Beliefs About the End of AIDS, Concerns About PrEP Functionality, and Perceptions of HIV Risk as Drivers of PrEP Use in Urban Sexual Minority Men: The P18 Cohort Study

Abstract

Using cross-sectional data from an ongoing cohort study of young gay, bisexual, and other men who have sex with men (N = 492), we examined the extent to which cognitive factors such as beliefs about the end of AIDS, concerns about the manner in which PrEP works, and perceptions about risk of contracting HIV, are related to PrEP uptake and use. While almost all participants indicted awareness of PrEP, a mere 14% had ever used PrEP. Those with lower concerns about the side effects of PrEP and greater belief that treatment and PrEP would eliminate AIDS were also more likely to have ever used PrEP. Our findings support the ongoing challenges of PrEP uptake as means of curtailing HIV in young sexual minority men, and suggest that beyond the structural factors, consideration must be given to further educating the population as a means of adjusting potentially faulty beliefs, concerns, and perceptions which may influence PrEP utilization.

Introduction

Although HIV is declining among several populations in the United States (U.S.), gay, bisexual and other men who have sex with men (hereto referred to as sexual minority men, SMM), as a group, continue to represent a disproportionate percentage of people living with the disease. In 2014, SMM constituted 83% of new HIV diagnoses among males age 13 years and older, and an estimated 54% of people diagnosed with AIDS in the U.S. [1]. Of those, 39% were Black, 32% were White, and 24% were Latino men. Moreover, young SMM (ages 12–24 years old) account for over 90% of new HIV diagnoses among all men in their age group, and over a quarter of new diagnoses among all SMM [1]. From 2010 to 2014, HIV diagnoses decreased a mere 2% among young Black, less than 1% among young White, and increased 16% among young Latino SMM [1]. Thus, young MSM as a population, and particularly young SMM of color, face staggering HIV-related disparities, and these epidemiologic patterns persist despite advances in HIV biomedical prevention and treatment.

The United States Food and Drug Administration (FDA) approved pre-exposure prophylaxis (PrEP) in the form of daily oral Truvada for HIV prevention in 2012 [2]. Recent patterns indicate an increase in awareness, willingness to use, and uptake of PrEP among SMM [3, 4], but use remains low. A cross-sectional online survey of SMM conducted between 2012 and 2015 found that 68% of respondents had heard of PrEP, 49.6% were willing to use PrEP, yet only 4.9% had used PrEP as of the last point of data collection [3]. Similarly, in New York City, less than half (41%) of SMM had heard of PrEP by 2014, and a mere 3% had used it [5].

Although PrEP has been proven to be an extremely effective method of preventing HIV among SMM [6], uptake has been slow, particularly among those populations that are considered most at risk for HIV [7, 8]. Differences exist across sociodemographic strata in relation to PrEP awareness, willingness to use, and uptake. White, higher income SMM with more years of schooling and who live in large urban centers indicated a higher likelihood of use and uptake than others [3, 4, 9]. Conversely, those who are lower income and/or racial and ethnic minorities have largely shown lower awareness and experienced slower uptake [7, 8, 10]. These disparities are likely driven by many of the same drivers of HIV [11].

Most of the extant literature has primarily focused on structural barriers to PrEP use, such as lack of access to prescribing providers and lack of health insurance [12–15]. In general, we know that as a population, racial and ethnic minority people face myriad health disparities, including lower rates of healthcare access and utilization [16–18], lower satisfaction with care [19], and poorer health outcomes in various domains [20]. Situated in the context of pervasive structural inequality [21], these disparities are partially attributed to medical mistrust stemming from historical and ongoing discrimination and racism [22], and can manifest as general mistrust of the healthcare system and providers, as well as in the form of “conspiracy beliefs” [23–25]. These phenomena have also been established among SMM of color in regard to PrEP. For example, Brooks and colleagues found that urban Black SMM who endorsed “conspiracy beliefs” reported a lower intention to uptake PrEP [26]. Similarly, in recent studies of PrEP uptake among SMM, researchers also found that Black participants reported medical mistrust [27, 28]. Thus, while the literature has documented PrEP and HIV-related mistrust among racial and ethnic minority men in particular, there is less known about what other kinds of beliefs, concerns and perceptions SMM may have surrounding PrEP and PrEP use.

