Pre-exposure prophylaxis (PrEP) with emtricitabine/tenofovir disoproxil is an effective HIV prevention strategy1–9 supported by high-quality data and multiple randomized control trials, many of which have focused on men who have sex with men (MSM). The PROUD study demonstrated a relative risk reduction in HIV seroconversion of 86% among MSM PrEP users,3 whereas the iPrEx study estimated PrEP efficacy at 92% if patients were adherent to therapy.4 In the open-label extension of the iPrEx trial,8 there were zero HIV infections per 100 person-years in participants taking 4 or more tablets per week. Modeling studies have estimated PrEP efficacy at 96% for 4 doses per week and at 99% for 7 doses.9 With emtricitabine/tenofovir disoproxil approved by the FDA for PrEP since July 2012, PrEP implementation and scale-up are now a central element of the US National HIV/AIDS Strategy.10
The burgeoning clinical and policy interest in PrEP has also brought renewed attention to another pharmacologic approach to HIV prevention, nonoccupational post-exposure prophylaxis (PEP). With guidelines issued by the CDC since 1998,11 PEP involves the short-term (28-day) use of antiretroviral therapy taken within 72 hours after a specific potential HIV exposure event. The evidence for PEP, however, is less robust than that for PrEP. Although several cohort studies have explored HIV seroconversion after PEP use,12–18 reported HIV incidence varies widely, from 1.27 to 7.6 per 100 person-years.16,17 Because PEP is already considered the standard of care to prevent HIV after potential exposure events based on the occupational health literature, it is unlikely that a randomized controlled trial of nonoccupational PEP will ever be conducted.19
Despite the more rigorous data for PrEP, clinicians or patients may opt for PEP if the risk of HIV exposure is infrequent19 or if taking a daily pill is seen as undesirable or impractical due to barriers such as poor health care access, cost, side effects, lack of provider or patient knowledge about PrEP, stigma, perceptions of promiscuity, and/or low perceived HIV risk.20–24 To date, no studies have assessed the likelihood of HIV seroconversion among those who have used PEP but never PrEP in a clinic-based setting where both are available. Understanding the risk of HIV seroconversion among PEP users with ongoing risk of HIV acquisition could inform counseling messages and programmatic approaches to PEP delivery.
San Francisco City Clinic (SFCC), San Francisco's only municipal sexually transmitted diseases (STDs) clinic, offers both PEP and PrEP alongside routine sexual health services and is uniquely positioned to measure outcomes among patients using each of these HIV prevention tools. In this study, we assessed the odds of HIV seroconversion among MSM with exposure to PrEP, PEP, or neither.
SFCC provides free and low-cost sexual health services, including HIV preventative pharmacotherapeutics. Our analysis focused exclusively on MSM (men who ever reported having sex with one or more male partners), who make up approximately 45% of the patient population at SFCC. Throughout the study period, MSM—like all patients seen at SFCC—were evaluated by a clinician who conducted a risk assessment, including questions about sexual practices in the past 3 months such as (1) total number of sexual partners, (2) total number of condomless partners by sexual act (eg, vaginal sex, insertive or receptive anal sex, or oral sex), and (3) timing of last exposure to HIV. HIV-negative MSM were asked if they had ever heard of PEP or PrEP, and whether they were currently taking PrEP.
PEP was offered to MSM patients not on PrEP who reported a potential exposure to HIV infection within the preceding 72 hours. Exposure events were discerned through clinician/patient interview and included condomless anal sexual intercourse and incidents of condom breakage. Patients who elected to initiate PEP received a 2-day starter pack of medications along with a prescription for the remainder of the 28-day course. They also met with a counselor who provided risk reduction and adherence counseling. From September 2012 through May 2016, SFCC prescribed emtricitabine/tenofovir alone for PEP. Emtricitabine/tenofovir plus raltegravir was used from May 2016 through August 2017. After this date, most patients were prescribed emtricitabine/tenofovir plus dolutegravir for PEP. PrEP was offered to any patient on request and to MSM who reported condomless anal sex, sexual intercourse with an HIV-infected partner, or injection drug use, as well as to those who anticipated future changes in behavior associated with HIV risk.
