Seasons of risk: anticipated behavior on vacation and interest in episodic antiretroviral pre-exposure prophylaxis (PrEP) among a large national sample of U.S. men who have sex with men (MSM)

Abstract

The current analysis evaluates interest in and acceptability of daily PrEP during short episodes of anticipated increased risk (i.e. Epi-PrEP). In 2013, U.S. members of an Internet-based MSM sexual networking site were invited to complete a survey about HIV prevention practices in the context of vacationing. 7,305 MSM responded to the survey. Of respondents who had vacationed in the past year, 25.6% reported condomless anal sex (CAS) with new male sex partners while vacationing. Most (92.6%) respondents agreed that having to use PrEP every day was a barrier to PrEP use and 74.3% indicated they would take PrEP if they knew it would be helpful for short periods of anticipated increased risk. MSM who reported increased CAS while on vacation in the past year were more likely to indicate that they would take PrEP if it were helpful when used for short periods than respondents who did not (aOR=2.02, 95% CI 1.59–2.56, p <0.001). Studies designed to evaluate uptake, adherence, and protective benefit of short PrEP courses are warranted.

INTRODUCTION

Men who have sex with men (MSM) represent nearly half of all Americans living with HIV (49%) and account for the largest number of new infections.(1, 2) While the incidence of new HIV infections has decreased among other groups in the U.S. (e.g., heterosexuals, injection drug users), annual incidence among MSM continues to increase.(1) Current approaches to HIV prevention, including condom promotion interventions, testing and linkage to care, and treatment as prevention, are not controlling the HIV epidemic among U.S. MSM.

Daily co-formulated tenofovir disoproxil fumarate and emtricitabine (TDF/FTC), FDA approved as HIV pre-exposure prophylaxis (PrEP), has been shown to decrease HIV incidence among MSM (3–5) and represents a promising new HIV prevention tool. Despite the efficacy of PrEP, levels of awareness and uptake among U.S. MSM remain low, (6–14) signifying a missed opportunity for prevention efforts. Moreover, due in part to concerns that pill burden, cost, and potential side effects associated with daily use may be a substantial barrier to PrEP uptake and adherence,(15–17) new studies are currently now exploring non-daily dosing regimens. Given the long intracellular half-life of tenofovir, non-daily dosing regimens of TDF/FTC as HIV PrEP are scientifically rational.(18) One study investigating efficacy of non-daily PrEP, iPERGAY, suggests event-driven dosing of PrEP (i.e. ‘on-demand PrEP’) may also be effective in preventing HIV in this population.(19) The ADAPT (HPTN067) study is comparing acceptability and tolerability of several PrEP dosing regimens including daily dosing, time-driven dosing (TDF/FTC twice weekly with a post-exposure boost) and event-driven PrEP (TDF/FTC before and after a potential exposure to HIV).(20) The current analysis evaluates interest and acceptability in PrEP taken daily over discrete episodes of highest-risk (i.e. episodic PrEP, or “Epi-PrEP”), such as vacations.

HIV risk among MSM is often episodic, with few MSM reporting continuous risk over time.(21) Episodic increases in condomless anal sex (CAS) have been reported among MSM who are on vacation or traveling away from home.(22–24) A better understanding of how adaptation of the daily PrEP regimen might support individuals whose risk is episodic may improve PrEP uptake, adherence, and reduce HIV incidence. To our knowledge, no studies have yet examined if Epi-PrEP is safe, efficacious, and acceptable. Utilizing one of the largest social and sexual networking websites for MSM in the U.S., the current study characterizes episodic risk (e.g. vacations) behaviors and attitudes towards Epi-PrEP among a large national sample of MSM.

