Table 4. The effect of COVID-19 pandemic on the Door-to-Device time was examined by multivariate regression analysis adjusted for the potential confounding factors.

Variables	Door-to-Device time, min		Door-to-Device time, ≤ 90 min
	Beta	p	OR	p
COVID-19 periods
Period-I	Ref	-	Ref	-
Period-II	-3.67	0.017*	2.34	0.095
Period-III	10.06	< 0.001*	0.30	0.002*
Sex, male to female	-3.77	0.065	2.44	0.053
CAD history	-3.49	0.138	-a	-a
Triage level, from 4 to 1	-2.31	0.042*	2.18	0.007*

GLM gamma regression was appropriately employed based on the goodness-of-fit test for non-normally distributed independent variables, while logistic regression analysis was performed for categorical data.

^a The CAD history variable was excluded due to no CAD history cases in the group with Door-to-Device time exceeding 90 minutes, leading to perfect separation and unstable model estimates.

CAD, Coronary artery disease; COVID-19, coronavirus disease 2019; GLM, generalized linear model; OR, odds ratio.