“…This was also the case for the analysis after excluding two studies [24,26] which used a time threshold other than 24 h from ICU admission for defining early/late intubation; i.e., there was no statistically detectable difference between compared groups regarding allcause mortality (10 studies [13, 14, 18-23, 25, 27]; 3870 deaths; 45.2% versus 38.1%; RR 1.06, 95% CI 0.99-1.15, p = 0.11, I 2 = 0%). Finally, in the sensitivity analysis using an alternate definition of early/late intubation, there was no statistically detectable difference on all-cause mortality between patients undergoing intubation without versus with a prior trial of HFNC/NIV (eight studies [13,14,18,19,22,24,25,27], 1128 deaths; 48.9% versus 42.5%; RR 1.11, 95% CI 0.99-1.25, p = 0.08, I 2 = 0%; Fig. 4).…”