Secondary Logo

Journal Logo

EPIDEMIOLOGY AND SOCIAL

Preexposure prophylaxis guidelines have low sensitivity for identifying seroconverters in a sample of young Black MSM in Chicago

Lancki, Nicolaa,b; Almirol, Ellena,b; Alon, Leigha,b; McNulty, Moiraa; Schneider, John A.a,b

Author Information
doi: 10.1097/QAD.0000000000001710

Abstract

Introduction

Preexposure prophylaxis (PrEP) has the potential to reduce new HIV transmission events [1]. Implementation of PrEP in high-risk populations is critical to reducing new HIV infections [2] and achieving National HIV/AIDS Strategy goals in the United States. These goals, however, are daunting given the considerable limitations in PrEP awareness and uptake among most at-risk populations such as young black MSM (YBMSM). PrEP awareness in several urban epicenters during PrEP implementation has been found to be very low among YBMSM with 40.5% in Chicago [3], 34% in New York [4], and 23% in Atlanta [5]. Much of this has been attributed to individual-level factors including age, race, lack of engagement in health care, and HIV/MSM stigma around PrEP use [5].

In 2014, the Center for Disease Control and Prevention (CDC) issued clinical practice guidelines formulated by an expert panel and data from several clinical trials to support broader implementation of PrEP[6,7]. In addition to the CDC standard guidelines, guidance has also been provided by the HIV Incidence Risk Index for MSM (HIRI-MSM) screening tool and Gilead's emtricitabine/tenofovir disoproxil fumarate (Truvada for PrEP) package insert. The CDC guidelines for PrEP use by MSM have criteria that include anal sex without condoms, diagnosis with a sexually transmitted infection (STI), and being in an ongoing sexual relationship with an HIV-positive male partner [8]. The HIRI-MSM consists of seven questions to assess HIV risk, validated on independent datasets [9,10] with a score at least 10 having a sensitivity of 84% in predicting incident HIV infection [11] and recommended as an indication to evaluate a client for PrEP [8]. Gilead indications for PrEP consideration include individual risk behaviors, as well as engaging in sexual activity in an area or social network of high-prevalence and structural factors, including ever being incarcerated, exchanging sex for commodities, and drug use [12].

In establishing these guidance documents, cohorts of YBMSM have not been included [6,7,9,10,13], and analysis has not been conducted during PrEP implementation. Accordingly, we use data from uConnect, a longitudinal cohort of YBMSM 16–29 years of age in Chicago, followed at three time points from 2013 to 2016. Our objectives for this analysis are to evaluate the sensitivity, specificity, and area under the receiver operating characteristic (ROC) curve (AUC) of three sets of existing guidelines, to examine individual and network factors associated with HIV seroconversion in this study cohort, and to determine dynamic PrEP awareness and use over this PrEP implementation period.

Methods

Setting

The baseline uConnect sample was generated in 2013 using respondent-driven sampling (RDS). The sampling scheme and sample generation have been described previously [14,15]. The study includes three time points for data collection: a baseline visit between June 2013 and July 2014 (wave 1) and two follow-up visits, one between April 2014 and May 2015 (wave 2) and another between February 2015 and February 2016 (wave 3). Follow-up visits were conducted 9 months apart over an 18-month period. Each study visit consisted of a structured interview and laboratory testing for HIV and Syphilis infection. This analysis uses survey responses and HIV testing from the baseline visit and the two follow-up visits, hereafter, referred to as wave 1, wave 2, and wave 3, respectively.

Eligibility criteria

Persons were eligible to participate in the study if they were self-identified as African-American or black, assigned male at birth, between 16 and 29 years of age, reported oral or anal intercourse with a man within the past 24 months, spent most of their time on the South Side of Chicago, and willing and able to provide informed consent at the time of the study visit. Participants were included in these analyses if they were HIV-uninfected at wave 1. We excluded participants who were HIV seropositive at wave 1. For participants without blood samples HIV seropositive status was verified by the Chicago Department of Public Health (CDPH) surveillance data following signed release of information. For the seroconversion analysis, we included participants who had documented seroconversion based upon change in serostatus or evidence of acute HIV infection at a follow-up study visit. Participants were included in the PrEP awareness and use analysis if they attended both wave 1 and wave 2 study visits and answered items on PrEP awareness and use at both time points.