Although somewhat limited, the literature suggests that PrEP-related beliefs, concerns and perceptions play an important role in understanding PrEP uptake among SMM. Further research is needed to explore how medication beliefs around PrEP may affect uptake; this is in light of the extant literature on antiretroviral-related medication beliefs which has shown that ART-related medication beliefs can impact adherence and motivation for adherence [29–33]. Similarly, concerns related to potential short-term side effects and long-term health impacts have been shown to be a barrier to uptake and use [12, 27, 34]. Among a sample of men who have sex with men and transgender women, Golub and colleagues found that barriers and facilitators predicted acceptability and motivation for adherence among those likely to use PrEP [35]. The most highly endorsed barriers were related to concerns over long-term physical impacts and short-term side effects of PrEP, and barriers and facilitators were more salient for Black participants than for White participants [35]. More research is needed to better understand how other concerns, such as those related to the efficacy of PrEP in preventing the acquisition of HIV, and the potential burden of having to take a pill everyday, may impact uptake and may vary by sociodemographic states including race, ethnicity, and income. Finally, these concerns must be considered in relation to an individual’s perceived self-risk of becoming HIV positive, as beliefs, concerns and perceptions likely affect PrEP-related decision-making.

Previous studies have found that perceived risk for becoming HIV positive may affect PrEP use [36, 37]. In a diverse sample of SMM surveyed at commercial sex venues, Gallagher and colleagues found that 78% of the participants did not perceive their HIV risk to be serious enough to use PrEP [38]. Yet, SMM also express concern about HIV. For example, Goedel et al found that 84.6% of SMM sampled in Atlanta were very or somewhat concerned about contracting HIV [8]. This disconnect warrants further exploration on how perceptions around HIV risk may be related to PrEP use and awareness, particularly among young SMM who have may have unique beliefs and concerns [39].

The overarching goal of this work is to further understand the challenges to the uptake of PrEP in emerging adult gay and bisexual men of diverse races, ethnicities, and socioeconomic classes who have come of age in the last decade. The developmental period of emerging adulthood is of particular interest as individuals are experimenting with lifestyle choices, establishing independence from their parents, and making healthcare decisions without the guidance of their parents [40]. Exploring individual-level facets, including beliefs, concerns and perceptions, is critical to understanding how various SMM subpopulations engage in decision-making in regard to PrEP uptake and use. Thus, we approached our work using the basic tenets of theory of reasoned action (TRA) and the health belief model (HBM) [41–43]. While this work is informed by constructs of these key social cognitive theories regarding health behaviors, in this case PrEP use, we are not testing the applicability of these theories. Rather, they provide a framing for understanding the role that PrEP and HIV-related beliefs, concerns, and perceptions of risk, play in PrEP use. In this view, we hypothesized that incorrect PrEP-related beliefs, heightened concerns and misperceptions undermine the uptake of PrEP in a population most at need for this intervention. Specifically, we (1) describe the awareness and use of PrEP in a sample of diverse sample of racial and ethnic minority men; (2) assess the association between use and awareness with demographic states; and (3) examine the extent to which PrEP use is related to beliefs about the end of AIDS, concerns about PrEP functionality, and perceived risk for HIV acquisition.

Methods

Sample

Data for these analyses were drawn from one wave of an ongoing cohort study of young SMM living in New York City and the surrounding metropolitan area. Details of this study have been published previously [39, 44]. In short, all potential participants were contacted and assessed for eligibility between March 2014 and March 2016. Of the 600 participants who had completed a prior version of the study, which began in 2009, 274 consented to participate in the continuation of the study. An additional 391 participants were newly recruited and enrolled, yielding a baseline sample of 665. At each study visit, participants completed Audio Computer-Assisted Self-Interviews (ACASI) to provide information on sociodemographic characteristics, individual psychosocial factors and their health-related beliefs and behaviors. HIV-negative participants were also administered a rapid finger prick HIV antibody test to determine HIV serostatus. The baseline sample was 665 and of these, 629 were determined to be HIV-negative.

The data utilized in these analyses were collected at the first follow-up visit that occurred 6 months post-baseline. Only participants who were HIV-negative at the 6-month visit were included in the final sample. Of the HIV-negative baseline sample of 629, 503 completed the 6-month follow-up visit, achieving an 80% retention rate. At this follow-up, 11 individuals tested HIV-positive and are therefore excluded from these analyses, yielding a final analytic sample of 492. Based on a sub analysis, these 11 individuals did not differ from the analytic sample along any of the key demographic factors. With the exception of participant demographics (i.e. race/ethnicity, foreign-born status, sexual orientation, annual personal income), which were assessed at baseline, all other measures were administered concurrently at the 6-month follow-up visit, which includes the most robust set of measures assessing beliefs, concerns and perceptions regarding PrEP. Data for the 6-month assessment were collected between 2014 and 2016, 2 years after the federal approval of PrEP in the United States. The protocol for this investigation was approved by the IRB of the institutions of the investigators and holds a Federal Certificate of Confidentiality.

