The possible true size of the Covid-19 infection-fatality ratio: 0.31%

A modeling study published online in Nature on August 26th (I don't have the link on me; only read a report on it published by a trade source but it should be easy to look it up) has some interesting projections about the Covid-19 outbreak in the United States throughout 2020 (no data for 2021 yet, so it doesn't take Alpha and Delta into account).

First, a warning: it's an epidemiological model, not hard data. Still, it seems well-done so it is not impossible that its conclusions are fairly accurate.

Looking back at data from all 3,142 counties in the United States including rate of positive Covid-19 tests, rate of positive antibody tests indicating prior infections, observed case numbers, and rate of migration between counties, the authors projected the following:

We only diagnosed 25% of the cases in 2020. This means that we missed 3 for each diagnosed case, a number I find to be credible, and a number that I speculated about in my previous posting.

This holding true, we would have had 31% of the population infected with the ancestral variants of the SARS-CoV-2 (again, this doesn't take into account Alpha and Delta given that data collection only goes up to 2020 when these variants weren't here and vaccines were just starting distribution to healthcare workers).

While at the time the case-fatality ratio was locally between 1% and 2% (meaning how many firmly diagnosed cases ended up dying), the infection-fatality ratio (including the undiagnosed cases, which also includes asymptomatic cases) went down from 0.77% in April 2020 to 0.31% in December 2020 - which the authors attributed to better treatments and possibly the virus becoming less lethal (again, no Alpha and Delta in these numbers).

This is still higher than the estimated rate for seasonal influenza (0.08%) and for the 2009 influenza pandemic (0.0076%).

We always knew that at one point we'd be able to better ascertain the full extent of the infection cases (including undiagnosed asymptomatic cases) and we knew that the infection-fatality ratio would necessarily be smaller than the case-fatality ratio. It's at the tail-end of a pandemic that we can ascertain the infection-fatality ratio.

------------

Thinking about 2021: probably if the new variants are not more lethal, the infection-fatality ratio will fall more because now we have vaccines and we have monoclonal antibodies (both still under-utilized but they do bring down the fatalities). Probably overall we're missing fewer undiagnosed cases because data for 2020 included the first few months of the pandemic when we had insufficient testing. So maybe now we're no longer missing 3 for each diagnosed case. As of now we have performed 575 million tests.

If we are missing, say, 2 for each diagnosed case, our 40 million cases are more like 120 million cases, that would give us 0.54% of infection-fatality rate - a bit worse than for the ancestral variants despite the vaccines and mabs, so maybe Alpha and Delta are more lethal, but let's remember that I'm just speculating with this guess of 2 missed cases for each diagnosed case.

------------

Finally, a warning: if the infection-fatality ratio is not bigger than 0.31%, do remember that death is not all that needs to be feared regarding the SARS-CoV-2, because a large number of survivors come out of it with permanent organ damage to hearts, brains, lungs, pancreas, kidneys, and the coagulation system. If someone recovers from Covid-19 then dies of a stroke or pulmonary embolism a couple of months down the road (a study showed significant mortality in the first 5 months following discharge from a hospital from a Covid-19-related stay), or develops fatal myocarditis/heart failure years later, those deaths will typically not be counted as Covid-19 deaths so the numbers may be a bit worse than this epidemiological model is indicating.

------------

So, my friends, get those jabs, and mask-up when indicated. Stay safe.

 

And that's factoring in, all mitigation measures to date.

 

Yes, of course. Numbers would have been lower if we had done a more homogeneous and thorough epidemiological containment, and would have been higher if we had done no epidemiological containment whatsoever.

 

Thanks for the info. I do have to point out, however, that you say "a large number of survivors come out of it with permanent organ damage to hearts, brains, lungs, pancreas, kidneys, and the coagulation system".

That is misleading in the sense that if you use your very same criteria, then there are also a large number of people who come out of being vaccinated with death, and other major medical events. The fact is, a very, very large number of people recover just fine from Covid and a very, very large number of the vaccinated have either few or no problems with the vaccine.

