4 ms·

Anchoring like that is a problem when it's used to dismiss something: "Eh, more people die from fentanyl, so X isn't so bad". Which disregards that X may be a

Anchoring like that is a problem when it's used to dismiss something: "Eh, more people die from fentanyl, so X isn't so bad". Which disregards that X may be a new source of increased deaths, not to mention that when X == COVID, the methods needed to mitigate the risk are antithetical to a mindset of "it's no big deal" because that mindset pushes back against taking even some minimal precuations. So, anchoring can help people understand the magnitude of something, but at the same time convey a misunderstanding, or short-circuit reasoning as well. This is precisely why it is a common sales tactic: Go into a jewelry store asking for a nice watch as a present, and you may be shown a $3k watch. Way over your budget, so when you ask to be shown a something else a $700 watch seems like a much better deal, even if it might still be a bit more than you wanted to spend. Maybe the third watch will be an ugly one for $400 to help convince you that you need to spend more, putting both an upper & lower bound on the purchase.

mlyle 5y ago
> Which disregards that X may be a new source of increased deaths, not to mention that when X == COVID, the methods needed to mitigate the risk are antithetical to a mindset of "it's no big deal" because that mindset pushes back against taking even some minimal precuations. I don't think anyone has said, "fentanyl is worse for 18-44, therefore COVID is no big deal". I think the message taken was "two sources of premature death shot way up, passing traffic accidents and suicide, which we all know are really bad in that age group". And maybe "wow, fentanyl is even bigger than COVID as a cause of death among young adults". It doesn't make me take traffic safety less seriously, either. Look, I ain't got time each year to compute loss-of-life expectancy numbers, and decide how preventable each and every cause is, and then come up with an analytical ranking of each cause and its "importance" (I did this at one point, but I'm not going to repeat it to understand trends). I don't work in public health, either. Approximation based on reasonable anchor points and understood risks is just fine.
- ineedasername 5y ago
  I don't think anyone has said, "fentanyl is worse for 18-44, therefore COVID is no big deal". My intent was to make a general statement about anchoring as a technique. But in any case, you're incorrect: my own father has made this sort of remark about opioids, and I've seen it pop up in message boards, and with other (not necessarily opioid) comparisons by folks who resist masks/social distancing/vaccination.
  - mlyle 5y ago
    I wonder if you're conflating arguments, though. I've never heard the "no big deal" one. I've heard another: If responses to COVID have worsened the opioid crisis, that's a cost that should be counted against COVID mitigation measures. (I still think this is faulty: I don't think responses to COVID have done much to change the opioid crisis. But at least this is a rational argument and open to debate).
    - ineedasername 5y ago
      Different social/family circles maybe? (And my own-- not always healthy-- habit of reading comments on political news site... I've mostly backed away from that for my own mental sanity) But I've seen the "no big deal" thing a fair bit. I remember one exchange nearly verbatim "Fentanyl kills more people but all the fearmongering media wants to talk about is a virus that gives you a cold". Heck, at the very beginning people saying that sort of thing by comparison to the flu. That line of thinking went away when deaths far exceeded annual flu deaths, but with the (possible) decrease in severe cases with Omicron I've seen hints of this argument again.
      - mlyle 5y ago
        > but with the (possible) decrease in severe cases with Omicron I've seen hints of this argument again. Well, at some point, it's a reasonable argument. Risk isn't going to go to 0. It looks like with vaccination my personal risk is 2-3x influenza, but I may be overstating it. At some point, we're going to just have to give up and assume risks. I'm not routinely wearing masks 3 years from now.
        ineedasername 5y ago
