Epidemiology probability question. - Page 3
Page 3 of 3 FirstFirst 123
Results 51 to 73 of 73
  1. #51
    RoadBikeReview Member
    Reputation:
    Join Date
    Oct 2013
    Posts
    941
    I am going to drop this here, gently, and tuck tail - it seemed on topic

    https://www.nytimes.com/2020/05/13/o...-accuracy.html

    "Just Because You Test Positive for Antibodies Doesn’t Mean You Have Them
    By Todd Haugh and Suneal Bedi
    Professors Haugh and Bedi are researchers who study judgment and decision-making




    In a population whose infection rate is 5 percent, a test that is 90 percent accurate could deliver a false positive nearly 70 percent of the time.

    The confidence that we should have in antibody tests depends on a key factor that is often ignored: the base rate of the coronavirus. The base rate is the actual amount of infection in a known population. In the United States, that appears to be between 5 percent and 15 percent.
    This simple fact is essential to understanding the accuracy of an antibody test. Yet overlooking this fact is also one of the most common decision-making errors made, so much so that it has its own name: the base rate fallacy.

    Here’s an example. If you took an antibody test that was 90 percent accurate, and it determined that you had coronavirus antibodies, how confident should you be that you actually have those antibodies?

    Most people say about 90 percent, with the average answer being above 50 percent. This makes sense. After all, 90 percent accuracy is pretty high.

    But the predictive value of an antibody test with 90 percent accuracy could be as low as 32 percent if the base rate of infection in the population is 5 percent. Put another way, there is an almost 70 percent probability in that case that the test will falsely indicate a person has antibodies..."

  2. #52
    We're no better than mice
    Reputation: EverydayRide's Avatar
    Join Date
    Sep 2008
    Posts
    7,080
    Wrong thread
    Last edited by EverydayRide; 2 Weeks Ago at 01:28 PM.

  3. #53
    We're no better than mice
    Reputation: EverydayRide's Avatar
    Join Date
    Sep 2008
    Posts
    7,080
    Wrong thread.
    Last edited by EverydayRide; 2 Weeks Ago at 01:29 PM.

  4. #54
    We're no better than mice
    Reputation: EverydayRide's Avatar
    Join Date
    Sep 2008
    Posts
    7,080
    Wrong thread.
    Last edited by EverydayRide; 2 Weeks Ago at 01:28 PM.

  5. #55
    Sooper Dooper Moderator!
    Reputation: il sogno's Avatar
    Join Date
    Jul 2002
    Posts
    19,407
    Eleventy.

  6. #56
    We're no better than mice
    Reputation: EverydayRide's Avatar
    Join Date
    Sep 2008
    Posts
    7,080
    Quote Originally Posted by QuiQuaeQuod View Post
    Chances are people did NOT learn this math in school. Not that it is above what people did learn in school, but the application is ... not intuitive. The only place I have see this kind of thing is in medical school and public health/epi curricula....
    I've always admired how your brain and explanations strive for MUCH further examinations and relavent theories.

    I appreciate it.

  7. #57
    We're no better than mice
    Reputation: EverydayRide's Avatar
    Join Date
    Sep 2008
    Posts
    7,080
    Dunno why my default is Freak Out Friday thread when I post here. Excuse me.

  8. #58
    RoadBikeReview Member
    Reputation: 9W9W's Avatar
    Join Date
    Apr 2012
    Posts
    2,724
    Wow. What a socially awkward snarky pedantic **** this guy is.
    "That pretty much sums it up. I'm 43 and my max is ~178-180. If that HR chart was mine or Froome's, we'd be on the verge of death. But for you it probably looks like a normal workout." -TLG

    LOLOLOL

  9. #59
    Boobies!
    Reputation: paredown's Avatar
    Join Date
    Oct 2006
    Posts
    8,130
    Quote Originally Posted by 9W9W View Post
    Wow. What a socially awkward snarky pedantic **** this guy is.
    I dunno--I like it when QQQ gets snippy. The problem is (as with that other thread)--everyone is an expert, without actually knowing much about what they are talking about--and argue opinion as fact, and certainly get much of the problems of assessing risk absolutely wrong.

    I heard a good clip of the Nobel Prize winner Daniel Kahneman talking about his recent personal dilemma--about whether or not he should go to Paris for a conference (despite the risk)--and admitting that he was just as likely as the next person to get this wrong.

    I had to google the results of the puzzle as well--but we did learn similar problems studying econometrics (applied statistics for economics)--but not my favorite part of the curriculum and I have long since forgotten the little I learned.
    Never attribute to malice that which is adequately explained by stupidity
    -Hanlon's Razor

  10. #60
    gazing from the shadows
    Reputation: QuiQuaeQuod's Avatar
    Join Date
    Jan 2003
    Posts
    26,973
    Quote Originally Posted by 9W9W View Post
    Wow. What a socially awkward snarky pedantic **** this guy is.
    BFD, person in the internet. Name calling? Did you get your ego boost for the day? I pity people who do that. Especially people who do ONLY that and don't make any other points.