METHODS

Participants and Procedures

In August 2013, U.S. members of an Internet-based MSM social and sexual networking site received an online invitation to complete a survey regarding HIV prevention practices. An anonymous message containing a description of the study purpose and a hyperlink to the study website was sent to all active users of the site located in the United States at the time of the study. On the study website, interested respondents who completed the study consent form were directed to the study questionnaire. Respondents were eligible for the study if they were HIV-uninfected (by self-report), were at least 18 years of age, and could read and comprehend English. The study website and survey could be accessed via desktop computer and was offered in English. A shorter version was accessible via mobile devices; however this version did not include many of the questions analyzed in this paper, and as such are not included in this analysis.

Of the 99,694 emails that were opened, 15,405 individuals clicked through to the survey (15.5%), and 9,179 (59.6%) started the survey. Of those who started the survey, a total of 7,305 (79.6%) respondents completed all questions without missing data, representing the analytic sample.

The study was approved by the Institutional Review Board at [redacted].

Measures

Participants completed survey questions on the following topics.

Vacation Practices

A series of yes/no questions were asked about vacation practices. Accordingly, respondents were first asked, “In the past year, did you go away from your home town for vacation or holiday?” To assess sex as a main goal of vacationing, respondents were asked, “In the past year, have you gone on vacation with a main goal of having anal sex with 1 or more new male partners?” Choosing a vacation location with the intent of having riskier sex was assessed by asking respondents, “Have you ever selected a vacation site with the expectation that you would be more likely to engage in condomless anal sex and/or anal sex with more partners than while at home?” Initiation of Epi-PrEP requires identification of a period of increased HIV risk in advance. Therefore, questions addressing sexual behavior intentions during the planning stage of vacationing increase understanding of the appropriateness of HIV prevention strategies requiring premeditation, like Epi-PrEP, for the respondent. Having any CAS with a new partner while on vacation in the past year was assessed by asking respondents, “In the past year, during a typical vacation, with how many new male sex partners did you have condomless anal sex?”

PrEP

The next set of yes/no questions asked about PrEP specifically. Accordingly, respondents were asked if they had ever heard about PrEP, defined as medication taken by mouth before a sexual encounter as protection against HIV, and how likely they would be to use it. They were asked if they currently or had ever taken PrEP, about experienced or perceived barriers to PrEP use, and about their preference for various PrEP dosing strategies: daily, peri-event dosing, or daily for short periods of risk/vacations (i.e. Epi-PrEP).

Sexual Behaviors

Respondents were asked how many male sexual partners they had in the previous 3 months and how many of these partners were anonymous (i.e., met in a public cruising area or bathhouse, or knew for less than 12 hours before having sex). Respondents were also asked about the numbers of partners with whom they had condomless insertive or receptive anal sex. Respondents were asked what substances they had used during sex in the past 3 months and responses were coded as any hard drug use (defined as stimulants [i.e., amphetamines, speed, etc.], crack/cocaine, ecstasy/molly, GHB, and/or heroin verses no hard drug use.

Healthcare Access

Respondents were asked about their health insurance status, which was coded as: no insurance, private insurance, Medicaid, Medicare, Tricare/Champus (healthcare programs of the U.S. Department of Defense Military Health System), VA coverage, and other or unknown insurance status.

Demographics

Respondents were asked to report their age, country of birth (coded as born outside of the United States versus born inside the United States), sexual orientation (bisexual, heterosexual/straight, gay), education status (coded as college graduate or above versus less than college graduate), and annual income (coded as <$6,000, $6,000 to $11,999, $12,000 to $17,999, $18,000 to $23,999, $24,000 to $29,999, $30,000 to $59,999, and $60,000 or above).

Statistical Analysis

Distributions (means and standard deviations for continuous variables and proportions for categorical variables) were calculated by whether or not individuals had been on vacation in the previous year. Separate bivariate and multivariable logistic generalized estimating equation (GEE) models were built to assess factors associated with 1) having been on vacation in the past year and 2) going on vacation with the intention of having sex. All candidate independent variables were included in multivariable models. A GEE approach was chosen to account for potential clustering within states.(25) Separate bivariate and multivariable GEE models were then built to assess the association between 1) sex as a main goal of vacationing and 2) having any CAS with a new partner while on vacation (independent variables) and the following dependent variables: 1) having taken PrEP; 2) having taken PEP; 3) experience that daily dosing of PrEP was a barrier to PrEP use; and 4) being willing to take PrEP if it would be effective for short periods of risk (i.e., vacations). Multivariable models were adjusted for age, sexual orientation, race/ethnicity, education, insurance coverage, income, and birthplace (inside or outside of the United States).