This is my gripe with the left, if someone posted that there are a large number of adverse events or deaths due to the vaccinations you would be all over them, saying that overall, those risks are very very small and that there a NOT large numbers of people having adverse reactions or deaths. And yet you totally flip flop the other way around and misleadingly say that there are a large number of people having permanent long term damage from Covid when the truth is those people are a very small minority because the huge majority recover just fine from Covid.

 

The numbers cited in the opening post are believable, but my question is who's dying?

Are these people who are dying really old, like the type who are convalescent in hospital beds and probably only would have had a few years left to live anyway?

We all know the elderly who are close to death can be taken out by even the common cold.

I applaud the OP post for at least an honest attempt at examining the numbers and analysis.

 

There's been a shift to younger deaths after Delta hit. I don't have the specific numbers on me. The CDC did show the demographics of deaths at one point but months ago; I don't know if they have updated the information. So the report that younger people now are more numerous in ICUs is anecdotal. Just today there was a study suggesting that Delta is twice more likely to result in hospitalization so it does seem like Delta is more lethal on top of being more infectious. Do realize that this is an evolving situation and hard data often come with a significant delay when officials and researchers look back at the evidence.

While of course we've seen worldwide that the older and the more infirm, the more likely someone is of dying from Covid-19, there are plenty of deaths in much younger individuals. I just posted in another thread the sad news of the death of a 30-year-old who left behind a grieving wife who is pregnant of his fourth child.

The death rate for children is small; about 0.01%, but of course when they do happen it's devastating. 0.8% of children with Covid-19 need hospitalization. Just two days ago a Houston boy with no co-morbidity died of Covid-19. I don't remember the exact number of pediatric Covid-19 deaths in the United States but IIRC they are in the 4 hundreds. I remember a pediatrician saying that if a new disease killed 300 children we'd be doing everything we could to protect our children; he said that in the context of lamenting that the FDA and CDC have not yet approved Covid-19 vaccines for children under 12. Remember, when there is an outbreak of meningitis, it's a big deal and those outbreaks kill about as many so although the numbers remain relatively small, it's justified to take action against pediatric Covid-19. It is extremely devastating for these 400+ families, and you are right to say that when a debilitated 88-year-old dies of Covid-19, the loss of remaining life is small, but when the disease kills a 10-year-old, it's a devastating blow and it curtails decades of life expectancy.

 

Ohio State study: 30% of student athletes [testing positive for Covid] have heart damage linked to COVID-19
https://www.fox2detroit.com/news/oh...athletes-have-heart-damage-linked-to-covid-19

 

Exactly, this is the Ohio State University study I've been talking about. Thanks for the link, I had misplaced the link.

 

You know we’ve been fascinated by this myocardial issue for a long time. I think it’s great we are testing these kids before they are cleared to compete.

My only concern is with some of the criteria. Using late gadolinium enhancement (LGE) as a predictor of damage from Covid wouldn’t be useful unless we at least knew the underlying rates of scarring/damage in the sports population and in uninfected individuals in that population. It’s my understanding there can be scarring/damage as a result of frequent extreme exercise/exertion that shows up with LGE and the cause can’t be identified. It could even be the result of some other previous viral infection. It seems like such pre-existing “damage” would not be uncommon in athletes. Do you know what rates of LGE detectable damage would be in uninfected athletes? Just curiosity, not a big deal.

Also, there used to be concern about overuse of gadolinium causing toxicity. Would repeated imaging of athletes be a health concern?

 

I do not know what the underlying rates of scarring/damage to the heart are in otherwise healthy athletes who never had a SARS-CoV-2 infection. Too far from my specialty. But it's a good point. It would be interesting to know this baseline, for comparison-sake. If expressive, it might impact on the interpretation of this study. But I just don't imagine that it would be as prevalent as what was seen in this study post-Covid-19 infection.

