Hacker News Reader

Comment by eximius | Hacker News Reader

This is mixed news. While this indicates the death rate may be lower and the infection rate higher among the general population, the demographic of individuals in the US military are going to be of significantly higher health than the general population.

So, it doesn't necessarily ease the risk for demographics already identified as 'at-risk'. If anything, their risk might be understated given the higher infection rate.

View on HN

Especially once you consider all of the pre-existing conditions that make you ineligible to be in the military.

this is a really good point. AFAIK, 100% of the canonical pre-existing conditions which are highly dangerous with COVID are disqualifiers for joining the military.

this cohort will give us a much better picture of how the virus impacts a totally healthy and young population. as with countless other pieces of evidence, it should also help to put the "it's just a flu bro" falsehood to bed.

Does the mortality rate across entire population (not particular subpopulations) exceed that of the flu?

Almost certainly. The US had 27k deaths in the past two weeks. That's more than some entire years of flu related deaths. Give it another few weeks and it will be more than any recent year.

Yes. Current understanding of the mortality rate is that it’s roughly 10x more lethal than the flu across all age demographics. This means you might only have a 0.2% of dying as a 20 year old, but you only had a 0.02% chance of dying of the flu.

The caveat is that mortality rate is intimately tied with the availability of certain types of care; if you need a vent and the hospital is full, you die even if otherwise you would’ve lived. Thus case mortality rate will change dramatically once the hospitals fill up.

I don't think that's true. covid has stronger age dependence. Just look at the ratio of covid vs. influenza deaths on the CDC's website (https://www.cdc.gov/nchs/nvss/vsrr/COVID19/index.htm). 3:1 covid:flu for over 85, but 1:1 for 25-35 and 1:10 for children. (relative ratios as we don't know total covid infections).

This is especially striking with children: if the US had the same number of covid infections as seasonal flu, you'd probably have lower pediatric deaths from covid than flu.

These statistics don’t support that conclusion, because these statistics are for total deaths. You can’t meaningfully compare the total deaths from a disease that’s done for the season (Flu) with one that’s still on going (COVID-19).

What you should be comparing is case fatality rates.

But can you not compare the ratios between age groups? That's all I'm doing here - showing deaths have a higher age correlation with covid than flu. - not making any statement about absolute CFR/IFR.

No, because you don’t know what the infection rate is by each demographic.

You have to compare just the case fatality rate. Comparing fatality ratios alone is too likely to measure unrelated factors.

It's more likely that cases/infections is skewed by age, because we test severe cases more than mild cases, than infections as a whole is skewed by age. Assuming infections are uniformly distributed, or rather have a similar distribution to the flu, seems entirely reasonable until we have good data suggesting otherwise.

There is no reason to believe that infections aren’t skewed by age. We took a wide range of preventative measures that are completely unique within living memory, such as closing down most schools and colleges. Until we’ve actually done statistical analysis on who got sick, any predictions based on historical trends are largely invalid.

Sure, but there is also no reason to believe that they are. There is however very good reason to believe that testing is skewed by age.

Fair enough, but would there be a reason it would be different? And even if it were, is that actually relevant?

Anyway, here's some IFR estimations for covid by Imperial college: https://www.thelancet.com/journals/laninf/article/PIIS1473-3...

You see significantly higher multipliers by age group than for flu. As one example, they are guessing teenagers have 1/17th the mortality rate of say a 40 year old. For flu (going by deaths), it's about 1/5th.

There is absolutely plenty of reason for the infection rate to be different by demographic: we closed schools down. We don’t do that for the flu.

We also don't Shelter in place for flu. Bay Area school closure date is 1 day before SIP.

Sweden and Iceland never closed schools for young kids. Data doesn't look any different.

That’s a good argument.

Sweden has a substantially higher case fatality rate for COVID-19 compared to other industrialized nations, so it's an interesting case.

"AFAIK, 100% of the canonical pre-existing conditions which are highly dangerous with COVID are disqualifiers for joining the military."

Including being "old".

Disqualified from signing up, but you can age surely?

I thought that smokers had elevated risks too, are smokers allowed in the military? As far as I can tell they are.

