|
Which horse film is your favorite? This poll is closed. |
|||
|---|---|---|---|
| Black Beauty |
|
2 | 1.06% |
| A Talking Pony!?! |
|
4 | 2.13% |
| Mr. Hands 2x Apple Flavor |
|
117 | 62.23% |
| War Horse |
|
11 | 5.85% |
| Mr. Hands |
|
54 | 28.72% |
| Total: | 188 votes | ||
|
I'm not convinced masks won't be gone again the second Delta has some downward momentum or we hit some new piddly X% of eligible population vaxxed bs benchmark. The MSM is already spinning up their "endemic means ditch NPIs" consent machine and there's some serious deja-vu.
Stickman fucked around with this message at 00:48 on Sep 14, 2021 |
#
¿
Sep 14, 2021 00:46
|
|
|
|
| # ¿ May 11, 2024 16:35 |
|
|
I just wish we'd waited for full approval of COVID before we infected 45% of Americans 
|
|
#
¿
Sep 15, 2021 09:30
|
|
|
You're overselling the difficulty by quite a bit. We had a massive toolbox built up over one 14 months: knowledge of what works and what doesn't (which could easily be tweaked for a more transmissible disease), better treatments, more mitigation infrastrucure, we knew Delta was coming months in advance, and critically, ~83% seroprevolence of antibodies from vaccination and prior infection. That last part alone puts Delta roughly between OG COVID and Alpha (roughly because there's no one-to-one analogue between R0+protection and lower R0 in a naive population). What we didn't have was political or individual will to control outbreaks, and we'd spent the previous couple of months convince everyone that the vaccine was all they needed and they were being ridiculous if they tried to avoid infection after vaccination. Fritz the Horse posted:The end goal of the vaccines was always to prevent disease, not transmission. I've seen this quite a few times (I think TWiV promotes it too), but I don't really buy it. They didn't design the vaccine to prevent disease rather than transmission, they designed it in a way to give a broad immunological response, and what we got happened to be the result. If protection against disease were truly the only component of interest, then the phase 3 trials should have been powered for VE against severe disease and it should have been the primary endpoint, rather than secondary. There should have been some effort to tease apart the conditional effect of protection against infection to better estimate long-term disease protection in the event that protection against infection waned. It's certainly not the endpoint health officials wanted, since they keep insisting that breakthroughs are "rare". The vaccines protecting primarily against disease is a disappointing outcome because it precludes the much, much more effective preventative effects of population vaccination (which ultimately would mean much better protection against disease overall). And in the end it makes ongoing NPIs more important, not less. Stickman fucked around with this message at 12:14 on Sep 15, 2021 |
|
#
¿
Sep 15, 2021 11:54
|
|
|
Platystemon posted:I’m not an immunologist, but my understanding is that for it’s better to get the shot while not actively immunosuppressed. People in this very topic have had doctors take them off certain drugs for a week or two to give the immune system a chance to train up before medication tamps it down again. The point is, talk to your doctor about it, don’t be content to wait till you feel like you check the boxes just right. This is what they were recommending in the first round so it seems like there's a decent chance that the doctor might prescribe a pre-immunosuppressant booster if asked.
|
|
#
¿
Sep 16, 2021 22:54
|
|
|
Besides the HCW studies that have already been discussed, there's this study showing declining antibody levels in HCWs. They seem to track with the similar antibody declines in the general population studies, suggesting that any HCW exposures aren't keeping antibodies up, at least. It does make sense because strict PPE use means they are likely aren't exposed more than say, your average service worker. The CDC also has VE estimates over time for HCWs, but the sample size is unfortunately pretty small, uses minimal covariate adjustments, and it's not possible to separate the effects of Delta from >5 month declines. There were moderate declines at 4 months, but the small sample size makes it impossible to say much. It also has this bit that I have no idea how it got past internal review: quote:...however the VE 95% CI were overlapping, indicating the difference was not statistically significant.  Charles 2 of Spain posted:Not sure if there's been studies on healthcare workers specifically. The long-term ones I've seen consist of mainly older people who would've got their vaccine earlier. I think nearly all of them show some waning for infection but hardly any change for hospitalisation. Most of what I've seen has shown waning VE vs infection, but have far too small of sample sizes to say anything definitive about hospitalization. So far I haven't seen any that offer good evidence that protection v severe disease doesn't decline either, just that it's probably not by as much as the more severe estimates of VE v infection decline (which would be consistent with the immunological results). I haven't had time to keep up with everything, though, so it's possible there's something I missed!
