I would trust sound data and transparent analyses showing age-stratified risk analyses if they existed. With all respect, what you just wrote is essentially hand-waving and talking past what I wrote. High effectiveness for elderly people (which I find plausible but also still deeply confounded by testing rate differences) in no way justifies mandates for people in their early 20s.
In some proportion of hospitals, when someone who is not vaccinated goes to hospital for any reason, they are tested for COVID-19. If found positive this is counted as a COVID-19 hospitalization. We have no access to the precise rate of these incidental hospitalizations. In some of these hospitals, it is the policy that vaccinated people who go to the hospital are NOT routinely tested unless they have symptoms. If this is the policy at a substantial number of hospitals, it could dramatically change the "effect size" of the measurements that we are talking about. The same issues essentially applies to COVID-19 deaths and cases.
Large effect size alone generally isn't convincing when you're using such fundamentally confounded sampling procedures, merged age-groups with wildly different risk profiles, and data aggregated across long time-windows with different population sizes.
The problem with this line of reasoning is that there is a huge difference between controlled trials (which are obviously imperfect themselves) and post-hoc analyses of uncontrolled real-world outcomes. There's a big leap of faith in trusting our public health authorities' ability to accurately measure and faithfully represent the real situation.
Just to name one small issue, our only measurements of vaccine efficacy are case counts, covid hospitalizations and covid deaths. Each one of these measurements is confounded by population-level differences in testing rate, testing polices (ie routine testing on entrance to hospital), PCR cycle count, and many other factors. There is no longer a monitored control group, so we can't ever account for any group-level differences or confounds. I've seen no attempt to address these issues.
We also have no access to reliable data about confirmed adverse effects. A year later, it is still very hard for a person to quantitatively assess his/her own age-stratified risk/benefit tradeoff, even with the confounded efficacy measures.
So, given all that, why isn't skepticism reasonable anymore?
My opinion is that we are seeing a mass depoliticization across many if not all communities. Paradoxically, people are becoming less capable of engaging in political discourse, while clinging ever more tightly onto simple pieties and tribalistic lines in the sand. Ironically, this rigidity is perceived as people being "too political", because it manifests as an increase in anger and communication breakdown. In fact it's the opposite of political discourse.
When it comes to the pandemic, the immediate cause for this change, I think people also come out on either side of one basic disappointment. On one side, there are people who feel they've been massively betrayed by their fellow citizens who have not been cooperative or compliant with the recommendations of experts. On the other side, there are those who feel massively betrayed by regulatory institutions and media they feel are untrustworthy and captured by corporate interests, unelected health experts who have made policy misjudgements, and a wealthy class that has profited from the crisis that their institutions have exacerbated.
Although it's possible to recognize truth in both of the above sides, it is not common in the wild. The division is very primitive, having to do with our responsibilities to our fellow citizens, our relationship to authority, fundamental rights over our bodies, and the expectation of transparency in our institutions. You'd think we would be able to start from some shared assumptions and common ground on these issues, but it's almost impossible today to have discussion about it without one or both parties instantly devolving into anger, fatigue, frustration. We've lost the collective ability to even articulate one another's positions about the pandemic response and other basic civic issues, which is, I think, one of the few things that marks political discourse as different from base tribalism.
Given how effective we know "influence campaigns" to be, I don't think it's an accident that we've arrived here, but that's another issue altogether.
These numbers are not even remotely comparable. The CDC number includes ages groups that are much more likely to have severe infections. The CDC number you cited is also based on hospitalized cases, which select for severity
To be clear: this is an overconfident assumption, not an evidence-based claim. It's not even plausible on the surface. Why would we assume these proteins would be distributed similarly in the infection case and the vaccination case? We know that very different physiological processes will be taking place, so why assume the inflammatory responses must be the same?
I am absolutely pro-vaccine, but can someone help me work through the relative risk calculations here?
Vaccine:
-?% reduction in likelihood of infection (there is currently no monitored control group and no transparent dataset that accurately accounts for baseline infection rates, geographical and demographic variation, etc.)
