Last week, a press released provided initial and promising results from a large scale trial of the corticosteroid dexamethasone in hospitalized COVID patients. Today the authors released a full report: https://www.medrxiv.org/content/10.1101/2020.06.22.20137273v1.full.pdf

As a reminder, dexamethasone was just one arm of the trial. The primary outcome measure was 28-day mortality.

I'm pleased to see that the subgroup analysis reported, namely effect of dexamethasone according to level of respiratory support, was pre-specified.

The randomization ended up with an age imbalance which was corrected. Though the correction was not part of the pre-registration, this seems entirely reasonable to me.

I do not have the expertise in trial design to comment on the decision to stop enrollment. That said, this seems appropriate and I do not see red flags.

In the press release I was surprised the survival difference between ventilators (40%/29% with/without dexamethasone), supplemental oxygen (25%/22%), and neither (13%/17%) was pretty narrow.

Here's a possible clue: ventilated patients were 10 years younger. I don't know why.

Secondary measures: Dexamethasone also showed small but significant benefits on time to discharge and probability of avoiding mechanical ventilation. These are overall differences; I do not believe these secondary measures are broken down by breathing support.

For me the take-home is that there are no big surprises relative to what we learned from the press release.

The trial gives a strong indication that dexamethasone is beneficial in hospitalized patients requiring respiratory support (O2 or vent).

As noted in the press release, patients not on respiratory support fared *worse* when treated with dexamethasone. The difference is not statistically significant, but concerning.

"There was no benefit (and the possibility of harm) among
patients who did not require oxygen."

As I noted before, the trial showed that dexamethasone helps the very ill. It's not for mild illness.

Taking it won't prevent you from getting COVID.

Nor, if you have COVID, will it prevent you from being hospitalized.

Nor is it a panacea even for the very ill.

Finally, @EricMeyerowitz has a more detailed analysis of the paper in this thread (h/t @CovidPath) that makes for good reading, particularly if you've read the paper and want a deeper dive. https://twitter.com/EricMeyerowitz/status/1275263218965979141