|
|
#
¿
Sep 17, 2021 04:27
|
|
|
Professor Beetus posted:I was skeptical as all hell but I am certainly not a subject matter expert by any means. The most critical I got to her face was that regardless of her confidence level, I recommend she still continue to use an appropriate level of caution. She's fairly low risk of covid, unfortunately she's a sport bike rider so uh covid might not necessarily be the worst risk assessment she currently has. Any parent can tell you that frequent exposure to disease just means you get sick more often 
|
|
#
¿
Sep 17, 2021 04:32
|
|
|
Charles 2 of Spain posted:The Kaiser study had one with about 3 million people, there was a recent one out of Portugal which I think had about a million as well. I mean the original waning freakout was based on Israeli observational data which wasn't exactly a huge sample size either. By "large enough sample size" I mean "large enough sample size of hospitalizations". There's always a lot fewer of those than cases! Israel's analysis is a great example because there's enough cases to have at least some evidence in decline vs detected infection, but not enough severe cases to really say much. Thanks for mentioning those two studies, I hadn't seen them. The Portugal study (is this the one you meant?) actually looks relatively decent for deaths sample-wise but unfortunately they only have temporal analysis of 80+ folks, presumably because that's the only cohort with sufficient follow-up periods. They estimated 74% VE vs death for 80+ at >98 days, compared to 86 for the first month after full vaccination, but unfortunately there's no discussion about how much of that period overlaps Delta. The hospitalization estimates are weird because they had 91 breakthrough deaths but only 43 breakthrough vaccinations, which I guess means that most of the cohort received care outside of hospitals? Maybe at care centers? Either way I don't think there hospitalizations metrics are likely to be a good proxy for "severe disease" in that cohort, and I'm not really sure how to interpret those numbers without knowing more about how they are defined. What is the Kaiser study? Nothing is popping up right away... Stickman fucked around with this message at 05:19 on Sep 17, 2021 |
|
#
¿
Sep 17, 2021 05:15
|
|
|
The Pfizer phase 3 team apparently published their 6-month analysis yesterday. VE against detected infection decreased from 96% [93% to 98%] across the 2 months following full vaccination to 84% [75% to 90%] for 4+ months (with ~6 months max follow-up). Unfortunately about half of each arm was lost to follow-up by 4 months, so randomization loses some of it's robustness. I suspect that much of the loss to follow-up was due to people becoming eligible for vaccination, meaning that older participants would have been lost first. There were only 23 "severe" cases in the control arm and 1 in the vaccine arm, so no analysis of VE over time was possible (but average VE was 96%).
|
|
#
¿
Sep 17, 2021 05:35
|
|
|
Charles 2 of Spain posted:The Kaiser one is behind a paywall, but I could get it from what appears to be Mexican government website. Seems to have around 12,000 hopsitalisations from a quick look: Thanks! My point is just that up to this Kaiser study, I hadn't seen a single analysis that could reasonably look at VE against severe disease over time, and most of the papers didn't even try. The Kaiser study at least seems like it has sufficient numbers and follow-up to say something, though. E: Looks like the Kaiser study has been around for two weeks, whoops! Charles 2 of Spain posted:Should note this study was mainly pre-Delta surge, although Beta was included as well which could be more vaccine-evasive if I remember correctly. Yeah, I meant to mention that. It's what makes this one pretty nice and rates more comparable! They sequenced South Africa cases and only found 9 Beta infections, so fortunately Beta was probably only around 1% total cases. Stickman fucked around with this message at 06:06 on Sep 17, 2021 |
|
#
¿
Sep 17, 2021 05:50
|
|
|
Fritz the Horse posted:This is going to be extremely pedantic, but your insurer actually can't bill you for $50,000 for stuff they cover (refusing coverage is a separate issue I'm not familiar with). The ACA-mandated out of pocket maximum is $8,550 for 2021. Which is absolutely still ruinous and lovely for working and middle class people. That's only in-network, unfortunately, and if you have a family plan that doubles even if it's just one person using it. Employer plans also have some more wiggle room and loopholes for larger OOP maximums, even in-network. The HHS put out some out-of-network surprise billing rules in July, but I have no idea how closely they are being followed or enforced, or if there are remaining loopholes.
