Are sexual minority stressors associated with young men who have sex with men’s (YMSM) level of engagement in PrEP?

Abstract

Sexual minority stressors (community homophobia, sexuality-related discrimination, and internalized homonegativity) are negatively associated with accessing HIV prevention services among men who have sex with men (MSM). Few studies have tested minority stressors’ associations with PrEP engagement among high-HIV risk young MSM (YMSM). Therefore, we assessed the associations between PrEP-indicated YMSM’s progression along the PrEP continuum and their experiences of minority stress. N = 229 YMSM completed a web-survey on PrEP-related behaviors and minority stress. Adjusted for covariates, we developed two partial-proportional odds models examining the associations between PrEP continuum progression and minority stressors, as a composite, and community homophobia, sexuality-related discrimination, and internalized homonegativity, respectively. Our multivariable model demonstrated minority stress levels to be negatively associated with PrEP continuum location (AOR = 0.76, 95% CI: 0.58–0.99). Broken down, discrimination was positively associated with reporting being at an advanced location along the continuum (AOR = 1.39, 95% CI: 1.06–1.82). Internalized homonegativity was negatively associated with continuum location between PrEP-aware participants with no intention to initiate and participants who intended to initiate PrEP (AOR = 0.45, 95% CI: 0.27–0.77) and between those who intended to initiate and those who had ever used PrEP (AOR = 0.39, 95% CI: 0.22–0.69). Our findings suggest that minority stress, especially internalized homonegativity, remains a barrier to PrEP among PrEP-indicated YMSM. Sexuality-related discrimination was associated with PrEP continuum progression, suggesting potentially well-developed, adaptive coping skills (e.g., ability to locate sexuality-affirming providers). Coupled with stigma reduction efforts, HIV prevention services aiming to promote PrEP should incorporate internalized homonegativity screenings and referrals into sexuality-affirming resources for PrEP-indicated YMSM.

Introduction

Scaled-up and sustained adoption of pre-exposure prophylaxis (PrEP) for HIV prevention among young men who have sex with men (YMSM) remains a public health priority given their disproportionate burden of HIV incidence in the United States.^1-3 Recent estimates suggest high, stable HIV incidence rates, yet youth under 25 years of age report low PrEP uptake and reflect only 11% of current PrEP users.^1,4 This is despite recent studies providing robust evidence of high PrEP awareness and willingness among YMSM.^5-8 Rates of PrEP uptake are continuing to rise among MSM as a whole with the assistance of many health systems, particularly in urban areas, that have been implementing initiatives to minimize financial barriers (i.e., medication costs).⁹ However, YMSM, in particular, experience ongoing structural and individual-level barriers to PrEP including transportation, PrEP stigma, perceived effectiveness, and concerns with long-term effectiveness.^5,10-16 These barriers are especially pronounced among racial/ethnic minority YMSM.^17,18

Intervention efforts to improve PrEP uptake must address the social and interpersonal contexts that shape YMSM’s engagement with services across the PrEP continuum. The PrEP continuum highlights steps of engagement spanning awareness, uptake, and adherence.¹⁹ Few researchers have given attention to measuring PrEP uptake as a spectrum, generally opting for binary constructions that distinguish those who take and do not take PrEP as an HIV prevention tool.^20-22 Cognitive processes around health decision-making (e.g., intention to use PrEP) reflect active involvement in one’s sexual health.²³ Assessing the social and cultural contexts that account for more nuanced conceptualizations of PrEP use, such as continuum-defined outcomes, will provide HIV prevention experts with a more in-depth understanding of barriers to care and PrEP engagement.

