Understanding how social support influences peer-delivered HIV prevention interventions among Ugandan female sex workers: a case study from HIV self-testing

Peer delivery of HIV prevention interventions has been proven an effective approach for reaching key populations at high HIV risk in diverse settings [1‐4]. In sub-Saharan Africa, nearly 20% of new HIV infections are estimated to be among female sex workers (FSWs) and their clients [5, 6]. FSWs are at increased risk of HIV infection for reasons including engagement with numerous sexual partners, inconsistent condom use, and structural barriers, such as stigmatization by self and healthcare providers, limited schooling, long distances to health facilities, high costs of traveling to healthcare facilities, and criminalization of sex work as well as high levels of alcohol and drug misuse [6‐9].

In Kampala, the capital city of Uganda, there are roughly 10,000 FSWs (~ 2% of all women ages 15–59); one in three are estimated to be living with HIV [5, 10‐13]. A critical entry point for HIV prevention and treatment services among populations at increased risk is HIV testing [8, 14]. A randomized implementation trial in this setting found that peer-delivered oral-fluid HIV self-testing (HIVST) significantly increased recent and repeat HIV testing among FSWs compared to standard facility-based testing [4]. Additionally, FSWs in this trial described peer-delivered HIVST as desirable because it increased privacy, convenience (including time and cost), sense of control, and confidentiality [15, 16]. Although this model of peer delivery has been found to be appropriate among FSWs in this setting, little is known about how FSW social support structures (e.g., informational, instrumental, and emotional support) influence the delivery and uptake of peer-delivered HIV prevention interventions.

Based on evidence that peer education for HIV prevention is significantly associated with increased HIV knowledge and condom use in a variety of settings [17, 18], we hypothesized that, delivering HIVST via trusted peer educators (themselves established members of the FSW community) would generate social support for—and lead to uptake of—HIVST among FSWs. Specifically, our hypothesis and intervention design were grounded in evidence that the likelihood of uptake of peer-delivered interventions increases when delivered by peer educators (PEs) who are trusted members of the target community [17]. We further based our hypothesis and intervention design on evidence that information delivered by PEs (as opposed to individuals outside of the target community) is often perceived by recipient peers not only as culturally appropriate and easy to understand but also as an open invitation to discuss sensitive topics, such as HIV [17].

In this study, we aimed to understand how FSW social support can influence (i.e., motivate or discourage) the uptake of peer-delivered HIVST in urban Uganda. Specifically, we aimed to understand how FSWs experienced social support within the context of the intervention and how this compared with the intended intervention design. We hope that a better understanding of FSW peer interactions in this setting can inform the design of future peer-delivered HIV prevention interventions among FSWs and other key populations to maximize intervention implementation and outcomes.

In February 2017, we completed five FGDs with 28 PEs (median age 33 years, interquartile range [IQR] 29–37 years). Between March and April 2017, we completed 30 IDIs with participants randomized to either the direct HIVST delivery arm (n = 19) or the coupon exchangeable for HIVST arm (n = 11). The median age of participants was 30 years (IQR 27–33 years), most (80%, n = 24) reported a primary or secondary level of education, and the majority (74%, n = 22) reported testing for HIV in the past year, Table 1.

In our data we identified three of the four types of social support: informational, instrumental, and emotional support [22]. We also modified the theory of social support by identifying how each of these types of support either motivated or discouraged participants from accepting the peer-delivered HIVST intervention. Motivating and discouraging social support factors, as experienced by participants, are mapped against the intervention design elements in Table 2.

Ugandan FSWs in this study supported each other informationally, instrumentally, and emotionally to engage in HIVST. While these forms of social support mostly encouraged intervention uptake and linkage to care, contextual factors and their indirect influence on social support sometimes discouraged FSWs from engaging in HIVST. FSWs felt motivated to accept HIVST when their PE provided accurate intervention information, directly delivered HIVST kits, and acted in their best interest, as well as when they felt emotionally supported by their PEs and study peers. In contrast, FSWs hesitated to use peer-delivered HIVST or coupons for HIVST when they heard negative experiences and misinformation about HIVST from peers (in the study and in their communities), when HIVST availability was limited due to product stock outs, and when they feared negative social consequences from their communities following involuntary HIV status disclosure.

