Table 3.

Multivariate logistic regression to determine factors associated with sleep quality¹

Variables	Univariate model OR (95% CI)	Multivariate model aOR (95% CI)	p-value
Age	0.99 (0.97, 1.01)	0.99 (0.97, 1.01)	0.233
Sex			0.121
Male	1.00 (reference)	1.00 (reference)
Female	0.70 (0.51, 0.94)	0.78 (0.57, 1.07)
Chronic comorbidity			0.015
No	1.00 (reference)	1.00 (reference)
Yes	2.22 (1.13, 4.87)	2.34 (1.17, 5.24)
Work environment			0.006
NFLHCW	1.00 (reference)	1.00 (reference)
FLHCW	1.62 (1.22, 2.16)	1.59 (1.16, 2.16)
Work time			<0.001
No change	1.00 (reference)	1.00 (reference)
Increased	1.76 (1.32, 2.34)	1.84 (1.37, 2.48)
Frequency of seeing critically ill patients dying			0.040
No change	1.00 (reference)	1.00 (reference)
Increased	1.76 (1.32, 2.34)	1.42 (1.03, 1.95)
Stress level			0.105
No change	1.00 (reference)	1.00 (reference)
Increased	1.48 (1.11, 1.98)	1.28 (0.93, 1.76)

OR, odds ratio; CI, confidence interval; aOR, adjusted odds ratio; FLHCW, frontline healthcare worker (involved in direct care and treatment); NFLHCW, non-frontline healthcare worker (including those who perform swabs for coronavirus disease 2019 cases and logistics work).

¹Multivariate logistic regression was performed using the enter method; All statistically significant variables were included in the model and adjusted for age and sex; All variables in the model had variance inflation factors in the range of 1.04–1.22, indicating no multicollinearity.