Page 1 of 2 12 LastLast
Results 1 to 10 of 17

Thread: Question for a statistician

  1. #1

    Question for a statistician

    I was doing some reading about case fatality rates where you divide the # of deaths by # of diagnosed cases to get your CFR. It occured to me that when we talk about death rates of flu, the number is artificially low because we don't generally test the entire population for flu. When this all shakes out in the end, I wonder if we'll find that flu and COVID had similar death rates? Maybe the crude mortality rate where you compare sick with the entire population would give a clearer picture? From the beginning, I've had a problem with the notation of testing well people. Although, it gives an accurate picture of COVID, it has enhanced fear mongering.

    Forgive me if I get my math reversed, I am kind of math dyslexic! I know how to do it, not explain it. LOL!

  2. #2
    well, in neither case is all the population tested.
    Well, people with a flu are not tested at all.
    The death rate for flu is different every year.
    We all have gotten "used" to the flu, so mainly it is not a problem. What you see is, that old and very old people whos immune system does not work well, die from flu.
    Here in Europe it is a bit different from what I have read about the US. Here deaths by covid-19 are only registered if there is proof. (like a PCR test).


    Now looking forward, lets say 10-20 years, Covid-19 may be a similar disese like flu.
    Looking back to the first 3 years of flu 50 Mio people died. Thats extrem a lot. But then the virus became less lethal
    This is what I expect to happen with Covid-19. (This is also what well known virologist say)

    The testing of well people makes sense since in many cases since you can spread Covid without being sick. But right now the incubation time of Omicron is short and it is less severe, testing does not help very much.
    does this help a little?

  3. #3
    So if the Son sets you free, you will be free indeed. John's Avatar
    Join Date
    Jul 2015
    Posts
    7,824
    Thanked: 7519
    Quote Originally Posted by Susan View Post
    I was doing some reading about case fatality rates where you divide the # of deaths by # of diagnosed cases to get your CFR. It occured to me that when we talk about death rates of flu, the number is artificially low because we don't generally test the entire population for flu. When this all shakes out in the end, I wonder if we'll find that flu and COVID had similar death rates? Maybe the crude mortality rate where you compare sick with the entire population would give a clearer picture? From the beginning, I've had a problem with the notation of testing well people. Although, it gives an accurate picture of COVID, it has enhanced fear mongering.

    Forgive me if I get my math reversed, I am kind of math dyslexic! I know how to do it, not explain it. LOL!
    I don't think we have good enough data sets to make any meaningful analysis. I know our governor ordered reported "covid deaths" to be re-examined to determine the "with covid" vs. "from covid" death numbers. With data it's garbage in - garbage out, I'm pretty sure everyone was stuffing numbers (garbage).

  4. #4
    John, what I save read here in several posts is, that hospitals get more money if it is a covid case. That can trigger people to change numbers.
    Maybe it is the same with deaths. That is bad, because I think we need good statistics to make choices.

  5. #5
    The CDC describes the model they use to estimate flu impact, and why they need to employ a model rather than issue exact numbers, here:
    How CDC Estimates the Burden of Seasonal Influenza in the U.S. | CDC
    Why CDC Estimates the Burden of Season Influenza in the U.S. | CDC

    The following excerpt describes how and why they believe that reliance on death certificates that specifically list flu would result in an undercount of the impact of the disease.
    Why doesn't CDC base its seasonal flu mortality estimates only on death certificates that specifically list influenza?
    Seasonal influenza may lead to death from other causes, such as pneumonia, congestive heart failure, or chronic obstructive pulmonary disease. It has been recognized for many years that influenza is underreported on death certificates. There may be several reasons for underreporting, including that patients aren't always tested for seasonal influenza virus infection, particularly older adults who are at greatest risk of seasonal influenza complications and death. Even if a patient is tested for influenza, influenza virus infection may not be identified because the influenza virus is only detectable for a limited number of days after infection and many people don't seek medical care in this interval. Additionally, some deaths – particularly among those 65 years and older – are associated with secondary complications of influenza (including bacterial pneumonias). For these and other reasons, modeling strategies are commonly used to estimate flu-associated deaths. Only counting deaths where influenza was recorded on a death certificate would be a gross underestimation of influenza's true impact.


    And the following excerpt gives a picture of the extent to which estimation plays a role in the published statistics. It is abundantly clear that it plays a very big role!
    Limitations of Influenza Burden Estimates
    First, rates of influenza-associated hospitalizations are based on data reported to the Influenza Hospitalization Surveillance Network (FluSurv–NET) that are current as of the time estimates are made. Final case counts may differ slightly as further data cleaning is conducted by FluSurv–NET sites. The most updated crude rates of hospitalization for FluSurv-NET sites are available on FluView Interactive (8).

