Coronavirus in the United States - news and thoughts

Why only look at deaths? What about hospitalization and follow up therapy which would have been almost entirely avoidable?

 

Don’t you know it’s not about them… they have to get it to protect everyone else.

 

The unvaxxed don't think it is a threat in the first place.

 

First, asymptomatic people can spread the disease. This is from the WHO, who claims the disease is spread from both symptomatic and asymptomatic. To just say let's ignore the asymptomatic spreaders because it completely ruins your argument is disingenuous. You can't just ignore something that happens. And if you don't like the WHO, here's a link about asymptomatic spread from the University of California medical system.

The study between vaccinated breakthroughs and unvaccinated looked at peak viral load. Which was the same. But that alone doesn't account for how contagious one is. The vaccinated rids the body of the virus days faster than the unvaccinated, meaning the vaccinated peak is more pointed, and the unvaccinated have more of a plateau peak. Therefore, the vaccinated breakthrough case is contagious for a significantly shorter period, therefore, they are less likely to spread COVID. Again, if you don't like the first link, here's another with a different study, same results, that show you are far less likely to spread COVID if you are vaccinated versus unvaccinated.

Vaccines aren't cures. If the vaccine was a cure, it would be called a cure, and not a vaccine. What the vaccine does is give a person protection from catching COVID in the first place, protection that if he/she does catch it, a better chance the case will be mild, and quicker time to rid the body of the virus, meaning the vaccinated is less contagious. This means, if everyone eligible were up-to-date with vaccinations, there would be less cases overall, less hospitalizations, and less deaths. So I ask again, why is any of this a bad thing?

 

Show Me State governor hid data showing that masks work

As the delta wave rose in Missouri last summer, much of the state remained unmasked. Four jurisdictions, though, restored their mask mandates, creating a natural experiment that was studied by the state’s Department of Health and Senior Services. It confirmed that, in cities and counties with mandates, masks significantly reduced infections and deaths from COVID.

Yet Gov. Mike Parson’s office, which had requested the data, kept it hidden from the public, according to a new report from the Missouri Independent.

The data was initially requested by Alex Tuttle, Gov. Parson’s legislative liaison for DHSS, on November 1, 2021. “Can you provide examples of local mandates and how those mandates impacted the spread of COVID in those areas?” he wrote.

In just 48 hours, DHSS had an answer for him. “I think we can say with great confidence reviewing the public health literature and then looking at the results in your study that communities where masks were required had a lower positivity rate per 100,000 and experienced lower death rates,” DHSS Director Donald Kauerauf wrote.

 

There isn’t data showing masks work. If there was it would be plastered everywhere. The Bangladesh study has been shot down once they finally released the data.

 

The data is as good as any. You can't say definitively that it was the masks but it certainly is a logical candidate. Also would be interesting to see how vax rates compare. The difference could actually be due to that.

In the past you have compared different states in different geographies in different parts of the cycle to assert that they don't work. When something closer to an apples to apples comparison comes along you immediately dismiss it.

Your schtick gets old.

 

Interesting story on the possible origins of Omicron.

The mystery of where omicron came from — and why it matters — NPR

 

Are you suggesting that the vaccine worked?

 

Let's go with the Bangladesh study then. So let's throw out everything else that was done(encouraging mask wearing, showing how to wear properly, giving incentives) and there was a grand total difference of 20 cases. So out of 340k people we had a 20 case difference, which means no statistical difference. So, yes, again, masks are PROVEN to not work. It's ok, jump on board. Better late than never.

 

Huh? Why/how would you throw out their treatment in a RCT? They randomized some additional treatments within the treatment group to test certain interventions (although they had materials and a scripted speech for Mosques for all treatment villages encouraging mask usage), but the big question was the change in behavior of each of those. And you certainly can't use the entire sample size for small subsamples with a randomization in treatment to see if it encourages mask usage better.

 

You aren't making sense. Simple fact is there is no statistical difference in the control and mask groups. The next time an RCT on masks shows that they work will be the first one.

 

Actually, yes there was.

AAAS

 

Nope. When a "study" doesn't release the data right away it should be a giant red flag. Some people believe anything I guess. Simple facts here, 1,086 vers 1106 case number after all that. Sucks, maybe next time sport.

A note on sampling biases in the Bangladesh mask trial

 

It's a pre-print study, not yet peer reviewed from South Africa, but the results show Omicron is 3X more likely to re-infect people who previously had COVID. No studies that I have found yet that are even in preprint that show vaccines efficacy against Omicron. Those studies are under way.

 

Please interpret that pre-print response in your own words, because the conclusions do not match what you claimed in your first post, even if we were to take them at completely face value.

Hint: in this interpretation, please account for the differences in effects between Covid symptoms (significant) and Covid Seroprevalence (significant in rate, but not raw figures) as DV.

 

I understand it's hard to fathom that the difference in the 2 groups was 20 cases over the whole trial. That's a tough pill to swallow. Nothing to interpret. The authors of the "study" didn't want to release the data at the time, wonder why LOL.

 

Okay, so let me post a true summary of what they posted:

The paper by the Pittsburgh, CMU, and Cal-Berkeley profs looks at the data and looks at one of their results: Seroprevalence Rate, which is calculated as Positive Tests/Total Population in the Village that took a test. The authors in the pre-print state that the issue was that there was an important distinction in the rates at which the households were declared "unreachable" between the control and treatment groups. They argue that this pushes the result by eliminating the rate DV and instead re-analyzing the data with raw positive tests. So essentially, the result may not be a true result but is potentially an artifact of a lack of true randomization.

So to put this in mathematical terms, the authors compared the rates of seroprevalence, which means in the treatment group, you saw a rate of 1086/170,497 = 0.637%, and in the control group, you saw a rate of 1,106/156,938 = 0.705%. That difference is significant. However, if you just compare 1,096 to 1,106, the result is not significant.

That is a smart point and is well put by the authors (unsurprisingly, because these are serious scientists as well). It could be pointed out that the difference in sample sizes makes the difference between positive tests a less appropriate statistic on its own, but it is a smart way to try to make the point that the denominators might be pushing the results.

Now, here is the other side of that argument from the initial study's authors:

1. This only looks at one of the DVs. The other DVs, most notably symptoms, show significant decreases (often with very substantial effect sizes) in both the original paper and in the re-analysis (although the authors of this response do not emphasize this for obvious reasons as replicating results is not as interesting) for treatment group. So when you get to interpretation, if the masks aren't preventing Covid, they must be preventing other Covid-like diseases with similar symptoms. It seems like a hard argument to make, especially if your desired interpretation is masks don't work in a general sense.
2. In order for the pre-print's authors' interpretation to really be true (in that the differences in "unreachable" rates pushed the results), you need to argue that classification of a household as unreachable was not random at the village level. If it was, it is a fundamentally unimportant point in terms of interpretation and is really just a small disagreement over methods.

In a response to the second point, the CMU professor states (correctly) that it is still an issue because RCTs are supposed to eliminate possibilities like this and that it would be appropriate to ask for a robustness check in which the randomization of "unreachable" households is tested.

To your last point, the authors were not hiding data and it is common not to release data until after publication, as different journals have different rules as to how data can be released. The authors of the pre-print specifically praise the authors for their willingness to release data at the first time that data for published academic work is generally released to the public (upon acceptance of the paper). The initial paper was recently accepted for publication at Science.

 

I'm waiting to see results from the vaunted double mask study.