6 ms·

Hoping someone on HN knows the answer to this. Can someone explain what "5X more likely to catch Delta" means? I'm vaccinated. If I'm around someone with a d

Hoping someone on HN knows the answer to this. Can someone explain what "5X more likely to catch Delta" means? I'm vaccinated. If I'm around someone with a delta infection, do I have 5X the amount of time to get away? Can I receive 5X the amount of viral particles with the same infection risk? Is my immune system response 5X as effective at preventing an infection? I'm assuming its a population level analysis and not at all attenuated to individual immune response performance (unlike the death multiple which I think you can draw individual conclusions from). But I'm wondering what function (if at all) the MRNA vaccines play in reducing COVID infection. Basically, if I engage in risky behaviors (no mask wearing, hanging around in big crowds) am I 5X less likely to pass the virus onto my son?

iammisc 5y ago
It means nothing as these studies rarely take into account confounding variables. For example the vaxxed are more likely to distance and stay at home
- jasonlaramburu 5y ago
  Or they’re more likely to take risks out of perceived protection. Well designed studies do take these factors into account and ensure a sufficiently large and diverse sample size.
  - iammisc 5y ago
    How can you possibly find the group that represents the intersection of both 'covid is no big deal and I'll continue to live my life as normal' and 'you should take the covid vaccine'? This is like trying to find a pro life obstetrician who performs abortion. Other than the mentally ill, no one holds those two beliefs honestly at the same time. If it does exist, it probably doesn't in numbers large enough to matter. We are approaching the limits of empiricism.
    - jasonlaramburu 5y ago
      Double blind technique should eliminate a lot of these confounders, eg if you don’t know whether or not you’ve received the vaccine, your behavior presumably wouldn’t change much. However that is admittedly harder with the MRNA vaccines since most people have some mild side effects after the second dose.
      - iammisc 5y ago
        Those distrustful of vaccination are unlikely to sign up for trials Or are the trials being conducted clandestinely?
- phren0logy 5y ago
  It’s an imperfect study, like all population-level research. That said, it’s not fatally flawed. So, unless you are aware of something better, it’s reasonable to say that this is is currently the best available data. An imperfect but reasonable study is not an excuse to throw everything out and jump to your own conclusions.
  - iammisc 5y ago
    Except it is. Statistical studies can show all kinds of weird correlations that disappear when larger context is taken into account. There is no reason to believe there is a large group of people who behaved as if covid were not dangerous and who also felt an urgent need to take the vaccine. The existence of such a group of people is vital to whether or not data can be used to make these claims about the covid vaccine. Until the existence of that group is shown I don't understand how behavior is controlled for. Everyone claims to be following the rules. case in point, my mother in law is absolutely freaked out about covid. Has the vaccine. Masks. Distances etc. Except she breaks the rules whenever convenient. But in her mind these are necessary. She'll still complain about others who break the rules similarly. Lots of people are this way. You can't trust self reports and you can't follow that many people Also, and more egregiously, there is a history of fake covid studies destined to make those on the 'side' of the anti vaxxers look bad. So my bayesian a priori estimate of the usefulness of studies on the other 'side' is shifted more to the distrust side.
- relativ575 5y ago
  That sounds reasonable on the face of it. Do you have data to back it up?
  - iammisc 5y ago
    https://news.gallup.com/poll/350666/americans-getting-cautiously.aspx https://news.gallup.com/poll/350666/americans-getting-cautio... Look at the table "Americans' Use of Face Masks, by Vaccination Status". 90% of the vaccinated still wear masks. 49% of the unvaxxed do. The study hinges on a magical group of vax recipients that act as if covid doesn't exist now and behaved the same before the vax . There is little reason to believe this group exists in large numbers.
    - robocat 5y ago
      That was the only figure in the whole article which split out vax recipients, versus no-vax. The implication is that all the other figures didn’t have such a spectacular difference. So why is mask wearing so correlated with taking a vaccine? I think you are suggesting one causation, yet the actual reason is probably the “hidden” correlate of partisanship. Lots of people use mask wearing to show party affiliation, and I would guess that party affiliation in the US is highly correlated with taking the vaccine or not. Also you didn’t answer at all your original point “It means nothing as these studies rarely take into account confounding variables. For example the vaxxed are more likely to distance and stay at home”. Instead your link has nothing to do with correlation of distancing or staying at home. Changing the topic is a classic technique.
      - iammisc 5y ago
        You don't think masking works? You don't think masking is correlated with distancing? You accuse me of not taking into account obvious correlations, while doing the exact same thing. Not changing the topic. I have little evidence. One of the pitfalls of empiricism is that when the only ones with the ability to publish data are also liable to narrative pushing (see the numerous retracted studies on COVID treatments due to fake data), it's difficult for laypeople to disprove it. I'm just pointing out that the vaxxed are more likely to follow other safeguards, like wearing masks, which I imagine is also correlated with distancing. This seems obvious to me, and informs my a priori assumptions. If you have data that shows that those who wear masks are also more likely to ignore social distancing (a counterintuitive result indeed), I'd be happy to change my assessment.
loonster 5y ago
There is an implied clause. > Unvaccinated are 5X more likely to catch delta (than vaccinated), 11X more likely to die (than vaccinated) Fixed the headline.
angelzen 5y ago
Possibly a side effect of biased data collection. "... I had gone to a house party in Montgomery County. There were 15 adults there, all of us fully vaccinated." "... Then, I started to hear that a few other people who had been at the party were getting sick. Then a few more. At this point, 11 of the 15 have tested positive for COVID." "... The CDC announced in May that it was only going to collect data on breakthrough infections that led to hospitalization or death, which are fortunately rare." https://www.baltimoresun.com/opinion/op-ed/bs-ed-op-0804-breakthrough-covid-20210803-t32trfpiwzdf5okfar45f64whi-story.html https://www.baltimoresun.com/opinion/op-ed/bs-ed-op-0804-bre...