Measures

Sociodemographic Characteristics

The race and ethnicity of the participants were ascertained through the use of a single survey item and categorized in these analyses as Hispanic/Latino, Black Non-Hispanic, Asian Non-Hispanic, Multiracial/Other Non-Hispanic, and White Non-Hispanic. Foreign-born status was determined by a dichotomous item asking “Were you born in the US?” Sexual orientation was determined using the Kinsey 7-item continuum ranging from (0) exclusively heterosexual to (6) exclusively homosexual [45]. Consistent with previous studies, this was then examined dichotomously as either exclusively homosexual (6) or not exclusively homosexual (0–5). Participants self-reported personal annual income; an original 12-category response set was recoded into three groups: less than $24, 999 per year; $25,000 to $54,999; $55,000 or more.

PrEP Use and Awareness

PrEP use and awareness were determined through the use of two dichotomously coded questions: “Have you heard of PrEP?” and “Have you ever taken PrEP?” Participants were provided with the following definition of PrEP when answering these questions: “an HIV-negative person taking a daily pill to prevent HIV.”

Beliefs About the End of AIDS

Given our geographic context and the recent statewide initiatives [46], two items were developed to assess beliefs about the end of AIDS. Participants were asked to rate the likelihood of two statements: “Some politicians have called for the elimination of HIV by 2020. How likely do you think this is?” and “How likely do you think HIV treatment for positive people and PrEP for negative people in combination will help to eliminate HIV by 2020?” Responses were self-reported on a 5-item Likert scale ranging from 1 = Very Likely to 5 = Very Unlikely.

Concerns About PrEP Functionality

These analyses employ six unique items, each addressing potential concerns regarding the use of PrEP (for a full list of items, see Table 3). Participants were asked to rate each item in terms of their level of concern. Responses were self-reported on a 5-item Likert scale ranging from 1 = Very Concerned to 5 = Very Unconcerned.

Table 3.

Beliefs, concerns and perceptions by participant characteristics

	Race/ethnicity					Born in the U.S.				Sexual orientation
	Hispanic/Latino	Black Non-Hispanic	Asian Non-Hispanic	Mixed/Other Non-Hispanic	White Non-Hispanic	p	No	Yes	p	Not exclusively homosexual	Exclusively homosexual	p
	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)		Mean (SD)	Mean (SD)		Mean (SD)	Mean (SD)
Beliefsa(pcrit = .025)
Some politicians have called for the elimination of HIV by 2020	2.85 (1.23)	2.71 (1.19)	3.21 (1.13)	2.84 (1.24)	3.35 (0.99)	< 0.001	2.92 (1.21)	2.99 (1.17)	0.63	3.05 (1.14)	2.90 (1.21)	0.16
HIV treatment for positive people and PrEP for negative people in combination will help to eliminate HIV	2.27 (0.99)	2.22 (1.04)	2.74 (1.16)	2.08 (0.99)	2.63 (1.08)	0.001	2.28 (1.03)	2.40 (1.07)	0.35	2.38 (1.02)	2.38 (1.11)	0.97
Concemsb(pcrit = .01)
PrEP can truly protect someone from getting HIV	2.77 (1.23)	2.49 (1.16)	2.49 (1.07)	2.86 (1.38)	3.16 (1.15)	< 0.001	2.88 (1.31)	2.79 (1.20)	0.56	2.71 (1.21)	2.88 (1.21)	0.12
The long-term effects of PrEP on my health	2.38 (1.21)	2.23 (1.21)	2.14 (0.96)	2.68 (1.38)	2.69 (1.16)	0.01	2.33 (1.89)	2.45 (1.21)	0.42	2.42 (1.89)	2.45 (1.22)	0.75
The side effects of PrEP	2.38 (1.16)	2.18 (1.18)	2.14 (1.13)	2.61 (1.15)	2.58 (1.10)	0.03	2.27 (1.15)	2.41 (1.15)	0.35	2.35 (1.12)	2.42 (1.18)	0.49
If someone becomes HIV-positive, certain medicines won’t work because he was taking PrEP	2.72 (1.25)	2.30 (1.19)	2.61 (0.90)	3.03 (1.48)	3.28 (1.15)	< 0.001	2.74 (1.24)	2.79 (1.26)	0.76	2.73 (1.28)	2.84 (1.23)	0.33
PrEP does not provide complete protection against HIV	2.60 (1.22)	2.23 (1.18)	2.11 (1.06)	2.63 (1.34)	2.72 (1.14)	0.01	2.54 (1.25)	2.49 (1.20)	0.80	2.47 (1.18)	2.53 (1.23)	0.63
Having to take a pill every day	3.02 (1.36)	2.52 (1.35)	2.53 (1.08)	3.53 (1.41)	3.45 (1.35)	< 0.001	2.63 (1.26)	3.09 (1.41)	0.01	2.94 (1.36)	3.09 (1.42)	0.24
Perceptionsc
How likely do you think it is that you will become HIV positive?	3.29 (0.87)	3.20 (0.92)	3.33 (0.74)	3.13 (0.78)	3.45 (0.76)	0.11	3.50 (0.76)	3.27 (0.85)	0.02	3.30 (0.77)	3.31 (0.88)	0.96