|
|
#
¿
Sep 17, 2021 06:32
|
|
|
Fritz the Horse posted:What is supposed to be "garbage" about those stats? I'm missing something here. Unfortunately, 90-day risk is useless without context in an epidemic that comes and goes in waves that infect double-digit proportions of the population. The important missing bit is which 90 days are being used as baseline risk of infection. Since pre-Delta hospitalization risk after infection for a healthy 30-year-old woman is ~1.6% (from multiple hospitalization risk models), I'm guessing that they're using pre-Delta conditional risk values and assuming 1% of healthy 30-yo women will be infected over 90 days. Given that we infected somewhere around 7% of the US population in the last two months and that Delta doubles hospitalization (and maybe death) rates, those assumptions seem pretty untenable. Plus, 90 days is a weirdly short time period to use when the real questions are "how often will I contract COVID under current trends and various mitigation strategies?" and "How serious will it be when I contract it?" Those are tougher questions to answer because there are a lot of unknowns, but the answers are almost certainly less rosy than the slide's overly-optimistic estimation of "90-day risk". I don't know the context of that slide, but I don't see how it's useful for anything except what it's being used for now - dumbass anti-vax propaganda. E: Spent too long typing and missed that the slide was from a dumbass anti-vaxxer too, lol. Stickman fucked around with this message at 21:20 on Sep 17, 2021 |
|
#
¿
Sep 17, 2021 21:09
|
|
that Dr. Eli David twitter account appears to be an anti-vaxxer or vaccine skeptic so be careful with your sources?
|
Discendo Vox posted:I'll be summarizing and reviewing Deer's book on this, eventually, and if I can get a used copy, I may do Wakefield's book as well. Used sales just pass up value by offsetting primary sales. Just  that heaping pile of poo poo. Or borrow it from an academic library (and never return it).E: The thing that I never understood about Wakefield's paper is that it was utter poo poo with zero evidence even if it wasn't fraudulent. The lancet never should have published it in the first place, and it never should have received any academic attention. Stickman fucked around with this message at 22:36 on Sep 17, 2021 |
|
#
¿
Sep 17, 2021 22:33
|
|
|
Taxes for the last couple of pages... I volunteer at a wedge-tailed shearwater colony and we just had our annual chick census; 275 of these bad boys and girls in one acre:  Three weeks ago:  Five weeks ago (different chick):  E: Lol, I'm waaay too slow at typing... Stickman fucked around with this message at 23:17 on Sep 17, 2021 |
|
#
¿
Sep 17, 2021 23:05
|
|
|
Thanks for posting your thoughts on the paper, Fritz!Charles 2 of Spain posted:Did they publish this table which was in the original preprint (maybe it's integrated into the figure somewhere)? It looks like the just replaced the explicit VE estimates for vaccination cohorts with "rates per 1000 over the study period". Since each vaccinated cohort (for a single age group) is being compared to the rate in the same unvaccinated cohort it is the same information as the VE table and statistical tests for one dataset are equivalent to statistical tests on the other. E: Or just "used instead of", since both the table and figures were in the original preprint. The tables are more easy to visually digest and the apparent "direction of effect" matches the temporal direction of waning immunity. Stickman fucked around with this message at 06:24 on Sep 18, 2021 |
|
#
¿
Sep 18, 2021 05:59
|
|
|
Discendo Vox posted:...do they control for familywise error? Are you talking about in the vaccination month cohort comparison? Because FWEA is absolutely not necessary there - the primary comparison of interest is going to be early v late and FWEA is going to be massively overly conservative. Probably would have been better with splines, though.
|
|
#
¿
Sep 18, 2021 06:35
|
|
|
Discendo Vox posted:No, as in they ran tests on several different entire populations and then excluded at least some of them from every part of their final publication. That's a layer above the cohort comparison, because they actually ran the test on a bunch of cohorts and (we know, from the slide deck) buried the ones without the results they wanted Which papers and which tests are you referring to here? There's several papers / slides and a bunch of tables so it's really not clear to me what you're talking about. E: Fritz has posted the slides (which link to the preprint of for waning effectiveness) and a paper for the booster data, which is a separate analysis from waning effectiveness. Is there a final paper for waning effectiveness that I've missed somewhere? Stickman fucked around with this message at 07:00 on Sep 18, 2021 |
|
#
¿
Sep 18, 2021 06:57
|
|
|
Discendo Vox posted:I'm referring to the matter discussed in this post. Oh come on. That isn't "running tests on several populations and excluding them from the final publication". For starters there isn't a final publication. The waning analysis is just the pre-print. It references the preprints on waning immunity because apparently they don't have those in press yet. The slides don't include the table, but the figures contain all of the data in the table, just presented as rates per 1000 instead of as VE estimates. Like, literally 1-1 in the first two charts that Fitz posted. There are no tests anywhere, just confidence intervals. And again, the comparison of most interest is the earliest vaccinations vs the latest vaccinations so there is no reason to do FWEAs for the cohort errors, and each age cohort is of a priori individual interest. There are no populations being excluded, there are no cohorts being swept under the rug, and there is no final paper that those things are happening in because the paper being discussed is a different analysis (booster analysis vs waning immunity for two-shot vax).
