Omicron, which Steven Byrnes pointed out. Seems like 12-20% (probably mostly vaccinated?) of Omicron infections vs 7-68% for previous variants (section 3.5, and I'm not sure to what extent vaccination status was considered here). I've also read that Delta was less likely to cause loss of sense of smell than earlier variants, too. If the probability of loss of sense of smell scales proportionally with other brain issues, then I'd guess Omicron is less severe than previous variants (assuming 2 doses of mRNA vaccines against Delta, and 3 against Omicron, with the last one recently). Maybe by a factor of ~2x, which is somewhat but not extremely reassuring.

Also, it's not clear that long COVID is less bad or less likely with Omicron than Delta (for a fully vaccinated person). IIRC, vaccines are less effective at reducing symptoms even conditional on catching COVID, and the supposed mildness of Omicron might be explained by which part of the respiratory system it attacks, but that might not have much bearing on how it affects the nervous system and brain in particular (if it does). Brain fog and chronic fatigue seem more like brain issues than respiratory issues, although maybe poor breathing could cause both?

There are studies for and against COVID infecting the brain (I haven't taken the time to evaluate them), and in particular olfactory regions, which could explain long term loss of smell. If it does have a route to the brain, that's very worrying and would likely rule out a purely psychosomatic explanation (although both could be involved), although I agree that the severity of some cases, the specificity of some symptoms, and long COVID progression (IIRC, it can have a fairly delayed onset) make a purely psychosomatic explanation unlikely already.

EDIT: Here are a few studies, but I wouldn't consider this a particularly representative sample, and again, I haven't looked through them in much detail.

For brain route:

1. Song et al., January 2021, Neuroinvasion of SARS-CoV-2 in human and mouse brain
2. Douaud et al., August 2021, Brain imaging before and after COVID-19 in UK Biobank (not yet peer-reviewed)

Against brain route:

1. Khan et al., November, 2021, Visualizing in deceased COVID-19 patients how SARS-CoV-2 attacks the respiratory and olfactory mucosae but spares the olfactory bulb

Thanks for such a thorough post. I'm basically trying to figure out if all this evidence holds when you have received a booster and what is the evidence for that. One question: for all the studies that you shared showing long covid negative impacts, what is the percentage of those studies that focus on getting COVID-19 (a) without vaccines (b) with "initial" vaccines (c) with a booster? Eg maybe 20% of the studies are about assessing the impact of COVID when you haven't received any vaccines, while 30% is about assessing the impact of COVID when you got the "initial" vaccines, and 50% all combined or something like that. I'd be curious about your rough assessment given that I'm sure this may take too long to do very much in detail.

One major concern I have with the tail of long COVID is its severity even if/when it does get better after a few years. If I have debilitating long COVID for 3 years and then I recover fully, how will my career be affected, and what kind of person will I be after that? I think it's reasonably likely that it would cause value drift away from effective altruism (in part because being connected to EA while feeling hopeless about my own future productivity seems psychologically painful, and if I become primarily preoccupied with my own recovery and wellbeing), so under my current values, a 3-year case of debilitating long COVID could cost me about half of my expected future altruistic value, which would be like a ~50% chance of my own death. My inside view, just considering my own psychology and ignoring external evidence of value drift, is that I would very likely get back to EA.

Getting COVID once seems like an around 0.2%-2% chance of something at least this bad for me (based primarily on Matt Bell's analysis for the high end, with downward adjustments for my good health, Omicron being less severe, and concerns about study bias for severity for the lower end). This is bad enough to me to worry about, but also not so bad that I'm not willing to make tradeoffs for exercise, mental health and productivity generally, based on gathering size, vaccine effectiveness and local case counts, etc.. It's mild enough that new information seems reasonably likely to update me towards no longer worrying about long COVID, and most of the considerations I've come across not accounted for in Matt Bell's analysis seem to point towards less severe. However, it's also not clear to me that each time you get COVID is less bad in expectation than earlier times, since immunity wanes, and you will be older and potentially accumulate more harms from COVID, increasing its severity when it does reinfect you.