    Second, national rates of influenza-associated hospitalizations and in-hospital death are adjusted for the frequency of influenza testing and the sensitivity of influenza diagnostic assays, using a multiplier approach (3). However, data on testing practices during the current season may not be available at the time of estimation. In such cases, we adjust rates using data from prior seasons. Burden estimates from a given season will be updated at a later date when data on contemporary testing practices become available.

    Third, estimates of influenza-associated illness are made by multiplying the number of hospitalizations by the ratio of illnesses to hospitalizations; estimates of medical visits are made by a similar process. These multipliers are based on data from prior seasons, which may not be accurate if patterns of care-seeking have changed.

    Fourth, our estimate of influenza-associated deaths relies on information about location of death from death certificates. However, death certificate data for a given season may not be available at the time of estimation. When this occurs, we use death certification data from prior influenza seasons where these data are available from the National Center for Health Statistics. Specifically, our model uses the frequency of influenza-associated deaths that have cause of death related to pneumonia or influenza (P&I), other respiratory or cardiovascular (other R&C), or other non-respiratory, non-cardiovascular (non-R&C) to account for deaths occurring outside of a hospital by cause of death. If these frequencies are not available from a given season at the time of estimation, we use the average frequencies of each cause from previous seasons.

    Fifth, estimates of burden were derived from rates of influenza-associated hospitalization, which is a different approach than the statistical models used in older published reports. This makes it difficult to directly compare our estimates for seasons since 2009 to those older reports, though the estimates from our current method are largely consistent with estimates produced with statistical models for similar years (9-10). However, it is useful to keep in mind that direct comparisons to influenza disease burden decades ago are complicated by large differences in the age of the US population and the increasing number of adults aged ≥65 years.


    In any event, the CDC-published data indicate tens of thousands of deaths in the US per year:
    Past Seasons Estimated Influenza Disease Burden | CDC
    2019-2020: 22K
    2018-2019: 34K
    2017-2018: 61K
    2016-2017: 38K
    2015-2016: 23K

    In the two years of Covid-19, we've lost about a million people, so we're talking about
    at least a factor of 10 increase in fatalities resulting from Covid-19.

  6. #6
    Quote Originally Posted by njtom View Post
    The CDC describes the model they use to estimate flu impact, and why they need to employ a model rather than issue exact numbers, here:
    How CDC Estimates the Burden of Seasonal Influenza in the U.S. | CDC
    Why CDC Estimates the Burden of Season Influenza in the U.S. | CDC

    The following excerpt describes how and why they believe that reliance on death certificates that specifically list flu would result in an undercount of the impact of the disease.
    Why doesn't CDC base its seasonal flu mortality estimates only on death certificates that specifically list influenza?
    Seasonal influenza may lead to death from other causes, such as pneumonia, congestive heart failure, or chronic obstructive pulmonary disease. It has been recognized for many years that influenza is underreported on death certificates. There may be several reasons for underreporting, including that patients aren't always tested for seasonal influenza virus infection, particularly older adults who are at greatest risk of seasonal influenza complications and death. Even if a patient is tested for influenza, influenza virus infection may not be identified because the influenza virus is only detectable for a limited number of days after infection and many people don't seek medical care in this interval. Additionally, some deaths – particularly among those 65 years and older – are associated with secondary complications of influenza (including bacterial pneumonias). For these and other reasons, modeling strategies are commonly used to estimate flu-associated deaths. Only counting deaths where influenza was recorded on a death certificate would be a gross underestimation of influenza's true impact.


    And the following excerpt gives a picture of the extent to which estimation plays a role in the published statistics. It is abundantly clear that it plays a very big role!
    Limitations of Influenza Burden Estimates
    First, rates of influenza-associated hospitalizations are based on data reported to the Influenza Hospitalization Surveillance Network (FluSurv–NET) that are current as of the time estimates are made. Final case counts may differ slightly as further data cleaning is conducted by FluSurv–NET sites. The most updated crude rates of hospitalization for FluSurv-NET sites are available on FluView Interactive (8).

    Second, national rates of influenza-associated hospitalizations and in-hospital death are adjusted for the frequency of influenza testing and the sensitivity of influenza diagnostic assays, using a multiplier approach (3). However, data on testing practices during the current season may not be available at the time of estimation. In such cases, we adjust rates using data from prior seasons. Burden estimates from a given season will be updated at a later date when data on contemporary testing practices become available.

    Third, estimates of influenza-associated illness are made by multiplying the number of hospitalizations by the ratio of illnesses to hospitalizations; estimates of medical visits are made by a similar process. These multipliers are based on data from prior seasons, which may not be accurate if patterns of care-seeking have changed.

