Table 2.

Binary logistic regression analysis for potential predictors of SARS-CoV-2 test-positive results.

	B	SEB	Wald	OR (95%CI)	p value
(When “contact at dormitory” is used as an explanatory variable)
(Constant)	− 3.750	0.150	626.912	0.02 (0.02–0.03)	< 0.0001
Age	+ 0.014	0.003	32.638	1.01 (1.01–1.02)	< 0.0001
Sex (Male)	− 0.073	0.116	0.390	0.93 (0.74–1.17)	0.5322
Close contact history	+ 1.247	0.140	79.157	3.48 (2.64–4.58)	< 0.0001
Contact at dormitory	+ 2.342	0.238	96.709	10.40 (6.52–16.59)	< 0.0001
(When “household contact” is used as an explanatory variable)
(Constant)	− 3.521	0.138	656.215	0.03 (0.02–0.04)	< 0.0001
Age	+ 0.014	0.002	31.970	1.01 (1.01–1.02)	< 0.0001
Sex (Male)	+ 0.065	0.113	0.328	1.07 (0.85–1.33)	0.5670
Close contact history	+ 0.955	0.145	43.343	2.60 (1.96–3.45)	< 0.0001
Household contact	+ 0.061	0.135	0.206	1.06 (0.82–1.39)	0.6500

Logistic regression analyses were performed for 4550 individuals with identified places of contact with COVID-19 patients. The upper half shows the results when contact in a dormitory, irrespective of the closeness of contact, is used as an explanatory variable. The lower half shows the results when household contact is used instead. The OR values are equivalent to exp(B). Wald χ² statistics (Wald) were calculated using the formula ${\left(B,/,S,E,B\right)}^2$, which is a marker of the significance of each coefficient in the predictive model.

B, unstandardized regression coefficient; CI, confidence interval; SEB, standard error of the coefficient; OR, odds ratio.