6 ms·

Is life expectancy measured from 5 years old onwards, or from 1 second old? It's frustrating to see numbers thrown around about how much life expectancy rose o

by yboris 4y ago

Is life expectancy measured from 5 years old onwards, or from 1 second old? It's frustrating to see numbers thrown around about how much life expectancy rose over the years in the past, only to discover that the average was so low because so many kids never made it past 5 years old. Why would including those deaths be in any way informative for people alive? That average tells me almost nothing about my life expectancy! :( The article never mentions it, but the graph says "at birth" is the starting point.

issa 4y ago
Maybe the article was updated since you read it? It goes into their methodology: "Arias and her colleagues calculated life expectancy using a technique called a period life table. This involved the researchers imagining a group of 100,000 hypothetical infants and applying the death rates observed for the real population in 2021 for each year of those infants’ lives. The result is not the life expectancy for a cohort of actual babies born in 2021 but rather a snapshot of how life expectancy rates would apply to various age groups at a specific point in time, Arias says."
aabaker99 4y ago
"Life expectancy" is short hand for life expectancy at birth, but other life expectancies such as life expectancy at age 65 are also measured. For instance, https://data.oecd.org/healthstat/life-expectancy-at-65.htm https://data.oecd.org/healthstat/life-expectancy-at-65.htm
- tomrod 4y ago
  The conditional distribution of time-to-death is how I interpret it after conversation with my actuary fellow friend. "Conditional on being in this well-defined population, how long until I die in expectation?"
phpisthebest 4y ago
Well that is a problem with all stats reporting. Infant Mortality is another one "journalists" love to beat the US health system over, never highlighting how the different stats are reported by different nations, or other factors that make the US infant Mortality rate naturally higher due to our reporting standards, and standard procedures. Life Expectancy is not really a good guide for anything, it is just a number various political groups can use to influence public policy to their agenda
- SQueeeeeL 4y ago
  I mean, there is still a large portion of the US who simply don't get any form of preventative cure, doctors becoming extremely overworked and stressed from the impossible amount of diagnostic and billing paperwork, and rural hospitals regularly closing. I've found there are few defenders of the US healthcare status quo who aren't insurance reps or ideologs who have never had a lapse in coverage
  - phpisthebest 4y ago
    You are correct that there is not much debate over the short comings of the US system, the debate is over what is the cause of those short coming, and what is the resolution. I, and many others, believe government regulations is the cause including everything from the tax code that irrevocably links insurance to employment, to regulations that transform "insurance" to prepaid health plans, to medicare massively under paying providers, and about 10000 other things... The "other side" believes government is the holy grail and will magically fix everything if we just make everything single payer medicaid for all...
    - SQueeeeeL 4y ago
      The main argument I've heard is that healthcare is simply something in which an individual carriers no bargainning power. Death versus payment is a system which is ideal for capitalism to thrive because you can basically charge as much as an individual can bare. In such a system, any non-guaranteed solution will be flawed because the desires of the market (low cost/high profits) will never align with the desires of patients (affordable care/access to large amount of care). The only way to break that paradigm is to guarantee access to care through some non revokable mechanism, which typically needs an entity like a state to provide (but could be run through a sufficiently large enough non-profit)
      - phpisthebest 4y ago
        >>Death versus payment is a system which is ideal for capitalism to thrive Which is a false dilemma, red herring and a strawman all in one. The vast majority of health services / decisions are not life or death, far far from it. We can see that in area's where insurance is does not normally cover or is very limited. Dental Services, Vision, Cosmetic Surgery, etc etc Even in "traditional" health care, the "urgent care" market is pretty competitive with providers having standardized price lists I can look up online and see which provider I want to go to. For example about a year ago I had a trip and fall, my shoulder was in pain and I have a High Deductible insurance so everything would be out of pocket. I looked at the various providers near me, got their standard prices and went to one that provided the best value to me. Market forces absolutely can be applied to 90% of medical care. For true Life and Death emergencies, a huge part of those will be covered under other insurances anyway like Home Owners, Car, etc. for the remainder will maybe that is a place for government regulations but that does not justify a full take over by the government
conductr 4y ago
I personally don’t see the average as useful anyway. It’s a stat meant for headlines and a surface level benchmark as it’s easy to consume. More interesting would be to see the mix perhaps by decade of age at death. It also needs to normalize for number of people alive in each cohort.
bena 4y ago
Life expectancy is a weird one, not going to lie. All the numbers are wrong from some perspective. You mention a good one. The fact that life expectancy was so low in the past was due to infant mortality. If you lived to like 15 (IIRC), your life expectancy got much longer. Like 60s to 70s (once again, IIRC). And that even holds today. The longer you live, the longer you can expect to live. If you're 70, you have a life expectancy of 14 (male) to 17 (female) years. If you're 80, you have a life expectancy of 7 (m) to 9 (f) years. If you're 90, you have a life expectancy of about 4 years. And all of that seems kind of weird on the face. If I'm 70, I can expect to live 14 years on average. Which takes me to 84, which is greater than 80. But if I'm 80, I can expect to live to 87, which is greater than 84. But it's the nature of averages. When I'm 70, I'm being average with other 70 year olds. Some of them die at 71, some at 93. Every year, more and more people drop off. If you make it to 80, all those who didn't make are no longer relevant.
NotYourLawyer 4y ago
Or measure it from conception, and it’s been falling like a rock for awhile now.
- amanaplanacanal 4y ago
  Depending on what you mean by "for a while now". It appears that abortions in the US peaked over thirty years ago and have been falling since.
  - NotYourLawyer 4y ago
    Interesting. I was just shooting from the hip, but it looks like you’re right. (And still right even if we look at the abortion rate or abortion ratio, instead of just the absolute number.)
TomSwirly 4y ago
The people who write these papers and their reviewers don't take such elementary facts into account. The actual answer is that US infant mortality has stayed almost constant during those 26 years, and was so low to start with that even reducing it to zero wouldn't have made much change in overall mortality. https://www.statista.com/statistics/1041693/united-states-all-time-child-mortality-rate/ https://www.statista.com/statistics/1041693/united-states-al... The linked article also goes into more details.

