“That person stopped being human”: Intersecting HIV and substance use stigma among patients and providers in South Africa

Abstract

Background:

South Africa has the largest number of people living with HIV in the world. Concurrently, problematic alcohol and other drug use (AOD) is prevalent in the country and associated with poor HIV treatment outcomes. Further, the high rates of stigma surrounding HIV and AOD contribute to poor HIV outcomes. Yet, how HIV stigma and AOD stigma together may affect HIV care has not been extensively studied in this context. Thus, we explored HIV and AOD providers’ and patients’ experiences of HIV and AOD stigma.

Methods:

We conducted 30 semi-structured interviews with patients living with HIV who were struggling with HIV medication adherence and problematic AOD use (n=19), and providers involved in HIV or AOD treatment (n=11) in Cape Town, South Africa to assess how HIV and AOD stigmas manifest and relate to HIV care.

Findings:

Two main themes around the intersection of HIV and AOD and their related stigmas were identified: (1) how patients use AOD to cope with HIV stigma; and (2) enacted/ anticipated AOD stigma from HIV care providers, which acts as a barrier to HIV care.

Conclusions:

Intersecting HIV and AOD stigmas exist at multiple levels and increase barriers to HIV care in this setting. Accordingly, it is important that future interventions address both these stigmas at multiple levels.

1. Introduction

Home to an estimated 7.7 million people living with HIV (PLWH), South Africa has the highest number of PLWH in the world (UNAIDS, 2019). Concurrently, problematic alcohol and other drug use (AOD)—frequent and/or heavy use that is likely to cause health or other problems (Humeniuk et al., 2010)—is prevalent among PLWH across the country, with alcohol-related morbidity and mortality highest amongst PLWH with low socioeconomic status (Probst et al., 2018). According to the World Health Organization (WHO), nearly 20% of South Africans over 15 reported heavy episodic drinking in the past 30 days, with the rate rising to almost 60% among people who reported drinking (WHO, 2018). Other substances, such as methamphetamine (locally “tik”) and cannabis (locally “dagga”) are also common, especially in the Western Cape province (Dada et al., 2016), which has the highest prevalence rate of past three-month illicit drug use among people over 15 (7.9%) (Peltzer and Phaswana-Mafuya, 2018).

If taken consistently, PLWH who are on antiretroviral therapy (ART) can reach normal lifespans. However, poor adherence to ART is associated with health complications (Bangsberg et al., 2001) and early mortality (García de Olalla et al., 2002). Suboptimal adherence also increases the likelihood of becoming treatment resistant, which is especially serious in resource-limited settings like South Africa, where only limited ART regimens are readily available (Moorhouse et al., 2019). Among PLWH, problematic AOD is associated with worse adherence (Azar et al., 2010).

In South Africa, having HIV and being identified as a person with AOD are both stigmatized identities (Magidson et al., 2019b; Simbayi et al., 2007; Sorsdahl et al., 2012), identities that are looked down upon, leading others to view people with such identities as inferior (Goffman, 1963). The Stigma Framework (Earnshaw and Chaudoir, 2009) has previously been used to conceptualize both HIV stigmas (Earnshaw et al., 2013) and AOD stigmas (Smith et al., 2016). It outlines three stigma mechanisms: enacted stigma, the lived experience of being stigmatized; anticipated stigma, the expectation of being stigmatized; and internalized stigma, self-belief about the stigma. Perceived community stigma, an individual’s perception of stigma in the community, has also been considered in other theories as a distinct yet related dimension of stigma (Turan et al., 2017a).

HIV stigma has been well-established as a barrier to HIV-related health globally (Katz et al., 2013). Although less studied, AOD stigma may also affect HIV health behaviors. There is evidence that high levels of anticipated AOD stigma are associated with worse ART adherence in the United States (Stringer et al., 2019), and that internalized alcohol stigma in the context of HIV may partially mediate the relationship between alcohol use and ART adherence in South Africa (Kalichman et al., 2020).

