3 ms·

You're right, it doesn't. The study posted by the parent is not trying to make claims about absolute risk to young people -- it's focused on the conditional ris

by timr 6y ago

You're right, it doesn't. The study posted by the parent is not trying to make claims about absolute risk to young people -- it's focused on the conditional risk, given that a young person is hospitalized in the first place.

Dylan16807 6y ago
"Conditional on being hospitalized for any reason" gives you an exceptionally useless number. There are many conditions that would give you a usable risk number, but not that one. Just for the obvious extreme example, if I say 0.01% of total hospitalizations are young people with ebola, that doesn't let you conclude anything about whether catching ebola is "quite low risk" for that group.
- timr 6y ago
  > "Conditional on being hospitalized for any reason" gives you an exceptionally useless number. It tells you what to expect if you're a doctor working with hospitalized patients, or if you run a hospital. This is why they did the study. > if I say 0.01% of total hospitalizations are young people with ebola, that doesn't let you conclude anything about whether catching ebola is "quite low risk" for that group. 1) that's not what this study calculated. You still don't understand it. The equivalent study would be "of young people hospitalized with ebola, what percentage of them die?" (and BTW, the answer would be: "a huge percentage of them"...unlike Covid.) 2) They weren't trying to say anything about catching the virus. I understand that you really want to use this paper to make the claim, but it's just not relevant. They took a population of young people hospitalized for Covid, removed the pregnancies, and looked at what happened to the rest. That's it. It doesn't tell you anything about "catching Covid". It does tell you that if you're young and not obese, not diabetic and don't have heart disease, you are overwhelmingly likely not to die from it.
  - Dylan16807 6y ago
    > 1) that's not what this study calculated. I know that. > 2) They weren't trying to say anything about catching the virus. I understand that you really want to use this paper to make the claim, but it's just not relevant. I know that, and I wasn't talking about catching, I was talking about the risk of death if you catch it. > The equivalent study would be "of young people hospitalized with ebola, what percentage of them die?" (and BTW, the answer would be: "a huge percentage of them"...unlike Covid.) > It does tell you that if you're young and not obese, not diabetic and don't have heart disease, you are overwhelmingly likely not to die from it. I'm not sure what your definition of "overwhelming" is so I'll put it this way: Of young people hospitalized, the percentage chance of death is still pretty high. The numbers you quoted were 10 percent! I'm pretty sure you quoted wrong and the real number is 3 percent, but that's still a very real chance of death that you should try very strongly to avoid. The original quote was "Young healthy people have never been at risk." Your amended version was "the risk is quite small indeed". Neither of those fits a 3 percent death rate, even after taking into account the risk factors and focusing on patients without them. That chance of death is something that might be acceptable once or twice in your entire life after great consideration. The only way to get a "small" risk here is to factor in a sufficient number of people that catch it but never end up in the hospital at all (if there are enough of them). Going purely by the hospital numbers it's not something to brush off at all. I didn't bother to explain that part because I didn't think you were actually suggesting that 343 out of 3222 is quite low risk! I assumed you were talking about a different number.

by timr 6y ago

Dylan16807 6y ago

"Conditional on being hospitalized for any reason" gives you an exceptionally useless number. There are many conditions that would give you a usable risk number, but not that one. Just for the obvious extreme example, if I say 0.01% of total hospitalizations are young people with ebola, that doesn't let you conclude anything about whether catching ebola is "quite low risk" for that group.

timr 6y ago
> "Conditional on being hospitalized for any reason" gives you an exceptionally useless number. It tells you what to expect if you're a doctor working with hospitalized patients, or if you run a hospital. This is why they did the study. > if I say 0.01% of total hospitalizations are young people with ebola, that doesn't let you conclude anything about whether catching ebola is "quite low risk" for that group. 1) that's not what this study calculated. You still don't understand it. The equivalent study would be "of young people hospitalized with ebola, what percentage of them die?" (and BTW, the answer would be: "a huge percentage of them"...unlike Covid.) 2) They weren't trying to say anything about catching the virus. I understand that you really want to use this paper to make the claim, but it's just not relevant. They took a population of young people hospitalized for Covid, removed the pregnancies, and looked at what happened to the rest. That's it. It doesn't tell you anything about "catching Covid". It does tell you that if you're young and not obese, not diabetic and don't have heart disease, you are overwhelmingly likely not to die from it.
- Dylan16807 6y ago
  > 1) that's not what this study calculated. I know that. > 2) They weren't trying to say anything about catching the virus. I understand that you really want to use this paper to make the claim, but it's just not relevant. I know that, and I wasn't talking about catching, I was talking about the risk of death if you catch it. > The equivalent study would be "of young people hospitalized with ebola, what percentage of them die?" (and BTW, the answer would be: "a huge percentage of them"...unlike Covid.) > It does tell you that if you're young and not obese, not diabetic and don't have heart disease, you are overwhelmingly likely not to die from it. I'm not sure what your definition of "overwhelming" is so I'll put it this way: Of young people hospitalized, the percentage chance of death is still pretty high. The numbers you quoted were 10 percent! I'm pretty sure you quoted wrong and the real number is 3 percent, but that's still a very real chance of death that you should try very strongly to avoid. The original quote was "Young healthy people have never been at risk." Your amended version was "the risk is quite small indeed". Neither of those fits a 3 percent death rate, even after taking into account the risk factors and focusing on patients without them. That chance of death is something that might be acceptable once or twice in your entire life after great consideration. The only way to get a "small" risk here is to factor in a sufficient number of people that catch it but never end up in the hospital at all (if there are enough of them). Going purely by the hospital numbers it's not something to brush off at all. I didn't bother to explain that part because I didn't think you were actually suggesting that 343 out of 3222 is quite low risk! I assumed you were talking about a different number.