Table 3.

The impact of opium use on the ICU rates, mortality, and need for intubation among hospitalized COVID-19 patients

Treatment	Outcome	Model	OR (95% CI)	P-value
Opium	ICU admission	PSM	1.87 (0.22, 2.91)	0.631
		Unadjusted	1.76 (0.36, 2.59)	0.467
		Full adjusted	–––	––
		Optimal model	1.42 (0.14, 2.24)	0.116
Opium	Mortality	PSM	2.38 (1.30–4.35)	0.005
		Unadjusted	2.19 (0.52, 3.74)	0.679
		Full adjusted	2.34 (0.17, 3.74)	0.697
		Optimal model	2.39 (0.13, 3.19)	0.571
Opium	Need for intubation	PSM	3.57 (1.38, 9.39)	0.009
		Unadjusted	2.01 (0.97, 4.17)	0.062
		Full adjusted	5.54 (1.97, 15.61)	0.001
		Optimal model	6.34 (2.59, 15.50)	< 0.001

A full adjustment method was implemented, considering all variables. The selection of the optimal model was carried out through a backward stepwise algorithm. In the propensity score matching (PSM) approach, variables exhibiting significant associations with outcomes (ICU admission or mortality or need for intubation) or opioid use were taken into consideration. Caliper parameters were set at 0.5 for the three response variables, while the ratio was established at 3, 3, and 4 correspondingly. Potential confounding variables were carefully matched between the group of individuals who received opioids and the group of individuals who did not receive opioids