When applied to health-related stigma, the intersectionality framework posits that to better understand groups facing multiple stigmas, these stigmas must be considered together (Bowleg, 2012, 2008; Crenshaw, 1989), as having multiple stigmatized identities may amplify the negative effects of the stigma (Bauer, 2014; Turan et al., 2017b). Despite HIV and AOD’s common co-occurrence, the intersecting effect HIV and AOD stigmas may have on HIV care has rarely been examined. This research gap suggests that we are missing important aspects of how these stigmas may affect each other and related treatment outcomes in this population. By examining multiple stigmas together, we can better elucidate how they intersect at multiple levels to impact health outcomes and health care utilization, providing important information to guide the design of health-promoting interventions (Turan et al., 2017b).

Therefore, the present study qualitatively explores intersecting HIV and AODs stigmas in South African HIV care. Through semi-structured interviews with patients and providers involved in HIV or AOD care at two primary care clinics, we aimed to assess how HIV and AOD stigmas manifest and relate to HIV care.

2. Methods

2.1. Setting

The present study took place in Khayelitsha, a peri-urban area in Cape Town, Western Cape, South Africa. Khayelitsha has the highest HIV prevalence in the Western Cape, with estimates ranging from 20% (Western Cape Provincial AIDS Council, 2016) to 33% (Shaikh et al., 2006). Although its official population is approximately 400,000 people (Statistics South Africa, 2013) some estimates put the population closer to one million (Brunn and Wilson, 2013; Cronje, 2014; Overdorft, 2019). Khayelitsha’s population is predominately Black African and isiXhosa speaking (Statistics South Africa, 2013). According to the 2011 census (Statistics South Africa, 2013), 38% of the labor force is unemployed, 55% of households live in informal dwellings, and 19% of households have no monthly income. AOD treatment options in this community include Matrix, a 16-week outpatient cognitive-behavioral treatment program (Gouse et al., 2016; Magidson et al., 2017a; Rawson et al., 1995; Shoptaw et al., 1995), and another outpatient program offered by the South African National Council on Alcoholism and Drug Dependence, which offers a variety of group and individual sessions (SANCA Western Cape, n.d.).

2.2. Recruitment

2.2.1. Patients.

As part of a larger qualitative study with primary aims around integrating AOD care into HIV care, patients were recruited from two HIV primary care clinics in Khayelitsha or the Matrix AOD treatment program co-located within one of the clinics. Clinic providers referred potentially eligible patients to a study field worker for eligibility screening. Eligibility criteria were: (1) HIV positive; (2) on ART; (3) 18-65 years old; (4) struggling with adherence (defined as having a viral load >1000 copies/mL or previously re-initiated on first-line treatment); and (5) WHO-ASSIST score of 4-26, indicating moderate to severe risk of AOD-related problems (WHO ASSIST Working Group, 2002). The WHO-ASSIST assesses severity of tobacco, alcohol, cannabis, cocaine, amphetamine, inhalant, sedative, hallucinogen, opioid and other drug use over one’s lifetime and the past three months, along with lifetime injection drug use, and has been validated in South Africa.

2.2.2. Providers.

Providers who were involved in delivering HIV or AOD care at either of the HIV primary care sites or the Matrix AOD treatment center co-located at one of the clinic sites were eligible. The medical officer at each site identified eligible providers.

2.3. Procedures

Research assistants (RAs) completed informed consent procedures with eligible participants, including describing the study in detail and stressing that participation was voluntary, confidential, and would not impact their job/treatment at the clinic. RAs also collected demographic and basic information from participants. Interviews occurred between October 2016 and February 2017 in private rooms at the clinics, were digitally audiotaped, and lasted approximately one hour. Ten providers were interviewed by a trained psychology PhD student in English. Patients and one provider who opted to be interviewed in isiXhosa were interviewed by one of two isiXhosa-English bilingual RAs who had prior experience conducting interviews in similar settings.

Interviews followed semi-structured interview guides comprised of open-ended questions and probes to further examine responses. Researchers developed separate guides for patients, HIV providers, and AOD providers. The patient interview guide was translated into isiXhosa and back-translated into English. Participants were compensated with a grocery store gift card (ZAR 150, approximately $11 USD during study). All procedures were approved by institutional ethics committees and the City of Cape Town.

