Antiretroviral treatment sharing among highly mobile Ugandan fisherfolk living with HIV: a qualitative study

Abstract

Antiretroviral treatment (ART) diversion from prescribed to non-prescribed users (i.e., sharing or selling) is an understudied dimension of HIV treatment adherence. We sought to explore ART diversion patterns in high-prevalence fishing communities on Lake Victoria, Uganda. We implemented a qualitative study in two fishing communities on Lake Victoria in south-central Uganda to identify facilitators of and pathways to ART diversion. We conducted 25 semi-structured interviews with HIV-positive fishermen (n=25) and female sex workers (n=10) covering personal and community experiences with ART selling/sharing, reasons for medication diversion, and potential solutions to reduce diversion. Data were analyzed using an adapted framework analysis approach. Participants reported frequent ART sharing within occupational networks, but no selling. Mobility was the principal driver of ART sharing and was associated with other barriers to treatment access including stigma, fear of negative health provider interactions, and transportation. ART sharing appears to emerge in response to short-term treatment interruptions in this setting. Future studies should explore characteristics and drivers of ART diversion in other high-burden settings and identify how these practices are correlated with key health outcomes like virologic failure and drug resistance.

Introduction

Patient adherence to antiretroviral treatment (ART) remains a complex challenge for HIV treatment and prevention efforts. Suboptimal ART adherence, including sporadic interruptions of HIV treatment or defaulting from care altogether, facilitates HIV disease progression and has been implicated in resistance to first- and second-line therapies (Beyrer & Pozniak, 2017). Poor ART adherence threatens UNAIDS 90-90-90 targets (UNAIDS, n.d.).

Diversion of ART from prescribed to non-prescribed users, through frequent or episodic drug selling or sharing, is an understudied dimension of ART adherence. Previous research from South Florida, United States has identified illicit marketplaces where ART is sold or traded (Buttram & Kurtz, 2016; Surratt, Kurtz, Cicero, O’Grady, & Levi-Minzi, 2013; Tsuyuki, Surratt, Levi-Minzi, O’Grady, & Kurtz, 2015). Reported reasons for ART diversion included substance use (Surratt et al., 2013; Tsuyuki et al., 2015), food and housing insecurity (Surratt et al., 2013), barriers to accessing HIV care (Surratt et al., 2013; Tsuyuki et al., 2015), and perceived efficacy of ART for HIV prevention (Buttram & Kurtz, 2016). Other studies from South Africa have reported ART use in a recreational drug cocktail, though evidence of ART presence in the compound has been mixed (Chinuoya et al., 2014; Grelotti et al., 2014). While drug sharing is widely documented for other prescription medications, we identified no other studies specifically examining ART drug sharing among people living with HIV. Research on the topic of ART diversion appears limited to these few aforementioned studies.

Fishing communities on Lake Victoria in south-central Uganda are HIV transmission hotspots, with adult HIV prevalence estimates of 41% (Chang et al., 2016). Occupational mobility and seasonal migration, substance use, and health service disengagement have been linked to suboptimal HIV care retention and health outcomes in these communities (Lubega et al., 2015). With high HIV burdens and some evidence of drug-resistant mutations in circulating HIV subtypes (Grabowski et al., 2018), these communities are potential settings for ART diversion, despite no evidence of ART drug sharing in this setting or among people living with HIV in Uganda. Documenting patterns of ART diversion in this setting may help identify vulnerabilities to inappropriate ART use and pathways to improving HIV treatment engagement.

Materials and Methods

We conducted a qualitative study in two fishing communities on Lake Victoria in south-central Uganda. From September 2017 to March 2018, 15 male fishermen and 10 female sex workers living with HIV participated in semi-structured in-depth interviews. Over half (n=15) reported current ART use, four had defaulted from HIV care, and six had never initiated ART. Individuals were recruited from Rakai Community Cohort Study (Chang et al., 2016) participants who indicated willingness to be contacted for future studies, and through snowball sampling from these initial respondents. Assisted by clinicians and peer mobilizers, experienced qualitative research assistants approached eligible individuals, explained the study’s purpose, and obtained written informed consent prior to conducting interviews in Luganda. Interviews covered personal and community experiences with ART selling/sharing, reasons for medication diversion, and potential solutions to reduce diversion. All interviews were audio-recorded, transcribed verbatim, then translated into English by the original interviewer. Audiotapes were erased after transcription.

Anonymized data were analyzed using an adapted framework analysis approach (Ritchie & Spencer, 2002). Ugandan and American researchers read interview transcripts, then discussed and developed codes capturing inductive and deductive themes. These codes were used to develop an analytical matrix, which included a summary of each participant’s responses in each category and relevant quotes. Matrices were populated for several transcripts, then refined to incorporate emerging themes. This approach allowed for meaningful comparisons between transcripts, facilitating pattern identification across salient participant characteristics (i.e., occupation, ART adherence status).