All eligible patients interested in initiating PEP or PrEP received baseline HIV testing with a rapid HIV antibody test (HIV1/2 STAT-PAK Assay; Chembio Diagnostic Systems Inc., Medford, NY); their drawn blood was sent to the San Francisco Department of Public Health's (SFDPH) laboratory for pooled HIV RNA testing (Abbott Diagnostics m2000 RealTime HIV-1 assay; Abbott Laboratories, Abbott Park, IL) to identify cases of acute HIV infection.25 Individuals who tested HIV-positive were offered linkage to HIV care. HIV-uninfected candidates for PEP or PrEP with poor access to health care, including the uninsured and underinsured, were aided with enrollment into patient assistance and copayment assistance programs.26 Finally, patients initiating PEP who reported ongoing risk of HIV infection were encouraged to transition to PrEP after finishing PEP.
We analyzed demographic characteristics, risk behaviors, STD history, and HIV seroconversion data of MSM seen at SFCC between September 1, 2012, and June 30, 2016. The study dates were selected because both PEP and PrEP were offered at the clinic during this time. All MSM aged 18 years or older who tested HIV-uninfected at their first clinic visit during the study interval were eligible for study inclusion.
The SFCC electronic medical record (EMR) and DPH HIV surveillance case registry, Enhanced HIV/AIDS Reporting System (eHARS), were used to identify patients who acquired HIV during follow-up (cases). HIV acquisition was defined as a positive HIV test with confirmation either at SFCC (with documentation in the EMR), or in the San Francisco or California eHARS after matching at the state and local levels between September 1, 2012, and April 27, 2017. We extended the dates of this matching process through April of 2017 to increase the follow-up time for capturing the outcome of interest (HIV seroconversion) in our study population. Because HIV is a reportable disease, outcome data were available for all study patients who remained in California and underwent subsequent HIV testing, regardless of whether these patients were followed up at SFCC.
Matched controls were selected using a 3:1 control to case ratio. For each HIV seroconverter identified, 3 control MSM patients who remained HIV-uninfected during follow-up were randomly selected and matched to the case on the basis of age (plus or minus 3 years), date of baseline visit within the study interval (within 6 months), and length of follow-up time (duration for controls had to be at least as long as that for cases).
All patients included in our analysis were classified into the following 3 mutually exclusive categories based on the primary exposure of interest during at least one SFCC visit within the study dates: (1) ever reported PrEP use or prescribed PrEP at SFCC (regardless of any prior or subsequent PEP use), (2) prescribed ≥1 course of PEP at SFCC without evidence of PrEP use by either self-report or chart review (with no upper limit on the number of PEP courses prescribed), or (3) never reported PrEP use or prescribed PEP or PrEP at SFCC.
The following participant characteristics were considered in our analyses: demographics (race, sexual orientation, and age at baseline), self-reported risk behaviors at the initial visit during the study interval [any anal receptive sex, number of total male sexual partners, and number of condomless anal receptive or insertive sexual partners (all in the past 3 months)], partner HIV status, and drug use in the past 12 months (injection drug use, methamphetamine use, and/or received or gave money or drugs for sex). In addition, the presence of documented non-HIV STDs (syphilis, chlamydia, and gonorrhea)—either before baseline or during the study interval—was assessed. Chlamydia and gonorrhea diagnoses were made based on nucleic amplification tests of urine and/or pharyngeal or rectal swabs. The diagnosis of syphilis was made by a newly positive non-treponemal antibody test [rapid plasma reagin (RPR) or venereal disease research laboratory (VDRL)] with confirmatory treponema pallidum particle agglutination assay (TPPA) or by a sustained fourfold rise in RPR titer from a prior value. Because the SFCC EMR and the SFDPH STD surveillance registry are integrated, we had complete ascertainment of all laboratory-diagnosed STDs for San Francisco residents regardless of whether they were diagnosed at SFCC.
For patients with missing sexual risk behavior data at their baseline visit, the most recent prior SFCC visit within 3 months was used to populate the missing variables where possible. Patients without documented responses to questions about drug use at baseline were considered not to have evidence of this behavior, as provider practice was often to leave this question set blank if the patient denied drug use.