2.4. Data Analysis

Independent, experienced, and trained bilingual translators whose first language was isiXhosa transcribed interviews into English verbatim. The translators consulted with each other and the US team; no specific translation challenges arose. Transcripts were analyzed with thematic analysis (Braun and Clarke, 2006) informed by grounded theory (Glaser and Strauss, 1967) to deductively analyze themes from the interview guide while inductively identifying additional themes. Final sample size was based on saturation for primary study aims. Two independent coders developed a codebook after open-coding five interviews using NVivo v.11. All transcripts were double-coded and coders met weekly with the study’s Principal Investigator (JFM) to iteratively develop the codebook. All coding discrepancies were resolved by discussion between coders. Interrater reliability was measured throughout coding, with a final Kappa score > 0.80.

3. Results

3.1. Participant Characteristics.

Patient and provider participant demographics are summarized in Table 1.

Table 1.

Patient and Provider Demographics

	Providers (N=11)	Patients (N=19)
Age – M (SD)	43.5 (7.6)	39.3 (8.1)
Recruited from HIV Clinic – N (%)	8 (73)	15 (79)
Gender – N (%)
Female	9 (82)	11 (58)
Race – N (%)
Black African	8 (73)	19 (100)
White	3 (27)	0
Substances – Patients Only (N=18a)
Alcohol – N (%)
	Any Use	18 (95)
	Moderate Use (> 11)	9 (47)
	ASSIST – M (SD)	21.8 (7.2)
Cannabis – N (%)
	Any Use	9 (47)
	Moderate Use (> 4)	3 (16)
	ASSIST – M (SD)	17.0 (11.4)
Sedatives – N (%)
	Any Use	1 (5)
	Moderate Use (> 4)	1 (5)
	ASSIST – M	23.0
Amphetamine – N (%)
	Any Use	2 (11)
	Moderate Use (> 4)	0
Inhalants – N (%)
	Any Use	2 (11)
	Moderate Use (> 4)	0
>1 substance endorsed – N (%)
	Any Use	10 (53)
	>1 Moderate Use	2 (11)
Injection Drug Use – N (%)
	Any Use	0
Other Characteristics – Patients Only (N=19)
Has seen someone to talk about emotions, mental health, or AOD use in past year – N (%)
	Social Worker	2 (11)
	Mother	1 (5)
	None	16 (84)
Hides HIV Meds – N (%)		6 (32)
Owns Cell Phone – N (%)		12 (63)

^aOne participant’s WHO-ASSIST scores are missing due to a malfunction on the tablet with which scores were collected.

Patient participants (n = 19) were 58% female and 100% Black African. While participants were asked about all classes of substances, none endorsed cocaine, amphetamine, hallucinogen, opioid, or injection drug use. Seventy-nine percent were recruited from HIV care, and the remainder from the AOD program. No participants recruited from HIV care had ever received AOD treatment.

Provider participants (n = 11) were 82% female, 73% Black African, and 27% white. Eight were recruited from HIV care and included an adherence counselor (n = 1), nurses (n = 3), medical officers (n = 2), and community health workers (CHWS; n = 2). Three were recruited from the AOD program and included a treatment director (n = 1) and addiction counselors (n = 2).

3.2. Findings

Two major themes arose from patient and provider interviews related to the intersection of HIV and AOD and their related stigmas: (1) AOD to cope with internalized HIV stigma among PLWH; and (2) enacted/ anticipated AOD stigma among HIV healthcare providers as a barrier to PLWH receiving HIV care.

3.3. AOD to cope with patients’ internalized HIV stigma

Some patients reported coping with internalized HIV stigma through AOD. For instance, when asked why he thought AOD was prevalent among PLWH, he recalled personally using because having HIV made him feel like he no longer mattered:

“…when I found out my [HIV] status. It seemed like – like the end of my life – that there was nothing else that I could do now. Like, I am a dead man walking. Sometimes, I feel like I don’t exist anymore. I am nothing on earth, you see?” – [Male, Alcohol/Cannabis, HIV Care]

Later, when this patient realized he was not alone, he was able to shift his outlook that PLWH are lesser: “there are many who are the same as I am…they still do the things they want to do, they feel like they can do them.”