On average, people who start out older tend to end up older, after a given period of time. That's my theory, at least.

> Disqualified from signing up, but you can age surely?

To an extent. The US military caps military service at 30 years (20 for reserve officers), last I checked. So if you enlisted at 35 (right at the cutoff for the usual maximum age to enlist), you'd be forced to retire at 65, whereas if you enlisted at 17, you'd be forced to retire at 47. 65 (IIRC) is right on the edge of when age starts becoming a significant risk factor; for the vast majority of military personnel, it ain't gonna be a factor.

There are of course exceptions to this based on rank or specialization; for example, had my grandpa been promoted to General, he would've been able to serve longer, but he wasn't selected for the promotion so he aged out as a Colonel (or at least that's how he explained it to me).

Majority on the ship are going to be under 40. Most people start their military careers between 18 and 22, and most people leave at it before 20 years of service.

Squadron commanders, senior medical officer, Dept heads and the CO and Admiral would most likely be in their forties or early fifties.

I think we should also consider that some members of the military depending on their duties get exposed to chemicals that the civilian pop don’t get exposed to at the same levels.

Or at all. Most of us will never come within contact with TNT during our life, while most military members will be exposed to its residue constantly.

Good point, yes, but I bet that Gulf war soldiers where also extra-healthy when they enlisted. Not so much when they returned home.

Army, is an activity with lots of health risks, chemicals and strange diseases exposure attached. Not the best choice if you expect to live for one hundred years.

That reminded me of an Army recruiter well known for his “tactics”. People close describes him as a “Car salesman “. Apparently, is very good at making sure the candidate doesn’t change his/her mind.

There are a lot of people who have pre existing conditions and have no idea, even in the military. There was a marine recruit who died during basic training because of a heart defect, it wasn't an issue his whole life and it didn't come up as a problem during the in depth physical you get before hand.

Sure but that is irrelevant. It is absolutely certain that the rate of most diseases associated with increased Covid19 risks is much lower in the active army than in the general population, even if it is not 0.

Existence doesn't exclude filtering beyond filtering every case

The former only holds if you assume that a large majority of asymptomatic carriers never develop symptoms. It’s entirely likely that an asymptomatic positive test is a precursor to developing symptoms.

Sure, but remember that symptoms range from mild to moderate in the majority of cases.

That’s a different argument altogether.

It would tend to add weight to the growing evidence that public health officials in most countries have been pursuing the wrong strategy all along. We could save more lives in the short term and avoid overwhelming the healthcare system by focusing our limited resources on providing safe voluntary quarantine spaces for the elderly and others at highest risk.

> the demographic of individuals in the US military are going to be of significantly higher health than the general population.

29 out of 33 pregnant women who tested positive at Columbia University Medical in NYC were also asymptomatic:

https://www.medpagetoday.com/infectiousdisease/covid19/85965

With other viruses, pregnant women are typically considered at elevated risk. For example, they had a significantly higher fatality rate with the Spanish Flu. So it's still unclear who is most likely to be asymptomatic based on the data we have so far.

Immune systems during pregnancy are vastly different than normal. It's quite possible that is a factor in the asymptomatic rate here. Many studies like this are using populations with inherent selection bias: that makes them interesting, for that population, but you cannot generalize from those results.

I don't quite understand why "more infectious but less lethal" is better than "less infectious but more lethal" -- both mean that lots of people are going to die.

Even with the earlier estimates of how infectious it is, it seems quite likely that a large proportion of the population is likely to catch it sooner or later. If that is so, then the less lethal it is, the better.

But it would indicate the we may be closer to herd immunity than we thought, which would be great news.

Let’s assume 100% in NY are or have been infected. 12,000 deaths / 20 million = 0.0006% fatality.

Then scale that up to US population outside N.Y.: 18,0000 deaths / 0,0006 = 3,000,000 infected out of 300,000,000 for an infection rate of... 1%.

Herd immunity takes about 66% to 70%. So: only 65% to 69% to go, or about 200,000 deaths.