For those working on short timeline-focused AI safety, long COVID could cost you most of your chance to contribute.

I think I would very likely get back to EA with only about 1 year of long COVID or less. I also think there's substantial probability that I'd get back to EA no matter how long before I'm mostly recovered, as long as I do recover, at least 20%.

As an aside, naively, psychedelics seem promising as a treatment for brain fog and chronic fatigue from long COVID (whether or not it's psychosomatic).

There's a new preprint with what seem to be good controls, although based on self-report and the participation rate was low, about 1/3:

Post-acute symptoms, new onset diagnoses and health problems 6 to 12 months after SARS-CoV-2 infection: a nationwide questionnaire study in the adult Danish population

Anna Irene Vedel Sørensen, Lampros Spiliopoulos, Peter Bager, Nete Munk Nielsen, Jørgen Vinsløv Hansen, Anders Koch, Inger Kristine Meder, Steen Ethelberg, Anders Hviid

medRxiv 2022.02.27.22271328; doi: https://doi.org/10.1101/2022.02.27.22271328

Methods We conducted a nationwide cross-sectional study including 152 880 individuals aged 15-years or older, consisting of RT-PCR confirmed SARS-CoV-2 cases between September 2020-April 2021 (N=61 002) and a corresponding test-negative control group (N=91 878). Data were collected 6, 9 or 12 months after the test using web-based questionnaires. The questionnaire covered acute and post-acute symptoms, selected diagnoses, sick leave and general health, together with demographics and life style at baseline. Risk differences (RDs) between test-positives and -negatives were reported, adjusted for age, sex, single comorbidities, Charlson comorbidity score, obesity and healthcare-occupation.

I recommend checking out the figures for risk differences for new self-reported health problems and new self-reported diagnoses. From figure 2, the risk difference for a new self-reported diagnosis of chronic fatigue syndrome (pooled across groups) was 2.5%=4%-1.5%, the largest of the diagnoses and statistically significant. 1.15% for anxiety, 1% for depression, 0.16% for PTSD and a non-statistically significant difference for fibromyalgia. From figure 3, the risk differences for new health problems they tested were pretty big, 17.27% (sleep problems) to 40.45% (physical exhaustion).

Figure 4 has risk differences for various health conditions by age group and gender, but not necessarily for new onset since COVID.

They highlight self-report and participation rates as the main limitations:

The main limitations of the study are the self-reporting of symptoms and the participation rate. With little over 1/3 of the invitees choosing to participate, we cannot rule out participation bias. The motivation for participation could be higher among those experiencing post-acute symptoms, but on the other hand, those with very severe symptoms might not have had the energy to participate. Still, response rates among test-positives and –negatives were similar. However, because of the size of the study and the marked risk differences between the case- and control groups, we believe that our results are valid.

And caution about CFS diagnoses:

The overrepresentation of CFS among test-positives must be interpreted with care due to variability in how this diagnosis is made and the risk of confusing CFS with other conditions when filling in the questionnaire.

They also looked at sick leave rates, but unfortunately no durations:

Among the test-positives 12.0% reported taking any sick leave 4 weeks after test and until filling in the questionnaire 6-12 months later, compared to 7.7% of test-negatives (RD=4.32%, 95% CI 4.00-4.64%). Full-time sick leave was reported by 9.4% of test-positives and 6.5% of test-negatives (RD=3.20, 95% CI 2.88-3.47%), whereas part-time sick-leave was reported by 4.2% of test-positives compared to 1.7% of test-negatives (RD=2.43%, 95% CI 2.25-2.62%).

