To The Editor:
Pre-exposure prophylaxis (PrEP), in which HIV-negative persons take antiretroviral medications before engaging in high-risk behaviors, is currently being evaluated in randomized controlled trials in Africa, Asia, South America, and the United States.1,2 These trials use tenofovir as the chemoprophylactic agent because of its long intracellular half-life;3 low prevalence of resistance;4 and promise in animal models, where it has been shown to prevent or delay HIV infection when administered before HIV exposure.5,6
Although promising, tenofovir has not been proven effective in preventing HIV infections. Because of considerable media coverage for clinical trials or because of “word of mouth” in some populations, persons may be aware of PrEP and may adopt PrEP in personal HIV prevention strategies before its efficacy is confirmed in randomized clinical trials. To determine PrEP awareness and whether HIV-negative persons had taken any antiretroviral therapy (ART) to prevent HIV infection, we surveyed attendees of 5 minority gay pride (MGP) events in 4 cities in 2004.
Data were collected from a convenience sample of attendees at MGP events in 2004 in Detroit, Michigan; Oakland, California (2 events); and Baltimore, Maryland as well as in the minority affinity sections of the San Francisco Gay Pride event. Attendees were approached by study staff; offered HIV rapid testing; and asked to complete a brief self-administered survey about prevention practices, HIV testing, and attitudes pertaining to trust of HIV prevention messages. Those who declined testing were still eligible to participate in the survey. Nonmonetary incentives (<$10 in value) were offered to those persons completing the questionnaire. The following 2 questions about PrEP were included: “Some people who are not infected with HIV have taken AIDS medicines BEFORE engaging in HIV risk behaviors because they hoped to reduce their chances of getting HIV. We don't know if this works. Have you ever heard of people who are HIV-negative using AIDS medicines BEFORE engaging in risky behavior?” and “Have you ever used AIDS medicines BEFORE engaging in risky behavior because you thought it would reduce your chances of getting HIV?” Respondents indicating that they had never heard of PrEP but reported prior use were excluded from analysis. Data were analyzed in STATA 8.0 (StataCorp LP, College Station, TX), and odds ratios (ORs) were generated. The project was considered to be part of HIV prevention programs and their evaluation, and, as such, review by the Centers for Disease Control and Prevention (CDC) Institutional Review Board was not required.
Of 2135 persons approached, 1041 responded to the questionnaire at 5 events (49% response rate). A total of 175 persons were tested; of these, 14 (8%) were positive by rapid HIV test. Respondents most identified as African American (43%), followed by Hispanic (19%), white (14%), and Asian/Pacific Islander (13%). Most respondents were male (89%); however, female (7%) and transgendered (3%) individuals also responded. Sixty-two percent identified as gay, with bisexuals (18%) and heterosexuals (18%) also responding.
In total, 248 (25%) respondents had heard of PrEP and 50 (5%) reported ever using PrEP (Table 1). Respondents from San Francisco were most likely to have heard of PrEP (29.0%), followed by respondents in Baltimore (19.8%), Detroit (19.0%), and Oakland (17.9%). Respondents from San Francisco reported the highest use of PrEP (7.0%), followed by respondents in Baltimore (6.8%), Detroit (2.0%), and Oakland (0%-2.9%).
African-American respondents were more likely to report PrEP use than persons of other race/ethnicity (OR = 2.1, 95% confidence interval [CI]: 1.0 to 4.3). There was no difference in the use or awareness of PrEP by sexual identity, sexual practice (reported oral, anal, or vaginal sex in the 5 years preceding the survey), or drug use (history of crystal methamphetamine or injection drug use in the 12 months preceding the survey). Persons reporting an HIV test in the 12 months preceding the survey were more likely to report use (OR = 3.2, 95% CI: 1.5 to 7.9]) or awareness of PrEP (OR = 1.8, 95% CI: 1.3 to 2.5), however, compared with those who had tested in the prior 12 months. Persons who were aware of PrEP were more likely to report meeting sex partners at sex clubs (OR = 1.5, 95% CI: 1.1 to 2.1), whereas PrEP users were less likely to report meeting sex partners at bars (OR = 0.4, 95% CI: 0.2 to 0.8), parties (OR = 0.5, 95% CI: 0.2 to 0.9), and through web sites (OR = 0.4, 95% CI: 0.2 to 0.9). PrEP users were less likely to trust HIV prevention information from health care providers (OR = 0.4, 95% CI: 0.2 to 1.0).
This is the first report describing PrEP awareness and use among persons at risk for HIV and suggests that despite unproven efficacy, PrEP is already known and used by some populations. Awareness of tenofovir clinical trials did not seem to have skewed results, because knowledge of PrEP documented in other cities was nearly as high as in San Francisco, where trials are underway; moreover, in Baltimore, where no such trials are occurring, PrEP use was similar to that reported in San Francisco.
Those reporting PrEP awareness and use had higher HIV testing rates in the year preceding the survey, suggesting an understanding of HIV prevention. PrEP users were more likely to distrust HIV prevention messages, however. This suggests that 25 years into the HIV epidemic, some persons with high-risk behaviors may be ignoring proven prevention messages in favor of unproven strategies. Our findings may represent the natural progression of prevention burnout or fatigue, in which persons at risk for HIV adapt available prevention modalities and messages to fit their needs.7
The implications of persons self-prescribing PrEP are serious. First, despite promising data in animal models, PrEP is not proven to reduce the risk of HIV infection in humans. If persons using PrEP abandon other proven risk-reduction methods (eg, condom use, reduction of partners), their HIV risk may actually increase. Second, the utility of PrEP is dependent on specific pharmacokinetic properties (eg, toxicity, intracellular half-life) that may or may not be present in the antiretroviral medications obtained opportunistically (ie, from friends, on the street). Third, breakthrough infections that occur while individuals are taking ART, especially if inadequate tissue drug concentrations are present, may be selected for resistance characteristics.
Respondents comprised a convenience sample of MGP attendees and may not be representative of all persons at increased HIV risk. Still, our estimates of proportions of persons aware of and using PrEP indicate that, at least among some members of at-risk populations, awareness and use of PrEP are surprisingly high. In addition, we did not ask respondents who had used PrEP whether they engaged in higher risk behaviors because they believed they were protected, the specific ART regimens used, or the frequency with which they employed this strategy.
PrEP is being used despite lack of proven effectiveness. Future studies should clarify the source and type of antiretrovirals used for PrEP and the extent to which PrEP is being used as a surrogate for proven prevention methods. Clinicians should be aware that PrEP use occurs and, where possible and feasible, should incorporate counseling on the risks of ART and the unproven nature of PrEP by HIV-negative persons. Persons living with HIV who are prescribed antiretroviral drugs may also be counseled not to provide their medications to uninfected persons for this purpose.
Scott E. Kellerman, MD, MPH*
Angela B. Hutchinson, PhD, MPH†
Elin B. Begley, MPH†
Brian C. Boyett, MS†
Hollie A. Clark, MPH†
Patrick Sullivan, DVM, PhD†
*The New York City Department of Health and Mental Hygiene, Bureau of HIV/AIDS Prevention and Control, New York, NY
†Centers for Disease Control and Prevention, National Center for HIV, STD, and TB Prevention
Division of HIV/AIDS Prevention-Surveillance and Epidemiology
Behavioral and Clinical Surveillance Branch, Atlanta, GA