    Fourth, our estimate of influenza-associated deaths relies on information about location of death from death certificates. However, death certificate data for a given season may not be available at the time of estimation. When this occurs, we use death certification data from prior influenza seasons where these data are available from the National Center for Health Statistics. Specifically, our model uses the frequency of influenza-associated deaths that have cause of death related to pneumonia or influenza (P&I), other respiratory or cardiovascular (other R&C), or other non-respiratory, non-cardiovascular (non-R&C) to account for deaths occurring outside of a hospital by cause of death. If these frequencies are not available from a given season at the time of estimation, we use the average frequencies of each cause from previous seasons.

    Fifth, estimates of burden were derived from rates of influenza-associated hospitalization, which is a different approach than the statistical models used in older published reports. This makes it difficult to directly compare our estimates for seasons since 2009 to those older reports, though the estimates from our current method are largely consistent with estimates produced with statistical models for similar years (9-10). However, it is useful to keep in mind that direct comparisons to influenza disease burden decades ago are complicated by large differences in the age of the US population and the increasing number of adults aged ≥65 years.


    In any event, the CDC-published data indicate tens of thousands of deaths in the US per year:
    Past Seasons Estimated Influenza Disease Burden | CDC
    2019-2020: 22K
    2018-2019: 34K
    2017-2018: 61K
    2016-2017: 38K
    2015-2016: 23K

    In the two years of Covid-19, we've lost about a million people, so we're talking about
    at least a factor of 10 increase in fatalities resulting from Covid-19.
    Here it comes! Justifying masks and social distancing and other measures throughout the cold and flu season! I was wondering how they were going to get there.

  7. #7
    The stats for deaths in 2017-18 are telling. Here in our rural area of small towns, that strain of flu hit hard. We had school closings, care homes on lockdown and people generally trying to avoid being around sick folk. Down the road a few miles from us, a single mom of 2 teens took ill and died within 48 hours. I caught it and was in a fight for my life to keep my bloodsugar under 350 while my body dumped cortisol in the blood stream. Finally after 4 days of high fevers I started to recover. That was rough.

  8. The Following User Says Thank You to FireBrand For This Useful Post:

    njtom (03-15-2022)

  9. #8
    Quote Originally Posted by Susan View Post
    Here it comes! Justifying masks and social distancing and other measures throughout the cold and flu season! I was wondering how they were going to get there.
    Helping to save other peoples' lives is never a bad idea, although I do recognize that there can be a tendency to go to extreme lengths that result in other harmful effects.

    In many Asian countries it is common to see people (young and old) on the streets wearing masks. When asked, they'll say they do it for following reasons:
    1) Allergies
    2) Pollution
    3) They have a cold and don't want to spread it to others.
    4) They don't have a cold but don't want to catch one from others.
    5) They didn't get a chance to put on their makeup.

    As long as mask wearing is legally voluntary, I have no problem with it. And if my doctor's office wants me to wear one because the waiting rooms have lots of elderly and/or immuno-compromised patients, I also have no problem wearing one.

  10. #9
    Quote Originally Posted by njtom View Post
    Helping to save other peoples' lives is never a bad idea, although I do recognize that there can be a tendency to go to extreme lengths that result in other harmful effects.

    In many Asian countries it is common to see people (young and old) on the streets wearing masks. When asked, they'll say they do it for following reasons:
    1) Allergies
    2) Pollution
    3) They have a cold and don't want to spread it to others.
    4) They don't have a cold but don't want to catch one from others.
    5) They didn't get a chance to put on their makeup.

    As long as mask wearing is legally voluntary, I have no problem with it. And if my doctor's office wants me to wear one because the waiting rooms have lots of elderly and/or immuno-compromised patients, I also have no problem wearing one.
    No, I won't argue with "saving lives" but will fight and refuse a mask mandate, especially when it is proven by science and doctors that it doesn't work and it did not in the case of covid.

  11. #10
    Quote Originally Posted by FireBrand View Post
    No, I won't argue with "saving lives" but will fight and refuse a mask mandate, especially when it is proven by science and doctors that it doesn't work and it did not in the case of covid.
    With daily Covid-19 fatalities in the USA down from 2000 to 300 in the past few weeks, I think we've seen the end of government-issued mask mandates. I do see (at least in my area) doctors' offices continuing to mandate masks, and I have no problem with this.

    As to the question of whether studies have shown that masks "work", I guess we'll have to agree to disagree. Studies I've read indicate that surgical-style masks (not cloth masks) do help.

    Surgical masks reduce COVID-19 spread, large-scale study shows | News Center | Stanford Medicine
    An evidence review of face masks against COVID-19
    Effectiveness of Mask Wearing to Control Community Spread of SARS-CoV-2 | Infectious Diseases | JAMA | JAMA Network

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
You can mitigate major repair costs with an extended service plan for your Jaguar. Many vehicle repairs can cost thousands of dollars in surprise expense, now is the time to consider an extended service plan for your vehicle.