For all descriptive statistics, we compared patient characteristics using contingency table analyses with χ2 or Fisher exact testing for proportions and the Wilcoxon test for medians. For the case–control analyses, the primary dependent variable was HIV seroconversion; the primary independent variable was exposure to PrEP, PEP, or neither. Unadjusted and adjusted conditional matched-pair logistic regression was performed to assess factors associated with HIV seroconversion. The multivariable (adjusted) model was initially constructed using all covariates that were significantly associated with HIV acquisition at the P < 0.05 level in the unadjusted models. A manual backward stepwise process was then used, and covariates were retained in the multivariate models at the P < 0.05 level. All analyses were prespecified at the study's outset and were performed using SAS software version 9.4.
In the event of discrepancies in the primary outcome (HIV acquisition) between the SFCC EMR and eHARS, the study's senior author (S.E.C., who is also the medical director of SFCC) was provided the patient's medical record to query the individual occurrence. This level of data sharing between SFDPH's HIV surveillance and SFCC STD programs is part of routine public health work. The University of California San Francisco's Institutional Review Board gave this study a designation of “not human subjects research.”
We identified 161 confirmed cases of HIV infection among MSM seen between September 1, 2012, and June 30, 2016, at SFCC. Cases were matched with 477 unique controls for a total study population of 638 MSM. Based on available data and matching criteria, most cases were matched at the planned 3:1 ratio; 2 cases had only one matched control, and 2 cases had 2. Characteristics of cases and controls can be found in Appendix 1, Supplemental Digital Content, https://links.lww.com/QAI/B344.
Among the 638 study patients, 137 (21.5%) reported ever taking PrEP, 98 (15.4%) were only ever prescribed PEP, and 403 (63.2%) had no evidence of taking either PrEP or PEP (Table 1). PrEP and PEP takers were younger than those who used neither strategy [median age 27.0 (PrEP) vs. 29.0 (PEP) vs. 32.0 (neither) years; P = 0.01]. Patients not using PrEP or PEP were more likely to identify as bisexual (11.0% vs. 7.1% vs. 16.0%) or as straight (0.7% vs. 0.0% vs. 6.0%; P < 0.01).
PrEP takers were more likely to have other non-HIV STDs (chlamydia, gonorrhea, or syphilis) during the study interval (72.3% vs. 55.1% vs. 42.4%; P < 0.01; Table 1). PrEP takers were also more likely to have been diagnosed with an STD before the study interval, although this difference was not statistically significant (51.1% vs. 38.8% vs. 43.2%; P = 0.14). PrEP users were also the most likely to report having receptive anal sex in the past 3 months (86.5% vs. 80.4% vs. 73.0%; P < 0.01).
PrEP and PEP takers both reported greater numbers of sexual partners in the 3 months preceding their initial study visit compared with their counterparts not using either strategy. This difference included median number of male sexual partners [4.5, interquartile range (IQR) 3–8 vs. 4.0, IQR 2–9 vs. 3.0, IQR 2–6; P < 0.01; Table 1] as well as median number of condomless receptive anal sex partners (1.0, IQR 0–2 vs. 1.0, IQR 0–2 vs. 0.0, IQR 0–1; P < 0.01) and median number of condomless insertive anal sex partners (1.0, IQR 0–2 vs. 1.0, IQR 0–2 vs. 0.0, IQR 0–1; P < 0.01). PEP takers were most likely to report having HIV-positive partners (30.3% of PrEP takers vs. 67.0% of PEP takers vs. 15.5% among the neither group; P < 0.01).
In the unadjusted conditional regression analysis, HIV acquisition was inversely associated with PrEP use and positively associated with number of male sexual partners (total partners as well as condomless receptive and insertive anal sex partners), other STDs before and during the study interval, having an HIV-positive partner, engaging in any anal receptive sex, methamphetamine use, injection drug use, and receiving money or drugs in exchange for sex. There was no difference in HIV seroconversion between PEP use and the use of neither PrEP nor PEP (Table 2).