General concerns which push me towards being much less cautious re: Long Covid than you, without getting into the specific details of studies:

1. We have studied COVID far more than other diseases, so we may be finding stuff that either isn't real or more likely, occurs with other pathogens (we just hadn't bothered to look that closely before.)  
   1. Upshot: worry less about Long Covid
2. In general, a strong bias against preclinical/in-vitro work because I think that kind of lit is not very predictive of real-world issues. 
   1. Upshot: worry less about Long Covid
3. I'm more pessimistic than you about almost everyone getting Covid. Even with theoretical new vaccines for Omnicron or next-strain that approximate the original vaccine efficacy against the Wuhan strain, which was maybe 95% against severe outcomes and ~ 70-90% against infection in general, it seems difficult to achieve the herd immunity required for the more contagious strains. It is hard to find exact values but this article implies an R value of 4 or 5. That would require herd immunity threshold of 75-80%, which might be achieved with full next-gen vaccine uptake in the whole population, but that won't happen in the US because mandates are not going to happen on a large scale.  
   1. But you could probably achieve herd immunity in a conscientious network of people, so perhaps this doesn't matter to you if you mostly socialize with those types of people. 
   2. Upshot: avoiding COVID forever seems quite challenging
4. Nobody who I know personally who had a mild (defined as not hospitalized)  case of COVID (at least 30 people I know well)  has had any long-term (more than 3 months) issues, and I have asked in detail. One person I know was in the ICU and does have long-term issues but that's expected for any ICU case from a respiratory disease. One person with a moderate case had a cough for 2 months after.  From my limited interactions with LW/EA/rationalist types, I would guess my network is significantly less neurotic than what I'm guessing is your network and so they are less likely to experience psychosomatic symptoms . 
   1. Upshot: worry less about Long Covid from personal experience

To balance that critique out, something that makes me think long-term issues from Covid are not so fanciful: a decent number of pathogens can cause weird long-term issues: EBV, measles, polio, Encephalitis lethargica, H. Pylori-->cancer. So COVID causing issues is not unprecedented.

Still, the idea of Covid causing accelerated aging in mild cases seems unlikely to me. I can't justify this intuition in-depth but my understanding is that when pathogens do cause long-term issues, it is relatively specific to a tissue or organ type, so I'm skeptical of broadly dysregulated issues from a mild infection.

G. Overall deaths from everything have been very unusually high at points in 2021, even in 15-64 age group

Yes, all-cause mortality is up. We should look at all-cause mortality because it's what we care about, but when reasoning about causes, we should break it down by causes. Not all causes are up. The insurance executive says that young people are dying of random things months after covid. I'm happy to attribute unexplained deaths to the long-term effects of covid. But first you have to check that they don't have obvious explanations. The numbers I've seen is that it's mostly fentanyl. I'm pretty sure that's not covid. It might be lockdown, but I think it's just the continuing expansion of fentanyl (though that had flattened out for a couple years before the pandemic).

Some other previous back of the envelope calculations (collected here [LW · GW]):

By AdamGleave [LW(p) · GW(p)] (2 shots and for Delta):

My new estimate from the calculation is 3.0 to 11.7 quality-adjusted days lost to long-term sequelae, with my all-things-considered mean at 45. 

By Connor_Flexman [LW · GW]:

That being said, we can still roughly estimate risk from definitely having Delta. A healthy 30yo probably has about 4x (3x-10x) less risk than before, due to vaccination, despite Delta causing higher mortality. It almost entirely comes from Long COVID. In absolute terms this is ~4 expected days of life, plus 1/200th of your future productivity and vitality. You can shorthand this to about 1-4 weeks of life lost if you expect to otherwise live a full life—obviously, it costs less if you expect to live less less time. This translates to microCOVIDs at roughly 1 hour of your life lost every 1k-5k uCOVIDs. Risk of death goes up by 3x for every decade in age, but Long COVID probably only scales at ~~1.5x per decade, so for people over 60 mortality starts becoming more relevant. All these calculations are "creative" so please don't take them as definites.

Good post, but there is one important angle you missed. It is not only the question if you get Covid, but also how often you get Covid over your lifetime which may lead to cumulative damage.

According to a massive controlled study published in Nature (more readable summary here). It also looks like they are saying that this is for non-hospitalized covid patients, though the paper is confusing to me.

I am also a bit confused by what this study is trying to show. In Figure 3 it shows that a lot of things occur more frequently in patients who tested positive for COVID than in their matched controls, more in those hospitalized than that, and yet more in those admitted to the ICU. Per their methods section, this was measured by comparing the number of patients first reporting that condition within the 6 month period following a COVID-19 diagnosis vs the 6 months following (not sure) for everyone else in the VA system.