by yboris 4y ago

issa 4y ago

Maybe the article was updated since you read it? It goes into their methodology: "Arias and her colleagues calculated life expectancy using a technique called a period life table. This involved the researchers imagining a group of 100,000 hypothetical infants and applying the death rates observed for the real population in 2021 for each year of those infants’ lives. The result is not the life expectancy for a cohort of actual babies born in 2021 but rather a snapshot of how life expectancy rates would apply to various age groups at a specific point in time, Arias says."

aabaker99 4y ago

"Life expectancy" is short hand for life expectancy at birth, but other life expectancies such as life expectancy at age 65 are also measured. For instance, https://data.oecd.org/healthstat/life-expectancy-at-65.htm https://data.oecd.org/healthstat/life-expectancy-at-65.htm

tomrod 4y ago
The conditional distribution of time-to-death is how I interpret it after conversation with my actuary fellow friend. "Conditional on being in this well-defined population, how long until I die in expectation?"

phpisthebest 4y ago

Well that is a problem with all stats reporting. Infant Mortality is another one "journalists" love to beat the US health system over, never highlighting how the different stats are reported by different nations, or other factors that make the US infant Mortality rate naturally higher due to our reporting standards, and standard procedures. Life Expectancy is not really a good guide for anything, it is just a number various political groups can use to influence public policy to their agenda

