COVID, masks and poor science

A CDC report that concludes
“face masks, evidence from randomized controlled trials of these measures did not support a substantial effect on transmission of laboratory-confirmed influenza”
is just WRONG.

It uses an incorrect fixed effects statistical model, thereby ignoring between study variations.Moreover its interpretation is plagued by the common null hypothesis statistical testing cognitive biases of dichotomania and nullism.

Proper analysis (see Figure below) shows average benefit could be as great as 56% reduction and next study could show a benefit of up to 82% reduction.

Unlikely any increased risk so more reasonable conclusion “MASK UP while awaiting more studies to refine the exact size of any benefit.”

Avatar
James (Jay) Brophy
Professor of Medicine & Epidemiology

I am a tenured (full) professor with a joint appointment in the Departments of Medicine and Epidemiology and Biostatistics where I work as a clinical cardiologist and do research in cardiovascular epidemiology.