Whether it is masks or meds consider π¬πππ© ππ π¬π ππ§π π§ππππ©, as well asπ¬πππ© ππ π¬π ππ§π π¬π§π€π£π
For benign interventions (eg., masks for COVID) if we say masks work & they donβt, consequence is small. If we say they donβt when they truly do, itβs tragic
For benign interventions (eg., masks for COVID) if we say masks work & they donβt, consequence is small. If we say they donβt when they truly do, itβs tragic
For any intervention we have to always consider the consequences of a Type I error relative to the consequences of a Type 2 error.
For many medicines, the consequences of a Type 1 error in terms of toxicity, harm, & cost usually outweighs those of a Type II error. We need RCTs.
For many medicines, the consequences of a Type 1 error in terms of toxicity, harm, & cost usually outweighs those of a Type II error. We need RCTs.
Sometimes there are interventions where consequences of erroneously concluding something is effective when it is not (Type I error) is small compared to concluding itβs not effective when it actually is (Type II error). Eg., hand washing to prevent COVID. Masks are like that.
Now with data from lab & clinical studies, hospital experience, as well as observational data, there is no question masks work. They protect others if you happen to have COVID. They also protect you from getting COVID, & decrease chance of serious disease by reducing viral dose.
β’ β’ β’
Missing some Tweet in this thread? You can try to
force a refresh
γ