RESULTS

Mean age was 43.2 years old (SD = 12.7), 82% identified as gay, most were white (85.7%), more than two-thirds (68.4%) of the sample were college graduates, and most (86.3%) had health insurance. Over half of the sample (58.7%) reported at least one episode of CAS in the prior 3 months. Table 1 lists frequencies of descriptive statistics of the sample. Our sample consisted of respondents from all 50 U.S. states as well as the District of Columbia. No state contributed more than 7.8% of the responses.

Table 1.

Demographics

Age (years; mean, SD)	43.2 (12.7)
Gender/sexual identity1
Homosexual/gay	4,966 (77.3%)
Bisexual	1,518 (23.6%)
Heterosexual/straight	152 (2.4%)
Transgender M to F	22 (0.3%)
Transgender F to M	5 (0.1%)
Gender queer	72 (1.1%)
Ethnicity
Caucasian/White	5,472 (85.7%)
African American/Black	249 (3.9%)
Hispanic/Latino/Chicano	477 (7.5%)
Asian/Asian American/Pacific Islander	171 (2.7%)
Native American/American Indian	143 (2.2%)
Multiracial	171 (2.7%)
Education
College graduate or above (vs. less than college education)	4,368 (68.4%)
Health insurance status1
No health insurance	685 (13.7%)
Private insurance	3,645 (72.8%)
Medicaid	138 (2.8%)
Medicare	390 (7.8%)
Tricare/Champus	78 (1.6%)
VA coverage	120 (2.4%)
Other/unknown insurance	373 (7.5%)

¹Participants could select more than one option; percentages may not add up to 100%.

Sex as a main goal of vacationing

902 (19.4%) of respondents reported having gone on vacation with a main goal of having anal sex with at least one new partner in the past year. In a multivariable model (Table 2), factors significantly associated with having gone on vacation with a main goal of having anal sex with at least one new partner included: non-white racial identity (aOR 0.67 for Caucasian- identified MSM, 95% CI 0.48 to 0.95, p < 0.05), number of sex partners in the past 3 months (aOR 1.03 per each additional sex partner, 95% CI 1.02 to 1.05, p < 0.001), any anonymous partner in the last 3 months (aOR 2.18, 95% CI 1.61 to 2.95, p < 0.001), and hard drug use during sex (aOR 1.98, 95% CI 1.36 to 2.86, p < 0.001). In a bivariate model (Table 3), factors significantly associated with having gone on vacation with a main goal of having anal sex with at least one new partner included: ever taken PrEP (aOR 3.71, 95% CI 1.98 to 7.14, p < 0.001), ever taken PEP (aOR 1.62, 95% CI 1.03 to 2.56, p < 0.05), and a willingness to take PrEP if the respondent knew it would be helpful for short periods of risk (aOR 1.45, 95% CI 1.19 to 1.78, p < 0.001).

Table 2.

Factors associated with A) sex as a main goal of vacationing and B) having had condomless sex while on vacation with a new male partner in the past year