Minority stress theory may be useful for understanding the social and psychosocial challenges that may inhibit PrEP engagement among YMSM.²⁴ Minority stress theory posits that, over their life course, sexual minorities are vulnerable to chronic, cumulative stress due to structural (e.g., societal homophobia) and interpersonal stigma (e.g., discrimination) uniquely attributed to their sexual identities,^25-28 which in turn may elicit increases in internalized homonegativity.^29,30 In prior studies, sexuality-related stigma was associated with decreasing primary healthcare utilization, accessing mental health care, and seeking HIV testing services among MSM.^31-35 Given the robust relationships between minority stressors and HIV treatment and prevention,^36-40 it is likely that these stressors may also affect PrEP engagement. However, few studies have examined whether minority stressors are linked to YMSM’s progression across the PrEP continuum (e.g., awareness, initiation, adherence).¹⁹

The goal of this study was to explore the association between minority stressors and PrEP continuum engagement in a sample of YMSM living in Philadelphia, Pennsylvania; Baltimore, Maryland; and Washington, D. C. – known as the Mid-Atlantic metropolitan corridor, which is one of the most heavily affected HIV regions in the United States.⁴ Using recommended guidelines regarding PrEP indication to characterize our sample, the first aim was to describe PrEP-indicated YMSM’s locations along the PrEP continuum.⁴¹ Guided by the minority stress theory, the second aim was to then examine whether YMSM’s composite and individual minority stress experiences would be associated with progression along the PrEP continuum. A composite, or additive, minority stress factor permits a greater understanding of the extent to which sexual minority stressors compound and potentially exacerbate negative outcomes. On the contrary, assessments of individual factors provide HIV prevention experts with evidence on identifying the most salient and strongest correlates of outcomes to inform the planning stages of intervention development. We hypothesized that minority stress (composite) scores would be inversely associated with YMSM’s PrEP continuum location. Lastly, we hypothesized that community homophobia, sexuality-related discrimination, and internalized homonegativity would be independently associated with less progression along the PrEP continuum.

Materials and methods

Procedures

Data comes from a cross-sectional study in which YMSM completed a behavioral health web-survey assessing PrEP care engagement. The study recruited participants from Grindr and Facebook (October 2018-February 2019) and were eligible if they identified as cis-gender men between the ages of 18–25 years old (inclusive), self-reported as HIV-negative/unaware, had sex with a man in the past 6months, and reported a residential zip code along the Philadelphia (PA), Baltimore (MD), or Washington, D.C. Mid-Atlantic corridor. Eligible participants provided informed consent online and then completed a 20–30-minute web-survey. Participants received a $10 Amazon incentive. An institutional firewalled server protected the data; the [masked] Institutional Review Board approved our study procedures.

Based on 1,287 attempted entries, we employed best practices to prevent duplicate and falsified entries by monitoring duplicate email and IP addresses, and crosschecking IP addresses with zip codes and residential addresses provided by participants.⁴² After removing ineligible (n = 112, 8.7%), duplicate (n = 93), and falsified entries (e.g., those who did not complete the screener/consent form, or consented, but did not complete the survey; n = 792), our study’s final sample consisted of N = 290 YMSM. We removed ten participants (3.4%) because they provided incomplete responses to the variables of interest and exhibited no statistically significant (p<.05) differences by sociodemographic characteristics. To comprehensively address individuals with high behavioral HIV risk, we focused the analysis on PrEP-indicated participants (n = 229; 81.8%) based on CDC guidelines and prior PrEP studies, including participants who reported at least one of the following criteria: any condomless anal intercourse in the past 30 days, any prior STI diagnosis, and any drug use before or during sexual intercourse.^41,43 PrEP-indicated YMSM were generally older (t_df=278 = −3.57, p < 0.001), more likely to be in a relationship (χ²_df=1 = 4.95, p = 0.026), and more likely to be full-time employed (χ²_df=2 = 7.27, p = 0.026) compared to participants not indicated for PrEP.

Measures

Outcome: progression along the PrEP continuum

Participants responded whether they had ever heard of PrEP (0 = No, 1 = Yes). Those who reported yes were asked to report their PrEP status (0 = Never used PrEP, 1 = Has taken PrEP, but is no longer on PrEP, 2 = Currently on PrEP). Participants who had never used PrEP were asked if they had any intention to seek PrEP in the next 3 months (0 = No, 1 = Yes). These three items were recoded into an ordinal variable classifying their progression along the PrEP continuum (0 = PrEP-Unaware, 1 = PrEP-Aware/No intention, 2 = PrEP-Aware/Intended to initiate in 3 months; 3 = Ever used PrEP). Participants who had discontinued PrEP were included with those currently on PrEP based on the assumption that they had successfully navigated/overcome any barriers to initiating PrEP care in the past.

