Comparing flu deaths to Covid deaths

[martyp]

This is an interesting look at the often stated figure that flu deaths are more than Corona deaths

https://blogs.scientificamerican.co...d4VGN3R5ObpmBWexCHzK6dgIWyaIIewF-bCSyde1cKqzQ

An excerpt

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. . . .
The CDC should immediately change how it reports flu deaths. While in the past it was justifiable to err on the side of substantially overestimating flu deaths, in order to encourage vaccination and good hygiene, at this point the CDC’s reporting about flu deaths is dangerously misleading the public and even public officials about the comparison between these two viruses. If we incorrectly conclude that COVID-19 is “just another flu,” we may retreat from strategies that appear to be working in minimizing the speed of spread of the virus.
The question remains. Can we accurately compare the toll of the flu to the toll of the coronavirus pandemic?
To do this, we have to compare counted deaths to counted deaths, not counted deaths to wildly inflated statistical estimates. If we compare, for instance, the number of people who died in the United States from COVID-19 in the second full week of April to the number of people who died from influenza during the worst week of the past seven flu seasons (as reported to the CDC), we find that the novel coronavirus killed between 9.5 and 44 times more people than seasonal flu. In other words, the coronavirus is not anything like the flu: It is much, much worse.

 

[MichaelB]

Blog author’s bio https://www.health.harvard.edu/blog/author/jfaust

Jeremy Samuel Faust, MD MS MA, is an emergency physician at Brigham and Women’s Hospital and an instructor at Harvard Medical School. He completed his residency training at Mount Sinai Hospital and Elmhurst Hospital Center in New York, after graduating from the Mount Sinai School of Medicine (MD, MS in Biomedical Sciences).

He is the cohost of FOAMcast, an award-winning free medical education podcast bridging core emergency medicine content with cutting-edge topics, which has been downloaded nearly 2 million times. He is a frequent contributor to Slate, has published essays in the New York Times, The New York Daily News, and Mother Jones and has appeared on the Rachel Maddow Show (MSNBC) and The Takeaway (Public Radio International/WNYC/WGBH). He serves on the editorial board of the Annals of Emergency Medicine and ACEP Now, and has served as a content editor for the New England Journal of Medicine Resident 360 Series.

 

[donheff]

Thanks. This article does put the numbers in perspective.

 

[devans0]

This is an interesting look at the often stated figure that flu deaths are more than Corona deaths

https://blogs.scientificamerican.co...d4VGN3R5ObpmBWexCHzK6dgIWyaIIewF-bCSyde1cKqzQ

An excerpt

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. . . .
The CDC should immediately change how it reports flu deaths. While in the past it was justifiable to err on the side of substantially overestimating flu deaths, in order to encourage vaccination and good hygiene, at this point the CDC’s reporting about flu deaths is dangerously misleading the public and even public officials about the comparison between these two viruses. If we incorrectly conclude that COVID-19 is “just another flu,” we may retreat from strategies that appear to be working in minimizing the speed of spread of the virus.
The question remains. Can we accurately compare the toll of the flu to the toll of the coronavirus pandemic?
To do this, we have to compare counted deaths to counted deaths, not counted deaths to wildly inflated statistical estimates. If we compare, for instance, the number of people who died in the United States from COVID-19 in the second full week of April to the number of people who died from influenza during the worst week of the past seven flu seasons (as reported to the CDC), we find that the novel coronavirus killed between 9.5 and 44 times more people than seasonal flu. In other words, the coronavirus is not anything like the flu: It is much, much worse.

The CDC has fundamentally changed under the Trump administration. I do not feel comfortable relying on complicated algorithms that have political spins and in this case, am not confident that the algorithms are not massaged.

There was no CV-18, so l am using previous years as a control and assuming that the increases cover all the effects of CV-19, such as secondary infections from CV-19 as well as other deaths that would not happen except for the prioritizing acute CV-19 impeding health care for other causes.

Our family has had other health care items cancelled as not being necessary. While this won't hurt us in all likelyhood, it still is a part of the total risks that matter.

I look at state and county demographics rather than country as a whole. YMMV.

 

[audreyh1]

Thanks for explaining the numbers. I had no idea that the comparative flu numbers given were estimates and not counted deaths.

 

[CardsFan]

The CDC has fundamentally changed under the Trump administration. I do not feel comfortable relying on complicated algorithms that have political spins and in this case, am not confident that the algorithms are not massaged.

For the record, the CDC has been estimating the flu deaths with algorithms for YEARS. This is nothing new, and certainly not something changed recently for political reasons.

The article fairly questions whether the CDC has been overstating flu deaths, and argues that maybe they should stop. But it is not new.

 

[MichaelB]

It’d be nice if we could keep politics out of the discussion.

 

[eytonxav]

Here is some info from CDC on Covid-19 death counts. There is a lag in data and it can be adjusted up or down from numbers we see reported through media sources:
https://www.cdc.gov/nchs/nvss/vsrr/covid19/index.htm

 

[Montecfo]

It is a little silly I think to complain that flu deaths are estimates when we know covid deaths are purposely overcounted since used as a default cause of death.

