Next generation real-time monitoring for PrEP adherence in young Kenyan women

Aim 1. Use the next generation Wisepill device in a cohort of young HIV-uninfected Kenyan women initiating PrEP to a) measure real-time adherence and b) determine the impact of SMS reminders on adherence. We will monitor adherence among all women with Wisepill. We will randomize half to receive daily SMS reminders with the option of switching to SMS reminders only as triggered by missed doses; the other half will receive no SMS reminders. There will be two primary outcomes for this aim. First, we will determine enacted adherence (i.e., pill taking) and persistence (i.e., continuation of PrEP over time) as determined by Wisepill from initiation of PrEP use (when adherence patterns are often established) through 24 months of use, comparing those randomized or not to SMS reminders. Second, we will compare dried blood spot (DBS) tenofovir-diphosphate levels between the two groups at multiple time points. With both outcomes, we will be assessing the synergy of SMS reminders coupled to real-time monitoring.

Aim 2. Determine the technical function, acceptability, cost, and validity of the next generation Wisepill device coupled to SMS reminders. We will measure data transmission reliability, battery life, and other technical features through the two-year cohort study. We will assess acceptability and usability of our strategy via quantitative surveys in all women, as well as qualitative interviews in about 50 women. We will additionally assess costs and ease of use with a time and motion study. Using these cost estimates and adapting an existing mathematical model of HIV transmission, we will estimate the cost-effectiveness of Wisepill plus the SMS reminders per HIV case averted. Adherence validity will be determined by comparing the Wisepill data with DBS tenofovir levels as measured in Aim 1. Discrepancies between the two measures will be explored through qualitative interviews at the conclusion of the study.

Aim 3. Triangulate Wisepill adherence data with weekly assessment of HIV risk behaviors and perceptions, to measure prevention-effective adherence. After the first 6 months of PrEP use (i.e., when initial decisions about PrEP may start to change), we will send all women a weekly SMS survey asking about recent sexual behavior, use of alternate HIV prevention tools, subjective HIV risk, and perception of their PrEP adherence. We will explore the overlap of perceptions and behaviors and PrEP adherence, through quantitative surveys in all women and qualitative interviews in about 50 women. Our goal is to measure dynamic changes in HIV risk and perception of risk or other motivators for HIV prevention that may lead to PrEP discontinuation (appropriately or inappropriately). We will use these SMS surveys to quantify prevention-effective adherence in young women at risk for HIV infection (individually and for the cohort overall). At exit from the study, we will administer a short questionnaire to understand future plans to use PrEP and experiences related to PrEP during the study.

Aim 4. To qualitatively determine why young women in Kenya who are at risk for HIV acquisition may choose not to enroll in a PrEP study or initiate PrEP as it relates to views on PrEP and the impact of technology on study involvement. We will interview up to 60 women who met eligibility criteria at screening but do not return for enrollment. We will also conduct at least 10 and up to 12 focus groups among five key populations across the two sites. These focus groups will include: key community leaders, male parents, female parents, young women (ages 18-24 who are not participating in the larger study), and young men (ages 18-24) from the community, assuming one focus group with each population at each site. We are allowing for one additional focus group per site in case of technical difficulties with the digital recorders or other data collection difficulties. Inquiry will be guided by domains from a prior conceptual model we have developed focused on understanding decision-making related to uptake of HIV treatment for prevention, as well as prior research on PrEP uptake and adherence. We will use an iterative approach over the course of data collection.