^a1 = highly likely to 5 = highly unlikely

^b1 = highly concerned to 5 = not at all concerned

^c0 = very likely to 4 = very unlikely

Perceptions of HIV Risk

To assess individual perceptions of HIV risk, we crafted a single item that asked “How likely do you think it is that you will become HIV-positive?” Responses were self-reported on a 5-item Likert scale 0 = Very Likely to 4 = Very Unlikely.

Analytic Plan

Descriptive statistics were computed for each of the key demographic factors (race and ethnicity, sexual orientation, and nation of birth). To establish comparability, the baseline sample was compared to the analytic sample, and no sociodemographic differences were found between the two groups (see Table 1). Sociodemographic factors were also analyzed using bivariable methods in relation to PrEP awareness and use, beliefs, concerns, and perceptions. Next, using independent sample t tests and one-way ANOVAs, we examined the relationships between PrEP use and beliefs, concerns and perceptions. Adjustments for multiple tests were made accordingly using Bonferroni correction. For the ANOVAs, we utilized Tukey comparisons for post-hoc testing. Covariates (beliefs, concerns and perceptions) significantly associated with PrEP use in the bivariable analyses were simultaneously examined with regard to their predictive power using a binary logistic model, controlling for demographic states. To avoid multicollinearity, significance at the 0.05 level was utilized to determine which variables to include in the logistic regression model based on the recommendations of Sarkar and Rana [47]. We then tested a multivariable model to explain ever having used PrEP. All analyses were conducted with SPSS version 23.

Table 1.

Demographic characteristics of P18 cohort study participants and analytic sample

	Total baseline sample n = 665		Analytic sample n = 492
	%	n	%	n
Race/ethnicity
Hispanic/Latino	32.2	214	31.1	153
Black Non-Hispanic	27.4	182	26.2	129
Asian Non-Hispanic	7.7	51	7.9	39
Mixed Non-Hispanic	6.0	40	6.1	30
Other Non-Hispanic	1.7	11	1.6	8
White Non-Hispanic	25.1	167	27.0	133
Born in the U.S.	84.3	560	84.5	415
Sexual orientation
Not exclusively homosexual	49.9	332	49.2	242
Exclusively homosexual	50.1	333	50.8	250
Annual personal income
$24,999 or less	74.1	493	72.8	358
$25,000 to $54,999	17.6	117	20.0	94
$55,000 or more	3.5	23	3.7	18

Results

Sample Characteristics

The analytic sample consists of 492 participants who were HIV-negative and completed the 6-month follow-up assessment, the time point at which we gathered data regarding PrEP-related beliefs, concerns, and perceptions. As previously noted, the baseline and analytic samples were comparable in terms of demographic characteristics of the participants (see Table 1). The analytic sample is largely composed of racial and ethnic minority SMM, with 31.1% (153) identifying as Hispanic regardless of race and 26.2% (129) identifying as Black Non-Hispanic. Due to the small sample sizes for participants identifying as Mixed Non-Hispanic 6.1% (30) and Other Non-Hispanic 1.6% (8), these two groups were combined for the subsequent analysis. The mean age of the analytic sample was 22.47 years old (SD = 0.63). The majority of the participants in the analytic sample were born in the United States, 84.5% (415). Using the Kinsey Scale, slightly more than half of the analytic sample identified as exclusively homosexual 50.8% (250).

PrEP Awareness and Use

Of the 492 participants, 96.1% (473; CI 94.4%, 97.8%) indicated that they were aware of PrEP. PrEP use was much lower than awareness. Only 14.4% (71; 95% CI 11.3%, 17.5%) of the analytic sample reported having ever used PrEP. There were no statistically significant differences in PrEP use based on demographic characteristics. There were differences in PrEP awareness by race/ethnicity (χ²(4) = 11.99, p = 0.02). Of the White Non-Hispanic men, 100% (133) indicated that they were aware of PrEP; however, only 95.4% (146) of Hispanic participants, 93.0% (120) of Black Non-Hispanic and 92.3% (36) of Asian Non-Hispanic men indicated that they were aware of PrEP. We identified no differences in awareness by sexual orientation, personal income, or nation of birth (see Table 2).

Table 2.