Perceived community homophobia

Participants reported their perceptions of people’s attitudes toward MSM in their area across 7 items (e.g., “Most people in my area think poorly of men who have sex with men”)⁴⁴ using a 4-point scale (1 = Strongly disagree, 4 = Strongly agree). We reverse coded positively worded items and then developed a mean sexual prejudice homophobia score with higher mean scores reflecting higher perceived community homophobia (α = 0.91).

Sexuality-related discrimination

We adopted 5 items from the Daily Heterosexist Experiences Questionnaire⁴⁵ (e.g., “Being treated unfairly in stores or restaurants because you are LGBT”), which participants completed using a 6-point scale (1 = Did not happen; 6 = It happened and it bothered me extremely). This scale yielded high internal consistency (α = 0.89). We developed a mean score, where a score of zero suggested no discrimination and higher scores suggested greater sexuality-related discrimination with high negative impact.

Internalized homonegativity

We used 9 items (α = 0.89) from the Internalized Homonegativity Scale to assess participants’ inward projections of negative attitudes regarding their same-sex attractions (e.g., “I often feel it best to avoid personal or social involvement with other men who have sex with men”).⁴⁶ Positively worded items were reverse coded, and then we calculated a mean score where higher scores implied greater internalized homonegativity.

Sociodemographic characteristics

Participants self-reported their age, sexual orientation (0 = Gay, queer, same-gender loving, or homosexual, 1 = Other sexual identity), relationship status (0 = Single, 1 = In a relationship), education level (0 = Less than a college degree, 1 = College degree or higher), employment status (0 = Unemployed, 1 = Part-Time, 2 = Full-Time), health insurance status (0 = Private Insurance, 1 = Uninsured or Government Assistance), and region (0 = Washington D.C., 1 = Philadelphia, 2 = Baltimore). YMSM’s race/ethnicity (0 = Non-Hispanic White, 1 = Racial/Ethnic minority) was dichotomized based on low variance among racial/ethnic minority participants (Non-Hispanic White [n = 149, 65.1%], Non-Hispanic Black [n = 21, 9.2%], Hispanic/Latino – All races [n = 30, 13.1%], and all Other Races/Ethnicities/Multiracial [n = 29, 12.7%]).

Data analysis

We generated descriptive statistics to describe our study sample’s PrEP continuum progression and assessed bivariate differences (t-tests, ANOVA, Spearman’s correlation) by minority stress variables and sociodemographic characteristics. To test the association between minority stress and engagement in the PrEP continuum, we first created a composite score using our three minority stress indicators. We treated sexual minority stress as a composite score, as opposed to a latent factor, to better understand an individual’s relative engagement with PrEP compared to other participants in the study based on disparate sexual minority stress experiences. Similar to studies that exhibit how compounding, negative psychosocial conditions (i.e., syndemics) produce additive HIV risk in YMSM,^47-49 researchers in prior studies have argued that the presence of multiple sexual minority stressors may be interconnected and exacerbate one another to produce negative health outcomes.^50,51 To create the composite measures and assess this hypothesis, we employed strategies similar to prior studies,^49,52,53 first by standardizing the mean scores for each minority stress variable and dichotomizing each variable to identify participants whose scores fell below the mean (0 = Below average [Z-scores less than 0]) or were at or above the mean (1 = Average/Above average Levels [Z-scores of 0 and above). We standardized the sexual minority stress variables to acknowledge the relative context in which minority stressors are experienced uniquely within this sample, as participants recruited in predominantly urban settings of the Mid-Atlantic United States. Roughly 55.9% reported average/above average perceived community stigma (Z-score range = −1.57–3.63), 60.3% reported average/above average sexuality-related discrimination (Z-score range = −1.05–3.68), and 57.6% reported average/above average internalized homonegativity (Z-score range = −1.10–2.71).