|
|
#
¿
Sep 18, 2021 07:08
|
|
|
Discendo Vox posted:My apologies, I got confused by the way it was presented because the paper is being cited to argue for cohorts not in the analysis. I still disagree about the lack of importance for FWEA, especially where presentation of data is selective. My apologies for getting grumpy here, too. FWEA is a whole debate because it's complicated and subtle (and largely arbitrary) to decide what a "family" should be. My general rule of thumb is that is hypothesis is being asked as part of a larger question that ties it together with other hypotheses then it makes sense to analyze them under the umbrella of a family (for instance "which of these many factors are important for X" or "which of these pairs of categories differ"). If it is of interest in it's own right then attaching it to a family depends arbitrarily on the other questions you happen to be asking at that particular point in time. We certainly can't reasonably expect to do FWEA across separate analyses, so if a hypothesis could reasonably be it's own independent analysis then it makes sense to exclude it from larger "families". In this case, we are not just interested in determining which of the age cohorts have waning effectiveness, we are interested in estimating the degree of waning in each in their own right. It's subtle and definitely open for debate, but it's extremely rare to see FWEA across major strata of interest for this reason. Stickman fucked around with this message at 08:23 on Sep 18, 2021 |
|
#
¿
Sep 18, 2021 08:02
|
|
|
Now that vaccines have full approval, my understanding is that there aren't any legal barriers to states or even the CDC recommending off-label booster use or doctors/pharmacists giving them. They're just extremely unlikely to do so, and ultimately the decision would still come down to doctors and pharmacists. E: It would be illegal for pharmaceutical companies to advertise off-label boosters, but they certainly don't need to directly do that.
|
|
#
¿
Sep 19, 2021 00:00
|
|
|
Suck Moredickis posted:I'm not even sure we know that the CDC was wrong. As far as I'm aware, we're still not sure the degree to which vaccinated people transmit the virus as opposed to unvaccinated. As we've seen, the delta variant has ripped through the unvaccinated and it isn't clear how much of that spread is due to the vaxxed vs the unvaxxed. N95s are certainly better for personal protection than a cloth mask, but if you're already vaccinated, an N95 may only offer you an insignificant amount of additional protection. The CDC was absolutely wrong. When they eliminated the mask requirement for vaccinated people, they claimed that vaccinated people were unlikely to transmit (and Walensky straight-up said that they "don't transmit"). They cited several papers to back that up, but none of them actually looked at the degree to which vaccinated people were contagious. At the time there was only a single preprint that directly addressed breakthrough transmissions, and it suggested that breakthrough cases likely had roughly half the transmission risk on unvaccinated cases, which is something but not super great for a virus as contagious as SARS-CoV-2. Unfortunately, this is a common problem with Walensky. Just two weeks ago she claimed that two studies published in the CDC's MMWR showed that "there was not increased disease severity in children" with Delta and "Instead, more children have COVID-19 because there is more disease in the community". However, neither of the two analyses were designed to distinguish the contributions of increased severity and increased community spread, and in fact the authors of both explicitly say that they cannot determine which is correct. If I had worked on those papers I would have been pissed at her spin. (The vaccination paper also has some serious issues with potential confounding that make it impossible to determine the actual contribution of vaccines, but that's a separate issue) Stickman fucked around with this message at 07:44 on Sep 19, 2021 |
|
#
¿
Sep 19, 2021 07:41
|
|
|
Fritz the Horse posted:edit: I really wish folks might discuss the actual data from Israel instead of bickering about Twitter personalities. In the other other COVID thread QuarkJets pointed out that Israel now has ~1500 severe cases <60 if you add in the August/September data. Hopefully they'll update their waning analysis so we can get an actual estimate for <60 + nail down >60. Same for boosters! QuarkJets posted:With error bars like those I don't think that you can make any serious conclusions Stickman posted:Okay, that makes more sense. Not being able to make conclusions about waning immunity in <60 was exactly my point - the waning immunity pre-print article was presumably limited to data available at submission, so it's just Delta cases through July 31st. The data you linked shows that there's more available now that Israel's been failing to contain Delta, so hopefully they'll release updated analyses soon. The analysis in the linked post needs a mountain of salt, though, because it doesn't have the ability to adjust for anything. The published booster analysis adjusts for finer age categories, demographics, date of second dose, and week of positive test (to account for massive variation in infection risk over the study). Stickman fucked around with this message at 10:07 on Sep 19, 2021 |