PrEP awareness and use by participant characteristics

	PrEP awareness n = 473			PrEP use n = 71
	%	n	p	%	n	p
Race/ethnicity			0.02			0.31
Hispanic/Latino	95.4	146		19.2	28
Black Non-Hispanic	93.0	120		14.2	17
Asian Non-Hispanic	92.3	36		5.6	2
Mixed/Other Non-Hispanic	100	38		15.8	6
White Non-Hispanic	100	133		15.0	18
Born in the U.S.	93.4	401	0.18	13.7	55	0.11
Sexual orientation			0.44			0.73
Not exclusively homosexual	95.5	231		15.6	36
Exclusively homosexual	96.8	242		14.5	35
Annual personal income
$24,999 or less	95.1	368	0.32	14.3	50	0.62
$25,000 to $54,999	97.9	95		15.1	14
$55,000 or more	100	18		22.2	4

Demographic Differences in Beliefs About the End of AIDS, Concerns About PrEP Functionality, and Perceptions of HIV Risk

We examined differences in PrEP-related beliefs, concerns, and perceptions by demographic characteristics (see Table 3). Differences arose by race and ethnicity and nation of birth. No differences were noted by sexual orientation. Beliefs related to the elimination of HIV by 2020 differed by race and ethnicity (F 4, 486 = 6.24, p < 0.001). Black Non-Hispanic participants and Hispanic participants endorsed this belief at a higher level than White Non-Hispanic participants (p < 0.001 and p = 0.002, respectively). When examining beliefs about HIV treatment and PrEP as an effective strategy to eliminate HIV, differences also arose by race and ethnicity (F 4, 487 = 5.07, p = 0.001). White Non-Hispanic SMM endorsed this idea less than Hispanic, Black Non-Hispanic and Mixed/Other participants (all p < 0.05). As for PrEP concerns, differences emerged by race across the following items: PrEP can truly protect someone from getting HIV (F 4, 460 = 5.71, p < 0.001), the long-term effects of PrEP on individual health (F 4, 464 = 3.39, p = 0.01), the side effects of PrEP (F 4, 463 = 2.69, p = 0.03), effectiveness of ART medications if someone seroconverts after having been on PrEP (F 4, 462 = 11.17, p < 0.001), PrEP does not provide complete protection against HIV (F 4, 464 = 4.01, p < 0.01) and having to take a pill every day (F 4, 465 = 10.25, p < 0.001). Overall, White Non-Hispanic participants reported being less concerned about these elements of PrEP functionality than any other racial or ethnic groups. Participants born in the U.S. were less concerned about having to take a pill every day (t (467) = 2.53, p = 0.01). The perception of HIV risk was higher among those born in the U.S. compared to those not born in the U.S. (t (488) =2.20, p = 0.03).

PrEP Use by Beliefs, Concerns, and Perceptions

Differences were examined in the use of PrEP by beliefs regarding the End of AIDS, concerns about PrEP functionality, and perceptions of HIV risk (see Table 4). Those who had ever taken PrEP indicated higher endorsement of both of the end of AIDS belief items (both p < 0.05). Those who had ever taken PrEP expressed less concerns with regard to effectiveness, side effects, and complete protection from HIV acquisition (all p ≤ 0.01). No differences in having ever used PrEP emerged with regard to perceived risk of seroconversion. Prior to running the model examining PrEP use, and as a means for considering the potential effects of multicollinearity, we computed correlations between the beliefs, concerns, and perception covariates which were to be included in the multivariable model (i.e, only those associated with PrEP use). All were highly correlated, ranging from .94 to .96 (p < 0.001).

Table 4.

Use of PrEP by PrEP-related beliefs, concerns and perceptions

	PrEP Ever Used				p
	No		Yes
	Mean (SD)	n	Mean (SD)	n
Beliefsa (pcrit = .025)
Some politicians have called for the elimination of HIV by 2020	3.03 (1.16)	401	2.72 (1.12)	71	0.04
HIV treatment for positive people and PrEP for negative people in combination will help to eliminate HIV	2.43 (1.07)	401	2.10 (1.06)	71	0.02
Concernsb (pcrit = .01)
Whether PrEP can truly protect someone from getting HIV	2.72 (1.15)	395	3.20 (1.47)	70	0.01
The long-term effects of PrEP on my health.	2.39 (1.15)	398	2.69 (1.45)	71	0.10
The side effects of PrEP	2.29 (1.07)	397	2.94 (1.42)	71	< 0.001
If someone becomes HIV-positive, certain medicines won’t work because he was taking PrEP	2.76 (1.23)	396	2.96 (1.39)	71	0.21
PrEP does not provide complete protection against HIV	2.43 (1.15)	398	2.87 (1.42)	71	0.01
Having to take a pill every day	2.97 (1.41)	399	3.30 (1.24)	71	0.07
Perceptionsc
How likely do you think it is that you will become HIV positive?	3.31 (0.85)	401	3.17 (0.86)	71	0.20

