4 ms·

"Material difference The study linked surgical masks with an 11% drop in risk, compared with a 5% drop for cloth..."

angelzen 5y ago
Consider taking it up with Lynne Peeples. The very subtitle of the piece is 'surgical masks are highly protective, but cloth masks fall short', I have no idea why she thinks '11% drop in risk' is 'highly protective', but '5% drop' is 'falling short'. Either way the numbers appear marginal at best. Edit: Fundamentally I have an issue with any definition of risk that bounds the measure to a short time frame. For all we know the mask intervention may slow the spread speed by a small factor, but were we to measure over a year (or even longer) we might observe the same total number of infections: the virus is going to infect everyone susceptible, only at a slightly slower pace.
- mike00632 5y ago
  1. Based on the size of the study there are thresholds for statistical significance. 2. Only a portion of the people were even wearing masks. A few people wore masks and the risk of infection dropped by a few percent. Declaring masks don't work because a couple people wore them yet the entire population didn't see a complete decline in infection is illogical and dishonest. 3. The study not being able to conclude in a statistically significant way that cloth masks prevent infection is completely different from concluding that cloth masks are worthless. Your aversion to masks is completely a reflection of your bias and not supported by this study.
  - timr 5y ago
    > A few people wore masks ...there were 54,000 people in the cloth mask villages, and 106,000 people in the surgical mask villages. Compliance was over 40%. That's more than "a few people".
timr 5y ago
The difference in the cloth arm wasn't even close to statistically significant (p=0.540). It was noise.
- mike00632 5y ago
  Yes, and as I wrote in another comment, that is completely different from concluding that cloth masks don't work. You're drawing a conclusion that the data doesn't support.
  - timr 5y ago
    > You're drawing a conclusion that the data doesn't support. I'm telling you the conclusions of the paper. Your faith in masks is an entirely separate question that is clearly unaffected by the data here, because it requires proving a negative. Is it possible that cloth masks actually work, and this huge study failed to find any significant effect? Anything is possible, but the results here are indistinguishable from noise, and don't support the claim. And with every RCT that gets published showing negligible effect from community masking, the probability that the claim is true gets smaller.

Consider taking it up with Lynne Peeples. The very subtitle of the piece is 'surgical masks are highly protective, but cloth masks fall short', I have no idea why she thinks '11% drop in risk' is 'highly protective', but '5% drop' is 'falling short'. Either way the numbers appear marginal at best. Edit: Fundamentally I have an issue with any definition of risk that bounds the measure to a short time frame. For all we know the mask intervention may slow the spread speed by a small factor, but were we to measure over a year (or even longer) we might observe the same total number of infections: the virus is going to infect everyone susceptible, only at a slightly slower pace.

mike00632 5y ago
1. Based on the size of the study there are thresholds for statistical significance. 2. Only a portion of the people were even wearing masks. A few people wore masks and the risk of infection dropped by a few percent. Declaring masks don't work because a couple people wore them yet the entire population didn't see a complete decline in infection is illogical and dishonest. 3. The study not being able to conclude in a statistically significant way that cloth masks prevent infection is completely different from concluding that cloth masks are worthless. Your aversion to masks is completely a reflection of your bias and not supported by this study.
- timr 5y ago
  > A few people wore masks ...there were 54,000 people in the cloth mask villages, and 106,000 people in the surgical mask villages. Compliance was over 40%. That's more than "a few people".

timr 5y ago

The difference in the cloth arm wasn't even close to statistically significant (p=0.540). It was noise.

mike00632 5y ago
Yes, and as I wrote in another comment, that is completely different from concluding that cloth masks don't work. You're drawing a conclusion that the data doesn't support.
- timr 5y ago
  > You're drawing a conclusion that the data doesn't support. I'm telling you the conclusions of the paper. Your faith in masks is an entirely separate question that is clearly unaffected by the data here, because it requires proving a negative. Is it possible that cloth masks actually work, and this huge study failed to find any significant effect? Anything is possible, but the results here are indistinguishable from noise, and don't support the claim. And with every RCT that gets published showing negligible effect from community masking, the probability that the claim is true gets smaller.