Another patient echoed the above sentiment when he explained to the interviewer that AOD use was common among PLWH because PLWH are viewed as subhuman. In order to cope with feeling lesser than human, PLWH use substances:

“It is like since the person is known that he she is HIV positive by some people, so it is like that person stopped being human, so from her his side this is why the person thinks that it is better to use drugs.” – [Male, Alcohol, HIV Care]

On the AOD side of care, another participant stated, “most of the people who use drugs are people who are [HIV] positive.” When asked why, she replied, “because they have stigma [issues]” and further elaborated that in the context of experiencing HIV stigma, AOD can lessen negative thoughts and make PLWH feel less lonely because AOD usually occurs in groups [Female, Alcohol, AOD Care].

Although providers were not directly asked about their patients’ experiences of HIV stigma, some responses alluded to the relationship between AOD and constructs that may reflect internalized HIV stigma (i.e., lack of acceptance of one’s HIV status, looking down upon oneself because of positive HIV status). For instance, providers shared that AOD use was prevalent among PLWH because “it’s better for them to be drunk instead of accepting that ‘I’m HIV positive’” [Female Provider, HIV Care]. Another provider shared, “[PLWH] would say that as a way of dealing with being HIV-positive, they would use drugs as a way of trying to forget that it hurts,” [Male Provider, AOD Care]. Indicating that their patients often turned to AOD because acknowledging that they had an HIV-positive identity “hurt”—or was difficult to accept—suggests they believed their patients looked down upon this aspect of their identity (i.e., internalized stigma).

3.4. Enacted and anticipated AOD stigma from HIV providers

Patients recalled experiencing enacted AOD stigma during interactions with HIV providers. For instance, one patient recalled staff shouting at him for AOD when he went to the clinic to pick up his ARTs. Instances of enacted AOD stigma likely contributed to anticipated AOD stigma, as other patients recalled fearing HIV staff due to their AOD stigma. For example, another patient stated that she lied to her HIV provider about her struggles with alcohol. When the interviewer asked why, she explained that she feared her provider’s negative reaction towards AOD:

“I had that fear…it’s like he/she is going to embarrass me [because I struggle with alcohol use]…” – [Female, Alcohol, HIV Care]

HIV providers, including CHWs, acknowledged that AOD stigma from providers was problematic in the HIV clinics while simultaneously expressing their own stigmatized views about AOD. For example, as one provider described that she was trying to be less judgmental, she also used stigmatizing language to describe patients with AOD:

“I want to not be judgmental…I sort of lead it like “I know people drink; I know people smoke; I know they use drugs”…so it is not judgmental that they feel like they can talk to me about it…Well, either they are stoned out of their minds they don’t know what the time of day is, or sometimes when someone’s really late, they are reluctant to come back. There is sort of a feeling that you get judged when you are late. We are trying to change but still sometimes it happens that they get scolded for being late. And then they just don’t come back. Or even when they come while high, or when they are absolutely drunk.” – [Female Provider, HIV Care]

Other HIV providers described patients who used drugs with negative adjectives including “lazy”, “reckless”, “outrageous”, “angry”, “naughty”, “rowdy”, “irritated”, and “rude”. They also used blanket statements to describe people who use AOD, asserting that they are “prone to criminal activities” [Female Provider, HIV Care] and are often “wandering around and robbing people’s phones and bags” [Female CHW].

Some HIV providers also drew a distinction between patients who use alcohol versus other drugs in their biases. As one CHW explained:

“A person that drinks alcohol…is not as naughty as someone that uses drugs. Those ones that use drugs are naughty…the behavior…is different from someone who uses alcohol only. You only see a person behaving well when drugs haven’t been used yet.” – [Female CHW]

AOD providers were also aware of enacted AOD stigma among HIV providers and expressed concern about it:

“So there is still a lot of stigma in the health profession and with health professionals—nurses and all of them around clients or people that are using substances. And very often, there would be threats made by the health professionals to these guys that, like, if you are using drugs, you are not getting your medication.” – [Male Provider, AOD Care]

3.5. Summary of findings

Figure 1 presents a summary of the findings. Altogether, the interviews suggest that PLWH use substances to cope with internalized HIV stigma, while concurrently facing enacted/anticipated AOD stigma from providers at the HIV clinic. This relationship may contribute to health care utilization and health outcomes, including potentially lower engagement in HIV care.

Figure 1. HIV and AOD Stigmas Contribute to Suboptimal HIV Care.