We then summed the three dichotomized/standardized minority stress variables (range: 0–3; see Table 1) to create a composite stress score. Unlike a latent variable approach where the variables are assumed to be highly inter-correlated, our composite stress score does not assume intercorrelation. Rather, the composite stress score indicates the amount of average/above average discrimination experienced by each participant, with higher scores reflecting greater amounts of experienced sexual minority stress relative to other YMSM in the sample. We then used this composite stress score to assess whether compounding experiences of minority stress were associated with engagement along the PrEP continuum. We adjusted for sociodemographic characteristics in our multivariable models (see Table 2).

Table 1.

Descriptive characteristics, N = 229 PrEP-indicated emerging adult MSM.

Variable	N (%)	m (sd)
Age, range 18–25		22.41 (2.07)
Race/Ethnicity
Non-Hispanic White	149 (65.1)
Racial/Ethnic Minority	80 (34.9)
Sexual identity
Gay, queer, same-gender loving, or homosexual	188 (82.1)
Other sexual identity	41 (17.9)
Relationship status	142 (62.0)
Single	87 (38.0)
In a relationship
Education level	92 (40.2)
Less than college degree	137 (59.8)
College degree or higher
Employment status	56 (24.5)
Unemployed	58 (25.3)
Part-time	115 (50.2)
Full-time
Health insurance status	201 (87.8)
Private insurance	28 (12.2)
Uninsured or government assistance
Metropolitan area	99 (43.2)
Washington, DC	86 (37.6)
Philadelphia	44 (19.2)
Baltimore
Composite minority stress score, range 0–3		1.38 (1.01)
0 (Very low minority stress)	56 (24.5)
1 (Low minority stress)	63 (27.5)
2 (High minority stress)	77 (33.6)
3 (Very high minority stress)	33 (14.4)
Perceived community homophobia, range 1–4		1.90 (0.57)
Sexuality-related discrimination, range 1–6		2.15 (1.08)
Internalized homonegativity, range 1 — 4		1.67 (0.63)
Progression along the HIV care continuum	13 (5.7)
PrEP-unaware	82 (35.8)
PrEP-aware, no intention to initiate	38 (16.6)
PrEP-aware, intend to initiate within 3 months	96 (41.9)
PrEP users

Table 2.

Bivariate associations between progression along PrEP continuum and sociodemographic characteristics, N = 229 emerging adult MSM.

Variable	m (sd)	t or F	p
Race/Ethnicity	2.05 (1.02)	2.20	0.029
Non-Hispanic White	1.75 (0.95)
Racial/Ethnic minority
Sexual identity	2.05 (0.98)	3.49	0.001
Gay, queer, same-gender loving, or Homosexual	1.46 (0.98)
Other sexual identity
Relationship status	2.06 (0.99)	2.11	0.036
Single	1.77 (1.01)
In a relationship
Education level	1.68 (0.94)	−3.46	0.001
Less than college degree	2.13 (1.01)
College degree or higher
Employment status	1.84 (1.01)	9.85	<0.001
Unemployed	1.53 (0.90)
Part-time	2.21 (0.98)
Full-time
Health insurance status	2.01 (0.98)	2.76	0.006
Private insurance	1.46 (1.04)
Uninsured or government assistance
Metropolitan area	2.10 (1.04)	2.27	0.106
Washington, DC	1.79 (0.93)
Philadelphia	1.91 (1.01)
Baltimore

In a subsequent analysis, we treated each form of minority stress as independent correlates in our multivariable analyses given their low inter-correlation (see Table 3). This approach allowed us to examine how each form of minority stress was associated with participants’ PrEP engagement when entered separately in our models.

Table 3.

Correlation table for progression along the PrEP continuum by minority stress and mental health variables, N = 229 emerging adult MSM.

