Skip to main content
Open Access Publications from the University of California

Optimizing the Delivery of HIV Pre-Exposure Prophylaxis (PrEP): An Evaluation of Risk Compensation, Disengagement, and the PrEP Cascade

  • Author(s): Hojilla, J. Carlo
  • Advisor(s): Carrico, Adam W
  • Vlahov, David
  • et al.


To inform the optimal implementation of HIV pre-exposure prophylaxis (PrEP), we conducted three studies that evaluated serodisclosure, identified correlates of PrEP disengagement, and characterized the PrEP cascade of care.


Study 1 examined the association of PrEP use with participant non-disclosure and lack of knowledge of partner HIV status in the iPrEx Open Label Extension (OLE). Study 2 examined the effects of stimulant use and binge drinking (i.e., ≥5 drinks on one occasion) on PrEP disengagement in a nested case-cohort in OLE. PrEP disengagement was defined as failure to show for a follow-up visit or tenofovir drug levels <700fmol/punch. Study 3 characterized the PrEP cascade (i.e., seeking services, initiating PrEP, and retention in care) and identified predictors of non-retention among those who started PrEP in a clinic-based cohort of men who have sex with men in San Francisco.


We analyzed data from 1,184 participants in Study 1. PrEP use was not significantly associated with non-disclosure or lack of knowledge of partner status in adjusted analyses (p-values≥.16). However, relationship characteristics were significantly associated with both outcomes. Of the 330 MSM and transgender women included in Study 2, 16% used stimulants and 22% reported binge drinking in the previous three months. Stimulant users had more than 5-fold greater odds of PrEP disengagement (adjusted odds ratio [aOR]=5.24, 95% CI [1.64-16.76]). Binge drinking was not significantly associated with PrEP disengagement after adjusting for stimulant use and baseline confounders (aOR=0.78 [0.33-1.88]). Of the 344 men who sought PrEP services in Study 3, 268 (78%) initiated PrEP. Cost was the most commonly cited reason for not starting. Among patients who initiated PrEP, cumulative incidence of non-retention at 13 months was 38%. Men with an STI diagnosis at enrollment had a 79% greater rate of non-retention (adjusted hazard ratio [aHR]=1.79 [1.06-3.01]).


We found no evidence to suggest that PrEP use is associated with non-disclosure or not knowing partner status. Comprehensive prevention approaches that enhance uptake, reduce stimulant use, and support PrEP adherence and persistence are needed to address the heightened risk of HIV infection of those who use stimulants and those with STI diagnoses.

Main Content
Current View