Figure 1 illustrates how both HIV and AOD sstigmas contribute to problematic AOD use and suboptimal HIV care in PLWH who use AOD. Participants stated that AOD is used by PLWH to cope with internalized HIV stigma (potentially driven by perceived community stigma). Problematic AOD use alone can lead to suboptimal HIV care as patients struggling with AOD may forget take their ARTs or attend their clinic appointments. However, interviews revealed that enacted and anticipated AOD stigma from HIV providers may also contribute to suboptimal HIV care. Based on past experiences of HIV providers who demonstrated AOD stigma, patients who anticipate their HIV providers stigmatizing AOD may lie to their providers or avoid the clinic altogether.

4. Discussion

The present study is among the first to describe how patient and provider HIV and AOD stigmas interact to increase barriers to HIV care for PLWH in South Africa. Two main themes emerged regarding the intersection of HIV/AOD and their associated stigmas: (1) AOD among PLWH to cope with internalized HIV stigma, and (2) enacted AOD stigma from HIV providers/patients anticipating this as a barrier to HIV care, as illustrated in Figure 1.

When asked why PLWH use substances, patients explained that they used because the HIV-positive identity made them feel worthless or “subhuman”. AOD was described as a coping mechanism for these negative feelings and a way to feel less lonely, as AOD is often a social activity. In line with a previous finding that PLWH who perceive more community stigma may experience more internalized stigma (Turan et al., 2017b, 2017a), one participant explained that internalized stigma occurs because “the person is known [to be] HIV positive by some people” in the community. Additionally, supporting evidence that knowing other PLWH may be associated with less internalized stigma (Prati et al., 2016), another participant described his internalized HIV stigma dissipating when he realized that others who did not seem lesser were also PLWH. Providers also explained that their patients used substances because it was difficult for them to accept having an HIV-positive identity, suggesting their shared view that internalized HIV stigma contributes to AOD among their patients.

Additionally, enacted/anticipated AOD stigma from HIV providers was identified as a barrier to HIV care. Patients recalled being scolded for AOD at HIV clinics and minimizing their AOD to providers to avoid anticipated provider AOD stigma. Both HIV and AOD providers were aware of provider AOD stigma and cited it as a potential barrier to patient engagement in HIV care. Yet, despite this acknowledgment and stated efforts to be less judgmental, HIV providers still used stigmatizing language to describe their patients struggling with AOD. Although not formally assessed given the small sample size, we did not identify any apparent differences in AOD stigma between HIV providers at different clinics, despite one clinic having a co-located AOD treatment program.

While this study has multiple strengths—including interviews with the hard-to-reach population of PLWH struggling with AOD in a resource-limited setting, a focus on an understudied area of AOD stigma in an LMIC, and a focus on the intersection of AOD stigma and HIV care outcomes—its limitations should be noted. First, patients were recruited from a clinic or treatment program, indicating that they were already engaged in care to some extent. Consequently, it is possible that these interviews did not capture the views of PLWH who are not engaged in care and may need the most support engaging. Yet, since we prioritized individuals struggling with adherence, we also did not capture perspectives of individuals who use substances who were adherent to ART. Further, our sample of providers was limited to two clinics and the available providers in each role. We also had limited variability in types of drugs used, which impeded our ability to understand how stigma may differ across different drug types among patients. Moreover, because this was a qualitative study, we did not aim to quantify the extent/frequency of HIV/AOD stigmas in the community. Additionally, final sample size was determined based on reaching saturation for primary study aims. As this study was a secondary analysis to capture the themes surrounding HIV/AOD stigmas that emerged in the inductive analysis, we did not assess saturation for stigma and probes focused on stigma and how stigma related to HIV/AOD were limited. Especially given the study findings emerged as part of an inductive analysis, it was challenging at times to disentangle stigma from other related constructs. For instance, when patients described feeling like “nothing” because of their HIV status, they could be referring to internalized HIV stigma (i.e., implying they view this aspect of their identity as inferior) or HIV-related distress. While these constructs can be difficult to separate, in order to advance our knowledge of intersectional stigmas in low-resource settings, we believe it is important to share the less-explicit narratives of stigma (Bowleg, 2008) and have tried to contextualize each quote (Bowleg, 2012). Finally, given that some interviews were conducted in isiXhosa and coded in English, there is a possibility that some themes or terms could have been lost or changed in the translation process.