Variable	I	II	III	IV
I PrEP continuum	–
II Age	r = 0.37 p < 0.001	–
III Perceived community homophobia	r = −0.18 p = 0.009	r = −0.20 p=0.001
IV Sexuality-related discrimination	r=0.08 p = 0.227	r = −0.03 p = 0.462	r=0.29 p < 0.001	–
V Internalized homonegativity	r = −0.27 p < 0.001	r = −0.22 p < 0.001	r=0.30 p < 0.001	r=0.09 p = 0. 156

Given the ordinal classification of the PrEP continuum variable, we conducted tests of parallel lines to determine that our independent variables adhered to the assumption of proportional odds.⁵⁴ Age (χ²_df=14 = 31.15, p = .005), relationship status (χ²_df=2 = 11.67, p = .003), education level (χ²_df=2 = 8.45, p = .005), and internalized homonegativity (χ²_df=42 = 64.61, p = .014) violated the proportional odds assumption. To account for these violations, we developed a partial proportional odds model using Version 9.4 of the SAS system⁵⁵ to assess location along the PrEP continuum among YMSM by minority stress variables (community homophobia, sexuality-related discrimination, and internalized homonegativity), adjusted for sociodemographic characteristics.

Results

Table 1 describes the sample’s demographic characteristics. Mean age was 22.41 (sd [standard deviation] = 2.07) years. Most participants identified as non-Hispanic White (65.1%), gay, queer or same gender loving (82.1%), single (62%), college graduates (59.8%) and reported having private insurance (87.8%). Regarding PrEP continuum outcomes, 5.7% were PrEP-unaware, 35.8% reported being PrEP aware, but did not intend on initiating PrEP, 16.6% intended on initiating PrEP, and 41.9% reported having taken PrEP. On average, participants reported a low-moderate amount of minority stress based on the calculation of our composite score (m = 1.38, sd = 1.01). Within minority stressors, mean perceived community homophobia was 1.90 (sd = 0.57), mean sexuality-related discrimination was 2.15 (sd = 1.08) and mean internalized homonegativity was 1.67 (sd = 0.63).

Tables 2 and 3 describe the bivariate associations between sociodemographic characteristics and progression in the PrEP continuum. Age was positively correlated with PrEP continuum progression (r = 0.36, p < .001). Non-Hispanic White YMSM were, on average, further along the continuum (t = 2.20, p = .029) compared to racial/ethnic minority participants. YMSM that identified as gay, queer or same gender loving were further on the continuum than other MSM, (t = 3.49, p = .001). YMSM who were single (t = 2.11, p =.036), college-educated (t = −3.46, p = .001) and had private insurance (t = 2.76, p =.006) reported more advanced locations across the PrEP continuum. YMSM who were employed full-time were significantly farther on the continuum (F(2,226) = 9.85, p < .001) than those who reported being unemployed or employed part-time. We observed no difference in PrEP continuum progression by metropolitan area. Regarding sexual minority stressors, perceived community homophobia (r = −0.18, p = .009) and internalized homonegativity (r = —0.27, p < .001) were negatively correlated with PrEP continuum progression. We observed no correlation between sexuality-related discrimination and our PrEP continuum variable.

In our first multivariable model (Table 4; Likelihood Ratio χ²_df=17 = 83.18, p < .001), greater cumulative minority stress was negatively associated with progression along the PrEP continuum (AOR = 0.76, 95% CI [0.58–0.99], p = .048). YMSM whose sexual identity was classified as other sexual identity exhibited lower odds of progression along the PrEP continuum (AOR = 0.37, 95% CI [0.19–0.74], p = .005) compared to gay-identified YMSM. Statistically significant, unequal slopes were observed by age between PrEP-aware with no intention and PrEP-unaware participants (AOR = 1.74, 95% CI [1.08–2.80], p = .023), those who were PrEP-aware and intended to initiate compared to those were PrEP-aware and did not intent to initiate (AOR = 1.34, 95% CI [1.12–1.62], p = .002), and those who had used PrEP compared to those who intended to initiate PrEP (AOR = 1.31, 95% CI [1.08–1.58], p = .006). Participants who were in a relationship were less likely to progress between intention groups (Threshold 3 versus 2; (AOR = 0.27, 95% CI [0.14–0.50], p < .001) and between those who had ever used PrEP and those who intended to initiate PrEP (AOR = 0.50, 95%CI [0.27–0.93], p = .027).

Table 4.

Partial proportional odds model for progression along the PrEP continuum by minority stress variables and sociodemographic characteristics, N = 229.