This study was approved by the Western Institutional Review Board, Uganda Virus Research Institute Research and Ethics Committee, and the Uganda National Council for Science and Technology.

Results

Participants generally stated that ART was not bought and sold in this setting. However, informal and episodic sharing of ART was widely reported. Many described lending ART to others living with HIV, with a smaller number reporting borrowing ART. ART diversion was broadly characterized as a mechanism for supporting adherence, and consequently, participants self-reporting adherence to ART regimens were more likely to describe personal experiences with ART sharing compared to participants who never initiated ART or defaulted from care. Key interview themes and supporting quotes are summarized in Figure 1.

Figure 1.

Themes, sub-themes, and supporting quotes from interviews with fisherfolk living with HIV

Occupational mobility was an underlying reason for ART sharing across participants. Fishermen described their prolonged voyages on Lake Victoria as an adherence challenge. The unpredictable duration of these trips, which varied according to fish catch, meant fishermen could run out of medication. High HIV prevalence among their fellow fishermen, however, allowed them to borrow ART from fishing companions with spare pills. “I have some extra medicine,” one fisherman recalled telling a colleague. “Let me help you with some as we go fishing.” Another fisherman explained how ‘clinic shopping,’ whereby patients would procure ART from multiple clinics for personal stockpiling, helped some fishermen avert potential treatment lapses.

Mobility similarly facilitated ART sharing among sex workers, who reported regional travel of unpredictable duration in search of clients. With limited transport money to get prescription refills from their home clinics, sex workers reported approaching or being approached by other sex workers to borrow ART until they could fill their prescriptions. Like fishermen, the high HIV prevalence among sex workers facilitated ART sharing within occupational networks.

Participants highlighted other factors facilitating ART sharing (Figure 1). These factors included healthcare inaccessibility, specifically having insufficient money for transportation to retrieve prescriptions refills; fear of negative attitudes from health providers, particularly among patients attempting to reinitiate HIV care after defaulting from treatment; and anticipated stigma associated with being seen attending HIV clinics. These factors were intertwined with occupational mobility and combined to create situations in which people felt they needed to borrow drugs rather than seek ART from health facilities.

ART sharing was not practiced indiscriminately. Participants described specific criteria for identifying ART suitable for sharing, specifically using tablet physical attributes (markings and color) as indicators of similar treatment regimens. As one sex worker explained, “The tablets are marked with numbers on top. They are labeled number seven, eight, nine, ten or thirty. If the numbers are the same, you are both taking the same ART.”

Participants’ feelings towards ART sharing varied from concern and rejection to passive tolerance and understanding. While most participants empathized with the need to borrow ART, some voiced concerns that both sharer and borrower could run out of medication and experience treatment interruptions. Others, like this fisherman, stressed the importance of ART sharing for facilitating adherence in the context of frequent mobility: “Sometimes you may have gone two days without taking this medicine simply because your medicine got finished. This forces you to get some medicine from a friend for the sake of your health.”

Discussion

Qualitative interviews with fishermen and female sex workers in fishing communities on Lake Victoria revealed frequent episodic ART sharing within occupational networks. The desire to remain adherent to medication, despite mobility and other occupational barriers, motivated short-term borrowing and lending. The high prevalence of HIV infection in these settings meant that participants were likely to know colleagues who were also living with HIV, providing them with greater opportunities to share or borrow ART.

Personal experiences with ART diversion were surprisingly concentrated in more adherent users and exclusively involved ART use for HIV treatment among people living with HIV. Previous studies have reported poor medication adherence in diverters (Surratt et al., 2013) as well as targeted diversion to seronegative at-risk groups for informal prophylactic use (Buttram & Kurtz, 2016). ART diversion in our study was additionally limited to episodic exchanges within occupational networks, as opposed to commercial diversion practices (Surratt et al., 2013; Tsuyuki et al., 2015) or exchanges within intimate partnerships (Buttram & Kurtz, 2016; Tsuyuki et al., 2015), which have been described previously. We identified occupation-associated mobility as a unique structural factor leading to ART diversion among fisherfolk.

In this context, ART sharing could have mixed implications. To the extent that patients are sharing identical regimens, ART diversion may be a local solution to widely-recognized barriers to ART adherence, like healthcare inaccessibility, judgement from healthcare providers, and stigma. However, it also carries risks of improper drug dosing, treatment interruption, and poor health outcomes, including virologic failure or drug resistance. Eliminating barriers to care engagement, through interventions like multi-month scripting for adherent patients, described by our participants could help reduce informal ART sharing.

Conclusions

While our qualitative approach and small sample size facilitated in-depth discussion of ART sharing, we were unable to measure the prevalence of sharing in this population. However, even in our small sample, sharing appeared common, and to our knowledge, this is the first report of ART sharing within occupational networks in sub-Saharan Africa. Future studies should explore factors associated with ART diversion in other high-burden settings and identify how these practices are correlated with key health outcomes like virologic failure and drug resistance.