    You do you, but you doing you looks pretty weak to me.
    .
    Stout beers under trees, please.

  11. #61
    gazing from the shadows
    Reputation: QuiQuaeQuod's Avatar
    Join Date
    Jan 2003
    Posts
    26,973
    Quote Originally Posted by il sogno View Post
    Eleventy.
    Best lounge answer there could be!
    .
    Stout beers under trees, please.

  12. #62
    RoadBikeReview Member
    Reputation:
    Join Date
    Oct 2013
    Posts
    3,888
    Personally, I just think he is confused and got his medication mixed up. He does have a point, but it really doesn't help.
    BANNED

  13. #63
    gazing from the shadows
    Reputation: QuiQuaeQuod's Avatar
    Join Date
    Jan 2003
    Posts
    26,973
    Quote Originally Posted by duriel View Post
    Personally, I just think he is confused and got his medication mixed up. He does have a point, but it really doesn't help.
    Accusations of mental illness? Yeah. I get it. You were wrong, tried to cover your butt, and now have descended with others to hurling insults while making no counterpoints about the POINTS you admit I made.

    The only meds I take are Advil, as needed, and a multi vitamin every day. FYI.

    You know how I treat people acting out on the net that I think have serious clinical issues? I ignore them, because they are mentally ill and I don't attack people who are ill.

    Idiots are another matter entirely.
    .
    Stout beers under trees, please.

  14. #64
    RoadBikeReview Member
    Reputation:
    Join Date
    Oct 2013
    Posts
    3,888
    One doesn't have to be 'mentally ill' to take medicine, almost all medications have side effects. I was just trying to make you appear more 'human', but that can be difficult.
    Is there something wrong with people on medication? Yes, but that's what they are taking medication for right?
    BANNED

  15. #65
    We're no better than mice
    Reputation: EverydayRide's Avatar
    Join Date
    Sep 2008
    Posts
    7,080
    Let's get out and ride.
    Last edited by EverydayRide; 2 Weeks Ago at 04:31 AM.

  16. #66
    gazing from the shadows
    Reputation: QuiQuaeQuod's Avatar
    Join Date
    Jan 2003
    Posts
    26,973
    Quote Originally Posted by duriel View Post
    I was just trying to make you appear more 'human', but that can be difficult.
    Keep on squealing.
    .
    Stout beers under trees, please.

  17. #67
    RoadBikeReview Member
    Reputation: 9W9W's Avatar
    Join Date
    Apr 2012
    Posts
    2,724
    Quote Originally Posted by paredown View Post
    I dunno--I like it when QQQ gets snippy. The problem is (as with that other thread)--everyone is an expert, without actually knowing much about what they are talking about--and argue opinion as fact, and certainly get much of the problems of assessing risk absolutely wrong.

    I heard a good clip of the Nobel Prize winner Daniel Kahneman talking about his recent personal dilemma--about whether or not he should go to Paris for a conference (despite the risk)--and admitting that he was just as likely as the next person to get this wrong.

    I had to google the results of the puzzle as well--but we did learn similar problems studying econometrics (applied statistics for economics)--but not my favorite part of the curriculum and I have long since forgotten the little I learned.
    Do you feel his domain knowledge is somehow redeeming of the disgustingly snide communication style he employs?
    "That pretty much sums it up. I'm 43 and my max is ~178-180. If that HR chart was mine or Froome's, we'd be on the verge of death. But for you it probably looks like a normal workout." -TLG

    LOLOLOL

  18. #68
    RoadBikeReview Member
    Reputation:
    Join Date
    Nov 2009
    Posts
    468
    Quote Originally Posted by 9W9W View Post
    Wow. What a socially awkward snarky pedantic **** this guy is.
    When I was in grad school, my roommate came up with something he called the D and A scales, that were applied to visiting scientists, guest lecturers, etc. The D scale was for dickfullness, and the A was for arrogance. A high score on one scale did not guarantee a high score on the other scale, but some people managed high scores on both ...

  19. #69
    gazing from the shadows
    Reputation: QuiQuaeQuod's Avatar
    Join Date
    Jan 2003
    Posts
    26,973
    Quote Originally Posted by 9W9W View Post
    Do you feel his domain knowledge is somehow redeeming of the disgustingly snide communication style he employs?
    Disgust? Well, I guess I am having a VERY powerful effect on you, huh?

    You people are way too soft. HTFU, as has been said in the Lounge for a very long time.
    .
    Stout beers under trees, please.

