03-20-2020, 04:31 PM
The evidence cited is laughable. I've been involved in an inpatient influenza clinical trial. The rule is "once randomized, always analyzed", and your outcome of interest is clinical status. We used a five-point clinical status (dead, ICU, hospitalized on oxygen, hospitalized without oxygen, discharged), but you could also use survival to hospital discharge as an outcome or something similar, or ventilator-free survival (or if you are trying it in a critical care sending, time to liberation from the ventilator).
So among the 26 patients, who were randomized to hydroxychloroquine, you had three go to the ICU and 1 death. This is broadly in line with what has been reported in the population of hospitalized patients with Covid-19 overall: (20% of symptomatic infectious patients need hospitalization, 1/4 of those need ICU care, and somewhere between 1/5 and 1/2 of those die).
It is definitely worth exploring, but nobody should go into this expecting that any given treatment will work. The PTS (probability of technical success) of a drug entering clinical trials is between 10% and 15% depending on indication, so expect 7 in 8 of these highly publicized treatments to fail.
BC
So among the 26 patients, who were randomized to hydroxychloroquine, you had three go to the ICU and 1 death. This is broadly in line with what has been reported in the population of hospitalized patients with Covid-19 overall: (20% of symptomatic infectious patients need hospitalization, 1/4 of those need ICU care, and somewhere between 1/5 and 1/2 of those die).
It is definitely worth exploring, but nobody should go into this expecting that any given treatment will work. The PTS (probability of technical success) of a drug entering clinical trials is between 10% and 15% depending on indication, so expect 7 in 8 of these highly publicized treatments to fail.
BC