Women's experiences with oral and vaginal pre-exposure prophylaxis: the VOICE-C qualitative study in Johannesburg, South Africa

Abstract

Background: In VOICE, a multisite HIV pre-exposure prophylaxis (PrEP) trial, plasma drug levels pointed to widespread product nonuse, despite high adherence estimated by self-reports and clinic product counts. Using a socio-ecological framework (SEF), we explored socio-cultural and contextual factors that influenced participants' experience of daily vaginal gel and oral tablet regimens in VOICE.

Methods: In Johannesburg, a qualitative ancillary study was concurrently conducted among randomly selected VOICE participants assigned to in-depth interviews (n = 41), serial ethnographic interviews (n = 21), or focus group discussions (n = 40). Audiotaped interviews were transcribed, translated, and coded thematically for analysis.

Results: Of the 102 participants, the mean age was 27 years, and 96% had a primary sex partner with whom 43% cohabitated. Few women reported lasting nonuse, which they typically attributed to missed visits, lack of product replenishments, and family-related travel or work. Women acknowledged occasionally skipping or mistiming doses because they forgot, were busy, felt lazy or bored, feared or experienced side effects. However, nearly all knew or heard of other study participants who did not use products daily. Three overarching themes emerged from further analyses: ambivalence toward research, preserving a healthy status, and managing social relationships. These themes highlighted the profound and complex meanings associated with participating in a blinded HIV PrEP trial and taking antiretroviral-based products. The unknown efficacy of products, their connection with HIV infection, challenges with daily regimen given social risks, lack of support-from partners and significant others-and the relationship tradeoffs entailed by using the products appear to discourage adequate product use.

Conclusions: Personal acknowledgment of product nonuse was challenging. This qualitative inquiry highlighted key influences at all SEF levels that shaped women's perceptions of trial participation and experiences with investigational products. Whether these impacted women's behaviors and may have contributed to ineffective trial results warrants further investigation.

Conflict of interest statement

Competing Interests: MTN-003C complies with all National Institutes of Health (NIH) data sharing policies. This does not alter the authors’ adherence to all the PLOS ONE policies on sharing data and materials.

Figures

Figure 1. Socio-ecological Model of Factors Affecting Adherence in VOICE, and Levels of Influence.

Figure 2. Female VOICE Participants Study Flow and Interview Mode.

Legend: Approximately 150 women were targeted to be enrolled into VOICE-C ancillary study. A total of 165 VOICE participants were randomly preselected into VOICE-C (15 additional women were selected to accommodate for refusals and losses to follow-up within VOICE). A participant was classified as “enrolled” if she successfully underwent screening within the past 2 weeks and provided written informed consent for VOICE-C study participation. In most cases, informed consent took place on the day of the (first) interview. There were 144 women screened, 106 enrolled, and 102 interviewed. This represents 68% of the original target of ∼150 VOICE participants to be interviewed. The primary reason for the smaller sample was that the FGD group sizes were lower than expected. The target number of approximately 90 women enrolled in FGD was based on an estimate of 8–10 women and 8 FGDs. In practice, only seven FGDs were conducted, and the mean number of women attending each group was 5.7. Note: EI = ethnographic interview. FGD = focus group discussion. IDI = in-depth interview. LTFU = lost to follow-up.