Glaucoma Characteristics and Influencing Factors during the COVID-19 Pandemic in the Huizhou Region

Table 5

Univariate and multivariate logistic regression model of the potential factors influencing the development of AACG during phase A.

Potential risk factors	Univariate model		Multivariate model^a
	Crude OR (95% CI)		Adjusted OR (95% CI)
Sex (ref: male)	4.308 (1.234–19.414)	0.018	2.609 (0.595–13.817)	0.257
Age	1.036 (0.999–1.079)	0.056
COVID-19 status (ref: positive)	0.037 (<0.001–0.260)	<0.001	0.089 (0.002–0.845)	0.030
Glaucoma started after testing positive for COVID-19 (ref: no)	10.168 (3.051–37.587)	<0.001	3.136 (0.786–13.809)	0.119
Adverse behavior (ref: no)	3.286 (0.173-inf)	0.467
History of medication use (ref: no)	3.715 (1.210–11.888)	0.019	1.024 (0.235–4.132)	>0.999
Monocular/binocular involvement (ref: monocular)	0.912 (0.305–2.805)	>0.999
NonCOVID-19 systemic diseases (ref: no)	0.445 (0.098–1.580)	0.272

Note. OR = odd ratio, 95% CI = 95% confidence interval. In the univariate model, only one candidate variable was entered into the logistic regression analysis, and the crude OR (95% CI) was calculated. Because the data were sparse, the lower limit could not be calculated. (^a) Variables with in the univariate model were selected as candidate influencing factors to construct a multivariate model to estimate the adjusted OR and 95% CI. Model evaluation: Hosmer–Lemeshow goodness-of-fit test, χ² = 5.509, , which indicated a good model fit. AACG, acute angle-closure glaucoma; N-AACG, nonacute angle-closure glaucoma.