Measures

Preexposure prophylaxi eligibility

Survey items from the wave 1 study visit assessing sexual behavior, sexual partner characteristics, and other risk factors were used to determine PrEP eligibility according to three sets of guidelines: the CDC's Clinical Practice Guidelines for MSM, the HIRI-MSM, and the Gilead indications for PrEP use. CDC's Clinical Practice Guidelines were followed to create a variable to reflect eligibility for PrEP [16]. This was operationalized using a combination of indicators of risk including report of sex with an HIV-infected partner and/or anal sex without condoms or diagnosis of an STI. The HIRI-MSM index was utilized and operationalized using respondent age, number of male partners, number of HIV-infected partners, occurrence of receptive anal intercourse, and use of poppers or amphetamines [8]. Risk scores were computed for respondents and those with a score at least 10 were classified as being eligible for PrEP. Gilead PrEP recommendations from the package insert provided PrEP indications for MSM at high risk of HIV acquisition [12]. Eligibility was operationalized using a combination of indicators of risk including report of an HIV-infected partner, and/or engaging in sexual activity in a high-prevalence area or social network with at least one of the following: anal sex without condoms, diagnosis of a STI, exchange of sex for commodities, use of illicit drugs or alcohol dependence, incarceration, or partner of unknown HIV status. Refer to Supplementary Table 1, http://links.lww.com/QAD/B200 for further details on variables used to define PrEP eligibility.

Laboratory testing and determination of acute infection

HIV infection status was determined by three assays applied to dry blood spot samples: ARCHITECT HIV Ag/Ab Combo, Multispot HIV-1/HIV-2 Bio-Rad, and Realtime HIV-1 RNA (Abbot). All samples collected included testing for HIV-1 RNA. Acute infection was documented in cases where viral load was detectable and fourth/third-generation tests were negative. We compared HIV serostatus at each study wave using laboratory test results from the wave 1, wave 2, and wave 3 visits; participants who tested negative at the wave 1 visit and had a seropositive result or who were acutely infected at a subsequent visit were classified as seroconverters.

Person-time for HIV incidence

For participants who remained HIV-negative throughout follow-up, we calculated their person-time as the difference between the date of the last study visit they attended (wave 2 or wave 3) and the date of the wave 1 study visit. For HIV seroconverters, we defined the date of seroconversion as halfway between the first visit when HIV was diagnosed and the previous visit. We assessed sociodemographic, behavioral, structural, and network factors associated with incident HIV during the study period, to evaluate what YBMSM client-level factors may improve existing PrEP guidelines.

Preexposure prophylaxis awareness and use

PrEP awareness was self-reported by respondents: ‘PrEP (preexposure prophylaxis) is when a person is given medicine before being exposed to HIV (like when they expect that they will be having unprotected sex with an infected partner) to prevent them from becoming infected. Before today, have you heard of PrEP?’ This question was asked at wave 1 (June 2013–July 2014) and wave 2 (April 2014–May 2015). Those respondents who reported having heard of PrEP were subsequently asked if they had ever used PrEP. Participants who were aware of PrEP at wave 1 and/or wave 2 were categorized as ‘PrEP aware’; those who were unaware at both wave 1 and wave 2 were categorized as ‘Persistently unaware’. Active PrEP linkage services became available in wave 3 of the study at which time all participants were provided detailed information on PrEP and offered linkage to PrEP care regardless of insurance status or ability to pay.

Statistical analysis

Gile's Sequential Sampling estimator was utilized to weight our population-based sample and generate probability estimates [17]. Weights were generated in Stata 14 (StataCorp, College Station, Texas, USA) [18]. To assess the performance of the different criteria in identifying respondents who were at increased risk for seroconversion, sensitivity, specificity, and area under the ROC curve (AUC) for the three PrEP eligibility criteria were calculated. Incidence-density rates and weighted bivariate and multivariate incidence rate ratios (IRRs) were computed for incident HIV infection over the study period, and sociodemographic, behavioral, and network factors associated with HIV seroconversion were determined using Poisson regression models. A sensitivity analysis was performed excluding participants who reported ever having used PrEP. Wave 1 characteristics of respondents were summarized and compared by PrEP awareness over the first two waves of the study using chi-square and t tests for categorical and continuous variables, respectively. Differences in eligibility by PrEP use and PrEP awareness were also compared using chi-square and t tests. Bivariate and multivariate logistic regression models were implemented to examine factors associated with PrEP awareness and use over the study period. All statistical analyses were performed in Stata 14.1 (StataCorp) [18].

