Stigma and discrimination experiences of HIV-positive men who have sex with men in Cape Town, South Africa

Abstract

Since the primary mode of HIV transmission in sub-Saharan Africa is heterosexual, research focusing on the sexual behaviour of men who have sex with men (MSM) is scant. Currently it is unknown how many people living with HIV in South Africa are MSM and there is even less known about the stigmatisation and discrimination of HIV-positive MSM. The current study examined the stigma and discrimination experiences of MSM living with HIV/AIDS in South Africa. Anonymous venue-based surveys were collected from 92 HIV-positive MSM and 330 HIV-positive men who only reported sex with women (MSW). Internalised stigma was high among all HIV-positive men who took part in the survey, with 56% of men reporting that they concealed their HIV status from others. HIV-positive MSM reported experiencing greater social isolation and discrimination resulting from being HIV-positive, including loss of housing or employment due to their HIV status, however these differences were not significant. Mental health interventions, as well as structural changes for protection against discrimination, are needed for HIV-positive South African MSM.

Keywords:

• South Africa
• men who have sex with men (MSM)
• HIV-positive
• AIDS-related stigma
• discrimination

Introduction

In the early 1980s in the Apartheid-ruled Republic of South Africa, the HIV epidemic affected mainly the White male homosexual population (van Hermelen et al., 1997). Thus, much of the initial research in South Africa focused on the sexual behaviour of this group in the context of the HIV epidemic and a considerable amount of literature on what it means to be gay in the face of the African AIDS epidemic was published.1

More recent same-sex research, conducted by Lane, Mcintyre and Morin (2006a), revealed that South African MSM engage in high-risk sexual behaviour. The study found that Black African South African MSM are highly vulnerable to HIV infection. In addition to this, Lane et al.'s (2006b) investigation also revealed that stigmatisation as gay and fear of being HIV-positive present barriers to making use of the available voluntary HIV testing and counseling services. Although South African MSM are recognized as at-risk for HIV/AIDS (South African Department of Health, 2000) this population remains marginalised and to a large extent neglected in current HIV/AIDS-prevention campaigns and research.

HIV prevalence and incidence among MSM in South Africa remains undocumented. We are also not aware of any research on the stigmatisation and discrimination experiences of MSM living with HIV/AIDS in South Africa. The current study was thus conducted to address this void by examining the stigma and discrimination experiences of MSM living with HIV/AIDS in Cape Town, South Africa. We hypothesized that HIV-positive MSM would report greater stigma and discrimination experiences than their non-MSM counterparts because of the added stigmas associated with homosexuality.

Methods

Participants

Anonymous surveys were completed by 92 MSM and 330 MSW, all of whom were living with HIV/AIDS. HIV-positive MSW were primarily sampled from local social service and healthcare providers offering services to people living with HIV/AIDS (PLWHA) in Cape Town, South Africa, whereas MSM were purposefully sampled from venues where gay men congregate, as identified by HIV-positive MSM key informants. In preparation for the fieldwork, antiretroviral roll out sites, service providers for PLWHA and established support groups were contacted and asked for their participation in this phase of the study. Thus, the sampling method used in this phase was convenience sampling.

Measures

Measures were administered in a seven-page survey that required approximately 15 minutes to complete. The questionnaire included demographic and health characteristics, HIV-risk history, internalised AIDS stigma, HIV/AIDS discrimination experiences, cognitive and affective depression, social support and substance use. The surveys were available in the three languages commonly spoken in Cape Town: Xhosa, English and Afrikaans.

Procedures

Eleven fieldworkers were recruited, 10 of whom were HIV-positive and openly living with their HIV status.

Before the administration of the questionnaire, participants signed informed consent forms. Respondents were asked not to indicate their names anywhere on the questionnaire. Since the questionnaire was designed for self-administration, fieldworkers reported that the respondents generally found it easy to complete, although there were a few cases where the fieldworkers administered the questionnaire verbally. Ethical clearance was obtained from the Human Sciences Research Council's Research Ethics Council (REC 3/13/10/04).