Altogether, despite the study limitations, the present findings suggest that several factors should be considered when implementing new efforts to keep patients struggling with AOD engaged in HIV care. For instance, HIV providers, and especially CHWs, expressed stigmatized opinions of patients with AOD, and suggested that they judged patients who used drugs more harshly than those who drank just alcohol. Although CHWs are meant to support patients by acting as a link between the community and the clinic, their stigmatized views towards AOD may distance patients from the clinic. Stigma-reduction training targeting CHWs may be beneficial for engaging patients in HIV care. Further, if an effective CHW training program is created, it will likely offer the potential for scale-up, as all clinics in this setting use CHWs, and CHWs are directly employed by NGOs or the Department of Health.

Offering an explanation for why a patient felt less stigmatized after realizing others he knew were PLWH, intergroup contact theory (Pettigrew et al., 2007) proposes that intergroup contact reduces prejudice. Having community members who are open about their positive status interact with the community more may increase the visibility of HIV, decreasing perceived community stigma and internalized stigma. However, such an intervention would have to be adapted for and studied within this setting, as such interventions have only been successful in reducing HIV stigma in some communities (Derose et al., 2016).

Individual patient-level efforts may address both HIV and AOD by focusing on decreasing internalized stigma while promoting health coping mechanisms. There is preliminary evidence that brief stigma reduction interventions delivered via phone can improve ART adherence in South Africa (Kalichman et al., 2019). Text-enhanced stigma reduction interventions for PLWH who use AOD also have shown promise in the United States (Batchelder et al., 2020). However, given that nearly 40% of patients we interviewed did not own a cell phone, and even patients with phones may not have a private space to take calls, in-person stigma reduction interventions should also be further explored. In particular, interventions that allow individuals come into contact with people with lived experience of the stigmatized identity who can provide examples of recovery and change may be beneficial (Thornicroft et al., 2016).

In many health/behavioral interventions in low-resource settings including South Africa, task-sharing—training lay health workers to safely deliver care that has traditionally been delivered by specialists—is essential (Magidson et al., 2019a, 2019c, 2017b). Our prior work has demonstrated that HIV/AOD patients perceive an intervention task-shared to peers—people with similar lived experiences to be less stigmatizing (Magidson et al., 2019b). Further, prior work involving PLWH in the implementation of stigma-reduction interventions has been successful in reducing stigma in the PLWH involved (Uys et al., 2009), suggesting that peers themselves might also benefit from such interventions. Thus, future interventions targeting PLWH or AOD in this setting should consider using peers as interventionists.

According to the intersectional stigma framework, the stigma-related themes identified should not be considered in a vacuum; themes were found specifically around PLWH who use substances (Bowleg, 2012, 2008; Turan et al., 2019). Consequently, to better understand how these stigmas may impact health behaviors, future work should also consider other potentially relevant identities. For instance, asking all participants about multiple dimensions that may further contribute to HIV care and AOD use—such as gender, race, substance type, and sexuality—may provide a deeper understanding into the role of stigma in this context. Additionally, directly asking such questions to both patients and providers would elucidate other stigmas that may exist in this context at multiple levels and potentially differ between HIV and AOD care. While key informant interviews have rarely explored intersectional stigma (Chambers et al., 2015), previous studies suggest it is possible (e.g., Rice et al., 2018). Further, comparing HIV/AOD patients who are and are not struggling with adherence may provide insight into ways to minimize harms and problems associated with AOD in the context of HIV care.

5. Conclusion

This research illustrates how HIV and AOD stigmas at multiple levels increase barriers to HIV care for PLWH struggling with AOD in South Africa. We hope that future research will build upon these findings to further understand how intersecting HIV and AOD stigmas may impact health outcomes. Future work is needed to better understand the multiple stigmas in this context, particularly to inform stigma reduction interventions that intervene at multiple levels—the patients, the providers, and the community.

Highlights.

• HIV and alcohol and other drug use (AOD) stigmas perceived as barriers to HIV care

• People living with HIV perceived to use AOD to cope with internalized HIV stigma

• Providers perceived to enact AOD stigma

• Anticipated provider AOD stigma perceived as barrier to patients accessing HIV care

• Future research needed to refine intersecting HIV/AOD stigma models in this context