Results

Study cohort

Among 618 participants who completed the wave 1 interview of the study, more than one-third (n = 225, 37.0%) of participants were HIV seropositive, and 52 (23.1%) were previously unaware of their infection. Of the remaining 393 participants, 343 had a documented negative HIV test at wave 1. Participants who did not have a documented HIV negative test from cohort labs or CDPH surveillance data were excluded from analyses (n = 50). Cumulative retention over the study period was 87% (n = 300) (Fig. 1); these individuals represented the final study sample for the seroconversion analysis. The final study sample for the PrEP awareness and use analysis consisted of participants who answered questions about PrEP awareness and use at the wave 2 study visit (n = 288).

Fig. 1
Fig. 1:
uConnect population with inclusion and exclusion criteria.

Indications for preexposure prophylaxi

Approximately half of the sample (49%) met criteria for PrEP use according to CDC guidelines, 72% according to HIRI-MSM Risk Index, and 86% met the Gilead indications for PrEP. Figure 2 shows the proportion of YBMSM eligible for PrEP according to the three guidelines by PrEP awareness, PrEP use, and seroconversion, respectively.

Fig. 2
Fig. 2:
Preexposure prophylaxi eligibility by awareness, use, and seroconversion with uConnect Chicago cohort from 2013 to 2016.

Performance of screening guidelines

During the 2-year follow-up, 33 incident infections were observed over 390.4 person-years for an incidence rate of 8.5 per 100 person-years [95% confidence interval (CI), 6.0–11.9 per 100 person-years]. The sensitivity, specificity, positive predictive values (PPVs), and area under the ROC curve (AUC) for the three guidelines in identifying participants at risk for seroconversion are shown in Fig. 3. The CDC guidelines were associated with a sensitivity of 52% (i.e. of those who became infected during the study period, 52% were eligible for PrEP by CDC guidelines), a specificity of 52% and an AUC of 0.51. The HIRI-MSM Score at least 10 was associated with a 85% sensitivity, 30% specificity, and an AUC of 0.54. The Gilead Guidelines were associated with a 94% sensitivity, 15% specificity, and an AUC of 0.57. The weighted sensitivity/specificity estimates were CDC guidelines (30%/59%), HIRI-MSM risk score (93%/22%), and Gilead guidelines (76%/36%).

Fig. 3
Fig. 3:
Sensitivity, specificity, area under the curve and positive predictive value of existing preexposure prophylaxi guidelines (N = 300).*Respondent-driven sampling weight using Giles Sequential Sampling estimator. **Positive predictive value.

Factors associated with HIV seroconversion

Table 1 provides weighted bivariate analyses of sociodemographic, behavioral, structural, and network factors associated with incident HIV during the study period. Individual-level risk factors associated with HIV seroconversion included being unemployed versus employed full time (IRR = 3.8, 95% CI, 1.2–11.5), having no health insurance (IRR = 2.4, 95% CI, 1.0–5.7), having an anorectal STI test performed in the past 2 years (IRR = 3.3, 95% CI, 1.3–8.5) and popper use [IRR = 3.3, 95% CI = 3.3 (1.0, 10.7)]. Having any partners at least 10 years older was a network-level predictor of HIV incidence (IRR = 4.4, 95% CI, 1.6–11.8). In multivariate analyses adjusting for age and sexual identity, having any partners at least 10 years older remained a significant predictor of seroconversion [adjusted Incidence Rate Ratio (aIRR) 4.6, 95% CI, 1.9, 11.0, P = 0.001]. The results did not change significantly after performing a sensitivity analysis excluding participants who reported ever having used PrEP. Popper use (IRR = 4.2, 95% CI, 1.3–14.0, P = 0.022) remained an individual-level factor associated with HIV seroconversion, whereas having no health insurance (IRR = 2.2, 95% CI, 0.9–5.5, P = 0.099) and being unemployed versus employed full time (IRR = 2.7, 95% CI, 1.0–7.2, P = 0.005) became marginally significant. Having an anorectal STI test performed in the past 2 years (IRR = 2.4, 95% CI, 0.8–7.2, P = 0.118) was no longer predictive. Having any partner at least 10 years older remained a network-level risk factor (IRR = 3.9, 95% CI, 1.5–10.3, P = 0.006). The HIRI-MSM Index (IRR = 3.2, 95% CI, 1.0–10.1, P = 0.052) was marginally significantly associated with HIV incidence in the sensitivity analysis.

