Comparative risk from Covid-19 by age
I ran across what I thought was a pretty helpful way to look at mortality risk related to Covid 19 in Bloomberg (while the article is paywalled, they allow a certain number of free articles).
As many have debated endlessly, figuring out fatality rates is challenging real time because of uncertainty about number of cases, lags in death relative to reporting of cases. While not perfect, focusing more on gross mortality rates (e.g. deaths per capita rather than deaths per case) means we only have to look at one number. The author did this by age category, then adjusted deaths for several different estimates for deaths by the end of the year (100,000; 200,000 and 750,000). The author also looked at how a death total of 200,000 would impact, mortality rates by age group.
It very starkly shows how little this impacts the youngest. Even at 750,000 deaths for those under 14, your chance of dying from Covid-19 are substantially less than your chances from dying from the flu.
What may surprise is the risks for those in the middle. At 45 to 54 we are rapidly reaching the point where more people have died from Covid-19 than died from car accidents. For those over 55, we’ve already surpassed it.
Another way to look at this is to compare what the theoretical increase in mortality rate would be if Covid deaths were additive to all deaths (while I can imagine people quibbling with the concept, if you focus on it merely as a way to compare across age groups, I think it is pretty illuminating. For those under 24, the number is well below 1 for 100,000 deaths (the author calculated for 200,000, I halved it). For 45 to 54 it is 3.2%. Not much below the 3.5% for those 85 and older.
in other words, while your overall odds of dying if you are 45 to 54 of any cause is way lower than your odds of dying from any cause if you are 85 or older, if you were to die in 2020, chances that it would be from Covid-19 are actually not much lower than the chances it would be for someone much older.
Published yesterday was a state by state graph that showed that an average of 40-50% of all deaths in many heavily hit states have occurred in nursing homes. Some states are in the 60% range, others (mostly lightly hit states so far) much less
That’s been the case across Europe and Canada as well. Enclosed indoor spaces with large numbers of vulnerable people living in close proximity are trouble.
Johnny Knoxville and Blink 182 are slightly more at risk. I am sad.?
The age that this virus inflicts doesn't matter. I have heard we will all be dead soon from Global Warming anyway.
Just like Covid-19, better, reasonable cost effective preparations today would pay huge dividends to reduce costs as the impacts of climate change continue to worsen,
Pre-existing conditions are omitted in this analysis (at least in your write-up).
All of the data is available on the New York state website:
Assuming New York state is fairly representative of the country. 90% of deaths have at least one pre-existing condition. 85% of deaths have been over age 60. A very detailed plan to protect these groups needs to be developed and implemented. For others the risk seems low. If you assume the pre-existing conditions are evenly distributed per age group, this would say of the 3,154 people who have died under age 60, there have been 315 without pre-existing conditions. I haven't been able to find data to compare, but I have to believe that's near or lower than flu levels. I'd appreciate any better data that anyone has on that though.
That's a nice link, thanks.
According to the CDC, nearly half of American adults have hypertension, so that's not great.
In fact, maybe that's why kids don't have much trouble with the corona--they hardly have any of these pre-existing conditions as a whole. Whereas the adult population---lots of hypertension, like 10% diabetes, etc.
Could be. It's frustrating to me - and I recognize it's easy to say this as a just a guy on the internet - that the medical community hasn't nailed down a likely reason for this yet. It's been known for going on 4 months now. I would assume by now with the data they have on this virus, how it's acting in the body in different age groups, and past viral infections, they would have a better answer.
I'm with the cytokine storm answer above. It seems to be the one thread that runs through all scenarios - elderly (much lower vitamin D than in the general poplulation), northern hemisphere have a worse effect (less vitamin D right now), darker skinned groups having worse outcomes (less vitamin D overall). And children may have immune systems that don't have this "over-response" yet. What do I I know, but this is where I've ended up.
I can't get completely on board the cytokine storm. Typically that occurs in people with a very healthy immune system, i.e. younger people.
Note many believe the Spanish flu was caused by a cytokine storm, and that hit people 20-40 especially hard.
If I had to guess I'd go with initial viral load and one of those pre-existing conditions as a factor in determining severity. I believe that's similar to flu.
As far as nailing down a reason....if it was one thing they would have figured it out by now. But if you get to two or three, the controls really can't be done in the real world. Too much you don't know about each individual, and too much randomness, for lack of a better word.
That's helpful to my understanding on both points, thank you.
I just feel like the data is there. If you fed it into supercomputer it would have told us by now. Too many people in the medical field - at least from my uneducated-in-medicine point of view - are like the old baseball scouts in Moneyball. They seem to make assumptions on feel and dismiss potential causality because it doesn't fit their personal experience.
One example from a neophyte perspective: cause of death. I'm astonished to find that it's very inaccurate even before the reality of COVD. I understand there are many difficulties here that I can't appreciate, but I've seen studies that say anywhere from 30-50% of cause of death may be incorrect. Now we have a system and policy where people with many conditions contract COVD, they die, and cause of death is COVD. I get that they may not be able to understand what exactly caused the death, but it seems to me they better to get to figuring that out quickly. I don't know how you can have a viable treatment when you don't know what the problem is (i.e. is it COVD, or COVD + hypertension, or 10% COVD and 90% hypertension and the COVD causes something that pushes the body over the edge in the elderly, etc).
