Thread by @cspramesh

The power of large, pragmatic randomized trials. Conducted by academic researchers using an adaptive trial design (#MAMS), and you’ve got a winner. Strong reasons why publicly funded research is so very important. Thread

The #RECOVERY trial has answered 3 important, clinically relevant questions about #COVID_19. It has shown that #Dexamethasone is beneficial in patients requiring oxygen/ ventilation & that HCQ and Lopinavir-Ritonavir are not useful.
#HCQ results are now out in preprint!

#RECOVERY randomized 1561 pts to #HCQ and 3155 to standard of care. Primary endpoint of 28-day mortality was 26.8% in the HCQ arm and 25% in the SOC arm. Lack of benefit consistent across all sub-groups. And remember, 28-day mortality in the same trial in the Dexa arm was 21.6%

Pts in the HCQ arm were less likely to be discharged alive from hospital at 28 days (60.3 vs 62.8%, 95%CI, 0.85-0.99); HCQ pts not on mechanical ventilation were more likely to reach composite endpoint of mechanical ventilation or death (29.8 vs 26.5%, 95%CI, 1.01-1.25)

What I like about the #RECOVERY trial is that it set out to ask some questions. And answer them reliably, so the question is not revisited again. And it has done precisely that. We’ve seen 3 press releases & 2 preprints with conclusive results adaptable immediately in practice.

This is the kind of scientific rigour it takes to either prove that something works (Dexa) or doesn’t (HCQ and Lopinavir-Ritonavir). With 200,000 new cases daily (globally) and 28,000 (India), there is no excuse to continue doing underpowered trials. Yes, even in a pandemic.