Table 1
Table 1:
Predictors of incident HIV infection among young black MSM in uConnect cohort, Chicago, 2013–2016 (N = 300).

Alternative guideline thresholds

We explored alternative thresholds for the HIRI-MSM Risk Index to determine if a different cutoff score would be optimal in this population. At a cutoff of eight, sensitivity increased from 85 to 91%, but specificity and AUC decreased (11% and 0.51, respectively). The ROC curve and sensitivity/specificity at different cutoffs are shown in Supplemental Table 2, http://links.lww.com/QAD/B200 and the Supplemental Fig. 1, http://links.lww.com/QAD/B200. Adding having a partner at least 10 years to the CDC criteria resulted in better performance of the CDC guidelines, with AUC = 0.61, sensitivity = 69%, and specificity = 52% (versus current sensitivity of 52%, specificity 52%, and AUC = 0.51).

Preexposure prophylaxi awareness and use

PrEP awareness increased from 32 to 64% from 2013 to 2015; 36% of participants were persistently unaware of PrEP over this time period. Table 2 demonstrates bivariate comparisons of participant characteristics from the wave 1 visit by dynamic PrEP awareness over the first two study waves. In the weighted multivariable regression model, participants were less likely to be aware of PrEP if they identified as bisexual versus gay [adjusted odds ratio (aOR) = 0.33, 95% CI, 0.19–0.58, P < 0.001] or reported being very close to the black community (aOR = 0.56, 95% CI, 0.32, 0.972, P = 0.041). Ever having participated in an HIV prevention program (aOR = 2.85; 95% CI, 1.40–5.78, P = 0.004) and having a higher proportion of social network members who were HIV positive (for every 10% increase: aOR = 1.26, 95% CI, 1.01–1.57, P = 0.041) were positively associated with being PrEP aware.

Table 2
Table 2:
Characteristics of HIV negative uConnect participants at wave 1 by dynamic preexposure prophylaxi awareness, Chicago, 2013–2016 (N = 288).

Self-reported PrEP use increased slightly over the study period, but remained low, from 4% at wave 1 to 6.6% at wave 2 and 10.1% by wave 3. Factors associated with PrEP use included having an STI diagnosed in the past 12 months [odds ratio (OR) = 2.29; 95% CI, 1.10–4.80, P = 0.027], having an HIV-positive partner (OR = 4.23, 95% CI, 1.78–10.07, P = 0.001), and having a partner of unknown HIV status (OR = 2.31; 95% CI,1.15–4.65, P = 0.019).

Discussion

This is the first study to evaluate the performance of existing PrEP guidelines in a cohort of YBMSM and provides important information to consider when determining PrEP eligibility. Accordingly, there are several important findings from this analysis. First, in the context of high HIV seroconversion rates with an incidence of 8.5 cases per 100 person-years, we identified one network predictor of seronconversion: having any sex partners at least 10 years older in this population. Certainly, this specific predictor may be helpful in refining future versions of PrEP guidelines. Most important to updating PrEP recommendations is our finding that nearly half of seroconverters would not have been eligible for PrEP according to CDC guidelines, the guidelines most often used by physicians in prescribing PrEP [19,20]. Although the Gilead guidelines had high sensitivity, the best AUC (although marginally), low specificity, and PPVs were evident across the three different guidelines.