Data analysis

For descriptive purposes, the frequencies of responses to the internalised stigma items and AIDS discrimination experiences were first examined. Men who have sex with men and HIV-positive MSW who endorsed each item were compared using logistic regression, reporting odds ratios with 95% confidence intervals. Three regression models were conducted that compared MSM and MSW on demographic characteristics, health status variables and stigma and discrimination experiences. The final model that tested stigma and discrimination experiences included covariates that controlled for differences on demographic and health characteristics that were either significant in the previous regressions or were conceptually relevant (e.g. years since testing HIV-positive). Statistical significance was defined as p<.05. Cell sizes vary as participants were included for all analyses in which they had non-missing values.

Results

Demographic characteristics

The demographic characteristics of the survey sample consisting of both MSM and MSW are shown in Table 1. As illustrated in Table 1, MSM were more likely to describe themselves as Coloured, Indian and White than of Black African origin. Only 4% of the MSW described themselves as White, compared to 17% of the MSM. Men who have sex with men were also less likely to have children (MSM 24%, MSW 65%). No significant differences between MSM and MSW were found on any of the remaining demographic characteristics reported on in Table 1.

Table 1. Demographic characteristics of HIV-positive MSM and MSW.

	MSM n = 92		MSW n = 330
Characteristic	n	%	n	%	AOR	95%CI
Age
18–20	13	14	19	6	Reference
21–25	33	36	67	21	0.86	0.17–4.26
26–35	30	33	133	42	0.45	0.11–1.91
36–45	12	13	70	22	0.79	0.19–3.34
46 or older	3	3	24	8	0.80	0.18–3.59
Race
African	30	33	233	73	Reference
White	15	17	13	4	2.29*	1.01–5.19
Coloured	28	31	35	11	0.27*	0.09–0.85
Indian	17	19	36	11	0.38*	0.16–0.93
Years of living in CT
Born in CT	63	69	131	41	Reference
More than 1 year	25	27	176	54	1.39	0.11–16.48
Less than 1 year	2	2	9	3	2.46	0.21–28.91
Only visiting CT	1	1	5	2	3.62	0.17–77.44
Education
No schooling	5	5	13	4	Reference
Primary	20	22	108	34	0.27	0.56–1.32
Secondary	58	64	169	53	1.17	0.39–3.49
Tertiary	8	8	31	10	0.67	0.24–1.86
Employed	32	35	91	28	0.77	0.40–1.48
Has children	22	24	209	65	0.19**	0.09–0.38
Married	11	12	87	27	1.25	0.92–1.71
Notes: *p<.05; **p<.01.

Health, treatment status and mental health

Controlling for all health status variables in the model, significantly more MSM reported positive perceptions of their current health status, than the MSW (Table 2). However, more MSM (42%) in the sample reported having two or more hospitalizations, compared to 28% of the MSW.

Table 2. Health characteristics of HIV-positive MSM and MSW.

	MSM n = 92		MSW n = 330
Characteristics	n	%	n	%	AOR	95%CI
Current perception of health
Excellent	11	12	26	8	Reference
Very good	43	47	80	25	0.14*	0.02–0.90
Good	35	38	185	57	0.12*	0.02–0.64
Poor-Very poor	2	2.2	33	10	0.30	0.06–1.52
HIV-related hospitalisations
Non-hospitalised	33	36	149	47	Reference
One hospitalisation	20	22	80	25	1.60	0.74–3.48
Two or more hospitalisations	38	42	89	28	1.08	0.49–2.38
Taking ARVs	25	27	188	58	1.02	0.77–1.35
HIV-risk history
Has had an STI	59	64	197	61	0.67	0.36–1.24
Received money	26	28	19	6	1.88	0.77–4.58
Given money for sex	26	28	34	11	0.98	0.43–2.22
Injection drug use	42	46	15	5	8.79**	3.70–20.89
IDU sex partner	40	43	19	6	3.51**	1.52–8.12
	M	SD	M	SD
HIV symptoms	7.3	2.7	5.8	3.7	0.96	0.85–1.08
Years since HIV-positive diagnosis	2.7	3.0	2.8	2.4	0.93	0.84–1.02
Depression	12.7	6.3	11.1	5.9	0.98	0.93–1.03
Social support	7.08	2.14	6.7	2.2	0.95	0.82–1.10
Notes: *p<.05; **p<.01.