Looking at their data, they observe that people with a positive COVID-19 test are 1.17x, 2.34x, and 3.44x more likely for non-hospitalized, hospitalized non-ICU, and ICU-admitted patients respectively, to report obesity in the 6 month period than the COVID-19 negative controls. They cite this as an excess burden of obesity of 3.37, 26.15, 46.9 people per thousand among those three groups, which gives a baseline risk of about 1.95% for the control group reporting obesity within their 6 month period, if I'm doing my math right.

So taken at face value, the average person in the VA system has about a 2% per 6 months chance of becoming obese, while the average person admitted to the ICU has about a 7% chance of becoming obese in the next 6 months. I can come up with two possible explanations for that:

1. People who are admitted to the ICU gain about 3.37x more weight, on average, than people who do not get COVID-19, and as such are about 3.37x more likely to become obese.
2. Getting COVID-19 has very little effect on your weight, but within the group of people who are not obese, the people who are closer to the threshold of obesity are more likely to be admitted to the ICU for COVID-19.

I don't have hard evidence one way or the other, but I did some searching for evidence that COVID-19 causes weight gain, and while I found a large number of articles talking about how people are gaining weight under lockdown, I didn't find any discussing weight gain from COVID-19 directly. Additionally, my prior is that controlling for confounding factors is really really hard, and so even for a study published in Nature I would not be surprised if the authors failed to perfectly control for those confounding factors.

I'd be interested to see what you get if you

1. Look at weight changes in a group of people who tested positive vs negative for COVID-19 in a specific narrow period, to see if the COVID-positive people experienced more weight gain over that period, controlling for weight at the start of that period (and if so, whether that weight gain was correlated with disease severity).
2. Reproduce this study, except look the relative risk of experiencing one of these negative sequelae in the 6 month period before a positive COVID-19 diagnosis (if this method successfully controls for confounding factors, doing this should show a hazard ratio very close to 1 for all conditions, assuming that getting COVID can't cause you to experience negative health outcomes in the past).

I remember from early in the pandemic studies suggesting that there's measurable long-term heart inflammation in many cases. That seems to be one plausible mechanism for long-COVID that's less subjective. Does anyone read the literature to have a more recent idea of this?

This is very interesting. I find this somewhat persuasive and am updating in the direction of being more afraid of getting covid. That being said, I think I'm still pretty far from your level of concern.  I can think of three places where I think we may disagree.

1. I find your example in K very odd, in that it requires 100/900 people without covid to get illnesses that are as bad as long covid, but only 10/100 people who had covid to have similarly bad symptoms from long covid. It would be weird if we lived in a world where people without covid are so much more likely to get other illnesses that they're more likely to have symptoms that could be confused for long covid then people who actually had covid (if this was the case for some reason, that would also make covid way less scary), and I think your assumption breaks down if the scenario was set up differently.
2. Judging by the effect size of vaccines on hospitalizations and deaths, my priors would be that they would be more than 50% effective against long covid (or at least the really bad long covid that seems most worrisome). The study in C makes me update a little towards 50% but not that much.  One potential issue is that e.g. if 25% of the unvaccinated long covid were psychosomatic, and vaccination does not affect the probability of psychosomatic long covid, then an observed reduction of 50% from vaccination, would mean a 67% reduction of non-psychosomatic covid.  I would guess Omicron is also less scary, given that it seems like it doesn't infect the lungs nearly as much, but I'm less confident on that.
3. This is somewhat specific to me, but I suspect a lot of the people reading this also fit into the group of <40 yo, healthy, and not obese, so maybe it applies to others as well. Given how big the differences are between demographics for covid, I find anything that looks at the population-level to be potentially misleading for me.  One thing that makes me feel less worried is looking at the results for professional athletes.  There's an obvious selection bias, there, but I think it's better than having a non-obvious selection bias in basically any other study, and I suspect my risk factors are more similar to professional athletes than to the average American (given that risk scales superlinearly with age, weight, and medical history, it's probably the case that a median American is closer to a professional athlete than to the mean American, and I'm younger and probably healthier than median). The other nice thing about professional athletes is that their health is tracked very closely (though unfortunately, I don't see much work being done to compile and study it altogether).  I feel like this is similar to the "How many people do I actually know that have been affected by this?" approach.  From what I can tell, >50% of athletes in the four major US sports seems to have had covid already, and I can only find a few cases of long covid per sport.  That seems approximately consistent with long covid risk being between 0.1%-1%, which isn't far off from some of the estimates being quoted. Notably, though, I'm having trouble finding any cases that have come from after vaccinations were widely available.  There could be all sorts of confounders here, the most obvious being that no one would yet be talking about a long covid cases from an Omicron infection they got a few weeks ago, but the fact that these leagues are not being devastated by long covid, despite having professionals who are very dependent on their lung function and energy levels makes me feel reasonably good about my own prognosis.