	A. Sex as a main goal of vacationing				B. CAS while on vacation
	OR (95% CI)	P	AOR (95% CI)	P	OR (95% CI)	P	AOR (95% CI)	P
Age (per year)	1.00 (0.99 to 1.00)	0.42	0.99 (0.98 to 1.00)	0.26	1.00 (1.00 to 1.01)	0.38	1.00 (0.99 to 1.01)	0.54
Homosexual/gay (vs other sexual identity)	1.23 (0.97 to 1.55)	0.09	0.95 (0.69 to 1.32)	0.77	1.43 (1.16 to 1.76)	0.001	1.19 (0.88 to 1.61)	0.26
Caucasian/white race (vs other)	0.73 (0.58 to 0.94)	0.01	0.67 (0.48 to 0.95)	0.02	1.14 (0.91 to 1.44)	0.25	1.03 (0.74 to 1.42)	0.87
Born outside of U.S.	1.20 (0.85 to 1.70)	0.30	0.98 (0.60 to 1.60)	0.93	0.93 (0.67 to 1.28)	0.65	0.69 (0.44 to 1.11)	0.13
College graduate or above	2.56 (2.19 to 2.97)	<0.001	0.94 (0.71 to 1.25)	0.69	1.21 (1.02 to 1.44)	0.03	1.08 (0.85 to 1.38)	0.53
Any insurance coverage	0.71 (0.55 to 0.92)	0.01	0.76 (0.53 to 1.08)	0.13	0.95 (0.76 to 1.19)	0.66	1.05 (0.76 to 1.45)	0.77
Income >$30,00/year	1.14 (0.92 to 1.43)	0.24	1.27 (0.92 to 1.75)	0.15	1.58 (1.31 to 1.90)	<0.001	1.92 (1.46 to 2.52)	<0.001
Number of sex partners in past 3 mo	1.05 (1.04 to 1.06)	<0.001	1.03 (1.02 to 1.05)	<0.001	1.05 (1.04 to 1.06)	<0.001	1.01 (1.00 to 1.03)	0.01
Any anonymous partner in last 3 mo	2.31 (1.82 to 2.93)	<0.001	2.18 (1.61 to 2.95)	<0.001	3.11 (2.51 to 3.84)	<0.001	2.34 (1.80 to 3.05)	<0.001
Any unprotected anal sex in last 3 mo	1.69 (1.37 to 2.09)	<0.001	1.15 (0.86 to 1.53)	0.35	5.60 (4.40 to 7.11)	<0.001	3.68 (2.70 to 5.03)	<0.001
Any serodiscordant unprotected anal sex in last 3 mo	1.81 (1.45 to 2.25)	<0.001	1.07 (0.81 to 1.40)	0.64	3.17 (2.64 to 3.81)	<0.001	1.80 (1.44 to 2.26)	<0.001
Hard drug use during sex	2.56 (1.90 to 3.44)	<0.001	1.98 (1.36 to 2.86)	<0.001	2.42 (1.86 to 3.14)	<0.001	1.63 (1.16 to 2.29)	0.005

Table 3.

Association between having taken PrEP or PEP, or barriers or facilitators to PrEP use and sex being a main goal of vacationing

	Odds Ratio1 (95% CI)	P-value	Adjusted Odds Ratio2 (95% CI)	P-value
Ever taken PrEP	4.14 (2.26 to 7.60)	<0.001	3.71 (1.93 to 7.14)	<0.001
Ever taken PEP	1.75 (1.11 to 2.75)	0.02	1.62 (1.03 to 2.56)	0.04
Having to take PrEP every day has been a barrier to PrEP use	1.82 (0.26 to 12.8)	0.55	1.49 (0.19 to 12.0)	0.71
Would take PrEP if knew it would be helpful for short periods of risk (i.e., vacation)	1.44 (1.18 to 1.76)	<0.001	1.45 (1.19 to 1.78)	<0.001

¹One model per outcome;

²Adjusted for age, sexual identity, race/ethnicity, education, insurance coverage, income, and birthplace outside of the United States