|
#
¿
Sep 19, 2021 10:04
|
|
|
Ignore the Natesayer
|
|
#
¿
Sep 19, 2021 23:54
|
|
|
That study is interesting and it definitely calls into question the comparability of hospitalization statistics without additional severity metrics. Some random thoughts: - I don't think SpO2 alone is necessarily sufficient to distinguish hospital admissions because of COVID from hospital admissions for other factors that happen to end up with a positive COVID test. BU's clinical admission algorithm also includes "dyspnea or increase respiratory rate (≥30 breaths per min)" and "clinical concern for outpatient failure due to risk factors" along with SpO2. I'm not sure what the rates of admission based on those different presentations are, though, and it's possible that admissions standards change over time or with hospital load as well. Some proportion of the detected COVID from other admissions might also meet one or both of those standards, so there's some overlap. Would dyspnea without progression to low SpO2 be considered mild or moderate COVID? - The hypothesis that a large portion of the non-severe COVID admissions are simply admissions for other reasons that test positive lines up with the decrease in admissions with low SpO2 in 2021. As cases declined and people started feeling like COVID was over hospitals had a huge backlog of non-COVID services that people had been putting off. The increase in the ratio of non-COVID- to COVID-caused admissions means that we would expect more non-COVID admissions that test positive. - It's also possible that shifting admission standards could increase the proportion with low SpO2. If hospitals have more space because cases are declining, perhaps they would start admitting less severe cases for observation. - I don't agree with Dr. Griffin's view that this is good evidence that vaccinated hospitalizations are less severe. The whole point of the study is that we don't know why the less severe cases were admitted. Since vaccinated folks are hospitalized at a lower rate than non-hospitalized folks, we would expect a greater proportion of their admissions to be due to incidental COVID detected in a non-COVID hospitalization (because the ratio of non-COVID to COVID health issues should be higher in the vaxxed group). It's possible that more vaccinated people were admitted under dyspnea or high-risk-factor guidelines, but if it's a small factor for unvaxxed it's probably a small factor for vaxxed too. If we actually wanted to show that outcomes for severe cases were better for vaxxed folks, we'd need to pick a baseline clinical threshold and then look at conditional outcomes from there. - If vaccinated folks do have a lower proportion of COVID-caused admissions, estimates of VE vs hospitalization will be biased low, but probably not by much since the issue affects both vaxxed and unvaxxed. Any VE vs severe disease based on clinical presentation wouldn't be affected. - They didn't adjust for any possible confounders, which is a little disappointing because it might help tease apart the "non-COVID hospitalizations" from "non-low-SpO2 COVID admissions". Of course the best way to do that would be to do their own data abstractions for a sample of cases; maybe they'll do that in a follow up? - I do like their suggestion for using "severe COVID" based on clinical definitions instead of hospital admissions as a proxy. Of course they're still relevant for hospital loading, though it would be important to have some way of distinguishing admission reasons. Stickman fucked around with this message at 04:01 on Sep 20, 2021 |
|
#
¿
Sep 20, 2021 03:57
|
|
|
It sounds like all of the cases they are calling "moderate-to-severe" should just be "severe" and the non-low-SpO2 admissions are likely "moderate". From one of the preprint comments: Margaret Chapman posted:I wonder why the team decided to deviate from the NIH definition of “moderate” covid (signs of respiratory infection but blood O2 at greater than 94%)? The cases they have categorized as moderate to sever would all be categorized as severe by the NIH guidelines. NIH clinical spectrum definitions: quote:symptomatic or Presymptomatic Infection:
|
|
#
¿
Sep 20, 2021 04:26
|
|
|
Fritz the Horse posted:thanks for giving it an actual read, shame on me a little bit I didn't dig into the underlying publication No shame, thanks for posting the article! It's very important (and somewhat depressing) for interpreting hospitalization statistics and how they shift. Hopefully they'll do some deeper follow-up so we can get a sense the breakdown between admissions for other causes with a non-serious COVID test and admissions for COVID with serious symptoms/risk-factors that happened to not have SpO2<94. Professor Beetus posted:Trick question; the dogs have already received treatment and the cat is still waiting to be seen. Probably another example of feline discrimination. That cat is looking pretty satisfied. Perhaps it is healthy but enjoys hanging out in vet clinics to watch dogs suffer?
