Background: The PrEPX demonstration study used existing health services to emulate the ''real world'' provision of HIV PrEP prior to government subsidisation in Australia in April 2018. We describe PrEPX participants who discontinued receiving study drug prior to the study ending, examine factors associated with discontinuation and describe these participants'' ongoing HIV and sexually transmitted infection (STI) risk.
Methods: Study drug dispensing data from pharmacy logs, HIV/STI testing and behavioural survey data from four study clinics participating in the Australian Collaboration for Coordinated Enhanced Sentinel Surveillance (ACCESS) system were extracted for the duration of PrEPX (26Jul 2016-30Apr2018). PrEPX participants were provided 90 pills per study drug dispensing event, and study discontinuation was classified as participants who were dispensed their last study drug before October 2017, seven months prior to study completion and missing at least two scheduled study prescriptions. Cox proportional hazards estimated covariates associated with discontinued study participation. HIV/STI diagnosis rates >100 days after last study drug dispensed are described and differences in HIV/STI positivity between study and post-study periods were assessed using Chi squared analyses.
Results: This analysis included 2451 participants; 515 (21.0%) discontinued study participation with a median time from last study drug dispensed to study end of 367days (IQR:272-499). PrEP naiveté (aHR1.67 95%CI: 1.11-2.48), age < 30 years (aHR1.65, 95%CI: 1.09-2.50), and reporting consistent condom use with casual partners (aHR1.52 95%CI: 1.01-2.30) at enrolment were associated with discontinuing study participation. Of these 515 participants, 130 (25.2%) accessed post-study testing at ACCESS sites; four participants (3.3%) were diagnosed with HIV during the observation period. Mean time between last study drug dispensed and HIV diagnosis was 338 days (range 140-466 days). STI positivity was similar between pre and post-study periods for chlamydia (8.5%, 8.3%, p=0.9), gonorrhoea (10.4%, 9.9%, p=0.9), and syphilis (0.5%, 1.3%, p=0.5).
Conclusions: Approximately 20% of participants in this analysis discontinued study participation. Four HIV diagnoses and similar STI positivity between study and post-study periods suggest ongoing HIV and STI acquisition risk and unmet HIV prevention need. Greater understanding of barriers to PrEP retention and factors affecting accurate risk perception are needed to maximise the HIV prevention benefits of PrEP.