iammisc 5y ago

It means nothing as these studies rarely take into account confounding variables. For example the vaxxed are more likely to distance and stay at home

jasonlaramburu 5y ago
Or they’re more likely to take risks out of perceived protection. Well designed studies do take these factors into account and ensure a sufficiently large and diverse sample size.
- iammisc 5y ago
  How can you possibly find the group that represents the intersection of both 'covid is no big deal and I'll continue to live my life as normal' and 'you should take the covid vaccine'? This is like trying to find a pro life obstetrician who performs abortion. Other than the mentally ill, no one holds those two beliefs honestly at the same time. If it does exist, it probably doesn't in numbers large enough to matter. We are approaching the limits of empiricism.
  - jasonlaramburu 5y ago
    Double blind technique should eliminate a lot of these confounders, eg if you don’t know whether or not you’ve received the vaccine, your behavior presumably wouldn’t change much. However that is admittedly harder with the MRNA vaccines since most people have some mild side effects after the second dose.
    - iammisc 5y ago
      Those distrustful of vaccination are unlikely to sign up for trials Or are the trials being conducted clandestinely?
phren0logy 5y ago
It’s an imperfect study, like all population-level research. That said, it’s not fatally flawed. So, unless you are aware of something better, it’s reasonable to say that this is is currently the best available data. An imperfect but reasonable study is not an excuse to throw everything out and jump to your own conclusions.
- iammisc 5y ago
  Except it is. Statistical studies can show all kinds of weird correlations that disappear when larger context is taken into account. There is no reason to believe there is a large group of people who behaved as if covid were not dangerous and who also felt an urgent need to take the vaccine. The existence of such a group of people is vital to whether or not data can be used to make these claims about the covid vaccine. Until the existence of that group is shown I don't understand how behavior is controlled for. Everyone claims to be following the rules. case in point, my mother in law is absolutely freaked out about covid. Has the vaccine. Masks. Distances etc. Except she breaks the rules whenever convenient. But in her mind these are necessary. She'll still complain about others who break the rules similarly. Lots of people are this way. You can't trust self reports and you can't follow that many people Also, and more egregiously, there is a history of fake covid studies destined to make those on the 'side' of the anti vaxxers look bad. So my bayesian a priori estimate of the usefulness of studies on the other 'side' is shifted more to the distrust side.
relativ575 5y ago
That sounds reasonable on the face of it. Do you have data to back it up?
- iammisc 5y ago
  https://news.gallup.com/poll/350666/americans-getting-cautiously.aspx https://news.gallup.com/poll/350666/americans-getting-cautio... Look at the table "Americans' Use of Face Masks, by Vaccination Status". 90% of the vaccinated still wear masks. 49% of the unvaxxed do. The study hinges on a magical group of vax recipients that act as if covid doesn't exist now and behaved the same before the vax . There is little reason to believe this group exists in large numbers.
  - robocat 5y ago
    That was the only figure in the whole article which split out vax recipients, versus no-vax. The implication is that all the other figures didn’t have such a spectacular difference. So why is mask wearing so correlated with taking a vaccine? I think you are suggesting one causation, yet the actual reason is probably the “hidden” correlate of partisanship. Lots of people use mask wearing to show party affiliation, and I would guess that party affiliation in the US is highly correlated with taking the vaccine or not. Also you didn’t answer at all your original point “It means nothing as these studies rarely take into account confounding variables. For example the vaxxed are more likely to distance and stay at home”. Instead your link has nothing to do with correlation of distancing or staying at home. Changing the topic is a classic technique.
    - iammisc 5y ago
      You don't think masking works? You don't think masking is correlated with distancing? You accuse me of not taking into account obvious correlations, while doing the exact same thing. Not changing the topic. I have little evidence. One of the pitfalls of empiricism is that when the only ones with the ability to publish data are also liable to narrative pushing (see the numerous retracted studies on COVID treatments due to fake data), it's difficult for laypeople to disprove it. I'm just pointing out that the vaxxed are more likely to follow other safeguards, like wearing masks, which I imagine is also correlated with distancing. This seems obvious to me, and informs my a priori assumptions. If you have data that shows that those who wear masks are also more likely to ignore social distancing (a counterintuitive result indeed), I'd be happy to change my assessment.

loonster 5y ago

There is an implied clause. > Unvaccinated are 5X more likely to catch delta (than vaccinated), 11X more likely to die (than vaccinated) Fixed the headline.

angelzen 5y ago

Possibly a side effect of biased data collection. "... I had gone to a house party in Montgomery County. There were 15 adults there, all of us fully vaccinated." "... Then, I started to hear that a few other people who had been at the party were getting sick. Then a few more. At this point, 11 of the 15 have tested positive for COVID." "... The CDC announced in May that it was only going to collect data on breakthrough infections that led to hospitalization or death, which are fortunately rare." https://www.baltimoresun.com/opinion/op-ed/bs-ed-op-0804-breakthrough-covid-20210803-t32trfpiwzdf5okfar45f64whi-story.html https://www.baltimoresun.com/opinion/op-ed/bs-ed-op-0804-bre...