Hacker News Reader

Comment by CoastalCoder | Hacker News Reader

Slight tangent, but does anyone know of a website that acts as a decent, up-to-date clearing house for what we do/don't know about COVID?

I sometimes see articles providing new (perhaps preliminary) info about how effective masks are in different circumstances, how vaccine efficacy changes over time, potential for vaccinated persons to spread the virus, etc. I'd love to see a single website that tracks and reconciles all of those developments into a current summary.

Just for context, my personal use case is trying to estimate the riskiness of having my family participate in various social activities in the next few months.

View on HN

Here you go: https://www.microcovid.org/

Great site. Has variables for mask type, vaccine status, etc

I’d add that immunity pretty clearly wanes post infection and vaccination. Especially by six months, but starting at 3.

Single biggest and easiest interventions that people don’t do:

* N95 vs other mask. Well fitted, about as comfortable. Just stupid we’re not using masks rated to filter. Or, a brace on a surgical mask works well too

* meeting outside

* Opening windows. Even a crack helps, if weather calls for AC or heating. People treat this as impossible, but really it just raises electricity costs marginally

* HEPA filter in your home. This is a good idea for other reasons anyway

* Saline rinse in nose when returning home. Again, also not a bad idea for other reasons

Anyone using this site should know that it is based on no validated method that I'm aware of. It's entirely speculative.

There are way too many people out there making "risk models" that are based on assumptions that haven't been validated, and methods that haven't been tested. A model is no better than the assumptions used to create it, and simply because a site tells you what they've done, doesn't mean that what they've done is valid.

Not having examined the site’s model, I’d indeed put more stock into the specific interventions I listed than modelling estimates.

We know certain things move the needle and can evaluate their cost. Knowing that is more useful than knowing the specific percent improvement.

Another layer of protection that can be used before a gathering is rapid tests day of (on top of other measures)

And the big idea from the site that risks rise as prevalence rises is true. The site could be off by a factor of ten in absolute terms, but the relative change in risk based on tests + positivity rate will be accurate.

> I’d indeed put more stock into the specific interventions I listed than modelling estimates. We know certain things move the needle and can evaluate their cost.

I don't really want to get into this debate, but at least some of what you listed falls under the category of "things people 'know', but which are unsupported by evidence".

Not the best example from the list, but to avoid the ones that are now hopelessly mired in political infighting, "saline rinse in the nose" is almost completely without supporting evidence. Lots of people say things about moistening the mucous membranes and whatnot, but it's basically superstition.

Take this letter, which cites a number of weak/equivocal studies on different viruses, then brings up an actual RCT which showed that a combination gargling / ultrasonic lavage treatment (i.e. not your standard neti pot) within 48 hours of symptom onset (good luck with that), had a ~mediocre effect on symptom duration (-1.9 days) and intra-household transmission (-35%) of unspecified "upper respiratory tract infections".

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7436790/

Based on this evidence, if someone asked me if daily saline rinse made any difference for Covid, the only answer I could possibly give is "we don't know".

Oh sorry, I should have qualified that one. It indeed has weakest evidence. I included it as it’s harmless, helpful for other reasons for many people, and has plausible physical mechanisms.

But yeah strike that one from the list of interventions we can be 100% confident about.

The rest though are completely certain to help given how aerosols work, so I wouldn’t go too broad with your comment.

You can physically measure both the effect of windows and HEPA on aerosols. With N95 masks they to through specific ratings for filtration of virus sized particles. And going outdoors is like opening a very large window. So physics would have to be wrong for any of those not to be helpful.

Saline in the nose does have well establish physical benefits in the nose. However, the nose’s biology is a complex system so pointing to the physical system is not as simple as measuring a reduction in aerosol particles. But these physical properties were in the back of my kind when I included it in the list. Nonetheless an oversight not to qualify it so thank you:

https://jamanetwork.com/journals/jamaotolaryngology/fullarti...

> The rest though are completely certain to help given how aerosols work

There's a lot in your comment that is past the available evidence.

Your comment started with the assertion that immunity "pretty clearly wanes post-infection". This isn't clear at all, and in fact, the majority of studies show a robust effect, especially with regard to mortality, but also for simple infection. The minority of studies that show a decline have generally shown minor declines, been extreme outliers, or have focused on meaningless metrics, like "antibody titers":

https://nitter.net/mugecevik/status/1430218372348878860?lang...

Then you talk about "N95 vs. surgical"...there are no studies to support your assertion, outside of pre-pandemic studies of flu in hospitals. Same for the comment about "brace on a surgical mask works well too". Plausible? Sure. Demonstrated amongst a real-world population? No.

Opening windows..."even a crack helps". Well, maybe. Ventilation is certainly more important than masks in general, but what level of ventilation? Is it really true that "even a crack" helps? Not really any data here. Mechanistically plausible, but lots of things are mechanistically plausible and make no difference.

"HEPA filter in your home." Absolutely no evidence of this. Again, mechanistic plausibility isn't the end of the story. There's such a vast array of things called "HEPA filters" that I couldn't begin to tell someone if their particular filter makes any difference at all.

In general, you're leaping from mechanistic theories and/or weak data to strong conclusions. I don't even disagree with most of them in principle, but I wouldn't draw the conclusions you have.

Take a look at https://www.reddit.com/r/Coronavirus/ and https://www.reddit.com/r/COVID19/.

The latter subreddit is the highest quality source of covid studies I know of. Great place, high discussion quality.

The (apparent) lack of a dashboard for information is one of the reason some of us are monitoring the articles :)

What some of us would have wanted is a frequently updated collection built, as a structure, over the commonsensical questions (e.g. "What is the average risk of transmission in a closed room of this size, this time of exposure and these other variables?"), and filled as studies are published, and annotated with the relevant references ("Virions to get infected: 300 (New Journal of Foo, August 1 etc.)").

Reply on news.ycombinator.com