CAS while on vacation

Of respondents who had taken a vacation in the past year, 1180 (25.6%) reported that in the past year, during a typical vacation, they had had CAS with at least one new male sex partner. In a multivariable model (Table 2), factors significantly associated with having had CAS with a new partner while on vacation in the past year included: income greater than $30,000/year (aOR 1.92, 95% CI 1.46 to 2.52, p < 0.001), number of sex partners in the past 3 months (aOR 1.01 per each additional sex partner, 95% CI 1.00 to 1.03, p = 0.01), any anonymous partner in the last 3 months (aOR 2.34, 95% CI 1.80 to 3.05, p < 0.001), any CAS in the last 3 months (aOR 3.68, 95% CI 2.70 to 5.03, p < 0.001), any serodiscordant CAS in the last 3 months (aOR 1.80, 95% CI 1.44 to 2.26, p < 0.001), and hard drug use during sex (aOR 1.63, 95% CI 1.16 to 2.29, p < 0.01). In a bivariate model (Table 4), factors significantly associated with having had CAS with a new partner while on vacation in the past year included: ever taken PrEP (aOR 3.87, 95% CI 2.14 to 7.03, p < 0.001), ever taken PEP (aOR 2.22, 95% CI 1.45 to 3.38, p < 0.0001), and a willingness to take PrEP if knew it would be helpful for short periods of risk (aOR 1.62, 95% CI 1.36 to 1.93, p <0.001). Among those MSM who reported CAS with a new partner while on vacation in the past year, 2.5% had previously used PrEP compared to 0.7% among those who did not report vacation CAS, c²(1, N = 4404) = 21.47, p < 0.001.

Table 4.

Association between having taken PrEP or PEP, or barriers or facilitators to PrEP use and having had condomless sex while on vacation with a new male partner in the past year

	Odds Ratio1 (95% CI)	P-value	Adjusted Odds Ratio2 (95% CI)	P-value
Ever taken PrEP	4.74 (2.50 to 8.96)	<0.001	3.87 (2.14 to 7.03)	<0.001
Ever taken PEP	2.33 (1.53 to 3.54)	0.001	2.22 (1.45 to 3.38)	<0.001
Having to take PrEP every day has been a barrier to PrEP use	2.46 (0.43 to 14.2)	0.31	5.65 (0.49 to 65.8)	0.17
Would take PrEP if knew it would be helpful for short periods of risk (i.e., vacation)	1.64 (1.38 to 1.96)	<0.001	1.62 (1.36 to 1.93)	<0.001

¹One model per outcome;

²Adjusted for age, sexual identity, race/ethnicity, education, insurance coverage, income, and birthplace outside of the United States

Interest in Epi-PrEP

Most (92.6%) respondents agreed that having to use PrEP every day was a barrier to PrEP use; 74.3% indicated they would take PrEP if they knew it would be helpful for short periods of risk. Those who indicated that they had selected a vacation site with the expectation that they would be more likely to engage in CAS and/or anal sex with more partners than while at home had greater odds of preferring less than daily use of PrEP (aOR=1.41; 95% CI: 1.16 to 1.72, p =0.001). Both MSM who reported increased CAS while on vacation and those who reported any CAS with a new partner while on vacation in the last year had greater odds of indicating that they would take PrEP if it were helpful for short periods compared to those who did not report such behavior (aOR=2.02, 95% CI 1.59–2.56, p <0.001, and aOR 1.62, 95% CI 1.36 to 1.93, p <0.001, respectively).

DISCUSSION

In this large sample of U.S. MSM using an Internet-based MSM social and sexual networking site, many reported vacations as an opportunity to engage new sexual partners with nearly 1 in 5 reporting they had selected a vacation site with the expectation of more CAS with more partners than at home. This is consistent with prior studies that found that the opportunity to have more sexual encounters is one of the most significant reasons why MSM are motivated to travel.(26–28) Studies of circuit partygoers and vacationing MSM indicate increased HIV risk behaviors have been associated with travel/vacation.(22–24) Even in instances where risk behaviors do not change significantly while on vacation, MSM may be vacationing in areas with far higher background community viral loads than are found in their home environment. Therefore the epidemiological context of their risk taking may change based on location with Epi-PrEP offering a solution to this problem. In the current study, more than 1 in 4 of those MSM who had vacationed in the past year reported some amount of CAS with a new partner while on vacation. In our sample, greater number of sex partners, anonymous sex, and hard drug use are significantly associated with both sex as a main goal of vacationing and having had CAS while on a vaction in the past year. Additionally, engaging in CAS and serodiscordant CAS in general were both significantly associated with CAS while on vacation. These findings suggest that MSM who are at greatest risk for HIV acquisition while at home are more likely to continue or increase those behaviors while vacationing. Given that interventions desigened to decrease HIV sexual risk among MSM have not met with a great deal of success, PrEP offers a way to reduce the HIV risk associated with these behaviors.(29)