Overall, our findings were consistent with other studies demonstrating that BMSM were less likely to meet indications for PrEP compared with other racial groups, despite their acute risk for HIV acquisition [21,22]. Low eligibility rates and lack of association between behavioral risks and seroconversion may be race specific. CDC guidelines are based on randomized controlled trials, which included predominantly white MSM [23]; these trials had limited numbers of BMSM in the study populations [1,9,10]. Studies have found seroadaptive behaviors that are protective for white MSM do not necessarily appear to be protective in black MSM [24]. Sullivan et al. identified factors involved in HIV seroconversion among black and white MSM from data collected within a cohort study in Atlanta initiated in 2010. Their data demonstrated that PrEP eligibility guidelines based on individual-level behaviors are likely to underestimate the risk for black MSM, with only 65% of seroconverters meeting eligibility guidelines [22]. Their study also identified two network factors to be significant predictors of incident HIV infection: having a partner at least 10 years older and the race of the partner. Similarly, our analyses found that having a partner at least 10 years older to be associated with HIV seroconversion. Notably in our cohort, important attributes unique to YBMSM (i.e. membership in house/ball community, criminal justice involvement) were not major HIV incidence predictors. Findings that many of the individual risk factors used in guidelines are not predictive of YBMSM seroconversion, indicate the need for more nuanced and locally relevant guidelines for PrEP use. Alternative guidelines that take into account network or other population-level factors may be more effective in guiding physicians and in turn preventing HIV in these communities.

Complicating the HIV prevention potential of PrEP was the relatively low level of PrEP awareness among this representative sample of YBMSM and concomitantly, the low level of PrEP use. The level of overall PrEP awareness increased over the study period in this cohort; however, one-third of study participants had never heard of PrEP, and 10.1% had ever used PrEP 3 years following US Food and Drug and Administration approval when providers in Chicago started first providing PrEP to patients [25]. Although some components of the guidelines (i.e. HIV-positive partner, STI diagnosis) were associated with PrEP use, PrEP use was not associated with overall eligibility by the guidelines, with only 10% of clients meeting CDC guidelines ever using PrEP. As demonstrated in previous work [3], PrEP awareness is associated with variables that indicate some level of engagement in HIV programing or in healthcare systems generally. Other studies have indicated that PrEP knowledge is associated with healthcare engagement such as experiencing recent HIV testing [5,26] or having had an STI diagnosed in clinic [27]. Increased engagement of YBMSM in HIV programing or health care will likely increase PrEP awareness and use in this population.

In addition to potential individual-level factors, provider-level barriers to PrEP initiation also likely exist. Karris et al.[28] found significant variation in infectious disease physician attitudes and practices around PrEP, which may in part explain the lack of association between PrEP eligibility and use. Other studies have demonstrated low levels of PrEP prescribing, particularly for young men of color. In one study, healthcare providers were more likely to believe prescribing PrEP would increase risk behaviors in BMSM than their white counterparts, resulting in decreased willingness to prescribe PrEP [29]. In the current analysis, we did not identify reasons for low PrEP use, and it remains unclear to what extent providers are contributing to limited awareness and uptake among YBMSM.

Our study had some important limitations. We used wave 1 factors to define indications for PrEP; however, an individual may have engaged in varied risk behaviors between study visits and behaviors may have changed over the 18-month study period. In addition, some of the questions used to operationalize PrEP indications were for the previous 12 months instead of in the previous 6 months, as most of the CDC guidelines outline. This, however, would likely result in an overestimate of those who were eligible from our study sample and result in reduced specificity. All models were estimated with and without the RDS sampling weights. Our conclusions and main findings were unchanged using the weighted and unweighted models. However, there was a relatively large discrepancy between the weighted and unweighted sensitivity for the CDC guidelines (30 versus 52%). The RDS weights were derived under a number of assumptions relating to RDS that have been described previously [3,15,30]. We believe the RDS assumptions were met, however, both weighted and unweighted estimates are presented for sensitivity, specificity, and PPV. Some participants only attended the wave 2 follow-up visit and did not have as much time to acquire HIV and be observed. If loss to follow-up was related to the predictors or seroconversion, these results may be biased. Finally, we had relatively small numbers of incident outcomes to evaluate predictors of seroconversion and PrEP use in the study cohort.

Having a standardized, effective PrEP eligibility decision aid for HIV vulnerable populations such as YBMSM may assist with PrEP implementation efforts. In 2014, the WHO recommended offering PrEP to populations with an HIV incidence of about three per 100 person-years or higher [31]. In our cohort, the incidence rate was 8.5 cases per 100 person-years. Public health programs and healthcare providers should consider demographics, local epidemiology, and network factors when considering provision of PrEP to patients. Additional efforts to better understand these networks may help further target prevention strategies [32].

