Factors associated with high HIV-related stigma among commuter populations in Johannesburg, South Africa

Figures & data

Table 1. Characteristics of the study participants by stigma level.

Factor	Total N	High stigma n/N (%)	Low stigma n/N (%)	p-value*
Gender
Female	579	309(52.8)	270(48.1)	0.112
Male	567	276(47.2)	291(51.9)
Age group
<25 years	366	192(32.8)	174(31.0)	0.512
$\ge 25$ years	780	393(37.2)	387(69.0)
Marital status
Single	801	402(68.7)	399(71.1)	0.056
Cohabiting/married	321	165(28.2)	156(27.8)
Divorced	24	18(3.1)	6(1.1)
Employment status
Unemployed	474	273(46.7)	201(35.8)	<0.001
Employed	672	312(53.3)	360(64.2)
Education level
Primary	324	144(24.6)	180(32.1)	0.008
Secondary	498	258(44.1)	240(42.8)
Tertiary	324	183(31.3)	141(25.1)
Sexual partnerships
None	108	51(8.7)	57(10.1)	0.629
One	681	354(60.5)	327(58.3)
More than one	357	180(30.8)	177(31.6)
Tested for HIV
No	246	123(21.0)	123(21.9)	0.711
Yes	900	462(79.0)	438(78.1)
Last HIV test
Never tested	220	103(17.6)	117(20.9)	0.097
Less than a year ago	639	321(54.9)	318(56.7)
More than a year ago	287	161(27.5)	126(22.4)
Preferred testing place
Clinic/hospital	414	198(33.9)	216(38.5)	0.101
Home/mobile outreach	732	387(66.2)	345(61.5)
HCT awareness				0.035
No	906	477 (81.54)	429 (76.47)
Yes	240	108 (18.46)	132 (23.53)
Perceived benefit of HIV testing				<0.001
No	738	456 (77.95)	282 (50.27)
Yes	408	129 (22.05)	279 (49.73)
Health worker effect				0.014
None	135	57 (9.74)	78 (13.90)
Doctors	243	111 (18.97)	132 (23.53)
Counsellors	657	354 (60.51)	303 (54.01)
Nurses	111	63 (10.77)	48 (8.56)

*Bivariate associations determined by chi-square tests at 5% significance level.

Table 2. Univariate and multivariate level analysis of factors associated with high stigma level.

Factor	Univariate analysis			Multivariate analysis
	Crude PR**	95% CI	p-value	Adjusted PR	95% CI	p-value
Gender
Female	1			1
Male	0.91	0.81–1.02	0.113	1.04	0.93–1.15	0.503
Age group
<25 years	1			1
$\ge 25$ years	0.96	0.85–1.08	0.509	0.97	0.87–1.08	0.567
Marital status
Single	1			1
Cohabiting/married	1.02	0.90–1.16	0.712	1.14	1.02–1.28	0.018
Divorced	1.49	1.17–1.90	0.001	1.38	1.07–1.78	0.012
Employment status
Unemployed	1			1
Employed	0.81	0.72–0.90	0.001	0.78	0.71–0.87	<0.001
Education level
Primary	1	–	–	1
Secondary	1.16	1.00–1.35	0.043	1.01	0.89–1.14	0.891
Tertiary	1.27	1.08–1.48	0.002	1.04	0.91–1.18	0.560
HCT awareness
No	1			1
Yes	1.17	1.00–1.36	0.044	0.85	0.75–0.97	0.016
Perceived benefit for HIV testing
High	1			1
Low	0.51	0.44–0.60	<0.001	0.69	0.60–0.79	<0.001
Perceived risk for HIV
High	1			1
Low	0.32	0.29–0.38	<0.001	0.36	0.31–0.42	<0.001
Health worker effect*
None	1			1
Doctors	0.80	0.65–1.00	0.045	0.97	0.77–1.22	0.769
Counsellors	0.95	0.80–1.13	0.565	1.10	0.89–1.34	0.379
Nurses	0.74	0.58–0.96	0.023	1.11	0.88–1.39	0.389

*Health worker responsible for reducing willingness to test for HIV, **PR-prevalence ratio.

Figure 1. ROC curve.

Figure 2. Deviance residuals.

Table 3. Assessment of multicollinearity.

Variable	Variance inflation factor	R-squared	Tolerance
Gender	1.12	0.1	0.9
Age group	1.13	0.11	0.89
Knowledge of health centres	1.04	0.03	0.97
Marital status	1.13	0.12	0.88
Education level	1.04	0.03	0.97
Employment status	1.21	0.17	0.82
Health worker stigma	1.02	0.02	0.98
Perceived benefit of HIV test	1.14	0.12	0.88
Perceived HIV infection risk	1.44	0.31	0.69