Additionally, non-Caucasian identity was associated with increased odds of having gone on vacation in the past year with a main goal of having sex with at least one new male partner. Differences in social stressors experienced by racial/ethnic minority MSM and Caucasian MSM may help explain this finding.(30) Sexual minority men who are also members of a racial/ethnic minority group are exposed to more stress and may have fewer buffering resources than Caucasian MSM.(31) It is possible that minority MSM may be more likely to prioritize sex as a main goal while traveling in response to fewer perceived stressors than when home (e.g., anonymity).(26, 32)

Respondents for whom Epi-PrEP may be well-suited were more likely than other respondents to indicate interest in PrEP over short periods of risk. While most respondents indicated that having to take PrEP daily was a barrier to use and nearly two-thirds indicated they would take PrEP if they knew it were helpful for short periods of risk, those who reported sex as a main goal of vacationing, those intending to engage in more CAS and/or sex with more partners on vacation than at home, and those who had CAS with at least one new partner while on a vacation in the past year each had greater odds of preferring less than daily use of PrEP. Despite this preference for less than daily use of PrEP, we are unable to quantify how many MSM in this sample engaged in HIV risk taking behaviors only while away and not at home, so it is difficult to determine how many MSM would benefit from daily PrEP or event-driven PrEP compared to Epi-PrEP.

Several limitations must be considered in interpreting these results. This was a convenience sample, recruited through one online social and sexual networking website. Although the sample size was large, the majority of individuals who opened the recruitment email did not consent to the study and complete the survey. It is unknown the degree to which individuals who did not respond to the survey differed from the responders as data was not collected about those individuals. In addition, attrition was significant over the course of the survey, however this was not dissimilar to that seen in other online samples of MSM.(33–35) This sample was generally White and of high socioeconomic status (SES), which may not be generalizable to the populations where HIV is spreading most rapidly in the United States, i.e. young Black and Latino MSM;(1) specifically, vacation practices among this sample of generally higher SES MSM may differ from lower SES MSM. Personal experience taking PrEP was low in this sample, potentially limiting the ability of respondents to draw conclusions about their future pill-taking behaviors. There is the potential for social desirability bias with self-reported measures. However, the survey was completed via an anonymous online survey, which may reduce bias as compared to face-to-face survey interviewing. Previous studies have demonstrated that audio computer-assisted self-interviewing (ACASI) reduces bias compared to face-to-face interviewing,(36) but studies assessing bias in anonymous, online survey administration compared to ACASI or face-to-face interviewing have yet to be done.

Despite these limitations, these data suggest there is value in investigating the feasibility of developing protective strategies for MSM during time-limited periods of episodic risk. While few studies have attempted to characterize attitudes regarding various PrEP dosing regimens (37–42) with some results indicating either no difference (39, 40) in acceptability or a preference for intermittent dosing,(37) this sample shows high general interest in Epi-PrEP with those who could benefit from it most expressing more interest. Indeed, long-term PrEP may be unnecessary for the many HIV-uninfected men who have episodic contextually defined high-risk periods, particularly when away from their home setting. Epi-PrEP for these HIV-uninfected men who take vacations in high-risk settings may be a useful tool in lowering HIV incidence rates. The low rates of PEP and PrEP use in this at risk sample indicate a missed opportunity in HIV prevention. Future work should be done to investigate the acceptability, perceived need, uptake and efficacy of short-term Epi-PrEP among MSM.