New study about mask effectiveness

[MoseyingAlong]

I know, I know. This subject has been hashed over repeatedly.
But this is a reputable source.

https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD006207.pub6/full

Recently someone asked for more recent study results.
This probably won't help them with their family but they may be interested. (I didn't find their original post.)

[LennStar]

This is not a recent study, it's a metastudy, and is now mis-used to say that masks don't help against Covid.

The study where the authors themselves say you can't use that as a base to say anthing about specific mask effectiveness in regards to Covid, not least because only 6 of the dozens of studies in this meta analysis are from Covid times.

btw. prior rounds of this metastudy have found masks to be effective. The only result you can get out of this study in regards to Covid is that masks mandates in the work place have no significant effect on a very easily transmittable desease if the people go in tens of thousands to big events where the coughing, feverish beer girl is proudly showing their positive rapid test on twitter saying "And still I work!"

[MustacheAndaHalf]

I know, I know. This subject has been hashed over repeatedly.
But this is a reputable source.

https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD006207.pub6/full

Recently someone asked for more recent study results.
This probably won't help them with their family but they may be interested. (I didn't find their original post.)

Can you explain why this is a "reputable source"?

The study also does not seem to be conclusive, given this quote from it:

Authors' conclusions
The high risk of bias in the trials, variation in outcome measurement, and relatively low adherence with the interventions during the studies hampers drawing firm conclusions.

And also:

The low to moderate certainty of evidence means our confidence in the effect estimate is limited, and that the true effect may be different from the observed estimate of the effect.

[BNgarden]

A summary of issues with the Cochrane review piece that I found useful:
https://theconversation.com/yes-masks-reduce-the-risk-of-spreading-covid-despite-a-review-saying-they-dont-198992?utm_source=twitter&utm_medium=bylinetwitterbutton

[MoseyingAlong]

I know, I know. This subject has been hashed over repeatedly.
But this is a reputable source.

https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD006207.pub6/full

Recently someone asked for more recent study results.
This probably won't help them with their family but they may be interested. (I didn't find their original post.)

Can you explain why this is a "reputable source"?

The study also does not seem to be conclusive, given this quote from it:

Authors' conclusions
The high risk of bias in the trials, variation in outcome measurement, and relatively low adherence with the interventions during the studies hampers drawing firm conclusions.

And also:

The low to moderate certainty of evidence means our confidence in the effect estimate is limited, and that the true effect may be different from the observed estimate of the effect.

@MustacheAndaHalf
Among the nursing and medical researchers I know, the Cochrane Library is considered a reputable outfit. I am not a researcher myself.
From https://library-guides.ucl.ac.uk/cochrane
"The Cochrane organisation started in the UK in the 1990s but is now a global network of researchers and Cochrane Reviews are one of the most highly regarded sources of healthcare evidence."

I was surprised by the results because I expected them to be much more conclusive that surgical masks and N95s are effective. My expectation was based on operating room experience. Surgical masks anytime the sterile field was open and N95s if TB, HPV or COVID were suspected/likely. I never looked into studies about it before because it just made sense to me and it's the standard everywhere I know about.

Honestly I thought the few attending anesthesiologists who pushed back were being big babies. Now I may have to revise that opinion. Maybe they were better informed. To be clear, the anesthesiologists were pushing back on their need to wear masks, not the surgeons or scrub techs.

[dang1]

masks maintain my man of mystery allure

[MustacheAndaHalf]

I know, I know. This subject has been hashed over repeatedly.
But this is a reputable source.

https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD006207.pub6/full

Recently someone asked for more recent study results.
This probably won't help them with their family but they may be interested. (I didn't find their original post.)

Can you explain why this is a "reputable source"?

The study also does not seem to be conclusive, given this quote from it:

Authors' conclusions
The high risk of bias in the trials, variation in outcome measurement, and relatively low adherence with the interventions during the studies hampers drawing firm conclusions.

And also:

The low to moderate certainty of evidence means our confidence in the effect estimate is limited, and that the true effect may be different from the observed estimate of the effect.

@MustacheAndaHalf
Among the nursing and medical researchers I know, the Cochrane Library is considered a reputable outfit. I am not a researcher myself.
From https://library-guides.ucl.ac.uk/cochrane
"The Cochrane organisation started in the UK in the 1990s but is now a global network of researchers and Cochrane Reviews are one of the most highly regarded sources of healthcare evidence."

I was surprised by the results because I expected them to be much more conclusive that surgical masks and N95s are effective. My expectation was based on operating room experience. Surgical masks anytime the sterile field was open and N95s if TB, HPV or COVID were suspected/likely. I never looked into studies about it before because it just made sense to me and it's the standard everywhere I know about.

Honestly I thought the few attending anesthesiologists who pushed back were being big babies. Now I may have to revise that opinion. Maybe they were better informed. To be clear, the anesthesiologists were pushing back on their need to wear masks, not the surgeons or scrub techs.

In my post I quoted "hampers drawing firm conclusions" and "The low to moderate certainty of evidence means our confidence in the effect estimate is limited".

You're drawing "firm conclusions" from a report that literally tells you not to draw "firm conclusions"?

[MoseyingAlong]

In my post I quoted "hampers drawing firm conclusions" and "The low to moderate certainty of evidence means our confidence in the effect estimate is limited".

You're drawing "firm conclusions" from a report that literally tells you not to draw "firm conclusions"?

I honestly have no idea what "firm conclusions" you are referring to. I was surprised that the results were not conclusive. How that translated into a firm conclusion I have no idea.

