Table 2.

Participant Characteristics Associated With Hepatitis B Infection in the Serosurvey: Binomial Logistic Regression Model

Characteristic	Univariate Odds Ratio (95% CI)	P Value	Multivariable Model, Odds Ratio (95% CI)a	P Value
Age, per y	1.03 (1.03–1.04)	<.001	a
Age group, y		<.001		<.001
0–14	0.09 (.04–.18)		…
15–29	Reference		Reference
30–44	1.82 (1.22–2.73)		1.80 (.87–3.72)
45–59	0.81 (.45–1.46)		1.49 (.56–3.98)
>60	0.74 (.35–1.56)		0.55 (.20–1.50)
Birth date		<.001
Born prior to vaccine	Reference
Born after vaccine	0.06 (.03–.11)
Sex, male vs female	1.46 (1.04–2.03)	.03	1.60 (1.06–2.41)	.02
Marital statusb		.08		.21
Single	Reference		Reference
Married	1.58 (1.00–2.49)		1.51 (.75–3.02)
Separated/divorced	2.73 (1.12–6.64)		2.89 (.96–8.67)
Widowed	0.96 (.39–2.36)		1.67 (.53–5.22)
Educationb		.79
Primary	Reference
Secondary	1.07 (.72–1.59)
Vocational	0.63 (.23–1.78)
University	1.44 (.57–3.61)
None	1.36 (.37–4.96)
Employmentb		.01
Unemployed	Reference
Student	0.93 (.40–2.16)
Self-employed	2.44 (1.28–4.64)
Paid employee	2.20 (1.11–4.38)
Unpaid family worker	1.30 (.62–2.72)
Retired	1.65 (.22–12.60)
Socioeconomic statusc		.49
Highest quintile	Reference
2nd highest quintile	1.21 (.62–2.37)
Middle quintile	0.73 (.36–1.49)
2nd poorest quintile	1.11 (.57–2.17)
Poorest quintile	1.10 (.56–2.17)

Multivariable model considers individuals aged ≥16 years and includes a cubic spline variable for age to account for change in prevalence with respect to age.

Education, employment, marital status, and socioeconomic status data are applicable to individuals aged >16 years.

Socioeconomic quintiles were derived for the 44 health surveillance areas in the serosurvey in which the participants resided.