SEARCH
A note on sampling biases in the Bangladesh mask trial
Related: ethics in medicine, gears of the machine, SARS-CoV-2 aka COVID-19 aka Coronavirus
Real science is never settled, and anyone who has certainty on such things is not qualified to discuss it.
There has been no credible scientific study on mask wearing having any influence on commnity spread of COVID—ever. Good quality masks are sure useful in some settings, but that’s really an entirely different case.
The Bangladesh mask trial was one of those “scientific” studies claimed to show that masks have a benefit. The media took it and ran with it. But like most studies it was total garbage from a scientific standpoint, with methodological fraud and numerous biases baked-in.
arXiV: A note on sampling biases in the Bangladesh mask trial
2021-12-02
...We suggest that the very large causal effects on consent rates and thus population denominators urge caution in interpreting the small differences we see in symptomatic seropositivity between treatment and controls, which are already not statistically significant according to standard non-parametric paired tests. Additionally, as the trial shows that the intervention studied can have large and highly significant effects even on unintended aspects of behavior, including staff surveying behavior, bias-susceptible endpoints that depend on subjective reports of symptoms from participants to staff should be used with care.
WIND: the Bangladesh mask trial was one of those frauds used by the media to promote the narrative of mask usage. Anyone on the peer review committee should lose all their scientific credibility, and be banned from publishing any research for the rest of their life. The authors of the farce study should be shamed out of academia.
The opinion below references the study above.
Coronamonomania Lives Forever
The inaptly named journal Science saw fit to publish the methodologically and analytically dreadful Bangladesh mask trial. In normal times this would never pass peer review, but Science is a completely politicized publication at this point. Breaks my heart to see this misuse of “science”.
In any event, here is one of the many good critiques of this study. People seem to forget that the primary outcome of the study was actually just whether the intervention increased mask wearing. The authors then decided to try to twist that into effect on cases. Just a grotesquely confounded and biased study, with a weird analytic approach designed to try to show an effect, after a standard analysis showed no effect. This paper, far better than I ever could, destroys the methods and reported results from this study, which of course, is getting lots of attention from the media. (Arvix Paper)
WIND: crap science abounds.
Mask science is best explained by Ralph.