The two groups differed significantly on risk factors such as injection drug use by either themselves or their sex partners, both of which were more common among MSM. Other markers for HIV-risk history included having received money in exchange for sex (MSM 28%, MSW 6%). Also, as shown in Table 2, there were no differences between MSM and MSW on the depression and social support scores.

Stigma and discrimination

We found that internalised AIDS stigma was high among this sample of HIV-positive men, with 57% of them reporting that they hid their status from others (see Table 3). Almost 47% of the sample felt guilty for being HIV-positive and 43% felt ashamed of being HIV-positive. Moreover, there were no differences between MSM and MSW for feelings of internalised stigma. Contrary to our hypotheses, therefore, there were no differences between MSM and MSW on the internalised stigma items.

Table 3. Stigma and discrimination of HIV positive MSM and MSW.

	MSM n = 92		MSW n = 330
Experiences	n	%	n	%	AOR	95%CI
Stigma
Difficulty of disclosure	59	64	213	66	0.90	0.38–2.10
Feeling dirty due to HIV-positive status	34	37	102	32	1.31	0.60–2.89
Feelings of guilt	46	50	148	46	1.00	0.45–2.23
Ashamed of HIV-positive status	41	45	139	43	0.44*	0.19–1.07
Worthless because of HIV-positive status	34	37	121	38	1.13	0.50–2.54
Self-blame for HIV-positive status	42	46	170	53	0.76	0.37–1.54
Hides HIV-positive status	51	55	188	58	0.94	0.42–2.10
Certainty of disclosing to sex partner	59	64	203	63	2.01	0.88–4.56
Discrimination
Talked to a friend about AIDS	53	58	239	74	0.43**	0.20–.089
Treated differently by friends & family	42	46	124	38	1.20	0.51–2.79
No more visiting after HIV-positive status	37	40	107	33	1.13	0.48–2.64
Job loss	41	45	72	22	1.82	.86–3.79
Discrimination experiences	58	64	131	40	1.11	0.51–2.39
Concealment of HIV-positive status	54	59	193	60	0.98	0.45–2.12
Notes: Model controlling for participant race, having children, perceived health, injection drug use and injection drug using partner histories, years living with HIV and HIV-related symptoms.Notes: *p<.06, **p<.01.

Men who have sex with men did report experiencing greater social discrimination resulting from being HIV-positive, including loss of job or places to stay, however, there were no significant differences found on discrimination experiences between MSM and MSW. Men who have sex with women seemed to have more supportive experiences, with 74% of them having had talked to a friend about AIDS compared to 58% of their MSM counterparts. There were no significant difference of being treated differently by friends and relatives between MSM and MSW.

Discussion

The current study is amongst the first investigations focusing on South African HIV-positive MSM and is the first to examine their experiences of stigma and discrimination. In this study we found that HIV-positive men experienced a considerable amount of internalised stigma.

In a study designed to assess the impact of internalised AIDS stigma in the US, Lee, Kochman and Sikkema (2002) found that 63% of HIV-positive persons sampled in two US cities indicated that they were embarrassed by their HIV infection and 74% stated that it was difficult for them to tell others they are HIV-positive. Lee et al. (2002) further showed that internalised AIDS stigma was related to depression symptoms over and above demographic characteristics, health status, symptoms of grief and various coping responses. These findings suggest that internalised AIDS stigma may play a crucial role in the emotional reactions and distress experienced by many people who are aware that they are living with HIV/AIDS. Internalised feelings of shame and guilt also have an adverse effect on the health status of the person living with HIV; so too can the level of social support provided also have an impact on their health status.

