Table 1.

Unadjusted and adjusted marginal relative hazards of in-hospital mortality

	Unadjusted HR (95% CI)	p-Value	Adjusted*£ HR (95% CI)	p-Value
	All patients
Control# (n = 92)	1.00		1.00
Hydroxychloroquine (n = 197)	0.43 (0.28, 0.64)	<0.001	0.66 (0.39, 1.11)	0.118
Hydroxychloroquine + Azithromycin (n = 94)	0.36 (0.21, 0.60)	<0.001	0.44 (0.24, 0.82)	0.009
	&Baseline PO2/FiO2 0-300
Control# (n = 41)	1.00		1.00
Hydroxychloroquine (n = 83)	0.52 (0.31, 0.87)		0.71 (0.37, 1.35)
Hydroxychloroquine + Azithromycin (n = 28)	0.46 (0.23, 0.93)		0.59 (0.26, 1.35)
				p-Value for interaction
				<0.001
	&Baseline PO2/FiO2 300+
Control# (n = 33)	1.00		1.00
Hydroxychloroquine (n = 100)	0.39 (0.15, 0.97)		0.49 (0.15, 1.63)
Hydroxychloroquine + Azithromycin (n = 60)	0.56 (0.21, 1.52)		0.62 (0.19, 1.97)

^*Adjusted for age, gender, number of comorbidities, CVD (yes/no), duration of symptoms, date of admission, CRP and censoring using IPW.

^£The overall estimate was also adjusted for baseline COVID-19 disease severity.

^#Heparin, immuno-modulatory drugs, HIV antivirals, combinations of these or no drugs at all.

^&45 patients missing baseline PO2/FiO2 not included in the stratified analysis.