Table 7.

Multivariate analysis on the merged cohort (Delta + Omicron)

	Death on day 28		ICU		Need for oxygen
	Best model	p	Best model	p	Best model	p
Variant: Omicron	0.53 [0.37-0.76]	0.001	0.19 [0.12-0.28]	<0.001	0.50 [0.38-0.67]	<0.001
Sex: male	1.66 [1.15-2.41]	0.007	1.37 [0.96-1.95]	0.084	1.65 [1.25-2.18]	<0.001
Age	1.07 [1.05-1.08]	<0.001			1.04 [1.03-1.04]	<0.001
Kidney failure	2.32 [1.46-3.67]	<0.001	0.41 [0.15-0.93]	0.050
HTA	1.28 [0.87-1.90]	0.211
Immunosuppressed	2.21 [1.30-3.66]	0.003	0.79 [0.31-1.83]	0.593	1.56 [1.05-2.33]	0.028
Previous SARS-CoV-2 infection	0.14 [0.01-0.67]	0.053	0.10 [0.01-0.50]	0.028	0.08 [0.02-0.26]	<0.001
Vaccination	0.45 [0.30-0.66]	<0.001	0.26 [0.17-0.39]	<0.001	0.20 [0.14-0.28]	<0.001
Solid cancer for <3 mo	5.61 [2.53-12.57]	<0.001	0.41 [0.09-1.31]	0.174
Overweight			1.72 [1.10-2.77]	0.021	1.63 [1.17-2.27]	0.004
At least one comorbidity			1.68 [0.84-3.45]	0.146	1.61 [0.97-2.67]	0.064
O2 treatment at home					2.86 [1.05-9.28]	0.053
Interaction vaccination × immunosuppresseda			2.95 [1.03-8.87]	0.047
Interaction vaccination × previous COVID-19b					5.60 [1.21-28.49]	0.030

^aIn the model for ICU, a significant and positive interaction was found between being vaccinated and being immunosuppressed, which means that the protective effect of vaccination on the risk of ICU admission decreases among immunosuppressed patients.

^bIn the model for O₂ need, a significant and positive interaction was found between being vaccinated and having a previous SARS-CoV-2 infection, which means that the effect of vaccination was decreased among patients having previously had SARS-CoV-2 infection. In addition, a trend towards the interaction between being vaccinated and being immunosuppressed was found, similar to the ICU model, but was not significant (p 0.08), which suggests an effect comparable to the ICU model. ICU, intensive care unit; HTA, Hypertension.