In our study, even though perceptions of current health were significantly different across the groups of men, HIV-positive MSM and MSW reported considerable levels of social support from their family and friends. In addition, moderate levels of cognitive-affective depression symptoms were found among both MSM and MSW, as with Lee at al. (2002) in their US study among PLWHA, but no significant differences were reported between the two groups. According to Hall (1999), receiving adequate social support, whether it is informational or in terms of a strong social network, was an influence on the psychological well-being of MSM living with HIV/AIDS.

All HIV-positive men in South Africa, irrespective of sexual orientation, experienced considerable internalised AIDS stigma, emotional distress and discrimination. However, HIV-positive MSM generally experienced more discrimination related to their HIV status than their non-MSM counterparts. It therefore appears that HIV-positive MSM suffer from double or multiple discrimination or superdiscrimination.

Results of this study should be interpreted in the light of its methodological limitations. Purposive sampling is subject to selection biases. In order to obtain the sample of HIV-positive MSM, specific locations were targeted where MSM congregate, thus the sample characteristics were biased towards MSM congregating at these specific venues. Another limitation of the study was that there was no procedure in place to ensure that only genuine PLWHA complete the survey.

However, it is unlikely that many HIV-negative participants would have done so, given the considerable stigma still attached to HIV-positive status in the communities surveyed, as well as the fact that participants were not made aware of the very small incentive until after completion of the questionnaire. In addition, the fieldworkers who conducted the survey were mainly from the same residential area and attended the same support groups as participants, so that most participants were known to them. Moreover, many surveys were collected at clinics, which offer closed support groups for people already identified as living with HIV/AIDS. However despite the above limitations, we believe that the current findings contribute new knowledge that could be useful in intervention planning for HIV-positive MSM.

These findings have important implications for services and interventions for all HIV-positive men in South Africa. Interventions are needed that can assist HIV-positive MSM to better adapt and adjust to their condition and the social environment. In particular, coping efficacy training targeted to address managing social stigma and reducing internalised stigma should be developed and tested among them, especially MSM.

In the development of risk-reduction interventions for HIV-positive MSM, a component focusing on reducing the use of injection drug use is also important in tailoring the intervention for HIV-positive MSM. In the US, using injection drugs and MSM are dual risks for HIV infection and contribute to the highest rates of infection among any risk group (CDC, 2000, cited in Kral et al., 2005). In an investigation conducted in a US city of HIV-positive gay and self-identified bisexual men injection drug users, Ibanez, Purcell, Stall, Parsons and Gomez, (2005) showed that this risk group engaged in more unprotected sex than non-injection drug users. They also found that HIV-positive gay and bisexual injection drug users experience more emotional distress than their non-injection drug using counterparts.

HIV-positive MSM may also benefit from interventions designed to broaden and strengthen their social support networks. For example, support groups, which are already common in South Africa, especially among MSM living with HIV/AIDS, may be used as a starting place for the development of social support interventions. However, the ultimate solution to AIDS stigma, especially among MSM, does not lie in the hands of HIV-positive men alone. Structural interventions are needed to change both the social climate of AIDS and sexual politics around sexual practices of MSM. Reducing the combined AIDS and MSM stigmas at the societal level could impact on the internalised stigmas that are clearly magnified in MSM living with HIV/AIDS.

Notes

1. Schurink and Schurink (1990) reported on the change of sexual behaviour of gay men in the context of the rising HIV epidemic. Pegge (1995), for example, described the lives of the lesbian and gay communities in South Africa with specific attention to the lives of gay men in Cape Town. In addition, Isaacs and McKendrick (1992) looked at the implications of the HIV epidemic for gay communities in South Africa.