^a1 = highly likely to 5 = highly unlikely

^b1 = highly concerned to 5 = not at all concerned

^c0 = very likely to 4 = very unlikely

The binary logistic model was tested using hierarchical entry with two blocks, controlling for race and ethnicity and nation of birth, given the associations of these demographic states with PrEP use. Only the beliefs, concerns, and perception covariates associated with ever having used PrEP (see Table 4) were entered in Block 2. There was significant fit for the model (χ²(9) = 34.81, p < 0.001), with Block 2 entry improving the fit over the demographic block 1 (χ²(4) = 25.52, p < 0.02). Results of the analyses are shown in Table 5 and indicate that less concern with the side effects of PrEP are associated with a higher likelihood of ever having used PrEP (AOR = 1.46, 95% CI 1.15, 1.85, p = 0.002).

Table 5.

Multivariable model explaining PrEP use

	Demographic block model			Full model
	AOR	95% CI	p value	AOR	95% CI	p value
Demographics
Race/ethnicity
Hispanic/Latino	1.50	0.77–2.91	0.23	1.60	0.80–3.20	0.19
Black Non-Hispanic	0.96	0.46–2.01	0.92	1.19	0.55–2.58	0.66
Asian Non-Hispanic	0.26	0.05–1.27	0.10	0.36	0.07–1.81	0.22
Mixed/Other Non-Hispanic	1.15	0.42–3.19	0.79	1.12	0.39–3.24	0.83
White Non-Hispanic	1.00	–	0.20	1.00	–	0.34
Born in the U.S.	2.06	1.03–4.12	0.42	1.96	0.94–4.06	0.07
Beliefs/concerns/perceptions
HIV treatment and PrEP will help to eliminate HIV	–	–	–	0.78	0.60–1.02	0.07
PrEP can truly protect	–	–	–	1.16	0.91–1.49	0.24
Side effects of PrEP	–	–	–	1.46	1.15–1.85	0.002
PrEP incomplete protection	–	–	–	1.11	0.87–1.42	0.41

Discussion

Understanding the nuanced beliefs around how PrEP can impact individuals is critical for increasing awareness and uptake in the populations most vulnerable to HIV. Our findings indicate that there are numerous cognitive factors associated with PrEP uptake in our sample. We examined how beliefs about the end of AIDS, concerns about PrEP functionality, and perceptions of HIV risk were related to ever having used PrEP. Despite the wealth of literature examining PrEP awareness, there is less research on how cognitive factors, particularly individual beliefs, concerns and perceptions regarding PrEP, affect one’s decision-making around PrEP uptake and use. Importantly, we approached this analysis using the basic tenets of the TRA and the HBM [41–43] as a guiding framework for contextualizing these relationships. These theoretical approaches highlight the relationship between people’s beliefs, concerns, perceptions and health behaviors, and thus provide an important lens for contextualizing these cognitive factors.

In the bivariate models, Black and Hispanic participants endorsed elimination-related beliefs at a higher level than White participants. Additionally, Hispanic, Black Non-Hispanic and Mixed/Other participants endorsed PrEP effectiveness and HIV treatment-related concerns at a higher level than White Non-Hispanic SMM participants. These findings differ from previous studies that have suggested that PrEP messaging is not reaching key demographics, i.e., racial and ethnic minority men who have sex with men [7, 14, 48]. Given the high levels of awareness among this sample, yet low levels of use, it appears that PrEP messaging is increasingly reaching racial and ethnic sexual minority men, but barriers to uptake and use remain. Our findings also suggest that young White SMM are still a key population for PrEP messaging, despite their relatively higher uptake and use than racial and ethnic SMM.

As with beliefs and concerns, perceptions of HIV risk also differed by subgroup. The perception of seroconversion risk was higher among those born in the U.S. compared to those not born in the U.S.; no differences emerged when looking at perceived risk of seroconversion and having ever used PrEP. However, previous research has suggested that risk perceptions are higher among SMM of color and older SMM [35], suggesting that further study is needed to understand the relationship between risk perception and PrEP use. Finally, bivariate models also showed higher levels of elimination-related beliefs and less concerns regarding PrEP effectiveness among those that had ever taken PrEP. While PrEP adherence entails its own set of challenges, interventions aiming to enhance PrEP use should leverage these positive beliefs and lower levels of concern to facilitate long-term use among this population.

