With Covid Studies, Quality of Evidence Matters

0
22
Labyrinthine Covid Reminder System Is The Real Reason For The Delays


ism The adventure of copper beech trees, Sherlock Holmes says: “The data! Data! Data! I cannot make bricks without clay. Recent claims of massive benefits of wearing masks and using ivermectin against Covid-19 depended mainly on poor quality clay.

Meta-analysis is a technique for pooling the results of many studies, but it cannot make silk bags from sow ears. A recent British medical journal The review looked at six fairly swine studies of mask wearing and found an impressive 53% reduction in risk. But the only randomized controlled trial estimated the smallest effect: an approximately 18% (-23% to 46%) reduction in Sars-CoV-2 infections. The “heaviest” studies, an analysis of American states and a survey of around 8,000 Chinese adults in early 2020, observed rather than experienced and its editorial highlights the risks of confounding variables influencing both the wearing of masks and infections and the inability to disentangle the effects of simultaneously fluctuating measures. Indeed, this review found an identical 53% reduction in handwashing.

Physical models and lab tests suggest that masks should have some effect, and to be fair, it’s hard to imagine rigorous randomized trials on the effects of wearing masks or any other behavior. But these are seen as both feasible and essential for evaluating medical treatments and therefore, in principle, it should be easier to evaluate ivermectin, a cheap antiparasitic drug heavily promoted as a neglected treatment for Covid. -19. A meta-analysis initially found that ivermectin reduced mortality from Covid-19 by about 56%, a huge effect. But then a preprint server removed the larger study and further concerns were raised about major errors and fraud in some ivermectin analyzes. After removing studies at high risk of bias, the estimated effect wore off: the decrease in estimated mortality was around 10%, but with a wide range of uncertainty (-42% to 43%) , therefore, no definitive conclusion can be drawn.

Huge efforts have gone into scrutinizing ivermectin research, even leading to the intriguing observation that the drug appears to work in countries with high roundworm prevalence, when what it takes, these are the kind of good quality studies that ultimately made claims about hydroxychloroquine for rest. Fortunately, these are underway for ivermectin.

David Spiegelhalter is President of the Winton Center for Risk and Evidence Communication in Cambridge. Anthony Masters is Statistical Ambassador of the Royal Statistical Society

This article was amended on November 28, 2021 to include a reference to “physical models and laboratory tests” with respect to mask wearing.

LEAVE A REPLY

Please enter your comment!
Please enter your name here