Chat about Novel Coronavirus - 2019-nCoV - COVID-19

Posted on
of 1,083
First Prev
/ 1,083
Last Next
  • These tweets from the COVID19actuary account are supporting what @badboybjorn has suggested - that the increase in deaths is connected to the prevalence of omicron.­s/1481305431444230154

    COVID-19 deaths in English hospitals increased by 43% week-on-week. With COVID prevalence so high this may include more who people who died of other causes, but with COVID, than usual. Purple estimates are based on the pattern of reporting delays in the last two months. 5/5­1481307462154301441

    Interesting to see in this regular update from @COVID19Actuary that whilst admissions have levelled off, there has been a big jump in deaths. Athough this could be partly due to more "with COVID" deaths with such high prevalence.

    Given this is a reasonable discussion to have with regards to the data, i would say @greenbank owes @badboybjorn an apology for his arrogant, insulting and dismissive tone.

  • Did I say they weren't? I was just pointing out that there are a lot of people who look at these stats for a day job and publish there opinions on it, there isn't a need to guess

    Also edited my last post to add another actuary

  • Thank you for sharing. My contention was with @greenbank and his ‘appeal to authority’ fallacy, not with you.

  • Maybe im beeing to fast when im trying to explain but im trying to be an effective courier whilst engaging in this conversation ;)

    I have not in any news or articles that i can remember seen discussed that even if the chart of deaths 28 days after infection numbers are rising the deaths from covid could be fewer than previously in reality.

    I have often heard that "yes we have more infections now but not more deaths but we Will have to wait and see if those numbers start climbing" but never a footnote or clarification that this number is expected to rise even if the actually number of deaths are The same or less with a more infectious variant.

  • I have not in any news or articles that i can remember seen discussed that even if the chart of deaths 28 days after infection numbers are rising the deaths from covid could be fewer than previously in reality

    I'm not sure it's possible to deduce a good % of "hey COVID is probably bless of a factor in deaths" right now based on the data.

    I'd imagine you can have a shot at it, but the data has to be there & the models to calculate otherwise it's guess work.

    Edit: For example: Do you take "people in hospital with 0 symptoms of COVID then dying?" "A group dying with Omicron not vaccinated at all VS delta"?

  • Sure agreed. I dont have a better solution either for reporting so its not about that but u could n perhaps should point out that a new variant which is more infectious will yield more deaths reported even if that is not the case in reality. Ppl look at those numbers to try and determine which way we are headed on their own and if its not a fair comparison that should be noted.

    Im sure experts and those responsible for our health are aware of this so its not that i think they have missed this, just its not in the message broadcasted so those worried who look at the charts could think its heading for disaster when infact its looking better than before. I have not looked actively if this has been discussed before in the news since its something that just popped into my head last night when thinking about that chart but im sure someone talked about it on twitter or similar but its not in the regular news channels to be picked up that i can remember anyways. But its also a rather new variant ofc and excess death will be much better info regardless.

  • Ppl look at those numbers to try and determine which way we are headed on their own and if its not a fair comparison that should be noted: I guess them we have the question of what is "fair".

    To me, it is an imperfect but at least consistent comparison. So if I see cases drop, deaths going up 2 weeks later, but then dropping again, I am thinking "right, we are over the peak, PHEW" so I am worried for a bit, then OK.

    But other people read "Omicron is mild" and VOILA let go of all the masks/being careful/party on/SUPERRRR and that's a personality trait, and we don't all get the same info, trust the same info, are on the same risk aversion scale...

    But it is also true that perhaps more context is needed, I do see explanations on the BBC (yeah don't trust them on some stuff, but covid news is usually OK) that explain a bit more.

    The sea of Twitter/random exports/social media confusion doesn't easy to get lost in "too much info" "bad info" and even on here it is not perfect with sometimes unclear posts and also we don't always agree, so what to do?

