The good side of all cause mortality is that it includes drug-caused deaths. Presumably, leronlimab patients will take fewer/less harmful drugs because they get better faster.

Drug-caused kidney damage might directly or indirectly cause a few deaths in this study, more likely in placebo patients given more desperate levels/treatments of dex.

I'm not sure what the medical phrasing is for "yeah, drug-related deaths, but not car crashes."