These findings highlight the complexity of utilizing word-of-mouth peer systems to disseminate intervention information, thereby emphasizing the need for intervention evaluations. While peer instruction on HIVST use was intended by the intervention, direct peer assistance conducting HIVST (e.g., timing, and interpreting HIVST results) was not intended and may pose risk of testing coercion or involuntary HIV status disclosure [15]. Furthermore, word-of-mouth peer networks were also prone to the quick spread of misinformation surrounding HIV risk and HIVST. While some of this misinformation around HIVST was shared by FSWs not enrolled in the study, we also found that misinformation shared between study peers persisted after PE intervention visits and was often only resolved through the sharing of positive HIVST experiences or role modeling. This indicates that while the intervention was experienced as successful regarding its distribution of verbal and pictorial HIVST information, some FSWs may benefit from having PEs or peers demonstrate HIVST use on themselves (i.e., role modeling) to mitigate fears surrounding adverse effects of HIVST. Thus, we recommend that future interventions enhance the informational support shared between study peers through role modeling, while maintaining awareness of the context in which the intervention is delivered and the misinformation which can be shared between FSW peers in the community. For example, future studies may utilize “champion” PEs (also referred to as “mentor mothers” in Ugandan prevention of mother-to-child transmission programs) who have had success with HIV prevention interventions, to share their experiences, and to address any concerns or misinformation about intervention side effects [32, 33]. We also recommend that future interventions intermittently evaluate the information being shared between FSW peers and adjust informational support accordingly to mitigate the spread of misinformation.

Furthermore, findings from this study emphasize the barriers Ugandan FSWs face while accessing facility-based care and the advantages of direct peer-delivered interventions. Like other FSW studies in this setting, [34] the PEs and peers in our study overcame barriers to intervention use by pooling resources (e.g., HIVST coupons, travel funds). In some cases, such assistance resulted in participants in the coupon-delivery arm receiving HIVST kits directly from their PEs, thus likening this intervention arm to the direct-delivery arm and indicating a general preference among Ugandan FSWs for direct delivery over coupon-delivery, a preference that was also confirmed in findings from the larger implementation trial [4]. We also speculate that PEs may have redeemed coupons to gain trust from their participants, remain well liked, and mitigate jealousy among participants in different study arms. Based on these findings, we recommend the direct peer delivery of future HIV prevention interventions, as this may better meet FSW preferences and further promote intervention uptake. Additionally, future FSW peer-delivered interventions should consider the delivery of multiple versus single intervention units (e.g., providing multiple HIVST kits at a time) to increase intervention penetration within the community, as demonstrated successfully among Ugandan men who have sex with men [3].

The findings that peer emotional support was a primary incentive to HIVST uptake among FSWs supported underlying assumptions in the intended intervention design. This research highlights the importance of who the PE is and the type of support they provide. PEs who are knowledgeable, maintain confidentiality of the peer’s HIV status, and provide reassurance should be recruited for future peer-delivered interventions and can be successfully identified by collaborating with established peer network organizations (including FSW NGOs) within the community [25]. Future interventions should collaborate with these selected peer educators at early stages of intervention development to ensure intervention appropriateness in the target population. Additionally, to enhance the advantages of peer emotional support, future peer-delivered interventions might incorporate the concept of an intervention partner (i.e., an individual to whom FSWs can be accountable or who can hold them accountable), which has been demonstrated to increase HIV pre-exposure prophylaxis (PrEP) and ART adherence in similar settings [32, 35, 36].

Our research finally recognizes a unique balance between the emotional support required for intervention uptake and the fear of social consequences from communities in which the intervention is delivered (e.g., social exclusion following involuntary HIV status disclosure) resulting from intervention uptake. While previous literature indicates stigmatization and social exclusion from FSW peer groups living and not living with HIV, we did not expect to find some participants misreporting an HIV-negative result for fear of social exclusion [16, 34]. The perception held among FSWs that most peers are living with HIV merits further investigation to ensure it does not dissuade FSWs from regular HIV testing or misreporting testing results.

While these findings provide novel insight into the role of social support on peer-delivered HIVST uptake, this study has some limitations. First, since we recruited participants through PEs who often worked within established, hierarchical structures (e.g., guest houses), FSWs who were not part of these structures (and were likely of lower social standing) may not have been recruited and thus may be underrepresented in this study [2, 37]. Second, HIVST misinformation may have emerged as a major theme, in large part, because HIVST was not commercially available in Uganda outside of research settings at the time this study was conducted. Finally, because this study used secondary data from a larger randomized HIVST trial that did not design the FGD or IDI guides with the dimensions of social support in mind, our findings may have overlooked some aspects of social support relevant to peer-delivered HIV interventions among FSWs [4, 38].

This qualitative study highlights how social support (e.g., informational, instrumental, and emotional support) among Ugandan FSW peers can work in ways that both motivate and discourage peer-delivered intervention uptake. Based on our findings, we suggest that future FSW peer-delivered HIV prevention interventions first work with community organizations to identify trusted and knowledgeable PEs, train these individuals on how to successfully deliver the intervention, and give them the opportunity to develop their own experiences with the intervention before having them directly deliver the intervention to their peers. We also encourage future interventions to consider the context in which the FSW peer-delivered intervention will be delivered and to develop strategies to mitigate anticipated challenges (e.g., community-level misinformation, product unavailability, and community-level stigmatization). These approaches may help maximize the uptake and impact of future peer-delivered HIV prevention interventions in Uganda and other similar settings.