A recent paper by Christopher Ruhm of the University of Virginia quantifies the value of various efforts to combat COVID-19 in the US. The headline result is a composite score for different states based on what kinds of restrictions they imposed, but I found the detailed national breakdown more interesting. Here are his estimates of how various interventions affected death rates:
By comparing excess death rates in different states with different restrictions in place, Ruhm estimates how effective each one was. The results are pretty simple:
- Vaccines work great.
- Mask mandates work OK.
- Prohibitions on mask and vaccine mandates are terrible.
Nothing else has much impact. The effects are all fairly small and in many cases aren't even statistically significant. Restaurant closures are minimally effective, but probably not worth it (though it's almost certainly a good idea for wait staff to wear masks).
This is one study, and its results are hardly the last word. Still, it's striking that an awful lot of research has now been done and it mostly finds that hardly anything makes much of a difference. Masks help a little bit, but aside from that it's vaccines or nothing.
Note, however, that there are several interventions that weren't measured in this study: social distancing, better ventilation, UV light, and remote work. All of these are things that likely have some impact.
I'm picturing Norm Macdonald in his SNL Weekend Update role going: "These findings are published in the latest issue of the scientific journal 'Duh!'"
I currently make about 6000-8000 dollars /month for freelancing I do from my home. For those of you who are ready to complete easy online jobs for 2-5 h every day from the comfort of your home and make a solid profit at the same time
.
.
Try this work______ https://dailypayment002.pages.dev
Again, the big differences in death rate were between nations - mostly Asian nations vs others - and there were big differences in measures taken. Trying to get a lot out of the different places in the US, which was pretty poor all around, is doing it the hard way.
From the paper:
You're kind of misinterpreting and misrepresenting the effects of mask mandates. School-based mask mandates worked just as well as school mandates on vaccinations for a reason: Difficult to get around compliance. Does that mean statewide mask mandates were just "OK"? No. It means compliance was determinative on excess deaths -- ignorant folks who refused to comply, often ended up dead. But without mask mandates, which we can interpret to be equal to a prohibition on mask mandates, was absolutely devastating. So, you're wrong and mask mandates were extremely beneficial.
Oh, and your title is wrong; it's not "what worked to stop the spread of COVID-19", it's "what worked to prevent COVID-19 deaths."
Well said!
It's actually pretty striking that school mask mandates were that effective. There are good a priori reasons to regard students as disease vectors, and some studies that showed them responsible for COVID spread, but in some quarters it continues to be held that kids are practically immune.
I wonder if school mask mandates are really just a proxy for something else - political leanings. Schools in most places were in distance learning prior to the initial vaccine rollout. Places with lots of vaccinated adults were also places that put in place school mandates when the kids came back.
Drum recently posted about a paper that claimed that surgical masks were useless - the very masks made available in schools. Did they magically start to work once placed on a child?
To be honest, I think most of it boils down to politics, because it is politics that determined how many people in an area chose to practice safety measures even when they wouldnt risk punishment, and those politics also show up in the official measures required by government.
Drum recently posted about a paper that claimed that surgical masks were useless - the very masks made available in schools. Did they magically start to work once placed on a child?
The whole "masks don't work" thing is really getting annoyingly stupid. Surgical masks have been in use in... umm... surgery for 150 years. For most of that time they used simple cloth masks that were less effective than what you can buy today at the drug store. If they're not effective does that mean that doctors and nurses have been using them all that time for no good reason?
I was told surgical masks are for a different purpose - splash and large droplet protection. Obvious that's mostly protecting the staff.
I don't believe they are effective filters for particles like the droplets SARS virus or others ride on, so any marketing of those masks as protection from them is probably unwarranted. I don't know if they have been definitely tested for SARS protection however.
Agreed, Kevin willfully misreads the effectiveness of masks since he is in the masks are marginally effective camp. But it was the second most effective measure.
Exactly!
Absent a mask mandate, spaces become difficult to enforce their own mask mandate. But they did! Hospitals, medical facilities, etc. They enforced masking even if there wasn't a mask mandate.
By examining where they were prohibited in doing so, you get a clear idea of what no masking would do.
There's a logical contradiction in saying that prohibiting mask mandates had a large positive effect on deaths, but having mask mandates had no effect.
The joys of nuance…
https://jabberwocking.com/the-lessons-of-covid-we-need-better-masks/
Nothing else has much impact.
It still seems like -- until vaccines become available -- even the less effective measures like shutdowns have a measurable, if comparatively small impact on reducing deaths.
The problem with any new pandemic will be that we won't know what we don't know. Every novel virus will have its unique characteristics, and its spreading mechanisms and any effective mitigating measures will first have to be discovered. Although it will likely be highly useful to have COVID data as a guide when the next contagious disease strikes, it would also be a mistake to interpret that data as definitive.
I think we have learned a lot about how respiratory viruses spread and how droplets work.
Pingback: VAX, MASK, CLOSE: – Brothers Judd Blog
We keep looking for specific tools and regulations, but the most effective thing may simply have been human willingness to limit contact with others.
Lots of countries with relatively low death rates were less strict with their school closures and/or mandates. Japan may not have implemented a lockdown, but many people modified their behavior as if it had.
The willingness to constrain behavior is part of political culture, and the US has extreme differences in political cultures. I suspect we're seeing a lot of correlations when we investigate mitigation measures in the US, not the actual usefulness of a particular measure.
Limiting contact is probably helpful but it seems like it doesn't work out in the long run or is just inherently unsustainable.
There are lots of stories of contact between isolated human bands and the wider world resulting in rapid debilitating disease in the previously isolated band, sometimes leading to their extinction. It shouldn't be a surprise to us that COVID worked something like that too.
The U.S. population was about 340 million at the time, so nationwide bar closures would have saved about 37,000 lives. That seems like a non-trivial impact to me, especially since if you issue a closure order you are probably going to close more than just bars. Closing both restaurants and bars nationwide would have saved more like 120,000 lives. I’m not saying that the benefits outweigh the costs, because business closures have significant negative effects.
Mask mandates seem like a winner here, with the potential to save something like 3/4 million lives.
Looking at the table on page S19 of supplement 1, it appears that the effect of restrictions on behavior was fairly small. For example, if I am interpreting the table correctly, mask usage went from 43.2% to 57.9%, though that is comparing states high restriction states to low restriction states, not states with mask mandates to states without mask mandates. This matches other studies on the effectiveness of masks, which show that attempts to get people to wear masks have only modest effects on behavior.
It's only modest because, as you say, some portion will mask already.
It's the comparison to where masking was disallowed that you find the key evidence: Because that's where the voluntary compliance is done away with and you can see the true effectiveness of masking.
Masks were clearly quite effective, Kevin's muddled statements are misleading.
Many of the items on this list like school closures, bans on gatherings, bar closures, etc...the design of this study will make it very hard to determine if they are effective or not. The first 6-9 months of the pandemic saw the most stringent and widespread measures being taken in the US and we also saw relatively few deaths until the first winter spike came in late 2020. Because the measures were widespread and deaths were low almost everywhere, these measures look ineffective. And maybe they were, but the study design may be designed to draw bad conclusions from this time period.
Also the study only looks at state level measures, ignoring county and city level measures.
According to Kevin's chart, leisure activities reduced COVID-19 deaths by 22%?
Laying about in hammocks will save us all! 🙂