	Model 1			Model 2
Variable	AOR	95% CI	P	AOR	95% CI	p
Equal slopes
Race/Ethnicity
Non-Hispanic White	REF	0.43, 1.31	0.319	REF	0.45, 1.40	0.428
Racial/Ethnic Minority	0.75			0.80
Sexual identity
Gay, queer, same-gender loving, or homosexual	REF	0.19, 0.74	0.005	REF	0.24, 0.99	0.049
Other sexual identity	0.37			0.49
Employment status
Full-time	REF	0.26, 1.36	0.081	REF	0.31, 1.36	0.254
Part-time	0.53	0.44, 1.82	0.771	0.65	0.53, 2.33	0.784
Unemployed	0.90			1.11
Health Insurance status
Uninsured or government assistance	REF	0.86, 4.32	0.112	REF	0.83, 4.27	0.129
Private insurance	1.93			1.88
Metropolitan area
Washington, DC	REF	0.37, 1.24	0.203	REF	0.36, 1.23	0.192
Philadelphia	0.67	0.46, 2.14	0.988	0.66	0.45, 2.28	0.968
Baltimore	0.99			1.02
Perceived community homophobia				0.67	1.06, 1.82	0.154
Sexuality-related discrimination
None				REF	1.06, 1.82	0.019
Any				1.39
Composite minority stress	0.76	0.58, 0.99	0.048
Unequal slopes
Age
Threshold 2 vs. 1	1.74	1.08, 2.80	0.008	1.75	1.06, 2.87	0.008
Threshold 3 vs. 2	1.34	1.12, 1.62	0.003	1.33	1.10, 1.61	0.003
Threshold 4 vs. 3	1.31	1.08, 1.58	0.029	1.30	1.07, 1.57	0.029
Relationship status: In a relationshipa
Threshold 2 vs. 1	1.07	0.29, 2.89	0.850	0.88	0.23, 3.35	0.850
Threshold 3 vs. 2	0.27	0.14, 0.50	<0.001	0.19	0.10, 0.39	<0.001
Threshold 4 vs. 3	0.50	0.27, 0.93	0.003	0.37	0.19, 0.71	0.003
Education level: College degree or higherb
Threshold 2 vs. 1	0.23	0.04, 1.42	0.133	0.24	0.04, 1.55	0.133
Threshold 3 vs. 2	0.58	0.27, 1.27	0.262	0.63	0.28, 1.41	0.262
Threshold 4 vs. 3	1.14	0.52, 2.51	0.581	1.26	0.56, 2.86	0.581
Internalized homonegativity
Threshold 2 vs. 1				0.52	0.19, 1.41	0.198
Threshold 3 vs. 2				0.45	0.27, 0.77	0.003
Threshold 4 vs. 3				0.39	0.22, 0.69	0.001

Note:

^aReferent Group = Single

^bReferent Group = College degree or higher; Threshold 1: PrEP-Unaware, Threshold 2: PrEP-Aware, No Intention to Initiate, Threshold 3: PrEP-Aware, Intend to Initiate in 3 Months, Threshold 4: Ever Used PrEP.

In our second multivariable model (Table 4; Likelihood Ratio χ²_df=21 = 91.69, p < .001), sexuality-related discrimination was positively associated with odds of higher locations along the PrEP continuum (AOR = 1.39, 95%CI [1.06–1.82], p = .019). Internalized homonegativity was negatively associated with reporting an advanced location between PrEP-aware participants with no intention to seek PrEP and PrEP-aware participants who intend to seek PrEP (AOR = 0.45, 95%CI [0.27–0.77], p = .003), and between those who intend to seek PrEP and those who had ever used PrEP (AOR = 0.39, 95% CI [0.22–0.69], p = .001). Older age was positively associated with odds of being further along the PrEP continuum. Specifically, older age was positively associated with the odds of being further along the PrEP continuum among YMSM who were PrEP-aware (AOR = 1.75, 95% CI [1.06–2.87], p = .008), YMSM who intended to initiate PrEP in the next three months (AOR = 1.33, 95%CI [1.10–1.61], p = .003) and YMSM who reported use of PrEP (AOR = 1.30, 95%CI [1.07–1.57], p = .029). Lower odds of progression were also found among YMSM whose sexual identity were classified as other sexual identity compared to gay-identified YMSM (AOR = 0.49, 95%CI [0.24–0.99], p = .049). No association was found between community homophobia and PrEP continuum progression.

