Background: HPTN 073 Study assessed the initiation, acceptability, safety, and feasibility PrEP for Black men who have sex with men (BMSM) in three US cities. Upon the PrEP initiation, levels of PrEP use were monitored using self-reported adherence.
Methods: HPTN 073 Study enrolled 226 HIV-uninfected BMSM in three US cities(
Los Angeles, CA, Washington DC, and Chapel Hill, NC, August 2013 and September 2014). All study participants were offered once daily oral FTC/TDF and client centered care coordination, and were followed for 12 months, with scheduled clinical visits every 13 weeks.
Results: Among the total 226 enrolled participants, 178 (79%) participants initiated PrEP. Proportions of self-reported high PrEP adherence (≥90%) ranged between 62% and 71%, while self-reported low PrEP adherence (< 50%) ranged between 13% and 19% during weeks 13 through 52. High adherence is associated with age ≥25, higher education, full time employment, no poly drug use and having a primary partner. Conversely, low adherence is associated with younger age, less education, non-fulltime employment, poly drug use, and no primary partner. Adjusted analysis shows that having a primary partner and no poly drug are highly associated with high adherence, whereas converse is true for low adherence.

 >= 90% Self-Reported Adherence< 50% Self- Reported Adherence
 OR(95% CI)AOR (95% CI)AOR P ValueOR (95% CI)AOR (95% CI)AOR P Value
Age >=252.08 (1.25,3.45)*1.46 (0.84,2.54)0.17820.48 (0.26,0.86)1.49 (0.76,2.95)0.2482
2 year degree or higher vs. HS or less Some college or vocational vs. HS or less2.48 (1.28,4.80)* 1.11 (0.59,2.08)1.59 (0.77,3.28) 1.02 (0.52,1.99)0.2090 0.95130.34 (0.15,0.74) 1.10 (0.55,2.20)0.46 (0.18,1.18) 1.06 (0.46,2.450.1057 0.8922
Employed FT vs. unemployed PT or self-employed vs. unemployed2.66 (1.39,5.11)* 1.04 (0.56,1.95)1.77 (0.85,3.70) 1.01 (0.52,1.99)0.1275 0.96630.34 (0.16,0.76)* 0.90 (0.45,1.81)0.76 (0.28,2.04) 1.09 (0.47,2.52)0.5801 0.8372
Poly drug use0.46 (0.22,0.94)*0.49 (0.24,0.99)0.04603.22 (1.36,7.60)*3.30 (1.37,7.96)0.0079
Primary partner1.71 (1.09,2.69)*1.75 (1.10,2.79)0.01790.44 (0.24,0.82)*0.42 (0.22,0.82)0.0104
* The factors included in the adjusted models are the factors that are <0.05 significance level in the unadjusted model. Analysis was done using generalized estimating equation with exchangeable covariance structure. The behavioral questions were asked for the past three months at each visit. Abbreviation: AOR, adjusted odds ratio.
[Table 1. Correlates of self-reported adherence]

Figure 1: Self-Reported PrEP Adherence
[Figure 1: Self-Reported PrEP Adherence]

Conclusions: Understanding the contextual factors that support and impede adherence (Figure 1) and targeting these in comprehensive intervention packages may maximize PrEP adherence and minimize lower adherence for BMSM. Our data support consideration of the need for addressing these factors as core elements for BMSM.