-?% chance of ?% reduction in severity of disease (there is no empirical evidence that has been consistent across time and location, especially as the virus mutates. All the data I've seen is aggregated across the entire vaccination campaign, never accounting for changing rate of vaccinated vs changes in baseline infection rate in each location)
-?% chance of acute vaccine side effects (VAERS unreliable, no transparent alternative dataset accessible to the public)
-?% risk of future unknown long-term effects of vaccination (including unpredictable outcomes like ADE)
-100% chance of assuming above risk of unknown long-term vaccine side effects (simply no data available by definition)
-?% chance of getting breakthrough infected with COVID-19 if exposed (I haven't seen any challenge studies, probably not possible for ethical reasons)
-?% risk of future long-term effects for recovered breakthrough infections (again, no data available. only time will tell whether breakthrough infections are less likely to lead to potential long-term effects than unvaccinated recovered infections)
-each additional vaccine booster dose adds ?% cumulative risk of both acute and long-term side effects, with no indication that boosters will ever end.
No Vaccine:
-0% chance of reduction in severity of disease
-0% chance of acute vaccine side effects
-0% chance of long-term vaccine side effects
-?% chance of getting infected and assuming a ?% risk of future long-term effects of recovered infection (my risk of getting infected is low and partially under my own control since I don't travel, work from home, self-isolate, etc)
Am I missing anything?
Even if I am relatively confident that the vaccine reduces the likelihood of infection and the severity of infection significantly and will continue to do so for variants (the data suggests this, though it less transparent than I would like, since there is no publicly accessible unaggregated dataset (e.g. infection rates for vaxxed vs unvaxxed starting AFTER 50% vaccination rate in a given region, there is no longer a control group and all "natural" control groups suffer from inherent geographical and demographic biases)
It seems to me that the number of unknowns makes the risk calculation much harder than people are making it out to be, but I would be very happy to to hear considered, thoughtful counterpoints to the above.
I'm not trying to deny that the vaccine has a positive effect, but this data as presented (aggregated across the entire year) is extremely misleading. Here's why:
1) The infection/hospitalization/death rates were highest last winter, and relatively few people were vaccinated during the peak of the infection
2) A large proportion of the people in the Unvaccinated column here were infected long before we reached the current vaccination rate.
3) Therefore, suggesting that the numbers in this table represent the outcome of vaccination is simply not correct. This data is aggregated across the entire vaccination campaign and is completely confounded by the timing of waves vs that of vaccination rates.
If the goal is to build trust and present data transparently and fairly, why not show the same data with aggregation starting in July or starting AFTER reaching 40% vaccination rate?
I'm not sure why you think I'm trolling. I want answers to my questions about the long term energy costs of handling the respective industrial waste products from each technology. I thought maybe someone would know or have some informative comment.
I have absolutely no personal agenda in asking, beyond a general concern about whether environmental risks are being fully considered. I support electric vehicles and want them to succeed.
You're just being rude. Maybe you're having a bad day. I'll give you the chance to apologize now, in front of everyone.
no need for that. I am asking these questions out of genuine curiosity. Are you certain that there is no extra heavy metal waste or plastic waste from the entire EV production chain? Can you nudge me towards that evidence?
First of all, as a premise, to me the only practical question of interest is, "which technology is better for the long term health of ecosystems". Greenhouse emissions are one critical factor among many that need to be considered when asking this question.
Given that, how does one accurately compare the relative environmental impact of greenhouse emissions versus that of heavy metals, plastics and other synthetic substances that result from the mass production and distribution of batteries and charging stations? I have never seen a convincing explanation of how this comparison could be done in a principled, empirical way.
Even putting aside the issue of being able to compare across pollutant types, are we able to confidently forecast the potential energy costs associated with the future remediation, reclamation, and/or storage of these substances?
Given that many of these materials and production biproducts (cadmium, mercury, lithium processing biproducts, PCBs, etc.) have catastrophic health effects on both humans and animals, how do we account for the energy costs of attempting to fix, manage or avoid those potential health and long-term ecosystem consequences?