Discussion

Our study sought to describe the PrEP care continuum in a PrEP-indicated sample of YMSM and examine the relationship between minority stressors and participants’ placement along the PrEP continuum. Less than half of PrEP-indicated participants indicated having taken PrEP, further supporting the abundance of literature demonstrating barriers to PrEP access among high-risk MSM.^5,10-16 Consistent with prior research, our findings indicated that YMSM remain burdened by minority stressors. We therefore examined whether the presence and levels of minority stressors were associated with YMSM’s location in the PrEP continuum.

YMSM who reported greater cumulative minority stress were less likely to be located further along the PrEP continuum, supporting prior research suggesting that minority stressors undermine HIV prevention by reinforcing negative cultural and community norms associated with same-sex sexuality and hegemonic masculinity.^56,57 Specifically, sexual minority stressors may compound and exacerbate YMSM’s readiness and self-efficacy to seek important sexual health services like PrEP. Thus, future multilevel research examining how to prevent and address the cumulative experiences of minority stress on YMSM’s PrEP engagement demands prioritization if we are to achieve the goals laid out in the Ending the HIV Epidemic initiative in the United States by 2030.⁵⁸

When we examined minority stressors as separate constructs, we found that internalized homonegativity was negatively associated with YMSM’s advanced location along the PrEP continuum. These findings align with prior research that suggests internalized homonegativity as an inhibiting factor on YMSM’s access to sexual health-promotive services (e.g., HIV testing).^38,59 This association is especially pertinent when comparing PrEP-aware participants who intended to seek PrEP to PrEP-aware participants who did not intend to seek PrEP, and those who reported having used PrEP compared to those who intended to seek PrEP. Thus, internalized homonegativity may play a significant role as YMSM are contemplating to initiate (or re-initiate) PrEP. Given the cross-sectional nature of this analysis, however, future research should extend our work and investigate the association between internalized homonegativity and PrEP engagement over time. From a practice stand-point, service providers may wish to encourage sexuality-affirming discussions with YMSM within clinical practice settings (e.g., medical and social service providers) and facilitate linkage to other sexuality-affirming resources (e.g., queer-friendly mental health services to counteract the potential impact of internalized homonegativity on PrEP engagement. Future intervention research focused on reducing internalized homonegativity as part of PrEP roll-out programs is warranted.

Sexuality-related discrimination was associated with further progression along the PrEP care continuum. This unanticipated finding was inconsistent with our hypothesis, as prior research has observed sexuality-related discrimination to be associated with decreased HIV prevention service utilization among MSM.^60,61 Although we were unable to assess participants’ coping skills, prior research on positive marginality has noted that many sexual minority youths may develop adaptive coping skills to deal with interpersonal discrimination.^62,63 For instance, PrEP campaigns have sought to represent this biomedical strategy as a symbol of sexual affirmation. Thus, it is plausible that YMSM who reported sexuality-related discrimination may be especially invested in using PrEP as a strategy to affirm their sexual practices and overcome HIV-and LGBT-related stigma. It is also possible that YMSM experiencing discrimination are more mindful of seeking healthcare providers who are nonjudgmental, sexuality-affirming, and more willing to promote PrEP.⁶⁴ Future research examining how YMSM overcome sexuality-related discrimination when accessing and adopting PrEP may offer opportunities to support resilience frameworks and enrich on-going strategies to roll-out PrEP.

