I have turned down several press requests to talk about Trump's treatment because I am not his doctor and don't know the situation. But the questions people asked made me realize there is still confusion about decision-making and treatment for #COVID19, so a general thread.

#COVID19 is diagnosed with a positive antigen test (the "rapid" test) or PCR test (hours-days). You are more likely to have a false negative test especially early in disease with the antigen, and more likely to have persistent positive late in the disease with the PCR.

Most people are not tested all the time, so they mostly get diagnosed when they have symptoms, or a few days after they have had a known exposure. Some are picked up with regular, routine screening like in schools or workplaces.

Because the actual date of infection is typically not known, when we describe the progression of disease, we usually start from first day of symptoms. For most people, the first few days of symptoms are unpleasant but not life threatening.

By contrast, risk of severe disease requiring hospitalization or worse is usually not until 5-10 days of symptoms (which in turn is even longer after the actual infection date because it takes a few days to develop symptoms).

How do we decide whether to hospitalize someone? A mix of their oxygen level (<95% start to worry), their blood tests (i.e. any evidence of damage to other organs), how far into the disease they are (see above), & their underlying risk (age, other illnesses, frailty).

Once we decide someone needs to be hospitalized, we decide on treatment. If persistent need for oxygen and >7 days symptoms, we use steroids (dexamethasone) because of the RECOVERY trial. (By the way it is a scandal we haven't done RCTs like this here.) https://www.nejm.org/doi/full/10.1056/NEJMoa2021436

Honestly, that's IT for strong evidence-based medicine. We will typically give remdesivir for those requiring oxygen, if available. At NYU, we also offer enrollment in a clinical trial to see if any of the other treatments work (convalescent plasma, blood thinners, immune meds).

Early on we gave a lot of antibiotics because we couldn't tell whether people had associated bacterial pneumonia or not - and they were so sick we didn't want to miss anything. Now, we have learned, and using antibiotics is uncommon.

Non-medication treatments include proning (keeping people on their stomachs instead of their backs) to help open up the lungs. For oxygen, we have a lot of ways to give people even very high amounts of oxygen and now put people on ventilators only if all of those don't work.

And since people ask me this a lot, there is no clear evidence for benefit of any vitamin supplement; not standard of care, but most of them are pretty harmless.