Comparing the best figures available seems reasonable, while noting weaknesses.

The CDC knows all about the lower figures the author cites, yet it does not cite them. The author does not explore this fact, but merely dismisses the widely cited CDC estimates

And of course, the author is not a medical researcher or statistician.

 

[audreyh1]

It is a little silly I think to complain that flu deaths are estimates when we know covid deaths are purposely overcounted since used as a default cause of death.

Comparing the best figures available seems reasonable, while noting weaknesses.

I think you could as easily argue that they are undercounted especially with the limited testing and sick people being denied tests.

 

[martyp]

It is a little silly I think to complain that flu deaths are estimates when we know covid deaths are purposely overcounted since used as a default cause of death.

Comparing the best figures available seems reasonable, while noting weaknesses.

The CDC knows all about the lower figures the author cites, yet it does not cite them. The author does not explore this fact, but merely dismisses the widely cited CDC estimates

And of course, the author is not a medical researcher or statistician.

If someones death was hastened by Covid then I think it is fair to count it as Covid. We are not talking about people who were going to die next month. If covid hasten their death bu 1 or 2 or 3 years then i would call that a Covid death. Statistically you can look at death rates going back many years and see the additional deaths for the last three months.

 

[Montecfo]

I think you could as easily argue that they are undercounted especially with the limited testing and sick people being denied tests.

You can argue it, sure. But confirmation of a death being from Covid19 does not require a positive test. Dr. Birx recently explained this and using Covid19 as default or presumed cause of death.

Also recall the reclassification of 4000 NY deaths as COVID a couple of weeks ago.

 

[Ian S]

It is a little silly I think to complain that flu deaths are estimates when we know covid deaths are purposely overcounted since used as a default cause of death

Do you have evidence for that claim? IIRC, for someone to have been reported as dying from COVID-19, they either have to tested positive or exhibit some of the identifiable signs of COVID-19 infection such as the telltale "ground glass" opacity found with a CT scan. Even then, at least in New York, there is a huge discrepancy between the prior years' average deaths + COVID-19 deaths and this years observed deaths.

 

[Aerides]

I think "purposeful over-counting" is an extremely pessimistic and skeptical view of it.

I look at it more like: Trying to get it right because come on we know people are dying that haven't gotten tested but common sense says they most likely had it.

 

[Montecfo]

Do you have evidence for that claim? IIRC, for someone to have been reported as dying from COVID-19, they either have to tested positive or exhibit some of the identifiable signs of COVID-19 infection such as the telltale "ground glass" opacity found with a CT scan. Even then, at least in New York, there is a huge discrepancy between the prior years' average deaths + COVID-19 deaths and this years observed deaths.

Dr. Birx statement on April 8. "We have taken a liberal approach to corona virus deaths".

"If someone dies with Covid19 we are counting that as a Covid 19 death"

https://youtu.be/zbFt6eO3wz0

CDC guidelines for reporting deaths does not require positive test.
https://www.cdc.gov/nchs/nvss/vsrr/covid19/tech_notes.htm

NY reclassifying almost 4000 deaths as COVID.
https://www.dailymail.co.uk/news/article-8219335/NYC-announces-3-700-probable-new-coronavirus-deaths.html

 

[statsman]

Dr. Birx statement on April 8. "We have taken a liberal approach to corona virus deaths".

"If someone dies with Covid19 we are counting that as a Covid 19 death"

https://youtu.be/zbFt6eO3wz0

CDC guidelines for reporting deaths does not require positive test.
https://www.cdc.gov/nchs/nvss/vsrr/covid19/tech_notes.htm

NY reclassifying almost 4000 deaths as COVID.
https://www.dailymail.co.uk/news/ar...es-3-700-probable-new-coronavirus-deaths.html

Under such guidelines, I am inclined to believe we've under-counted flu deaths in the years prior to this one.

 

[Montecfo]

I think "purposeful over-counting" is an extremely pessimistic and skeptical view of it.

I look at it more like: Trying to get it right because come on we know people are dying that haven't gotten tested but common sense says they most likely had it.

How is it "pessimistic" to suggest CV cases are being overcounted?
Let's look at some underlying facts, they are interesting.

The government is clearly making Covid19 the default cause of death, per Dr. Birx. (Link above but widely reported).

And heart attack and stroke cases are down, according to the NY Times.

https://www.nytimes.com/2020/04/06/well/live/coronavirus-doctors-hospitals-emergency-care-heart-attack-stroke.html

 

[Aerides]

How is it "pessimistic" to suggest CV cases are being overcounted?

It's a pessimistic view of federal health officials, government, etc. Because, why would they want to add to to the pile? if there were 50k or 75k deaths would we not be in the exact same situation? I don't understand what the motivation would be to make things up or inflate them. There's no upside.

eta: if numbers were being fudged, wouldn't it be more likely that the federal government would prefer to have a lower count than higher? To open the economy sooner?