Perceived community homophobia was not associated with progression along the PrEP continuum. Minority stress theory posits that community homophobia may distally shape health outcomes and behaviors among MSM,¹⁷ thus absence of an association between community homophobia and PrEP engagement may be due to the presence of more proximal stressors (e.g., discrimination, internalized homonegativity) in the multivariable models. Our bivariate-level findings support this interpretation, as perceived community homophobia had a strong correlation with both discrimination and internalized homonegativity, but negative correlation with location along the PrEP continuum. Future research should assess whether proximal minority stressors mediate the relationship between perceived community homophobia and PrEP engagement, as it may inform multilevel LGBT-inclusive diversity efforts (e.g., mass media campaigns) and disrupt the negative associations between proximal minority stress indicators and engagement in the PrEP continuum.^65-67

Several sociodemographic differences were associated with YMSM’s PrEP continuum engagement. Older YMSM were more like to progress along the PrEP continuum, underscoring the importance of on-going efforts to ensure PrEP access among MSM in early adulthood.⁶⁸ Additionally, participants who reported being in a relationship were less likely to be further along the PrEP continuum. Unfortunately, we did not ascertain for how long participants have been in a relationship, how they characterize their relationships (e.g., monogamous versus non-monogamous), and whether sexual partner agreements are in place regarding risk reduction practices. Given that these components may inform PrEP eligibility,⁶⁹ future research exploring how YMSM in relationships integrate PrEP eligibility, uptake, and maintenance within the contexts of interpersonal or romantic relationships is warranted. We also observed that participants who reported another sexual identity (as opposed to gay, homosexual, or same-gender loving) were less likely to be further along the PrEP continuum, supporting the need to create PrEP messaging focused on behaviors rather than sexual identity when seeking to increase individuals’ perceptions of HIV risk.⁷⁰ Formative research is necessary to explore and inform effective PrEP messaging to elicit PrEP uptake among MSM who do not identify as gay.

Our findings provide additional considerations for future research as well. Prior studies have expanded and applied alternative stressors (e.g., concealment of sexual minority identity, anticipated sexuality-related rejection, and violent victimization) to better understand the role of minority stress on health behaviors and outcomes.^27,71,72 These constructs may inform how YMSM seek sexual health services as well. Moreover, minority stressors may not comprehensively address the role that social adversity plays in PrEP uptake among YMSM. Prior studies have expanded the utility of minority stress to incorporate within-group differences among sexual minorities, addressing stress attributed to compounding, intersectional oppressions by race and class (i.e., multiple minority stressors).^73,74 Lastly, given that the minority stress theory originated to assess mental health disparities that burdened sexual minority communities, future studies should account for potential mental health outcomes (e.g., depression, post-traumatic stress) that may mediate minority stressors’ influences on PrEP engagement.

Our study has several limitations. First, our findings are not generalizable to YMSM outside of our sample. Participants were recruited across the Mid-Atlantic region via Facebook and Grindr. Although this region exhibits high rates of HIV prevalence, our analyses may yield different results if applied to rural areas or other regions with fewer sexuality-affirming resources. Though we observed no differences in PrEP continuum outcomes by race/ethnicity, our findings may exhibit Type II error since we were not powered to assess racial/ethnic differences beyond non-Hispanic White and racial/ethnic minority distinctions. Future research should aim to replicate our findings with alternative community samples of YMSM, especially among racial/ethnic minority YMSM given historically lower rates of PrEP adoption compared to their Non-Hispanic White counter-parts.^17,75 Similarly, the measurement for sexual identity, conflating gay, queer, same-gender loving, and homosexual as one category, potentiates a weakened understanding of how sexual identity and sexual minority stressors arise in relation to PrEP use. Ongoing studies on sexual minority stress should give greater attention to disparate categories on sexual identity. Furthermore, engagement along the PrEP continuum may be a cyclical process similar to engagement along the HIV care continuum given previously identified barriers to access, adoption, and adherence, respectively;⁷⁶ yet our cross-sectional design assumes progression as a linear trajectory. Prior research has also found many YMSM discontinue PrEP use; however, the extent to which sexual minority stressors play a role remains poorly understood.⁷⁷ Longitudinal analyses may better inform engagement along the PrEP continuum and attend to how changes in progression are affected by minority stress variables over time.

Conclusion

Sexual minority stressors remain a social burden as well as a hindrance to HIV prevention efforts including PrEP. YMSM could greatly benefit from PrEP given their disproportionate rate of annual HIV infection, yet this community continues to be burdened by barriers to PrEP access like minority stressors. To increase YMSM’s PrEP engagement, efforts that assist those at high-risk to overcome experiences of homophobic stigma must be prioritized.