SQueeeeeL 4y ago
I mean, there is still a large portion of the US who simply don't get any form of preventative cure, doctors becoming extremely overworked and stressed from the impossible amount of diagnostic and billing paperwork, and rural hospitals regularly closing. I've found there are few defenders of the US healthcare status quo who aren't insurance reps or ideologs who have never had a lapse in coverage
- phpisthebest 4y ago
  You are correct that there is not much debate over the short comings of the US system, the debate is over what is the cause of those short coming, and what is the resolution. I, and many others, believe government regulations is the cause including everything from the tax code that irrevocably links insurance to employment, to regulations that transform "insurance" to prepaid health plans, to medicare massively under paying providers, and about 10000 other things... The "other side" believes government is the holy grail and will magically fix everything if we just make everything single payer medicaid for all...
  - SQueeeeeL 4y ago
    The main argument I've heard is that healthcare is simply something in which an individual carriers no bargainning power. Death versus payment is a system which is ideal for capitalism to thrive because you can basically charge as much as an individual can bare. In such a system, any non-guaranteed solution will be flawed because the desires of the market (low cost/high profits) will never align with the desires of patients (affordable care/access to large amount of care). The only way to break that paradigm is to guarantee access to care through some non revokable mechanism, which typically needs an entity like a state to provide (but could be run through a sufficiently large enough non-profit)
    - phpisthebest 4y ago
      >>Death versus payment is a system which is ideal for capitalism to thrive Which is a false dilemma, red herring and a strawman all in one. The vast majority of health services / decisions are not life or death, far far from it. We can see that in area's where insurance is does not normally cover or is very limited. Dental Services, Vision, Cosmetic Surgery, etc etc Even in "traditional" health care, the "urgent care" market is pretty competitive with providers having standardized price lists I can look up online and see which provider I want to go to. For example about a year ago I had a trip and fall, my shoulder was in pain and I have a High Deductible insurance so everything would be out of pocket. I looked at the various providers near me, got their standard prices and went to one that provided the best value to me. Market forces absolutely can be applied to 90% of medical care. For true Life and Death emergencies, a huge part of those will be covered under other insurances anyway like Home Owners, Car, etc. for the remainder will maybe that is a place for government regulations but that does not justify a full take over by the government

conductr 4y ago

I personally don’t see the average as useful anyway. It’s a stat meant for headlines and a surface level benchmark as it’s easy to consume. More interesting would be to see the mix perhaps by decade of age at death. It also needs to normalize for number of people alive in each cohort.

bena 4y ago

Life expectancy is a weird one, not going to lie. All the numbers are wrong from some perspective. You mention a good one. The fact that life expectancy was so low in the past was due to infant mortality. If you lived to like 15 (IIRC), your life expectancy got much longer. Like 60s to 70s (once again, IIRC). And that even holds today. The longer you live, the longer you can expect to live. If you're 70, you have a life expectancy of 14 (male) to 17 (female) years. If you're 80, you have a life expectancy of 7 (m) to 9 (f) years. If you're 90, you have a life expectancy of about 4 years. And all of that seems kind of weird on the face. If I'm 70, I can expect to live 14 years on average. Which takes me to 84, which is greater than 80. But if I'm 80, I can expect to live to 87, which is greater than 84. But it's the nature of averages. When I'm 70, I'm being average with other 70 year olds. Some of them die at 71, some at 93. Every year, more and more people drop off. If you make it to 80, all those who didn't make are no longer relevant.

NotYourLawyer 4y ago

Or measure it from conception, and it’s been falling like a rock for awhile now.

amanaplanacanal 4y ago
Depending on what you mean by "for a while now". It appears that abortions in the US peaked over thirty years ago and have been falling since.
- NotYourLawyer 4y ago
  Interesting. I was just shooting from the hip, but it looks like you’re right. (And still right even if we look at the abortion rate or abortion ratio, instead of just the absolute number.)

TomSwirly 4y ago

The people who write these papers and their reviewers don't take such elementary facts into account. The actual answer is that US infant mortality has stayed almost constant during those 26 years, and was so low to start with that even reducing it to zero wouldn't have made much change in overall mortality. https://www.statista.com/statistics/1041693/united-states-all-time-child-mortality-rate/ https://www.statista.com/statistics/1041693/united-states-al... The linked article also goes into more details.