In the final-adjusted conditional regression model, HIV seroconversion remained inversely associated with PrEP use, with an adjusted odds ratio (OR) of 0.24 [95% confidence interval (CI): 0.13 to 0.46]. The other variables retained in the final multivariate model, all of which were positively associated with HIV acquisition, were as follows: other diagnosed STDs during the study interval (adjusted OR 2.58, 95% CI: 1.63 to 4.10), condomless receptive anal intercourse in the past 3 months (adjusted OR 1.15, 95% CI: 1.05 to 1.27), and methamphetamine use in the past 12 months (adjusted OR 3.73, 95% CI: 2.02 to 6.88; Fig. 1 and Table 2). PEP use had no impact on HIV seroconversion compared with use of neither PrEP nor PEP.
To further evaluate the patients who only took PEP without evidence of PrEP use, we reassigned our study cohort into 4 groups, categorized as (1) PrEP-only, (2) PEP-only, (3) PEP and PrEP, and (4) neither PrEP nor PEP. In this separate analysis, there were 96 (15.1%) PrEP-only patients, 98 (15.4%) PEP-only patients, 41 (6.4%) patients who used both PrEP and PEP, and 403 (65%) patients who used neither. In unadjusted analyses and with the neither PrEP nor PEP group as the referent (see Appendix 2, Supplemental Digital Content, https://links.lww.com/QAI/B344), the odds of HIV seroconversion were lower in the patients using PrEP-only (OR 0.50, 95% CI: 0.25 to 0.93) and those using both PrEP and PEP (OR 0.38, 95% CI: 0.11 to 1.02), but not among those using PEP alone (OR 1.35, 95% CI: 0.82 to 2.17).
This study assessed the comparative effectiveness of using PrEP or PEP compared with neither PrEP nor PEP as strategies to prevent HIV acquisition. In our case–control study of 638 patients, MSM who had ever used PrEP were 76% less likely to seroconvert compared with those who used neither PrEP nor PEP. PEP use alone, on the other hand, regardless of the number of times used, made no difference in HIV seroconversion in the long run compared with the use of neither strategy for HIV prevention.
These findings are particularly compelling given that PrEP takers reported higher-risk sexual behaviors (including anal intercourse and condomless receptive anal sex) and greater numbers of male sexual partners in percentages at least as high if not higher than those seen in their PEP-using counterparts. MSM who report higher-risk sexual behaviors should thus be referred for PrEP services; those using PEP should be offered PrEP on completion of PEP. Episodic or “on-demand” PrEP (taken just before and after specific sexual encounters rather than once daily) has also been shown to be an effective HIV prevention strategy27 and is now being offered as an alternative to daily PrEP to patients at SFCC who face adherence challenges and/or decline to take a once-daily pill.
Like previous studies, we found that HIV acquisition was associated with condomless receptive anal intercourse, methamphetamine use, and acquisition of other STDs.28–30 MSM who report these behaviors or have STD histories should be offered PrEP along with adherence support tools and outreach to support their long-term retention in PrEP care.31–33 Given the strong association between methamphetamine use and HIV acquisition, MSM using these drugs should additionally be referred to drug treatment programs and counseled on harm reduction. PrEP should be recommended for MSM using PEP who have ongoing risk factors for HIV seroconversion; these individuals should be provided support to ensure their successful linkage to PrEP on completion of PEP.
On most metrics of sexual risk collected at SFCC, PrEP users reported risk behaviors equal to or greater than those using PEP, as discussed above. The only exception was that PEP takers reported having HIV-positive partners in greater proportion than those using PrEP. However, because SFCC's PEP intake form specifically prompts clinicians to query patients regarding partner HIV status, and because 20% of these data were missing from PrEP patients, our results may underestimate the proportion of PrEP takers with HIV-positive partners.
Strengths of our analysis included our ability to match our cohort to eHARS, thus ensuring complete ascertainment of all cases of HIV seroconversion occurring among study participants who remained in the state of California and underwent subsequent HIV testing. The integration of the SFCC EMR and SFDPH STD registry systems similarly assured complete ascertainment of reportable STDs diagnosed anywhere in the city of San Francisco.