    How do you asses a risk in that sense? When is info "so wrong" that "you cannot trust it at all?" In the past we just accepted government/healthcare info, now that's all gotten a lot more blurred.

    I guess I am a simple person...wear a mask (wear a really good one not cloth if you want to have more protection spreading your covids to others), keep your distance, get jabbed, be super careful visiting vulnerable people.

    The rest interesting as-is, I cannot work out even though I like statistics, I know I am not good at it due to lack of training/education :)

  • I apologise to @badboybjorn for my tone last night, posting after coming home from the pub is never a good thing.

    Maybe I should flip it round, I can see no evidence that Covid deaths are actually falling and that any rise in them is solely down to increased prevalence. A few tweets that say things "could" be limited to the increased prevalence doesn't constitute proof. Are there any where any experts are willing to be more definitive?

    I don't doubt that there is some increase in the "28 days" death stats due to increased prevalence, but that doesn't explain the corresponding increases (albeit lagged) in the "covid on death certificate" death stats, which is the argument put forward by many of the experts (when I've finished work for the day I'll try and dig out where I read that.)

    So, again, apologies for my tone, but that doesn't change the fact that I disagree with the notion that genuine Covid deaths are falling and that any increase in the stats are solely due to the increased prevalence making more non-Covid related deaths being flagged as Covid.

  • @greenbank - very fair comment. I would think prevalence is contributing to the spike, but I would also think we are seeing deaths increasing due to omicron.

  • Apology accepted no worries. Been there.

    The only thing i was trying to convey is that math tell us that a more contagious virus will results in more deaths on the 28 day after covid test chart than a less contagious one even if the latter is much more dangerous in total. I find this interesting and its also shows that comparing new variants with very different parameters than before with old statistics could be misleading if trying to draw the same conclusions as one were when trying to compare the same variant at different stages in the pandemic.

    Wether or not we have more deaths right now than previously or not and what that is down to if so i have no real opinion off. I live in sweden n here the hospitals are full of all kinds of ppl . Last i heard 4 out of 5 beds were not occupied of covid patients. but we are a few weeks behind u guys on the curve. Time will tell if we are moving towards something better or not with this new variant.

    As for death by certificate i dont know enough bout this and how it differs from the 28 day after death count to draw any conclusions of my own or even think anything about. really. I guess the accuracy of such data would be down to how eager or not those who make out the certificate is to label unknowns as covid or not. If it corresponds well with excess death i would think they are doing a pretty good job determining the cause of death.

  • Oh i missed this post. Always nice to see someone thought the same thing. That in the very least means that any condition i may have is shared by a few others .)

  • Nice that the bbc are following this thread­91

  • Novax Djokovic further up the thread, very good.

    Moving on, is it 2 much to ask for ppl to rite proper English rather than using txt speech on the forum? Just askin 4 a friend.

  • I don't think @badboybjorn is a native speaker of English based on his previous posts re his location.

  • Well there u go. cheers

    Now it will be discussed which is what i thought was strange that it had not been.

    And sorry for language, i try my best. But Im a swede so its not going to be perfect.

  • Your English is excellent, that’s my point!

  • For the Covid-horror-fans (no one mentioned, no one forgotten), the new Omicron BA2 mutation is taking over in Denmark over the 'original' BA1. From 2% at the end of December, till 28% a week ago.

  • Ah but not good enough to understand your point the first time. Thank you.

  • Providing there isn’t a new variant we might have scrapped isolation by March/ April perhaps?
    Has anyone seen any predictions/ modelling on this?

  • I should have been more precise with my language. Providing there isn’t a variant with more severe outcomes.
    There will continually be variants of course.

  • Your English is better than some and better than our seeedish

  • Post a reply
    • Bold
    • Italics
    • Link
    • Image
    • List
    • Quote
    • code
    • Preview

Chat about Novel Coronavirus - 2019-nCoV - COVID-19

Posted by Avatar for deleted @deleted