|
|
#
¿
Sep 21, 2021 22:05
|
|
|
The study's estimate of the 95th percentile for the incubation period of Delta is ~12 days, not that far off from pre-Delta SARS-CoV-2. If they're concerned about quarantine infections, then I could see 21 days being sensible. That probably around the 95th percentile of a quarantine infection, which are likely relatively rare themselves. It would also have the added benefit of making travel less attractive without having to explicitly control it. The rest of it sounds not great, though I would personally take national zero-COVID control over local vaccination if it could be worked into regional/global vaccination strategy. Stickman fucked around with this message at 05:24 on Sep 22, 2021 |
|
#
¿
Sep 22, 2021 05:22
|
|
|
Platystemon posted:??? I'm assuming rotating cohorts were people with roughly the same intake period are quarantined on the same level/zone/whatever and if you get a secondary case then you can keep everyone else for some extra days. If everyone is just thrown in together willy-nilly then yeah, it wouldn't apply. Smeef posted:Can you clarify that last point about taking zero Covid over local vaccination? What is the end game for a zero covid strategy that does not include local vaccination? I don't see these as mutually exclusive in the least. Sure, I just mean that there is no global or even regional vaccination strategies, but given limited supplies the most sensible collective approach would be to first use those resources to alleviate disease burden in regions that cannot feasibly maintain zero-COVID controls, ie the opposite of how vaccination nationalism actually played out. Of course, a globally directed response could also feasibly build zero-COVID economic bubbles, which would make individual nation's zero-COVID strategies much easier to maintain until sufficient vaccine supplies were available. It's a pipe dream because there's no political will for international cooperation on that level and capital-driven countries like the US have fostered a culture deathly afraid of collective action on that level. Smeef posted:My expectation in 2020 was that strict controls with zero Covid targets would be eased as vaccination thresholds were met, as treatments improved, and as systems and infrastructure were strengthened to be more resilient and responsive to Covid. Easing controls would not mean stopping testing or shutting down systems that allow for adaptation to new outbreaks caused by mutations, etc. In the paper "incubation period" is the time from infection to symptom onset while "latent period" is the time from infection to becoming infectious, which they measured by time to detectable PCR. That seems ~8 days for the 95% (as opposed to 12 for incubation period), but if you're trying to minimize the risk of lockdowns you'd want something higher than the 95th percentile! If they're doing daily testing in quarantine, it does seem like that should be sufficient to catch index cases that might produce the secondary infections, though, so I guess that might throw out that argument for the longer quarantine. They could just extend the quarantine for potentially exposed people around the index case.
|
|
#
¿
Sep 22, 2021 06:48
|
|
|
Professor Beetus posted:Hey uh, I am pretty sure I know the answer, but those things people wear around their necks for "air filtration" are just stupid hokum right? The packaging literally sounds like snake oil. "Just clip this to your clothing or wear around your neck and you'll have a 4 ft shield around your person that will stop deadly viruses!"  Just don't use lasCOVID.
|
|
#
¿
Sep 23, 2021 00:26
|
|
|
That Itch.io link is the project page so it should always have the latest version! E: or do you mean a direct download link? Stickman fucked around with this message at 01:46 on Sep 23, 2021 |
|
#
¿
Sep 23, 2021 01:44
|
|
|
E not Q 
|
|
#
¿
Sep 23, 2021 01:46
|
|
|
quote:The committee voted against recommending a booster for people ages 18 to 64 who have a high risk of being exposed to the virus at work, including health care workers and teachers. The FDA approved boosters for this group - it's absolutely ridiculous that the CDC would not recommend them. Given the immediacy of necessity I would personally agree with the FDA that the similarity of antibody responses between >60 (for which we have confirmed effectiveness) and <60 (for which we don't) is sufficient evidence that boosting will likely significantly increase protection v infection, which also protects contacts and family of high-risk workers. E: Annoyed that Israel continues to sit on thousands of cases without updating their waning analysis.
|
|
#
¿
Sep 23, 2021 23:06
|
|
|
The CDC just makes recommendations but those recommendations usually have a lot of informal power to influence individual doctor's decisions / state's policies on vaccination distribution / etc. Had they recommended boosters for frontline workers most states would have started gearing up the infrastructure to get them boosted; without a recommendation it'll probably be much more patchwork.