There were certain limitations inherent in the design of this study. First, because our study was performed at a single site and was limited to MSM, our study results may not be fully generalizable to other geographic or demographic groups, including heterosexual or transgender individuals. Nevertheless, our sample does represent patients seen at an urban population-based STD clinic.
Second, it is notable that 20 (12.4%, see Appendix 1, Supplemental Digital Content, https://links.lww.com/QAI/B344) of the HIV seroconverters had taken PrEP during the study interval. We were unable to assess PrEP or PEP adherence during this study, or to confirm whether patients filled their PrEP or PEP prescriptions as intended. However, we do not suspect that cases of HIV seroconversion among PrEP takers represent treatment failures, as multiple randomized controlled trials have demonstrated the efficacy of PrEP in the setting of high levels of adherence.4,8,9 It is instead more likely that patients had stopped using PrEP or were nonadherent at the time of their HIV exposure. Similarly, we suspect that our PEP seroconverters were not actively being treated with PEP at the time of their HIV acquisition. Our goal was not to examine the efficacy of PrEP and PEP from a biologic perspective but rather to explore the relationship between these 2 alternative approaches and HIV acquisition in a real-world setting.
As a third limitation, HIV seroconversion could have been missed, particularly among study patients who left California (and are therefore not in the local or state eHARS database) and/or never underwent subsequent HIV testing after their SFCC visit. However, cross-matching the study patients with eHARS—which has a validated HIV case completeness rate of 99%34—allowed us to maximize ascertainment of HIV acquisitions and minimize detection bias. In addition, the proportion of persons living with HIV who are unaware of their infection (ie, those who never underwent subsequent HIV testing) is low (6%) in San Francisco,34 further reducing detection bias. Misclassification of exposure to PEP or PrEP could also have occurred if a participant took one of these therapies after their initial visit, but subsequently either (1) did not return to SFCC or (2) did not report their PrEP use at a subsequent clinic visit.
Fourth, the fact that PrEP users were more likely to have other non-HIV STDs (chlamydia, gonorrhea, or syphilis) during the study interval may reflect ascertainment bias, as PrEP users are more likely to be screened routinely for STDs. Nevertheless, acquisition of other STDs remained independently associated with a HIV seroconversion (OR 2.6) in the multivariable model (regardless of use of PrEP, PEP, or neither), underscoring the importance of linking MSM with STDs to PrEP care. Fifth, patients without documented responses to questions around drug use were considered not to have evidence of this behavior based on common provider charting practices at SFCC; we may thus have underestimated the prevalence of drug use in the population. However, this potential misclassification of drug use would have biased our results toward the null, and it is unlikely that there was differential misclassification based on PEP or PrEP use.
Finally, although the CDC now recommends the use of 3-drug regimens in PEP therapy,35 SFCC had previously used a 2-drug PEP regimen as its standard of care through May 2016. Further research would be needed to establish whether there are differences in HIV seroconversion rates among the 2- vs. 3-drug PEP patients at SFCC.
Our results suggest that HIV seroconversion is less likely among MSM who have ever used PrEP than those who used neither PEP nor PrEP. On the other hand, PEP use alone, regardless of the number of PEP courses prescribed, does not affect the odds of HIV seroconversion in the long term compared with use of neither PEP nor PrEP. MSM at STD clinics should be offered PrEP—particularly those who have a history of other STDs, and/or report methamphetamine use or condomless receptive anal intercourse. MSM patients using PEP should be continued on PrEP after completion of their PEP course. At SFCC, all patients initiating PEP are now counseled about the availability of PrEP at the beginning of their treatment course. PEP patients also receive a check-in call 3 weeks into their course, at which time they are encouraged to return to clinic before they complete PEP so that they can be transitioned to PrEP without a gap if they have ongoing risk factors for HIV acquisition. Further research is needed to identify barriers to PrEP use as well as gaps in the PEP to PrEP transition to more appropriately target the most vulnerable and at-risk populations. In the meantime, health care providers and public health outreach campaigns should continue to promote PrEP as a highly effective therapy for HIV prevention and work to optimize PrEP retention while linking PEP users to PrEP care.