The cause of death is basically like a typical virus as far as I know. It attacks the lungs and then all sort of shit can happen, depending on the person. Obviously pneumonia. Inflammation due to a health viral response makes the heart work harder, less oxygen in blood, etc.
A supercomputer is only as good as the inputted data, and that's the issue. We already have trends such as pre-existing conditions and age. I wonder about viral load, for example, but there's people with a ton of virus in them that are asymptomatic. So you just need a real large, broad population, for each thing you're looking at, and you're only getting detailed data from some hospital patients.
Also, the medical front-liners that treat the patients aren't the ones trying to figure this out on a broad scale, that's more your scientist/cdc types.
To your last point, of course, it would be silly to think nurses are taking your blood pressure and then reviewing variables in the data. I'm talking about CDC and related.
There are challenges of course, it just doesn't feel at all precise.
I can assure you they don't go by feel.
Actually, many of us "medical front-liners" treating patients are trying to figure this out. Critical care physicians (intensivists and emergency physicians) are compiling data and brain storming. There are discussion groups developing different hypotheses regarding the disease process and possible treatments.
If the rona causes me to have pneumonia, and then the pneumonia kills me, rona was the cause of my death. Same with the flu, since absent the virus that caused the pneumonia I would still be alive.
But one of the very obvious problems with that is inconsistency, and in just one example in relation to the seasonal flu. Has what you described been the standard for years with the flu? Very doubtful. I assume pneumonia was given as the cause of death quite often. Now are primary comparator is misstated vs covd, which has an enormous impact given we shut down for the latter and not the former.
I'm not arguing that we shouldn't have shut down, I believe it was a prudent decision and actually implemented incredibly well across the country (credit to the leaders of partisans on both sides, though there were many mistakes). Just that the inaccuracy on cause of death data is and always has been staggering.
This dr suggests that the flu is misstated in the opposite direction you think. Also, I’m pretty sure that the rationale for the stay at homes was not a comparison to the flu, but rather modeling showing hospitals being overwhelmed.
“The 25,000 to 69,000 numbers that Trump cited do not represent counted flu deaths per year; they are estimates that the CDC produces by multiplying the number of flu death counts reported by various coefficients produced through complicated algorithms. These coefficients are based on assumptions of how many cases, hospitalizations, and deaths they believe went unreported. In the last six flu seasons, the CDC’s reported number of actual confirmed flu deaths—that is, counting flu deaths the way we are currently counting deaths from the coronavirus—has ranged from 3,448 to 15,620, which far lower than the numbers commonly repeated by public officials and even public health experts.
There is some logic behind the CDC’s methods. There are, of course, some flu deaths that are missed, because not everyone who contracts the flu gets a flu test. But there are little data to support the CDC’s assumption that the number of people who die of flu each year is on average six times greater than the number of flu deaths that are actually confirmed. In fact, in the fine print, the CDC’s flu numbers also include pneumonia deaths.”
Median age of death from C19 is between 75 and 80 depending on where you look. What’s so difficult to understand that old people with no immune systems are going to die if they catch almost anything.
50% of deaths are in nursing homes! If we didn’t test for this i wonder if we would have even noticed it passing through the system?
yes, we would notice. seasonal flu and cold season goes thru nursing homes every year, as employees and visitors bring it in. every 75-80 year old does not just die b.c they got a cold (which is what it sounded like you implied). and even if that were somehow true, during the peak in southeast michigan u of m filled their ICUs with patients on ventilators who were and are on them for weeks and were pulling nurses from other areas of the hospital to help staff. this is very noticeable - hospitals don't do that just for kicks. it isn't all about the death rate - people being really sick on ventilators for weeks because of a novel illness (so it's in addition to what our expected capacity is, i believe u of m runs at 95% capacity normally) is a big strain on our health care system. i don't want an overstretched, burned icu team taking care of me that is using general care nurses b.c they don't have the icu nurses. and i say that as a general care nurse. we can convert hospital beds into ICU beds, we can build field hospitals, we can ramp up ventilator production. we can't create ICU nurses or doctors or respiratory therapists or lab techs or radiologists or x-ray techs or the cleaning team who now has to do at least double the work due to new protocols or all the other people who keep a hospital running with their specialized knowledge and experience.
Too much logic for the ignorant.
Yeah, I'm sure if it wasn't for Covid, nobody would notice the nursing home deaths, because nobody cares about them, right?
Yeah, I'm sure you guys were all super concerned last year when 50% of our 3,000,000 normal deaths were elderly folks. Where you guys demanding a shut down for them?
I appreciate your edit. I also agree that further analysis adding in pre-existing conditions would be useful. As others have noted, many people under 65 have risk factors (being one of those, that is part of the reason I’ve spent time understanding this better).
538 did an analysis suggesting
“Overall, nearly 40 percent of all Americans under 65 have a condition that makes COVID-19 more dangerous for them.”
That's a great link too. Tho it looks like they should have said 40% of all Americans between ages 18-64.