If there is an agreed upon scale for comparing the environmental risk of plastic and heavy metal waste to that of emissions, where can I read about it? How was it determined, by which organizations, and what are the standards? Is there a "time preference" included as a parameter in the risk model? Do we account for worst case potential effects of bioaccumulation? and so on...
tldr: great news for EVs, greenhouse emissions are a serious issue, but how can we be certain that we have the right relative risk model?
one frightening prediction I've heard is that we will have parallel economies. One for UBI consumer serfs who receive a monthly stipend that expires at the end of each month, another for people who work and receive payment perhaps in other independent currencies, and another, which already exists, for the very wealthy who hold wealth primarily in tangible assets like real estate and fine art. This will either fail spectacularly or it will be the beginning of an even more rigid caste system.
Very worrying for a system like this to be produced and executed by institutions who have proven time and again that things like human dignity, civil liberties and a chance at upward mobility are no longer relevent.
Well, we won't know if the "anti-science movement" can actually be dealt with until we fix the corruption first, will we? Why assume that things can't be improved? You wouldn't try to secure a second funding round without a viable product or even a concrete business plan, would you? Bad example maybe...
Regardless, simply telling people they need to listen obediently to experts is going to do much more harm than good. If we take concrete steps to fix the massive problems in our institutions and there are still legions of anti-science activists holding us back through political means, we can at least point to our own reformed system with integrity and say that we're doing our best to get it right. Until then, we have no solid ground to stand on, and we'll continue to lose our own base of supporters, myself included. Scientists themselves will look elsewhere for a system that supports them.
When you come into a conversation not knowing your interlocutor's intentions, I believe it is deeply unethical, inefficient and really a complete waste of everyone's time and energy to assume based on some weak superficial heuristic rhetorical cues that they are simply repeating a divisive talking point that you've heard elsewhere.
Again it's this a priori assumption that the other person is dumb that has gotten us into this situation. It's also just bad manners. All else equal, I do not see how you can possible read what I wrote as specifically promoting anti-vaxers. I didn't say a word about vaccination. You're applying your own baggage to a simple message. Anti-vaxers are for the most part just frightened people who deserve compassion, not derision. They've simply overgeneralized a very rational fear of institutions that have indeed made some very careless decisions for profit.
If you treat every conversation as adversarial by default, you will continue making all your conversations worse. I recommend following the Gricean Maxims in your conversations. Be productive and cooperative with others until you are certain that they're not being productive and cooperative. That's when you leave the conversation. It's not as complicated as you're making it out to be.
As someone with an undergraduate degree in engineering and some professional experience as an engineer, I would counter that the training of engineers absolutely does deprioritize the kind of high level critical thinking I'm talking about. I'm not saying every engineer needs to be an ethics and politics scholar, but there is definitely room for improvement.
Nuance in communication and dealing intelligently with ambiguity is also a really big issue in all of STEM. Engineering is no exception. This is in fact something that COULD be improved. It's just not as profitable.
Yes, I'm familiar with this idea, and I think it really does get close to one of the roots of the problem. I hope more people start listening carefully to individuals like Mr. Weinstein before it's too late to change course.
I don't always agree with everything he says on other issues, but we are lucky to have him in a position where he's afforded the right to speak freely. It's a shameful embarrassment to the state of our public discourse that it takes a billionaire hedge fund backing to secure that right.
I totally agree. I hope people like you will keep fighting for a vision like this. It's the only hope we have, and this problem is far more serious than most people are willing to admit. Unfortunately being outspoken about this viewpoint can be a big liability in a scientific career.
Right, but what's your point? If it's an old problem we shouldn't try to address it? I think the evidence points to it getting worse and more importantly becoming more and more of a liability for our future ability to compete on the world stage.
In some proportion of hospitals, when someone who is not vaccinated goes to hospital for any reason, they are tested for COVID-19. If found positive this is counted as a COVID-19 hospitalization. We have no access to the precise rate of these incidental hospitalizations. In some of these hospitals, it is the policy that vaccinated people who go to the hospital are NOT routinely tested unless they have symptoms. If this is the policy at a substantial number of hospitals, it could dramatically change the "effect size" of the measurements that we are talking about. The same issues essentially applies to COVID-19 deaths and cases.
Large effect size alone generally isn't convincing when you're using such fundamentally confounded sampling procedures, merged age-groups with wildly different risk profiles, and data aggregated across long time-windows with different population sizes.