Stickman fucked around with this message at 00:03 on Sep 24, 2021 |
|
#
¿
Sep 23, 2021 23:59
|
|
|
Oh good, this loving bullshit again.
|
|
#
¿
Sep 24, 2021 03:01
|
|
|
It's "seasonal" in the sense that outbreaks occur during seasons like winter, spring, summer, fall, wet, and dry.
|
|
#
¿
Sep 24, 2021 11:00
|
|
|
Slow News Day posted:The problem with the Israeli study about waning immunity is one of looking at statistics in a vacuum, without intuition from the real world. I've talked about this before but Simpson's paradox isn't actually a problem with the Israeli study, or at least not to the degree that you're implying. Epidemiologists are well aware of the potential for confounding (the larger principle underlying Simpson's paradox; reference just for anyone who might not know the term) and design their sampling and analyses to address them as much as possible. Israel's waning study only looks at infections in the Delta-dominant period from July 11-31 to avoid differences between Delta and Alpha infection. The analysis is stratified by age category, and further adjusted for specific age, gender, COVID testing propensity, major demographic group, and week of infection. That doesn't mean that the possibility of confounding is entirely eliminated, of course - that's an ever-present risk in observational research. For example, there may be behavioral differences between people in the same age/demographic/risk category who vaccinated early vs those who vaccinated later (though this would likely bias results towards an appearance of increasing effectiveness, since those with the greatest time since vaccination are also likely the most behaviorally risk-adverse). I would have liked to see adjustment for comorbidities, but that's more likely to affect serious disease analysis and probably not much of an issue for detected infection. Overall Israels' analysis does a decent job of addressing the primary known sources of confounding. Dowdy's tweet is pretty old so it didn't have the benefit of the multiple corroborating studies that have been published/pre-printed since late August, but it's a prime example of exactly the thing you are talking about - it's dangerous to make assumptions about underlying trends from aggregate data because they may be obscured by confounding effects. Dowdy's trend could easily explained by increasing (relative) risk of exposure in young children as schools/preschools/etc reopened offsetting declining protection against infection in older adults. I haven't been able to find a nice graph of rates stratified by age in LA County or California so it's just a theory, but something is surely confounding the relative rates because Kaiser Permanente Southern California released an extensive analysis of SoCal vaccination effectiveness through August 8th and found evidence of significant decline in effectiveness vs infection (Lancet preprint page and non-paywalled manuscript):   Waning effectiveness has been found in every study that I'm aware of: - Pfizer's phase 3 follow-up (randomized controlled trial). Vaccine efficacy vs infection declined from an average of 96% two weeks to two months post-vaccination, to an average of 84% 4 to 6 months post-vaccination. All of the follow-up was pre-Delta. - UK prospective cohort study with monthly testing. Pfizer VE vs infection during the Delta wave declined from ~85% to ~73% over the course of three months post-full-vaccination. Effectiveness and longevity was better for those <35.   - Qatar (test-negative case control). Pfizer VE vs Delta infection (any or symptomatic) declined to ~50% after 4 months. There are a couple more studies (like Mayo Clinic's Minnesota study) but I'd need to look through them again to remember if they adequately control for the potential effects of Delta. Stickman fucked around with this message at 23:20 on Sep 24, 2021 |
|
#
¿
Sep 24, 2021 23:12
|
|
|
Yeah, that makes sense too. Older folks in multi-generational homes would still be affected by increased secondary attack rates, but that just going a be a proportion compared to all kids 0-4. Hopefully they'll be some tracing analysis to break it down, but it's definitely clear that something is increasing the baseline infection risk of kids even more than the baseline risk of adults!
|
|
#
¿
Sep 25, 2021 04:42
|
|
|
Qatar estimated VE approching zero ~5 months out but it's an outlier. Most studies have been ~40-60% at 4-5 months v Delta (Israel, Southern California, UK, Minnesota). It's entirely reasonable for a average 40-60% VE to end up with a 90% attack rate in a prison under constant exposure, though (or outside of a prison under similar aggregate exposure over a longer period of time). E: Even a 90% effective vaccine will ultimately have very high cumulative breakthrough rates if measures are insufficient to control outbreaks. Stickman fucked around with this message at 05:27 on Sep 28, 2021 |
|
#
¿
Sep 28, 2021 05:20
|
|
|