In summary, we have demonstrated that several client-level factors listed in existing guidelines to determine PrEP eligibility may exclude a substantial proportion of most-at-risk-persons who could benefit from PrEP in the United States. PrEP with its demonstrated efficacy is susceptible to implementation failure and concerted efforts should be made to increase awareness and use among those at highest risk for HIV. We recommend that all YBMSM in high-prevalence areas should be initially offered PrEP regardless of self-reported risk behaviors, with continued follow-up and shared decision-making [33] around when PrEP cessation should occur.

Acknowledgements

The current work was supported by NIH grants R01DA033875, R01 DA039934, and R01AI120700. We would also like to thank all study participants for the time and effort required to recruit their network members and take part in the interview.

uConnect Study Team: Ishida Robinson, Eve Zurawski, Elc Estrera, Billy Davis, Dexter Voisin, Steve Muth, Michelle Taylor, Ethan Morgan, Aditya Khanna, Britt Skaathun, Rebecca DuVoisin, Iman Little, Keith Green, and Billy Davis.

Conflicts of interest

There are no conflicts of interest.

References

1. Grant RM, Lama JR, Anderson PL, McMahan V, Liu AY, Vargas L, et al. Preexposure chemoprophylaxis for HIV prevention in men who have sex with men. N Engl J Med 2010; 363:2587–2599.
2. Kelley CF, Kahle E, Siegler A, Sanchez T, Del Rio C, Sullivan PS, et al. Applying a PrEP continuum of care for men who have sex with men in Atlanta, Georgia. Clin Infect Dis 2015; 61:1590–1597.
3. Khanna AS, Michaels S, Skaathun B, Morgan E, Green K, Young L, et al. Preexposure prophylaxis awareness and use in a population-based sample of young Black men who have sex with men. JAMA Intern Med 2016; 176:136–138.
4. Gupta S, Lounsbury DW, Patel VV. Low awareness and use of preexposure prophylaxis in a diverse online sample of men who have sex with men in New York City. J Assoc Nurses AIDS Care 2017; 28:27–33.
5. Eaton LA, Driffin DD, Bauermeister J, Smith H, Conway-Washington C. Minimal awareness and stalled uptake of pre-exposure prophylaxis (PrEP) among at risk, HIV-negative, Black men who have sex with men. AIDS Patient Care STDS 2015; 29:423–429.
6. Pinkerton SD, Abramson PR. Effectiveness of condoms in preventing HIV transmission. Soc Sci Med 1997; 44:1303–1312.
7. Menza TW, Hughes JP, Celum CL, Golden MR. Prediction of HIV acquisition among men who have sex with men. Sex Transm Dis 2009; 36:547–555.
8. Centers for Disease Control and Prevention (CDC). Preexposure prophylaxis for the prevention of HIV infection in the United States – 2014 clinical providers’ supplement. 2014.
9. Bartholow BN, Buchbinder S, Celum C, Goli V, Koblin B, Para M, et al. HIV sexual risk behavior over 36 months of follow-up in the world's first HIV vaccine efficacy trial. J Acquir Immune Defic Syndr 2005; 39:90–101.
10. Koblin B, Chesney M, Coates T, Team ES. Effects of a behavioural intervention to reduce acquisition of HIV infection among men who have sex with men: the EXPLORE randomised controlled study. Lancet 2004; 364:41–50.
11. Smith DK, Pals SL, Herbst JH, Shinde S, Carey JW. Development of a clinical screening index predictive of incident HIV infection among men who have sex with men in the United States. J Acquir Immune Defic Syndr 2012; 60:421–427.
12. Gilead Sciences Inc. TRUVADA [package insert]. Foster City, CA: Gilead Sciences; 2014.
13. Flynn NM, Forthal DN, Harro CD, Judson FN, Mayer KH, Para MF, et al. Placebo-controlled phase 3 trial of a recombinant glycoprotein 120 vaccine to prevent HIV-1 infection. J Infect Dis 2005; 191:654–665.
14. Morgan E, Skaathun B, Michaels S, Young L, Khanna A, Friedman SR, et al. Marijuana use as a sex-drug is associated with HIV risk among Black MSM and their network. AIDS Behav 2016; 20:600–607.