Your other point is well-taken too. Gadolinium is not harmless. The most notable risk is nephrogenic systemic fibrosis. Manifestations of NSF are thickening of the skin, and fibrosis of the dermis and deeper structures including the muscle, fascia, lungs, and heart. It is however very rare, with some 200 cases described in the literature. The exact pathogenesis is not completely understood, but it is related to fibrocyte overproduction of collagen in response to gadolinium deposition in tissues. The thing is, it tends to only happen in patients with severe (stages 4 and 5) chronic kidney disease who have delayed clearance of gadolinium after administration (which is one of the reasons why many radiologists want a creatinine screening test before they proceed with the exam). However one would think, what happens to the healthy guys who are repeatedly exposed to gadolinium? Given that NSF can be fatal, it's a valid concern.

Not as serious as NSF are other side effects that are often transient and include injection site pain, nausea, itching, rash, headaches and dizziness. But some can linger. A 2016 study found headaches, bone and nerve pain, and skin thickening as the most commonly reported reactions in patients that were presumed to have gadolinium toxicity. In the study of 42 people with symptoms, brain fog and headaches lasted for more than three months in 29 people.

Like any other parenteral administration of a substance, there is also the risk of allergic reactions to gadolinium. They are also rare, to the point of being called a negligible risk.

 

Don't you agree that anyone doing these studies would consider these concerns?

In physics, that's why physicists do the science. We don't have to hope and pray we thought of the obvious. And we know how to do real science. We don't need advice from people on the internet who haven't even formerly studied the subject.

 

Thanks. I just remembered bits and pieces of imaging technology and heart tissue because I don’t deal with it every day. I appreciate you fleshing it out a bit.

I suspect we should be seeing a lot more studies concluding on myocardial damage and other sequelae in the next year. I’m especially interested to see what Covid does to the very young. We know things like development of type 1 diabetes and even obesity can be influenced/caused by viral infection. I hope we don’t find that is the case with asymptomatic cases in children. I’ve always hoped a couple asymptomatic natural infections and maybe a couple vaccinations in kids would set them up with a pretty good level of immunity for their adult life. But if we see stuff like type 1 diabetes etc. (just as an example) resulting from childhood infections it changes the dynamics.

 

If you were to actually read the research letter you would not have to ask such questions. The baseline of scarring from adaption is mentioned clearly as a limitation of the study. The letter claims the observed rate is higher than some previously observed baseline but does not provide any further information. Thus the question posed to the person here best able to answer it.

You could learn a lot by listening to people who have studied these subjects their entire educational and professional career. If you want people to value your knowledge/posts on physics, you should reciprocate by valuing the knowledge of those educated in and practicing biological sciences.

 

Yes, there's been cases of damage to the beta pancreatic cells from the SARS-CoV-2 triggering new onset diabetes.

 

Yes, of course. Here, from the study:

"Study limitations include lack of baseline CMR imaging and variable timing of CMR imaging from a positive COVID-19 test result. Athletic cardiac adaptation could be responsible for these abnormalities; however, in this cohort, mean (SD) T2 in those with suspected myocarditis was 59 (3) milliseconds vs 51 (2) milliseconds in those without, favoring pathology. Additionally, the rate of LGE (42%) is higher than in previously described normative populations. To conclude, while long-term follow-up and large studies including control populations are required to understand CMR changes in competitive athletes, CMR may provide an excellent risk-stratification assessment for myocarditis in athletes who have recovered from COVID-19 to guide safe competitive sports participation."

However, like @557 has mentioned, the authors did not disclose the rate of LGE for normative populations, which is why he asked, and I replied that I do not know what this rate is - but obviously the number is known given that it's been "described."

We are not "advising" the people who did the study. We are merely commenting on the findings.

Look, I started a thread about the EURO 2020, the European soccer championship that features some of the best soccer players on Earth and some of the most knowledgeable coaches. I do not possess their level of skills. However it doesn't stop me from commenting on the games, and even negatively criticizing some of the athletes and the coaches; and some of my criticism is actually valid and shared with some very knowledgeable analysts such as former players and coaches.

While I'm not a cardiologist or cardiac imaging radiologist, I'm still familiar with the science involved in Medicine, something I've done professionally for 41 years, to be able to issue comments after I read a study, regarding methodology and limitations. I did highlight, though, that I did not know the answer to 557's questions, but his points were valid.

I don't really understand why you decided to pick a fight with this.