[PeteD01]

This Cochrane review is just another example of using the tools of metaanalysis in a mindless way while accepting the arbitrary notion that randomized controlled trials are always the gold standard for evidence. This is a consequence of uncritically accepting Evidence Based Medicine (EBM) dogma while being blissfully unaware of the nature of the subject at hand. Of course, this is followed with yet one more absurd call for more large randomized controlled trial to provide the evidence deemed admissible by EBM criteria.
Actually, approaching the issue with this mindset virtually guarantees that there will never be evidence at the level acceptable to the authors as these trials would be inappropriate, expensive and never forthcoming - thus freezing the Cochrane mask report in perpetual inconclusiveness while the rest of us uses masks with great benefit as required for the tasks at hand.
 
If there is any lesson to be learned from that review (it is not a study or even good journalism) is that knowing something about EBM and the tools of metaanalysis are no substitute for actually understanding the questions at hand on the subject matter level.
Come to think of it, the text really does have a whiff of this LLM chatbot imbecility combined with self-assuredness to it.

My verdict is that this particular Cochrane review reaches the level of medical disinformation due to ignorance of the subject matter on the part of the "researchers".

Here is a decent summary of the criticisms directed at the review that leaves no doubt that the Cochrane authors lack expertise:


https://theconversation.com/yes-masks-reduce-the-risk-of-spreading-covid-despite-a-review-saying-they-dont-198992

[MustacheAndaHalf]

In my post I quoted "hampers drawing firm conclusions" and "The low to moderate certainty of evidence means our confidence in the effect estimate is limited".

You're drawing "firm conclusions" from a report that literally tells you not to draw "firm conclusions"?

I honestly have no idea what "firm conclusions" you are referring to. I was surprised that the results were not conclusive. How that translated into a firm conclusion I have no idea.

In your mind, all database searches by every group of researchers will give the same conclusive result?  It's also possible they included apples & oranges in their comparisons - surveying random people asking if they wore a mask for the past few months is fairly low quality data.  Monitoring medical workers in a hospital, wearing masks, would be higher quality data (but even then, hospital workers may put on masks in some locations and not others, exposing themselves to air particles).  Combining a group of studies is tricky, so I wouldn't put as much faith in one meta analysis study.

If you see someone claim that randomized, double blind trails are not the gold standard, they are going against the medical science that has brought us many of the discoveries we rely on.  Simply stating they disagree is not sufficient to counter decades of best practices in medical research.

That said... how do you do a "double blind" experiment with masks?  Someone knows if they're wearing a mask or not, so that can't be hidden.  The studies of mask wearing by their nature can't be double blind studies (where neither doctor nor patient knows if an active drug or sugar pill is being given to the patient).  And someone can take off a mask for an hour in a high risk environment, which is significant.  Adhering to the study protocol is more difficult than simply taking medicine daily.  So there's lots of ways for studying mask wearing to be imperfect, and a good study design will figure out ways to minimize those problems.

[blue_green_sparks]

That study doesn't impress me at all.
Since I do live in the South, I compared the death rates of conservative "let's just go about our business" states to liberal states and decided the N95 precaution would generally work towards me staying alive. Mississippi and Louisianna did not fare well compared to Connecticut and Mass by a 2x factor per capita. The preachers around here, claiming divine protection were dropping like flies. Now I do realize there are other mitigating general health factors but hey, wearing a mask is easy enough but I was worried about getting beat up for it at times.

[cool7hand]

I can't believe people still talk about this

[PeteD01]

...

If you see someone claim that randomized, double blind trails are not the gold standard, they are going against the medical science that has brought us many of the discoveries we rely on.  Simply stating they disagree is not sufficient to counter decades of best practices in medical research.

That said... how do you do a "double blind" experiment with masks?  Someone knows if they're wearing a mask or not, so that can't be hidden.  The studies of mask wearing by their nature can't be double blind studies (where neither doctor nor patient knows if an active drug or sugar pill is being given to the patient).  And someone can take off a mask for an hour in a high risk environment, which is significant.  Adhering to the study protocol is more difficult than simply taking medicine daily.  So there's lots of ways for studying mask wearing to be imperfect, and a good study design will figure out ways to minimize those problems.

Randomized controlled trials (RCTs) have their place but only for very specific questions.
The effectiveness of N95 respirators in the infection control for airborne diseases is so well established that clinical equipoise in settings in which an RCT could conceivably be conducted won´t be present, neither with the researchers nor in the study subjects.

Here is a study showing the lower excess mortality among practicing physicians during the pandemic when compared with the general population. This is despite of the high risk environment the physicians were exposed to.
These findings support the notion that N95 respirators and other personal protective equipment when used by well informed and trained individuals is effective Covid infection control:

https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/2800889

We know that only fitted respirators, and not surgical masks, adequately protect from SARS-CoV-2, especially from the later variants, and we also know that the physicians studied had N95 respirators properly fitted and were likely compliant at least at their workplace.
This is about the best evidence you are going to get that personal protective equipment including masks in combination with all other infection control measures against airborne threats are highly effective.

Conducting a RCT in this environment (and the healthcare setting is the only setting this could even conceivably getting done) would run into the problem that there is no equipoise among healthcare workers regarding masks, so it would not be possible to find investigators or study subjects, not to even mention the ehtical issues.

So it really is just common sense one needs to understand that asking for a RCT is an impossibly high bar when it comes to evaluating personal safety equipment and is also unneeded.
 
Ultimately, N95 respirators are just another type of personal safety equipment, such as hard hats, goggles, steel toe boots, seat belts etc, things for which RCTs were not performed to establish effectiveness. The impact of these measures and devices when used properly is often simply too great to submit a control group to the risk, precluding RCTs and leaving mostly epidemiological analyses as the appropriate tools.

[GilesMM]

I don’t have time to read studies, but we haven’t seen a mask in a long time.  They are gone pretty much at this stage.