4 ms·

You have moved the goalposts from "not hospitalized" to "asymptomatic or very minor symptoms". About half the people I know who got covid (and knew about it) ha

by dTal 5y ago

You have moved the goalposts from "not hospitalized" to "asymptomatic or very minor symptoms". About half the people I know who got covid (and knew about it) had, or are currently having, a very unpleasant time.

throwaway22032 5y ago
The study itself states that in patients that are not hospitalized the risks of various heart conditions increase on average by ~1.5-2x. That's for the entire cohort of non-hospitalized individuals, it's an average. If you assume that they're uniformly distributed (e.g. if you assume that the relative risk of future myocarditis is identical for someone who has no symptoms, and for someone who's bedridden for a week with breathing difficulties but non-hospitalized), the figures show that if you're not hospitalized your risk of myocarditis and other heart issues is multiplied by about 2. If the study were binned more finely I think you'd see a 5x multiplier if you have mild illness (mild in the clinical sense which simply means non-hospitalized but exhibiting symptoms) or no discernable effect if asymptomatic. I agree that there is no specific data for this, I'm extrapolating the curve. > About half the people I know who got covid (and knew about it) had, or are currently having, a very unpleasant time. We're not debating this point, this is a paper showing the correlation between the short term and long term effects of the disease.
- dTal 5y ago
  You say "only" a doubling of the risk of rare heart related issues, over the course of a single year. I say gee, it sure sounds like the virus massively fucked with their system. Large statistical anomalies are measurable an entire year later. Whether the baseline risk of such issues is high or low is irrelevant, since the main takeaway is that everyone basically gets Long Covid, to a greater or lesser extent. Remember that this isn't the only measurable effect either - it's just the one this paper is talking about, and it fits into a wider pattern of research indicating systemic damage even in so-called "minor" cases. Again, speaking from both personal anecdata and the wider scientific consensus - if you have symptomatic covid, it takes you a long time to feel "normal" again.
  - throwaway22032 5y ago
    The data doesn't imply that ~everyone has long term effects because we don't know the distribution in the non-hospitalized cases. > if you have symptomatic covid, it takes you a long time to feel "normal" again. The non-hospitalized group in this study includes both symptomatic and asymptomatic. This comment thread is a bit weird, you seem to be attacking me on an emotional basis. I'm interested because I think this data shows a good correlation between severity of disease and severity of long term prognosis, that's all. I have an interest in knowing this, as we all do, because like basically everyone else, I've had coronavirus, and if the data shows that this now means I have an elevated risk of heart issues I should be careful with things like heavy exercise. I don't think it does based on the correlation shown (I didn't have severe coronavirus), so it's actionable information. If you disagree with my assessment that a 2x multiplier on a rare event is not worrisome, that's fine - that's just like, my opinion, man. :)
    - davorak 5y ago
      > This comment thread is a bit weird, you seem to be attacking me on an emotional basis. I'm interested because I think this data shows a good correlation between severity of disease and severity of long term prognosis, that's all. Your original comment reads as a much stronger statement: > The meat of this seems to be in confirming that if you're not hospitalized the long-term effects are minimal. Not _zero_, but very small. I think also 'the long-term effects are minimal' reads too broad since this study only covers a narrow set of long-term effects while your original comment can read as applying to all long-term effects. > If you disagree with my assessment that a 2x multiplier on a rare event is not worrisome, that's fine - that's just like, my opinion, man. :) I would want to look at the number of people effected a year before and after the change in order decide if it was worrisome or not.
  - timr 5y ago
    > Large statistical anomalies are measurable an entire year later. Doubling of a very small number is not a "large statistical anomaly". Figure 5 shows that the difference is skewed strongly toward the hospitalized, and that the largest difference observed for the non-hospitalized cohort is measured at maybe 10 per 1,000 (1%) and most well-below that (single-digit difference per 1,000). https://www.nature.com/articles/s41591-022-01689-3/figures/5 https://www.nature.com/articles/s41591-022-01689-3/figures/5
timr 5y ago
> You have moved the goalposts from "not hospitalized" to "asymptomatic or very minor symptoms". About half the people I know who got covid (and knew about it) had, or are currently having, a very unpleasant time. And 100% of the people I know who had it (measured in the tens; including myself) have had no lingering symptoms at all. Anecdotes are not science. It is not "moving the goalposts" to note that lingering symptoms correlate with severity. The authors of the paper have already shown this, but didn't break it down enough.