  20. #70
    gazing from the shadows
    Reputation: QuiQuaeQuod's Avatar
    Join Date
    Jan 2003
    Posts
    26,973
    Quote Originally Posted by tomato coupe View Post
    When I was in grad school, my roommate came up with something he called the D and A scales...
    I really wish you folks could come up with better material. But if it gets you all off in some way to have your circle jerk in such a simple and pedestrian manner, please keep going with juvenile name calling.

    I've see way better in my time here. I've seen worse too, but not often.
    .
    Stout beers under trees, please.

  21. #71
    RoadBikeReview Member
    Reputation: 9W9W's Avatar
    Join Date
    Apr 2012
    Posts
    2,724
    Quote Originally Posted by QuiQuaeQuod View Post
    Disgust? Well, I guess I am having a VERY powerful effect on you, huh?

    You people are way too soft. HTFU, as has been said in the Lounge for a very long time.
    Ever since RBR fell out a few years ago the skeleton crew comprises of a bunch of self-aggrandizing geezers in a circle jerk formation.

    RBR's single redeeming quality is the superb technical advice from select members.
    "That pretty much sums it up. I'm 43 and my max is ~178-180. If that HR chart was mine or Froome's, we'd be on the verge of death. But for you it probably looks like a normal workout." -TLG

    LOLOLOL

  22. #72
    RoadBikeReview Member
    Reputation:
    Join Date
    Oct 2013
    Posts
    3,888
    As I stated earlier, this example was just to make you feel like a da, and someone feel like #1. Kinda like someone else in the news today. Only in this case, this person my actually know something.
    But what does he know, in reality numbers, not actual information. As he applies his numbers, they are false, there are no "numbers". No one knows what the numbers are, so the math is pretty useless, other than showing how useless it is. ... ergo, sh*t in, sh*t out.
    BANNED

  23. #73
    gazing from the shadows
    Reputation: QuiQuaeQuod's Avatar
    Join Date
    Jan 2003
    Posts
    26,973
    Quote Originally Posted by duriel View Post
    No one knows what the numbers are, so the math is pretty useless, other than showing how useless it is. ... ergo, sh*t in, sh*t out.
    We (meaning people who bother to look) know a lot of numbers. For example, we know the Abbot Labs test (used in the White House) has an error rate that is WAY above 1%.

    Researchers at the Cleveland Clinic tested 239 specimens known to contain the coronavirus using five of the most commonly used coronavirus tests, including the Abbott ID NOW. The ID NOW has generated widespread excitement because it can produce results in less than 15 minutes.

    But the ID NOW only detected the virus in 85.2% of the samples, meaning it had a false-negative rate of 14.8 percent, according to Dr. Gary Procop, who heads COVID-19 testing at the Cleveland Clinic and led the study.

    ...

    Based on his study, Procop said his hospital has stopped using the test to screen patients being admitted for care. The hospital also stopped using another test, called the DiaSorin Simplexa, because it only detected 89.3% of infections in his study, Procop said.

    ...

    The Cleveland Clinic plans to rely on other tests that performed better in the study, Procop said. That includes the test developed by the Centers for Disease Control and Prevention, which detected 100% of positive samples. Another test, made by Roche, detected 96.5% of positive samples. The fifth test in the study, made by Cepheid, detected 98.2% of infected samples, Procop said. The Cepheid test produces results in less than an hour.
    https://www.npr.org/sections/health-...-covid-19-test

    Just because you don't know the numbers does not mean there are not numbers known. Those CC study numbers are false negative rates. False positives freak people out (until re-test), but false negatives lead to more infections. So WAY worse for the spread of a disease.

    BTW, that Abbott test? From yesterday...

    As first reported on NPR, as many as 15 to 20 out of every 100 tests may produce falsely negative results. A study released this week indicated that the test could be missing as many as 48% of infections.
    https://www.npr.org/sections/coronav...ronavirus-test

    So, the numbers are not down to the last decimal place, but the numbers are good enough to say that we should NOT be using that test. For reasons that should be obvious.
    .
    Stout beers under trees, please.

Page 3 of 3 FirstFirst 123

Similar Threads

  1. EBAY question "question"
    By PaulCL in forum General Cycling Discussion
    Replies: 2
    Last Post: 08-23-2006, 02:33 PM
  2. Replies: 4
    Last Post: 07-22-2005, 02:49 PM
  3. Replies: 1
    Last Post: 07-05-2005, 05:11 AM
  4. Social Security question (actual question)
    By Tri_Rich in forum Politics Only
    Replies: 1
    Last Post: 02-21-2005, 10:27 AM
  5. bontrager question race x lite tire question?
    By andy02 in forum General Cycling Discussion
    Replies: 1
    Last Post: 03-19-2004, 07:55 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

THE SITE

ABOUT ROADBIKEREVIEW

VISIT US AT

© Copyright 2020 VerticalScope Inc. All rights reserved.