 

[Montecfo]

It's a pessimistic view of federal health officials, government, etc. Because, why would they want to add to to the pile? if there were 50k or 75k deaths would we not be in the exact same situation? I don't understand what the motivation would be to make things up or inflate them. There's no upside.

Ask Dr. Birx. They are clearly assuming COVID as the cause of death if someone has COVID regardless of cause of death. She called this a "liberal" approach. Link above.

And CDC guidelines clearly do not require a positive test. Link above.

I am just looking at published reports. I will leave the policy reasons to others. I am not suggesting ill intent. But the article was about flu counts being estimated (and inflated) while COVID deaths are assumed to be accurate.

I questioned both of those premises by the author.

 

[eytonxav]

eta: if numbers were being fudged, wouldn't it be more likely that the federal government would prefer to have a lower count than higher? To open the economy sooner?

I'm sure everyone would prefer less death, but the federal government is only rolling up the data they receive from local/state officials, including public and private hospitals. Data gathering at lower levels or even fed rollup levels may have issues with accuracy/errors.

 

[Amethyst]

I agree, because it is a written forum regulation to stay away from politics. The trouble in this case is how to distinguish actual COVID-19 facts and numbers, from the dense political thicket that seeks to conceal them.

It’d be nice if we could keep politics out of the discussion.

 

[Sunset]

I can see flu deaths being over estimated, as it fits with the annual goal of encouraging flu shots, and I've always wondered who are these people who die of flu, since I've never known one in many decades of paying attention to health.

As for Covid-19, it's really the desire of any gov't to look like they are doing a good job to report low numbers of deaths of Covid-19.
I think this is why the estimated death models to come out lately have all pointed to extremely low death rates of less than 100,000.
Given that we have a lack of testing still going on, it seems that Covid-19 deaths will be missed as there is no way to test some dead body due to lack of resources.

Since we are already over 66,000 deaths from Covid-19 and it will continue on for another year, I fully expect the death count will be higher than any flu in modern times (excluding the Spanish flu).

Hopefully with a surplus of testing that is promised to come, we can test every person that dies to make a more accurate count.

 

[MichaelB]

I agree, because it is a written forum regulation to stay away from politics. The trouble in this case is how to distinguish actual COVID-19 facts and numbers, from the dense political thicket that seeks to conceal them.

The thread topic is inherently not political, it’s about contrasting views of COVID risk. Politics creeps in when people begin associating motive and political purpose to data and weaving it into the more technical discussion. Pointless but disruptive.

We have incomplete information and knowledge about COVID, so the comparison and effort to equate the two (COVID vs Flu) is faulty. What we need, and don’t have, is a more detailed understanding of the risk of COVID and how that applies to individual situations that we each can relate to. Until we know more we will need to deal with uncertainty. Looking harder for that one expert opinion is not helpful. To quote Caitlin Rivers, (epidemiologist and senior scholar at Johns Hopkins)

If experts tell you something is unknowable, don’t keep asking new people until you get a straight answer. Because in doing so you haven’t found the truth, you have found someone who wants your ear.

 

[Ian S]

Dr. Birx statement on April 8. "We have taken a liberal approach to corona virus deaths".

"If someone dies with Covid19 we are counting that as a Covid 19 death"

https://youtu.be/zbFt6eO3wz0

CDC guidelines for reporting deaths does not require positive test.
https://www.cdc.gov/nchs/nvss/vsrr/covid19/tech_notes.htm

NY reclassifying almost 4000 deaths as COVID.
https://www.dailymail.co.uk/news/ar...es-3-700-probable-new-coronavirus-deaths.html

I notice you neglected to explain the huge number of excess deaths this year. In NYC alone, we're talking over 4,000 deaths that can't be accounted for by adding the "liberal" # of COVID-19 deaths to the normal number of deaths for the time period (which is amazingly consistent from year to year). We could be looking at an undercount on the order of 25%! https://www.nytimes.com/interactive/2020/04/27/upshot/coronavirus-deaths-new-york-city.html

 

[Montecfo]

I notice you neglected to explain the huge number of excess deaths this year. In NYC alone, we're talking over 4,000 deaths that can't be accounted for by adding the "liberal" # of COVID-19 deaths to the normal number of deaths for the time period (which is amazingly consistent from year to year). We could be looking at an undercount on the order of 25%! https://www.nytimes.com/interactive/2020/04/27/upshot/coronavirus-deaths-new-york-city.html

Neglected? No. And I have no need to explain everything about flu and covid deaths. I do think it requires analysis beyond the scope of a few posts on a public retirement board.

The article fails to identify the portion of the apparent discrepancy NOT explained by seasonal factors (such as the record severe flu season Gov Cuomo reported on in February), or by pure seasonality or monthly fluctuations. Should we do the work to identify the reasons?

It also makes clear that the reason for the apparent discrepancy is unknown, though that is also obvious from reading the article.

I think what I would like to see is cause of death data by source for a period of time required for necessary context. I have not seen an article that does that, but perhaps you have and can post it.