This is obviously a factor. It still seems like age is a bigger factor tho. Leads me to think a strong immune response is helpful, especially if you have are at risk.
Harbaugh approves.
Of not winning championships? Yeah, you are right about that.
Michigan was the B1G East co-champion in 2018.
And they've got the trophy to prove it!
Yeah... That really has served them well.
I would not go around toting that as an accomplishment.
I didn't. I used it to easily refute your asinine comment.
All I know is that I’m staying the fuck off Facebook for the foreseeable future. Apparently Bill Gates is the new George Soros.
This is so weird. I've been a Mac user since 1986, but the dude deserves a Nobel prize for what he's doing.
I would really love to see a few of these multi-billionaires get together, donate a billion each and really kickstart the testing infrastructure. Gates, Cuban, and Bezos could do wonders for their reputations. They could go down in history as the Carnegies and Getty’s of our era. Likewise, if Bloomberg and Steyer really want to improve matters, donate massive funds to NGOs supporting capacity improvements in the testing infrastructure.
Gates is (and has been for quite a while) putting his money where his mouth is. The William and Melinda Gates Foundation has been funding improving health, ending poverty, increasing education, and many other initiatives in some of the world's poorest countries. He's probably done more with his money in that regard than many countries. It's pretty impressive.
So, when he talks about immunization, pandemics, and so on, I consider him someone who is better informed than the average Joe Bagadonuts. While the models he's funded at IMHE have been high, low, and everywhere in between, at least he's not sitting on his ass while the world is on fire.
IHME
That's funny and true.
Ironically, on other places in the internet, he's a capitalist pig.
We had a conversation about speed limits and I think it’s an interesting example of society compromising needs with risk. We accept a certain level of risk by setting the speed limit at 50-65. Deaths are directly proportional to speed. If we accepted no deaths, the speed limit would be 10 everywhere but it’s a huge trade-off for productivity.
There’s a parallel to COVID that is difficult to broach. At some point, society will need to define what an acceptable level of risk is to re-open.
I’m in a utilitarian mindset with this. I would like the most good for the most people and don’t know what that translates into policy-wise. So I stay relatively silent and consider experts’ proposal as they are presented. I do have a default stay-at-home and mandatory masks opinion because it provides the best outcomes medically for those at high-risk. I do wonder if we are going to cause more harm with this though. Economists probably are the best people to figure it out - if they have reliable medical data to use.
I am an emergency physician and I agree that we need to weigh the costs/benefits with any approach. We need input from economist and business people not just health care experts. And we need to listen to the general populace. As you point out, we accept a certain amount of risk in daily life in exchange for freedom and a productive life. We can't stay shut down forever and will need to make some hard choices, hopefully with open mindedness and less polarization.
Some sanity to both you and CFraser on these COVID topics.
I like the vitamin D aspect. I am also a strong believer that the virus doesn't transmit as well under more sunlight when people are outdoors. Take Florida as a good example, a state with a huge Senior age population. I heard that in the Villages; a huge 65 and over city, as big as a county north of Orlando has had no COVID deaths thus far. So I ask, what's up with that? My explanation; the old folks are outside play a lot of shuffleboard.
This could, meaning one or more reasons, explain why states like FL, TX, AZ & CA have had a lower number of cases per 100,000.
I'm with you on the "let's all agree to wear masks in places like stores, shops, restaurants, and houses of worships." That would be a reasonable compromise to get the economy moving while balancing risk.
Sadly, there are no shortage of people who want to believe that this is all some control conspiracy by the government/Illuminati/Bill Gates/Pharma companies. So they flat out refuse to wear one.
Maybe I can take them down to Florida and point out the sign that says "No swimming. Alligators," and find out if they are still willing to "Stick it to the man!"
Generally agree with your points, but wearing a mask in a restaurant just doesn’t work for me.
Doesn’t for me either. I think there will be quite a few restaurants (Florida is doing this) where sitting in the dining room is strongly discouraged, but outdoor cafes are extremely popular.
I think that's a small minority of the country, but you like to focus on them because your hatred of them is something very important to you.
What about the many, many people recovering (aka not dying) with longer term health issues from C-19? Those not count in the data analysis?
Little kids have their own risks from COVID. This happened to the daughter of one of our friends (she’s 2).
Doctors think she will be OK, but this is so new it’s hard to know. And the cardiac issues might persist for the rest of her life.
Correct. But trying to explain to millennials and younger who want their ‘beach time’ is useless.
At the end of that linked article it says a link to Covid is suspected but not yet proven. And even more to the point, there's this:
[Dr. Audrey] "John said: In general, families do not need to worry about this. I doubt that this is really new. I think it's just really newly recognized.John said: In general, families do not need to worry about this. I doubt that this is really new. I think it's just really newly recognized."
Interesting data. It appears that one's risk of death is still significantly greater for causes other than Covid-19 for all age ranges. It does provide some perspective. Nevertheless, I'm still fearful of Covid-19 no matter the numbers. It still increases risk significantly for anybody who is older.
But most older people probably already have some preexisting condition. If you are a healthy senior, your odds may be pretty damn good for survival.