        I hope it does end up being true of Omicron, a bit too soon to tell though. It's a little early to tell: We're a month on from the end of November when Omicron really hit our radar and it took a few more weeks to see the really big spike we're in right now. On the national (US) level I don't see much of an uptick in deaths yet (although my own state has a bit of one) but deaths are also a lagging indicator so it may be another month before we can really tell for sure. Another confounding variable are testing rates: Mandatory testing is in place to a significant degree more than last year, so it is difficult to tell how many more positive tests this year are simply due to mandatory testing. For example, they're required for air travel and about 3x more people traveled this year for the holidays. This is speculation though: Relevant data would be at least in part include asymptomatic case rates from year to year, but I couldn't find a good source for that. Information like [0] are very promising but still indicate that it's too early to tell. [0] https://www.reuters.com/business/healthcare-pharmaceuticals/hospitalisation-risk-omicron-around-one-third-delta-uk-analysis-shows-2021-12-31/ https://www.reuters.com/business/healthcare-pharmaceuticals/... >At some point, we're going to just have to give up and assume risks. I'm not routinely wearing masks 3 years from now. Mostly agreed, though as variants come & go I won't be surprised if some variants create "mask seasons" as they rise & fall. In retrospect even pre-covid it seems odd that it was deemed acceptable to go into work when you were sick with a cold or flu-- not necessarily even for severe health risks, but just as a matter of common courtesy. Of course changing that behavior in the long term would also require society (again, I'm in the US) to rethink things like what constitutes a reasonable amount of paid sick time. That is an especially problematic area when workplaces mandate 5/10/14 days of isolation but don't cover it with paid leave. My workplace does not cover that time if an employee is already out of paid sick leave. In the past, when going in to work when you were sick was somewhat optional, that made a little more sense. Now, telling someone they're losing 4% of their annual income due to a mandatory 14 day isolation period w/o pay is... I'm not sure I have a good word for it, but it feels wrong.
evv555 5y ago
>"Eh, more people die from fentanyl, so X isn't so bad" The increase in deaths is being driven by lockdown/social isolation so it looks more like medical triage. Who do you save, upperclass/baby boomers who generally don't have these problems or disenfranchised young?
- mlyle 5y ago
  > The increase in deaths is being driven by lockdown/social isolation so it looks more like medical triage. This is an argument many make, but it's not a really good argument. - Fentanyl deaths were trending up before 2020. - The degree to which fentanyl is lacing other street drugs is unprecedented and seemingly independent of lockdown. - Look, we don't really have "lockdown" anymore.
  - evv555 5y ago
    >- Fentanyl deaths were trending up before 2020. They were trending up but there was a very sharp increase once the epidemic started. >- The degree to which fentanyl is lacing other street drugs is unprecedented and seemingly independent of lockdown. There was precedent before 2019. Seemingly independent growth that's also parabolic during 2020? Unlikely >- Look, we don't really have "lockdown" anymore. The genie is out of the bottle once people relapse. Maybe you don't realize how dependent recovering addicts are on rehab programs and social connections to stay clean. For some people giving them a steady unemployment income and forcing them to isolate is basically a death sentence. This was a predicted outcome during the start of the isolation.
    - mlyle 5y ago
      > They were trending up but there was a very sharp increase once the epidemic started. Opioid deaths increased 1014% from 2013 to 2019. This is an average compound growth rate of 46.8% per year over that span. Opioid deaths increased by about 50% from 2019 to 2020. It looks like they increased by less than 40% from 2020 to 2021. So over the past 2 years we have had the same increase rate as from the entire span. The data don't support your assertions.
raxxorrax 5y ago
But there needs to be a discussion if that short cuircuit is our behavior in light of this. Of course the number could be higher if we didn't react, but I would guess destroying the basis of living for so many people is also quite significant and another "new" thing. And if this sales tactic would be applied to any security issue you can more or less forget about civil liberties.