15. Schneider J, Cornwell B, Jonas A, Lancki N, Behler R, Skaathun B, et al. Network dynamics of HIV risk and prevention in a population-based cohort of young Black men who have sex with men. Netw Sci 2017; 5:381–409.
16. Department of Health and Human Services. US Public Health ServiceCenter for Disease Control and Prevention, Preexposure prophylaxis for the prevention of HIV infection in the United States: 2014 clinical practice guideline. Atlanta: 2014.
17. Gile K. Improved inference for respondent-driven sampling data with application to HIV prevalence estimation. J Am Stat Assoc 2011; 106:135–146.
18. StataCorp. Stata: release 14. Statistical software. College Station, TX: StataCorp LP; 2015.
19. Calabrese SK, Magnus M, Mayer KH, Krakower DS, Eldahan AI, Gaston Hawkins LA, et al. Putting PrEP into practice: lessons learned from early-adopting U.S. providers’ firsthand experiences providing HIV pre-exposure prophylaxis and associated care. PLoS One 2016; 11:e0157324.
20. Smith DK, Mendoza MC, Stryker JE, Rose CE. PrEP awareness and attitudes in a National Survey of Primary Care Clinicians in the United States, 2009–2015. PLoS One 2016; 11:e0156592.
21. Hoots BE, Finlayson T, Nerlander L, Paz-Bailey G. National HIVBSSG. Willingness to take, use of, and indications for preexposure prophylaxis among men who have sex with men-20 US cities, 2014. Clin Infect Dis 2016; 63:672–677.
22. Sullivan PS, Rosenberg ES, Sanchez TH, Kelley CF, Luisi N, Cooper HL, et al. Explaining racial disparities in HIV incidence in black and white men who have sex with men in Atlanta, GA: a prospective observational cohort study. Ann Epidemiol 2015; 25:445–454.
23. Schneider JA, Bouris A, Smith DK. Race and the public health impact potential of pre-exposure prophylaxis in the United States. J Acquir Immune Defic Syndr 2015; 70:e30–e32.
24. Golden MR, Dombrowski JC, Kerani RP, Stekler JD. Failure of serosorting to protect African American men who have sex with men from HIV infection. Sex Transm Dis 2012; 39:659–664.
25. U.S. Food and Drug Administration. Truvada for PrEP fact sheet: ensuring safe and proper use. 2012.
26. Rucinski KB, Mensah NP, Sepkowitz KA, Cutler BH, Sweeney MM, Myers JE. Knowledge and use of preexposure prophylaxis among an online sample of young men who have sex with men in New York City. AIDS Behav 2013; 17:2180–2184.
27. Dolezal C, Frasca T, Giguere R, Ibitoye M, Cranston RD, Febo I, et al. Awareness of post-exposure prophylaxis (PEP) and pre-exposure prophylaxis (PrEP) is low but interest is high among men engaging in condomless anal sex with men in Boston, Pittsburgh, and San Juan. AIDS Educ Prev 2015; 27:289–297.
28. Karris MY, Beekmann SE, Mehta SR, Anderson CM, Polgreen PM. Are we prepped for preexposure prophylaxis (PrEP)? Provider opinions on the real-world use of PrEP in the United States and Canada. Clin Infect Dis 2014; 58:704–712.
29. Calabrese SK, Earnshaw VA, Underhill K, Hansen NB, Dovidio JF. The impact of patient race on clinical decisions related to prescribing HIV preexposure prophylaxis (PrEP): assumptions about sexual risk compensation and implications for access. AIDS Behav 2014; 18:226–240.
30. Gile KJ, Handcock MS. Respondent-driven sampling: an assessment of current methodology. Sociol Methodol 2010; 40:285–327.
31. World Health Organization. Policy Brief PrEP. 2015.
32. Schneider J, Cornwell B, Jonas A, Lancki N, Behler R, Skaathun B, Laumann E. Network dynamics of HIV risk and prevention in a population-based cohort of young Black men who have sex with men. Network Science 2017; 5:381–409.
33. Peek ME, Lopez FY, Williams HS, Xu LJ, McNulty MC, Acree ME, et al. Development of a conceptual framework for understanding shared decision making among African-American LGBT patients and their clinicians. J Gen Intern Med 2016; 31:677–687.
Keywords:

HIV prevention; national guidelines; preexposure prophylaxis; young black MSM

Supplemental Digital Content

Copyright © 2018 Wolters Kluwer Health, Inc.