Potato Salad posted:At this point you probably need to start citing where you're getting your information on these assertions that you're making, with the understanding that this will provide the thread the opportunity to point out where you are misinterpreting conclusions. The prison outbreak infected 74% of prisoners in an 80% fully vaxxed prison in one month, with very high attack rates for both vaccinated and unvaccinated. That's definitely sufficient to say that this level of vaccination was insufficient to control the prison outbreak. There's a decent chance that no level of vaccination would have controlled it, though that depends on unknown dynamics. Vaxxed + previous infection seemed to have lower attack rates, but there enough uncontrolled confounders that it's tough to say anything beyond the population-level "vax rate was high but most were still infected" and "rates of serious disease were probably lower than they would have been otherwise". HelloSailorSign posted:What's fun about this one is that OC43 (a betacoronavirus to note, not a sarbecovirus as SARS CoV-2 is, though the distinction is mostly nil for our purposes) is suspected to have been responsible for a late 19th century pandemic which killed a few million people worldwide when it made its crossover from cattle. OC43 is significantly less virulent than SARS-CoV-2 in our current population, but it's definitely one of the nastier viruses we call "common colds" and has significant morbidity burden. While it's tough to find good information on its epidemiology and outcomes, one study looking at ~1,400 hospitalization for respiratory infections across 4 years in NY (mostly older patients) found that OC43 accounted for ~4-5%, compared to ~10% RSV and ~12% influenza A and B. They also note that the actual comparative hospitalization burden may be higher since they surveyed across the peak flu season, which didn't seem to coincide perfectly with OC43's seasonal pattern. While it's certainly possible that SARS-CoV-2 might end up eventually where oc43 is in terms of virulence, but I don't think we're at the point where we can assume that will happen or know how quickly it might occur. I assume that if oc43 is now significantly reduced in virulence that it was a combination of acquired immunity from repeated exposures (seasonal prevalence is currently ~5-10%) and mutation. Given that vaccinated breakthrough SARS-CoV-2 infections are still significantly more dangerous than seasonal influenza for older folks (based on estimated protection v severe disease and unvaccinated severity) we're definitely not there with vaccines alone, and probably not with infection-acquired immunity either. Unfortunately, I haven't found any good studies of the relative severity of reinfection - most just estimate protection vs reinfection but not severe disease. There's one study that claims to but it only looks at confirmed reinfections, which is pretty worthless for determining overall severity since less severe cases are not likely to be confirmed. Since we're talking about oc43, I got a hearty sad lol out of this anti-prescient paper from March 2020 that I came across while looking for oc43 sources: SARS-CoV-2: fear versus data the abstract posted:SARS-CoV-2, the novel coronavirus from China, is spreading around the world, causing a huge reaction despite its current low incidence outside China and the Far East. Four common coronaviruses are in current circulation and cause millions of cases worldwide. This article compares the incidence and mortality rates of these four common coronaviruses with those of SARS-CoV-2 in Organisation for Economic Co-operation and Development countries. It is concluded that the problem of SARS-CoV-2 is probably being overestimated, as 2.6 million people die of respiratory infections each year compared with less than 4000 deaths for SARS-CoV-2 at the time of writing. They then proceed to abuse the heck out the differences between CFR and IFR while ignoring study population bias effects! It looks like it didn't get a print publication until May - that must have been embarrassing.
|
|
#
¿
Sep 28, 2021 21:43
|
|
|
Professor Beetus posted:Seriously. The first conclusion isn't unreasonable, but I think the second two are just kinda baseless speculation without a larger sample size. The second one seems pretty reasonable, too. It's is an example of 80% full vaccination failing to prevent very high attack rates in a prison setting, so it wouldn't be surprising to see similar results in comparable environments. Stickman fucked around with this message at 21:49 on Sep 28, 2021 |
|
#
¿
Sep 28, 2021 21:45
|
|
|
|
| # ¿ May 11, 2024 16:35 |
|
|
Platystemon posted:It’s interesting that OC43 appears to have mellowed out quickly, but there’s a lot about the behavior of coronaviruses that we don’t understand. The recent evidence that the 1890 pandemic was OC43 is pretty cool, too (in a horrifying way). The molecular clock argument has been around for a while: the divergence of OC43 from it's closest bovine relative was estimated to be ~1890 back in 2005. The new evidence is that it's clinical presentation seems closer to SARS-CoV-2 than to influenza. quote:Contemporary medical reports from Britain and Germany on patients suffering from a pandemic infection between 1889 and 1891, which was historically referred to as the Russian flu, share a number of characteristics with COVID-19. Most notable are aspects of multisystem affections comprising respiratory, gastrointestinal and neurological symptoms including loss of taste and smell perception; a protracted recovery resembling long covid and pathology observations of thrombosis in multiple organs, inflammation and rheumatic affections. As in COVID-19 and unlike in influenza, mortality was seen in elderly subjects while children were only weakly affected. Contemporary reports noted trans-species infection between pet animals or horses and humans, which would concur with a cross-infection by a broad host range bovine coronavirus dated by molecular clock arguments to an about 1890 cross-species infection event. E: Where'd the bunny go? Stickman fucked around with this message at 01:01 on Sep 29, 2021 |
|
#
¿
Sep 29, 2021 00:57
|
|