Our multivariable model indicates that PrEP use can be predicted by an individual’s country of origin, belief that HIV treatment and PrEP will help eliminate HIV, and beliefs about the side effects of PrEP. To our knowledge, this is the first study to examine PrEP use by country of birth as a predictor variable, as 14% of our sample was foreign-born. In particular, this study suggests that individuals born in the U.S. were more likely to report PrEP use. Recent studies have reported country of birth in the participant demographic information but noted that foreign-born participants only accounted for approximately 7% of the study population [8, 49, 50] and as a result were not included as a variable in the multivariate models. A 2015 study found that 65.7% of PrEP prescriptions filled by NYC Medicaid enrolled patients where White [51], although nation of birth was not assessed. This PrEP prescription data, along with our finding that PrEP use is more likely among U.S. born participants, is interesting when examined in the context of the HIV Continuum of Care, as foreign-born individuals report higher rates of both linkage to and retention in HIV care [52]. This suggests that public health programming for foreign-born populations may focus on HIV testing and treatment and not HIV prevention.

Participants in our study were less likely to believe that PrEP in conjunction with HIV treatment could lead to the end of AIDS. Our findings confirm previous studies about patient beliefs about PrEP effectiveness [53, 54]. A study by Philbin et al. found that patients did not understand how to interpret PrEP effectiveness rates and that when effectiveness rates were properly understood, patients were still reluctant to initiate a medication that was not 100% effective at preventing HIV transmission [15, 53]. Similarly, providers also question the real-world effectiveness of PrEP in light of sub-optimal medication adherence [55]. Questions persist among patients [53, 56] and providers [49, 55, 57, 58] about the effectiveness of PrEP to help prevent HIV transmission and thereby end the AIDS epidemic, and pose a considerable barrier to current HIV prevention policies.

Finally, this study also revealed that participants who were less concerned about side effects were more likely to use PrEP. The relationship between concerns over side effects and lower levels of PrEP use are documented in the extant literature [35, 49]. For example, a study of PrEP uptake among men who have sex with men and transgender women in NYC found that 80.5% of participants who were not likely to take PrEP and 59.8% of participants who were likely to take PrEP were concerned about side effects [35]. Similarly, a study of SMM who declined to participate in a PrEP study found that concern about side effects was the second most common reason for non-participation after concerns about the cost of PrEP [59]. Thus, despite widespread efforts to increase PrEP awareness and knowledge, specific messaging around potential side effects appears to be inadequate. Taken together, these findings suggest that beliefs, concerns and perceptions around PrEP and PrEP use are salient factors in how people think about and make decisions around uptake. However, it is also necessary to situate people’s beliefs, concerns and perceptions in the context of the pervasive structural inequality that shapes their access to healthcare, the quality of care they receive, and their interactions and relationships with healthcare providers. Thus, from the perspective of a socioecological framework [60], these individual level factors must be addressed in conjunction with interventions and policies that aim to reduce structural inequality.

With these findings in mind, it is also crucial that the results be contextualized with regard to the age of the participants and the potential generational effects, which may shape findings. The participants are emerging adults in their early 20 s who came of age after the development and uptake of ART in 1996 and some three decades after the initial diagnosis of HIV in the United States. The experiences of this generation referred to by some as the Queer Generation [61] are markedly different from those young sexual men who came of age in the 1980s and 1990s and who have come to be known in the literature as the AIDS Generation and whose lives were shaped by the losses and trauma inflicted by the epidemic, as well as those who came of age in the 1950 s and 1960 s; (the Stonewell Generation) [61–64]. Unlike their older counterparts, the lives of younger generation of sexual minority men are not solely defined by the trauma of AIDS due to the management and treatment of the disease. In fact, data suggest that while concerns about HIV is evident in the lives of these men, there are other equally pressing concerns related to to finances and career shaped in part by the economic crash experienced worldwide in the first decade of the 21st century [39]. This is also evidenced in the differences in sexual behaviors of gay men across the decades (e.g. before 1970, 1970–1979, 1908–1989, 1990 and later) that demonstrate heightened risk for those who came of age in the 1990s, but also a decrease in both HIV and STI infection with each subsequent cohort [65]. While risk taking may be more evident among younger men than their older peers, these young men also benefit rom the advances bestowed by pre exposure prophylaxis (PrEP) and treatment as prevention (TaSP), the latter which ahs been associated with a decrease in community viral load and in turn a decrease in the transmission of HIV [66]. However, while studies have also noted increases in testing for HIV among younger MSM, prevalence in some geographic areas such as Baltimore, which are populated by a large proportion of racial minority MSM, continue to demonstrate increases in infections [67]. In light of our findings, the results of these aforementioned studies suggest an evolution in the management of HIV across generations. However, we also note an inconsistency in the progress made which are reflected our findings including the more limited knowledge of PrEP in communities of color and greater distrust of the potential held by biomedical advances, all of which must be considered alongside the disproportionate burden of HIV in sexual minority men who are Black and Hispanic.