1. Guidance on Pre-exposure Oral Prophylaxis for Serodiscordant Couples, Men and Transgender Women Who Have Sex with Men at High Risk for HIV: Recommendations for Use in the Context of Demonstration Projects. World Health Organization; 2012. Available at: https://www.ncbi.nlm.nih.gov/books/NBK132010/
. Accessed June 5, 2019.
2. Spinner CD, Boesecke C, Zink A, et al. HIV pre-exposure prophylaxis (PrEP): a review of current knowledge of oral systemic HIV PrEP in humans. Infection. 2016;44:151–158.
3. McCormack S, Dunn DT, Desai M, et al. Pre-exposure prophylaxis to prevent the acquisition of HIV-1 infection (PROUD): effectiveness results from the pilot phase of a pragmatic open-label randomized trial. Lancet. 2016;387:53–60.
4. Grant RM, Lama JR, Anderson PL, et al. Preexposure chemoprophylaxis for HIV prevention in men who have sex with men. N Engl J Med. 2010;363:2587–2599.
5. Thigpen MC, Kebaabetswe PM, Paxton LA, et al. Antiretroviral preexposure prophylaxis for heterosexual HIV transmission in Botswana. N Engl J Med. 2012;367:423–434.
6. Baeten JM, Donnell D, Ndase P, et al. Antiretroviral prophylaxis for HIV prevention in heterosexual men and women. N Engl J Med. 2012;367:399–410.
7. Fonner VA, Dalglish SL, Kennedy CE, et al. Effectiveness and safety of oral HIV preexposure prophylaxis for all populations. AIDS. 2016;30:1973–1983.
8. Grant RM, Anderson PL, McMahan V, et al. Uptake of pre-exposure prophylaxis, sexual practices, and HIV incidence in men and transgender women who have sex with men: a cohort study. Lancet Infect Dis. 2014;14:820–829.
9. Anderson PL, Glidden DV, Liu A, et al. Emtricitabine-tenofovir exposure and pre-exposure prophylaxis efficacy in men who have sex with men. Sci Transl Med. 2012;4:151ra125.
10. National HIV/AIDS Strategy. What is the National HIV/AIDS Strategy? U.S. Department of Health and Human Services; 2017. Available at: https://www.hiv.gov/federal-response/national-hiv-aids-strategy/overview
11. Centers for Disease Control and Prevention. Management of possible sexual, injecting-drug-use, or other nonoccupational exposure to HIV, including considerations related to antiretroviral therapy. Public Health Service statement. MMWR Recomm Rep. 1998;47:1–14.
12. Jain S, Oldenburg CE, Mimiaga MJ, et al. Subsequent HIV infection among men who have sex with men who used non-occupational post-exposure prophylaxis at a Boston community health center: 1997–2003. AIDS Patient Care STDS. 2015;29:20–25.
13. Schechter M, do Lago RF, Mendelsohn AB, et al. Behavioral impact, acceptability, and HIV incidence among homosexual men with access to postexposure chemoprophylaxis for HIV. J Acquir Immune Defic Syndr. 2004;35:519–525.
14. Thomas R, Galanakis C, Vézina, et al. Adherence to post-exposure prophylaxis (PEP) and incidence of HIV seroconversion in a major North American cohort. PLoS One. 2015;10:e0142534.
15. Heuker J, Sonder GJ, Stolte I, et al. High HIV incidence among MSM prescribed postexposure prophylaxis, 2000–2009: indications for ongoing sexual risk behaviour. AIDS. 2012;26:502–512.
16. Pierce AB, Yohannes K, Guy R, et al. HIV seroconversions among male non-occupational post-exposure prophylaxis service users: a data linkage study. Sex Health. 2011;8:170–183.
17. Whitlock G, McCormack C, Fearnley J, et al. High HIV incidence in men who have sex with men attending for postexposure prophylaxis: a service evaluation. Sex Transm Infect. 2016;93:214–216.
