"Within 48 hours of diagnosis" indicates nothing about disease severity, nor how long patients were sick prior to diagnosis. > Wait, where does it say that in

"Within 48 hours of diagnosis" indicates nothing about disease severity, nor how long patients were sick prior to diagnosis. > Wait, where does it say that in the study? The study _supports_ the idea that sicker patients get hydroxychloroquine more frequently than less-sick patients. The study's raw figures show that sicker patients get hydroxychloroquine more than healthier patients. The study states that besides age and BMI, "all other [control] data were treated as categorical variables in the model." It's clear the study's authors don't sensitively control for patients' angiotensin-renin-aldosterone system health. There's strong subject selection bias in "initial disease severity" as well when some subjects die at home and others go to the ER mistaking mild Covid for heart attack symptoms.