by dTal 5y ago

throwaway22032 5y ago

The study itself states that in patients that are not hospitalized the risks of various heart conditions increase on average by ~1.5-2x. That's for the entire cohort of non-hospitalized individuals, it's an average. If you assume that they're uniformly distributed (e.g. if you assume that the relative risk of future myocarditis is identical for someone who has no symptoms, and for someone who's bedridden for a week with breathing difficulties but non-hospitalized), the figures show that if you're not hospitalized your risk of myocarditis and other heart issues is multiplied by about 2. If the study were binned more finely I think you'd see a 5x multiplier if you have mild illness (mild in the clinical sense which simply means non-hospitalized but exhibiting symptoms) or no discernable effect if asymptomatic. I agree that there is no specific data for this, I'm extrapolating the curve. > About half the people I know who got covid (and knew about it) had, or are currently having, a very unpleasant time. We're not debating this point, this is a paper showing the correlation between the short term and long term effects of the disease.

dTal 5y ago
You say "only" a doubling of the risk of rare heart related issues, over the course of a single year. I say gee, it sure sounds like the virus massively fucked with their system. Large statistical anomalies are measurable an entire year later. Whether the baseline risk of such issues is high or low is irrelevant, since the main takeaway is that everyone basically gets Long Covid, to a greater or lesser extent. Remember that this isn't the only measurable effect either - it's just the one this paper is talking about, and it fits into a wider pattern of research indicating systemic damage even in so-called "minor" cases. Again, speaking from both personal anecdata and the wider scientific consensus - if you have symptomatic covid, it takes you a long time to feel "normal" again.
- throwaway22032 5y ago
  The data doesn't imply that ~everyone has long term effects because we don't know the distribution in the non-hospitalized cases. > if you have symptomatic covid, it takes you a long time to feel "normal" again. The non-hospitalized group in this study includes both symptomatic and asymptomatic. This comment thread is a bit weird, you seem to be attacking me on an emotional basis. I'm interested because I think this data shows a good correlation between severity of disease and severity of long term prognosis, that's all. I have an interest in knowing this, as we all do, because like basically everyone else, I've had coronavirus, and if the data shows that this now means I have an elevated risk of heart issues I should be careful with things like heavy exercise. I don't think it does based on the correlation shown (I didn't have severe coronavirus), so it's actionable information. If you disagree with my assessment that a 2x multiplier on a rare event is not worrisome, that's fine - that's just like, my opinion, man. :)
  - davorak 5y ago
    > This comment thread is a bit weird, you seem to be attacking me on an emotional basis. I'm interested because I think this data shows a good correlation between severity of disease and severity of long term prognosis, that's all. Your original comment reads as a much stronger statement: > The meat of this seems to be in confirming that if you're not hospitalized the long-term effects are minimal. Not _zero_, but very small. I think also 'the long-term effects are minimal' reads too broad since this study only covers a narrow set of long-term effects while your original comment can read as applying to all long-term effects. > If you disagree with my assessment that a 2x multiplier on a rare event is not worrisome, that's fine - that's just like, my opinion, man. :) I would want to look at the number of people effected a year before and after the change in order decide if it was worrisome or not.
- timr 5y ago
  > Large statistical anomalies are measurable an entire year later. Doubling of a very small number is not a "large statistical anomaly". Figure 5 shows that the difference is skewed strongly toward the hospitalized, and that the largest difference observed for the non-hospitalized cohort is measured at maybe 10 per 1,000 (1%) and most well-below that (single-digit difference per 1,000). https://www.nature.com/articles/s41591-022-01689-3/figures/5 https://www.nature.com/articles/s41591-022-01689-3/figures/5

timr 5y ago

> You have moved the goalposts from "not hospitalized" to "asymptomatic or very minor symptoms". About half the people I know who got covid (and knew about it) had, or are currently having, a very unpleasant time. And 100% of the people I know who had it (measured in the tens; including myself) have had no lingering symptoms at all. Anecdotes are not science. It is not "moving the goalposts" to note that lingering symptoms correlate with severity. The authors of the paper have already shown this, but didn't break it down enough.