18. Roland ME, Neilands TB, Krone MR, et al. Seroconversion following nonoccupational postexposure prophylaxis against HIV. Clin Infect Dis. 2005;41:1507–1513.
19. Jain S, Krakower DS, Mayer KH. The transition from postexposure prophylaxis to pre-exposure prophylaxis: an emerging opportunity for biobehavioral HIV prevention. Clin Infect Dis. 2015;60:S200–S204.
20. Hannaford A, Lipshie-Williams M, Starrels JL, et al. The use of online posts to identify barriers to and facilitators of HIV pre-exposure prophylaxis (PrEP) among men who have sex with men: a comparison to a systematic review of the peer-reviewed literature. AIDS Behav. 2018;22:1080–1095.
21. Marks SJ, Merchant RC, Clark MA, et al. Potential healthcare insurance and provider barriers to pre-exposure prophylaxis utilization among young men who have sex with men. AIDS Patient Care STDs. 2017;31:470–478.
22. Zhang HL, Rhea SK, Hurt CB, et al. HIV preexposure prophylaxis implementation at local health departments: a statewide assessment of activities and barriers. J Acquir Immune Defic Syndr. 2018;77:72–77.
23. Whitfield THF, John SA, Rendina HJ, et al. Why I quit pre-exposure prophylaxis (PrEP)? A mixed methods study exploring reasons for PrEP discontinuation and potential re-initiation among gay and bisexual men. AIDS Behav. 2018;22:3566–3575.
24. Wood BR, McMahan VM, Naismith K, et al. Knowledge, practices, and barriers to HIV pre-exposure prophylaxis (PrEP) prescribing among Washington state medical providers. Sex Transm Dis. 2018;45:452–458.
25. Patel P, Klausner JD, Bacon OM, et al. Detection of acute HIV infections in high-risk patients in California. J Acquir Immune Defic Syndr. 2006;42:75–79.
26. San Francisco City Clinic. Pre-exposure Prophylaxis (PrEP). San Francisco: Department of Public Health; 2017. Available at: http://www.sfcityclinic.org/services/prep.asp
. Accessed June 5, 2019.
27. Molina JM, Capitant C, Spire B, et al. On-demand preexposure prophylaxis in men at high risk for HIV-1 infection. N Engl J Med. 2015;373:2237–2246.
28. Riddell J, Amico KR, Mayer KH. HIV preexposure prophylaxis: a review. JAMA. 2018;319:1261–1268.
29. Koblin BA, Husnik MJ, Colfax G, et al. Risk factors for HIV infection among men who have sex with men. AIDS. 2006;20:731–739.
30. Rodger AJ, Cambiano V, Bruun T, et al. Sexual activity without condoms and risk of HIV transmission in serodifferent couples when the HIV positive partner is using suppressive antiviral therapy. JAMA. 2016;316:171–181.
31. Dobleck-Lewis S, Liu AY, Feaster DJ, et al. Patterns and correlates of participant retention in a multi-city pre-exposure prophylaxis demonstration project. J Acquir Immune Defic Syndr. 2018;79:62–69.
32. Chan PA, Mena L, Patel R, et al. Retention in care outcomes for HIV pre-exposure prophylaxis implementation programmes among men who have sex with men in three US cities. J Int AIDS Soc. 2016;19:20903.
33. Arnold T, Brinkley-Rubinstein L, Chan PA, et al. Social, structural, behavioral and clinical factors influencing retention in pre-exposure prophylaxis (PrEP) care in Mississippi. PLoS One. 2017;12:e0172354.
34. Population Health Division. HIV Epidemiology Annual Report 2017. San Francisco Department of Public Health; 2018. Available at: https://www.sfdph.org/dph/files/reports/RptsHIVAIDS/AnnualReport2017-Green-20180904-Web.pdf
. Accessed June 5, 2019.
35. Updated Guidelines for Antiretroviral Postexposure Prophylaxis After Sexual, Injection Drug Use, or Other Nonoccupational Exposure to HIV—United States, 2016. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention; 2016. Available at: http://www.cdc.gov/hiv/pdf/programresources/cdc-hiv-npep-guidelines.pdf
. Accessed June 5, 2019.