Lower the assumed infection rate in NY to anything realistic and it only gets worse. For (still high) 20% you’re looking at a million deaths to achieve herd immunity.

that would be 30 million, not 3 million

It does, the more young healthy people are immune, the less likely the virus will make it to them.

Being asymptomatic is not the same as being immune. It means that more people will feel fine and spread the virus more because they don't know they have it.

Well, for what it's worth, if they get the virus while on a long tour of duty, by the time they get back they should be good to contribute to the herd immunity.

Only if they form antibodies. Studies have been suggesting that people who get it in asymptomatic form are not all forming antibodies - leaving them wide open to be future communicators of the disease.

That sounds really interesting, could you please link to more info about this?

I'll check back in here later with links - didn't save it at the time I was reading through.

Wouldn't that mean that a vaccine is close to impossible?

My layman understanding is that a vaccine relies on the body creating antibodies.

It’s possible and desirable for a vaccine to stimulate antibody production where a “natural” infection would not, which is one of the things that will be tested during their development. An extreme example is infection with the tetanus-causing bacterium clostridium tetani, which doesn’t provide natural immunity at all.

A known failure mode of vaccines is if the person who is being vaccinated lacks the right B cells for creating antibodies matching the antigens on the surface of a pathogen. If the B cells a person has lack the right B Cell receptors, they can't generate the correct antibody response to an antigen, because they can't bind with/identify an antigen in the first place. Additionally, they may lack a sufficient number of antigen naive T cells, which haven't yet become differentiated as T memory cells and effector T cells.

But that doesn't mean that it is impossible to create a vaccine: this problem exists already for successful vaccines. For example, the Hepatitis B vaccine has a relatively high failure rate, but is still provides immunity 90+% of the time. It just means that some smallish percentage of the people you vaccinate aren't actually contributing to population immunity.

See below for more info:

https://en.wikipedia.org/wiki/B_cell

https://en.wikipedia.org/wiki/Memory_B_cell

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3760154/ --> this one is particularly good. A lot of good info in here on how vaccination and "memory" interact.

https://www.sciencedirect.com/science/article/pii/S0264410X0...

EDIT: this one covers the major routes of vaccination failure --> https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4962729/#!po=60...

There are no studies that have suggested anything like this. There simply has not been sufficient time to perform a study of the type needed to prove such a claim: you would need to gather a cohort of people that you know had the virus, wait for them to recover, then test for antibodies.

Moreover, even if the percentage of such people were shockingly high (i.e. over a fraction of a percentage), you would need to conduct a huge survey to distinguish it from error rates on the tests themselves.

Or maybe it was a false positive.

Forming antibodies (antibody driven adaptive immunity) seems like the most likely way patients are recovering. This would be typical of recovery from Coronavirus infection.

https://www.nature.com/articles/s41591-020-0819-2

I can’t find any research suggesting that patients (symptomatic of not) are recovering in some other way.

And in any case, recovery implies some form of immunity, either innate or adaptive. The virus doesn’t just give up after 14 days.

To everyone suddenly thinking about antibodies vs T-cells vs whatever else, the podcast "This Week In Virology" discusses some of the difficulties at https://www.microbe.tv/twiv/twiv-602/ (discussion with Stanley Perlman). A number of other episodes also hit on coronavirus. Listen while you're walking the dog.

Another fun seminar is this one by Prof Pamela Bjorkman at Caltech; it's useful to watch the slides with at least one eyeball so not for dog-walking. https://www.youtube.com/watch?v=OBcc_dk9Q9U

It's not clear to me that asymtomatic doesn't mean immune. If the primary symptoms are due to over-replication of the virus and not fighting it enough resulting in cell death in lungs and other airways, then asymptomatic infection indicates the immune system is fighting it enough to fend it off.

Now, this rests on symptoms being directly caused by higher virus cell counts rather than toxic byproduct cascades of some sort. I'm not sure that's true

This is devolving into semantics. Being a asymptomatic means you are infected, which means you are not immune. You can still develop symptoms later and/or still transmit the disease.

You would be considered immune if your body produced antibodies that prevented infection. This immunity can be gained by successfully fighting the infection or by receiving a vaccination that triggers an immune response.

Reply on news.ycombinator.com