| dc.description.abstract | Introduction: Pre-Exposure Prophylaxis (PrEP) is recommended as an additional HIV prevention measure for female sex workers (FSW) and other high-risk populations. However, although uptake of PrEP by FSW has increased over time, adherence to and retention on PrEP have remained sub-optimal, spelling a threat to realizing the benefits of this novel strategy. This study aimed to determine the rates of PrEP adherence and retention and associated factors among FSW in Kigali, Rwanda, to inform the design of interventions intended to improve PrEP outcomes.
Methods: This study utilized two study designs: a retrospective cohort and a cross-sectional study design. In the retrospective cohort analysis, health facility records of 309 FSW were extracted from five (5) health centers using a data abstraction form. Data were extracted for two similar time-periods in two consecutive years: April-June 2020 and April-June 2021. The time-period April – June 2020 was designated as the PrEP initiation period while the subsequent time-period (April-June 2021) was designated as the retention period. A FSW was considered to have been retained on PrEP if she honored all the scheduled PrEP visits to carry out procedures and obtain a drug refill during the period under consideration. We used Kaplan-Meier survival analysis to estimate retention rates at months 1,3,6,9, and 12 post-PrEP initiation, and Cox regression to determine the factors associated with 12-month retention. In the cross-sectional study, primary data were collected from 486 FSW who returned for PrEP refills between April - June 2021. Data were collected using a structured, interviewer-administered questionnaire. Data were collected on socio-demographic characteristics, sexual behavior, and PrEP adherence. PrEP adherence was defined as optimal for FSW who reported to have taken all their PrEP medication in the last 3 days prior to the interview and as sub-optimal for those that had alternative responses. Factors associated with optimal PrEP adherence were determined using a modified Poisson regression. Data analysis was conducted using STATA (version 14.0).
Results: In the retrospective cohort study, out of the 309 FSW whose records were reviewed, data for 268 (87%) FSW were found to be complete and, therefore, used in subsequent analyses. Fifty percent (50%, n=133) of the participants were within the age bracket of 25-34 years, majority had never been married (51%, n=136), only completed primary level as their highest level of education (73%, n=196), not engaged in any other job other than sex work (69%, n=184), and living alone (88%, n=236). Kaplan-Meier estimates revealed that retention on PrEP was high (81%, n=216) at month-1, 72% (n=193) at month-3, 67% (n=179) at month-6, 59% (n=157) at month-9 and dropped to 53% (n=143) at month-12. The sharpest drop in retention was at month 1, with continuing drops in retention at 3-, 6-, 9-, and 12-months post-PrEP initiation, albeit at lower rates. Multivariable Cox regression revealed that FSW desired to have two or more additional children (adjusted Hazard Ratio [aHR] = 1.654; 95% Confidential Interval [95%CI]: 1.008, 2.713), and using hormonal (aHR = 2.091, 95%CI: 1.181, 3.702) or no method of contraception instead of condoms (aHR = 2.036, 95%CI: 1.006, 4.119) were factors positively associated with PrEP retention; while FSW accessing PrEP from urban clinics (aHR = 0.290; 95%CI: 0.183, 0.458), never using (aHR = 0.329; 95%CI: 0.149, 0.726) or inconsistently using condoms with each sexual encounter (aHR = 0.413; 95%CI: 0.228, 0.749) were factors negatively associated with retention on PrEP. For the cross-sectional study, of 486 FSW interviewed, forty-one percent (n=198) were below 30 years, ninety-one percent (n=440) did not have any other form of employment besides sex work, while sixty-six percent (n=319) were living in households of 3-4 members. Overall, eighty-six percent (86%, n=418) of the FSW interviewed reported ‘optimal’ adherence. Multivariable modified Poisson regression revealed that taking more than 3 alcoholic drinks per week, i.e., 75mls or more (adjusted Prevalence Ratio [aPR] = 0.908, 95%CI: 0.850, 0.970) and living in a household with 3-4 persons (aPR = 0.901; 95%CI: 0.836, 0.970) were factors associated with sub-optimal adherence to PrEP.
Conclusions: Overall, our study found high initial (1-month) retention on PrEP among FSW followed by a continuous decline in retention with only slightly more than half being retained on PrEP at 12 months. Majority of the respondents showed optimal adherence. FSW that desired to have 2 or more additional children and those using hormonal, or no method of contraception were likely to be retained on PrEP while FSW from urban clinics and those that inconsistently use or never use condoms were less likely to be retained on PrEP. Furthermore, FSW that lived in households of 3-4 members and those consuming excessive amounts of alcohol, were less likely to adhere to PrEP. To improve PrEP retention among FSW, PrEP enrolment priority should focus on those that desire to have children soon while strengthening community monitoring in urban PrEP clinics and among inconsistent condom users. Measures to improve adherence among FSW, need to focus on addressing excessive alcohol consumption and stigma that emanate from household crowding. | en_US |