That being said this report that you referenced suggests the odds ratios between demographics are not as extreme for long covid as they are for hospitalizations or deaths, which would mean maybe I'm not that much safer than an average person.  That could potentially also be explained by a fairly uniform distribution of psychosomatic cases, though.

The key point for me is the Value of Information of waiting some more. Many people will get Omicron soon and we will have more data.

A French study found that long covid is barely associated with having had covid according to an antibody test, yet associated with believing one has had covid (which itself is unrelated to the antibody test results).

Sorry, are you claiming that "belief in covid" is uncorrelated with antibody test results? I think:

• That's a wild claim (people are seriously no better than random at determining if they've had covid?).
• As far as I can tell, the study doesn't claim that.
• At a glance, the most important result of the study seems to be that antibodies are correlated with long covid as long as you don't control for belief-in-covid; but antibodies are not associated with long covid once you do control for belief-in-covid. That makes a lot of sense if antibodies and belief-in-covid is very correlated with each other, but is harder to explain otherwise.

Relatedly, your story for why the french study doesn't imply that covid is psychosomatic is way more extreme than it needs to be. Claiming that you have long covid is totally associated with having covid, as long as you don't control for belief-in-covid. See e.g. this post [LW · GW].

Edit: See also the comment I wrote here [LW(p) · GW(p)].

G. Overall deaths from everything have been very unusually high at points in 2021, even in 15-64 age group

This is wrong. Excess death increased significantly compared to the baseline only in the 45+ age groups. OWID mislead you by lumping in a bunch of age groups together.

https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm (select "Weekly Number of Deaths by Age")
https://www.euromomo.eu/graphs-and-maps

If everyone believed that covid was bad enough to get, it would die out in a month (possibly with terrible economic downsides for a month).

The same thing could be achieved without the economic downsides if everyone wore respirators [LW · GW].

In general, I agree with your view that one should try not to get Covid (Unless the individual costs for trying not to get it are very high, which would depend on the individual life circumstances. Since here in Germany many things are closed down anyway, for me the costs are smaller than the benefit of playing ist safe). Having said that, short remarks about two of your points:

G. Here it is interesting to look at different countries. For me the default in the tool was UK and Germany, and the differences in excess mortality for the age group 15-64 were huge. I am not sure whether that can only be explained by different Covid rates in different countries.

Q. On a short time-scale one important factor could be how long ago you were boostered (assuming you have got 3 vaccinations). With a booster very recently (maybe a couple of weeks) it could be better to be infected now than in 3 months. (I still think it would be even better not be infected now or in 3 months, but depending on the individual costs of not getting infected the timing of the booster could be a relevant factor for some people).

Are there other useful things that could be funded to get more evidence?

One thing that might be possible would be funding a larger survey, maybe with a more random sample of the population.

Is there any work going on that seems at all likely to decrease risk of getting long covid conditional on getting infected, beyond current vaccines? E.g. alternative vaccines with a story to suggest why they might decrease risk, evidence of Paxlovid or other treatments decreasing risk, other targeted research programs