There are limitations that need to be considered in relation to these findings. As with all cross-sectional data, significant correlations may be spurious in nature. Still, the findings align with the social cognitive paradigms in which the constructs are rooted and thus provide theoretical support for the finding. Second, we note that our sample is a convenience sample. However, we used active recruitment methodologies to engage a diverse set of participants in regard to race, ethnicity, nation of birth, and socioeconomic status. While our analytic sample differs in size from our baseline sample, the two sets of participants are equivalent in terms of key demographic factors. The PrEP-related data were collected at the 6-month assessment; data for this assessment were collected between December 2014 and October 2016. All P18 participants receive rapid HIV testing and post-test counseling protocols, which include providing information on PrEP. Therefore, participation in the P18 Study may bias the sample towards higher levels of PrEP knowledge than the general population. This pattern also aligns with the relatively high viral suppression of the HIV-positive individuals in our sample [68]. This was a period of significant coverage in the media with regard to PrEP. Moreover, numerous local efforts of community-based organizations, departments of health, and health service agencies sought to increase awareness and uptake. This is most boldly evidenced in the Ending The Epidemic efforts of New York State [46]. While the usage of PrEP was relatively low, it is likely that this is a highly evolving variable. In future analyses, and upon the completion of our cohort study, we will be able to undertake growth modeling to determine the extent to which uptake patterns changed. Finally, as with all self-reported data, social desirability must always be considered. However, it is highly unlikely that this bias is introduced in the data given the excellent rapport of our research team with the participants, the use of computer assisted administration for the PrEP items [69], and the fact that participants are asked to provide much more sensitive data regarding sexual behavior and drug use. Finally, it is important to note that all of the participants in this sample are young, emerging adult men of a generation that Halkitis has called The Queer Generation [62]. We recognize that cohort differences may exist with regard to uptake given that the most recent data suggest that PrEP users are on average in their mid to late 30 s [70]. This may contribute to the relatively low uptake in our sample. However, given that all of the participants in our sample were drawn from a cohort study in which they were the same age, we cannot test for any cohort effects.

Efforts have been enacted to increase PrEP uptake as means of ameliorating the spread of HIV in the United States and globally. However, and despite these efforts, the uptake of PrEP has remained relatively low as we show in these analyses of sexual minority men. Studies have delved into the social and structural barriers to PrEP as a means of understanding this phenomenon. Yet few studies have focused on how cognitive factors in the form of beliefs, concerns, and perceptions have shaped the roll out of this highly effective biomedical intervention. Like oral contraceptives which were introduced several decades ago, uptake was not immediate [71]. Thus it is not surprising that similar patterns have emerged with regard to PrEP. Like with any new technology, there are few early adopters, while the majority of those who will ultimately use PrEP will grow in number the longer it is on the market. In addition, PrEP use will become increasingly normalized, countering the concerns that many, especially many young SMM, continue to hold. Such growth in uptake will also occur as less ambivalence regarding PrEP is expressed by healthcare providers [72]. Until then, efforts must continue to educate the public, thus alleviating the resistance associated with PrEP uptake in young SMM. However, generic messaging for the entire population will likely prove futile in addressing the needs of those most at risk. Rather, targeted educational efforts must be enacted for those most at risk for acquiring HIV, namely young SMM of color, with a recognition of the complex realities and worries that shape the lives of these emerging adults [39].

Conclusion

Like so many health behaviors, PrEP use is complex. Decisions about one’s management of health are surely directed by structural inequities, but as we have shown, cognitions play a critical role in these decision-making processes. For young SMM who are grappling with numerous competing concerns [39], HIV may not be the primary presenting problem as it was for previous generations and in light of the differential access to care and burdens experiences by SMM of color. Taken together, our findings suggest that efforts must continue to be directed to the education of the SMM emerging adult population in order to ameliorate concerns and address misinformation around PrEP, particularly among communities of color who may not benefit form these advances aw much as White SMM. In the end, this work will only be effective if we truly understand how young SMM make sense of HIV in their lives, particularly in relation to the myriad concurrent challenges a new generation of SMM face and to the more heightened biomedical, psychosocial and structural burdens experienced by men of color. To this end, our approach to HIV prevention cannot be informed by the sensibility regarding the disease that was evident in the pre-ART days; rather, HIV must be nested alongside, but not different from, the numerous other STIs and other health and psychosocial challenges that compromise the well-being of young sexual minority men.