mlyle 5y ago

> Which disregards that X may be a new source of increased deaths, not to mention that when X == COVID, the methods needed to mitigate the risk are antithetical to a mindset of "it's no big deal" because that mindset pushes back against taking even some minimal precuations. I don't think anyone has said, "fentanyl is worse for 18-44, therefore COVID is no big deal". I think the message taken was "two sources of premature death shot way up, passing traffic accidents and suicide, which we all know are really bad in that age group". And maybe "wow, fentanyl is even bigger than COVID as a cause of death among young adults". It doesn't make me take traffic safety less seriously, either. Look, I ain't got time each year to compute loss-of-life expectancy numbers, and decide how preventable each and every cause is, and then come up with an analytical ranking of each cause and its "importance" (I did this at one point, but I'm not going to repeat it to understand trends). I don't work in public health, either. Approximation based on reasonable anchor points and understood risks is just fine.

ineedasername 5y ago
I don't think anyone has said, "fentanyl is worse for 18-44, therefore COVID is no big deal". My intent was to make a general statement about anchoring as a technique. But in any case, you're incorrect: my own father has made this sort of remark about opioids, and I've seen it pop up in message boards, and with other (not necessarily opioid) comparisons by folks who resist masks/social distancing/vaccination.
- mlyle 5y ago
  I wonder if you're conflating arguments, though. I've never heard the "no big deal" one. I've heard another: If responses to COVID have worsened the opioid crisis, that's a cost that should be counted against COVID mitigation measures. (I still think this is faulty: I don't think responses to COVID have done much to change the opioid crisis. But at least this is a rational argument and open to debate).
  - ineedasername 5y ago
    Different social/family circles maybe? (And my own-- not always healthy-- habit of reading comments on political news site... I've mostly backed away from that for my own mental sanity) But I've seen the "no big deal" thing a fair bit. I remember one exchange nearly verbatim "Fentanyl kills more people but all the fearmongering media wants to talk about is a virus that gives you a cold". Heck, at the very beginning people saying that sort of thing by comparison to the flu. That line of thinking went away when deaths far exceeded annual flu deaths, but with the (possible) decrease in severe cases with Omicron I've seen hints of this argument again.
    - mlyle 5y ago
      > but with the (possible) decrease in severe cases with Omicron I've seen hints of this argument again. Well, at some point, it's a reasonable argument. Risk isn't going to go to 0. It looks like with vaccination my personal risk is 2-3x influenza, but I may be overstating it. At some point, we're going to just have to give up and assume risks. I'm not routinely wearing masks 3 years from now.
      - ineedasername 5y ago
        I hope it does end up being true of Omicron, a bit too soon to tell though. It's a little early to tell: We're a month on from the end of November when Omicron really hit our radar and it took a few more weeks to see the really big spike we're in right now. On the national (US) level I don't see much of an uptick in deaths yet (although my own state has a bit of one) but deaths are also a lagging indicator so it may be another month before we can really tell for sure. Another confounding variable are testing rates: Mandatory testing is in place to a significant degree more than last year, so it is difficult to tell how many more positive tests this year are simply due to mandatory testing. For example, they're required for air travel and about 3x more people traveled this year for the holidays. This is speculation though: Relevant data would be at least in part include asymptomatic case rates from year to year, but I couldn't find a good source for that. Information like [0] are very promising but still indicate that it's too early to tell. [0] https://www.reuters.com/business/healthcare-pharmaceuticals/hospitalisation-risk-omicron-around-one-third-delta-uk-analysis-shows-2021-12-31/ https://www.reuters.com/business/healthcare-pharmaceuticals/... >At some point, we're going to just have to give up and assume risks. I'm not routinely wearing masks 3 years from now. Mostly agreed, though as variants come & go I won't be surprised if some variants create "mask seasons" as they rise & fall. In retrospect even pre-covid it seems odd that it was deemed acceptable to go into work when you were sick with a cold or flu-- not necessarily even for severe health risks, but just as a matter of common courtesy. Of course changing that behavior in the long term would also require society (again, I'm in the US) to rethink things like what constitutes a reasonable amount of paid sick time. That is an especially problematic area when workplaces mandate 5/10/14 days of isolation but don't cover it with paid leave. My workplace does not cover that time if an employee is already out of paid sick leave. In the past, when going in to work when you were sick was somewhat optional, that made a little more sense. Now, telling someone they're losing 4% of their annual income due to a mandatory 14 day isolation period w/o pay is... I'm not sure I have a good word for it, but it feels wrong.

evv555 5y ago

>"Eh, more people die from fentanyl, so X isn't so bad" The increase in deaths is being driven by lockdown/social isolation so it looks more like medical triage. Who do you save, upperclass/baby boomers who generally don't have these problems or disenfranchised young?

mlyle 5y ago
> The increase in deaths is being driven by lockdown/social isolation so it looks more like medical triage. This is an argument many make, but it's not a really good argument. - Fentanyl deaths were trending up before 2020. - The degree to which fentanyl is lacing other street drugs is unprecedented and seemingly independent of lockdown. - Look, we don't really have "lockdown" anymore.
- evv555 5y ago
  >- Fentanyl deaths were trending up before 2020. They were trending up but there was a very sharp increase once the epidemic started. >- The degree to which fentanyl is lacing other street drugs is unprecedented and seemingly independent of lockdown. There was precedent before 2019. Seemingly independent growth that's also parabolic during 2020? Unlikely >- Look, we don't really have "lockdown" anymore. The genie is out of the bottle once people relapse. Maybe you don't realize how dependent recovering addicts are on rehab programs and social connections to stay clean. For some people giving them a steady unemployment income and forcing them to isolate is basically a death sentence. This was a predicted outcome during the start of the isolation.
  - mlyle 5y ago
    > They were trending up but there was a very sharp increase once the epidemic started. Opioid deaths increased 1014% from 2013 to 2019. This is an average compound growth rate of 46.8% per year over that span. Opioid deaths increased by about 50% from 2019 to 2020. It looks like they increased by less than 40% from 2020 to 2021. So over the past 2 years we have had the same increase rate as from the entire span. The data don't support your assertions.

raxxorrax 5y ago

But there needs to be a discussion if that short cuircuit is our behavior in light of this. Of course the number could be higher if we didn't react, but I would guess destroying the basis of living for so many people is also quite significant and another "new" thing. And if this sales tactic would be applied to any security issue you can more or less forget about civil liberties.