Gay and Bisexual Men Who Report Anal Sex Stigma Alongside Discomfort Discussing Anal Sex with Health Workers Are Less Likely to Have Ever Received an Anal Examination or Anal Swab

Abstract

Purpose:

We sought to determine whether stigma toward anal sexuality was associated with having ever received an anal examination or anal swab among gay and bisexual men (GBM).

Methods:

In 2017, we conducted a cross-sectional online survey with 1513 adult cisgender GBM living in the United States. We used structural equation modeling to test whether the Anal Sex Stigma Scales (a validated measure comprising provider stigma, self-stigma, and silence) was negatively associated with lifetime receipt of anorectal examination or anal swabbing by a medical provider. The model assessed mediation by respondents' comfort discussing anal sex practices with health workers and adjusted for possible confounders.

Results:

As hypothesized, anal sex stigma was associated with less comfort discussing anal sex (β = −0.44, 95% confidence interval [CI]: −0.50 to −0.38, p < 0.001), and greater comfort was associated with greater likelihood of screening (β = 0.28, 95% CI: 0.19 to 0.37, p < 0.001). The model demonstrated good fit (root mean square error of approximation = 0.045, comparative fit index, and Tucker–Lewis index each = 0.99) and adjusted for everyday discrimination, social support specific to anal sex, age, income, education, medical coverage, outness, and ethnic/racial identification. Collectively, model variables accounted for 48% of the variance in screening (p < 0.001).

Conclusion:

GBM who endorsed less anal sex stigma reported greater comfort discussing anal sex with health workers and were more likely to have ever received anal health screening by a medical provider. To improve anal health and cancer prevention among GBM, anal sex stigma and related discomfort discussing anal sex with health workers are targets for intervention.

Introduction

Gay and bisexual men (GBM) are at increased risk for sexually transmitted infections (STIs) including anal human papillomavirus (HPV) infection and associated anal cancer.^1–4 Anal STIs and cancers often remain undetected and therefore are diagnosed at a more advanced stage.^5,6 Late diagnoses occur, in part, because STIs and cancers are often asymptomatic in the anorectal tract, particularly early in their natural history.^6–8

Unless health professionals routinely screen for infections and cancer by performing anorectal examination and swabbing for disease detection, many GBM will continue to remain undiagnosed.^2,9 According to one estimate conducted across 42 STI clinics in the United States, had only urethral samples been tested, 72% of rectal STIs would have remained undetected among GBM.⁹

To this point, several policy initiatives, including the federal plan for Ending the HIV Epidemic and recommendations from the International Anal Neoplasia Society, the Centers for Disease Control and Prevention, the New York State Department of Health AIDS Institute, the HIV Medicine Association of the Infectious Diseases Society of America, and the National Coalition of STD Directors, now recommend that sexual health screening among GBM include anorectal examination and swabbing, even among GBM who are not living with HIV.^10–14 Together, these procedures could curb synergistic epidemics of HIV and other STIs among GBM, and increase earlier detection of anal cancer to reduce mortality in this at-risk population.^10–12

GBM are underengaged in anorectal examination and swabbing, and the causes are not well understood. In the American Men's Internet Survey 2017, among 10,049 sexually active GBM, only 13% had received rectal swabbing in the preceding year.⁵ These anal health screening procedures remain low even within HIV clinics, where they should be standard clinical practice among providers caring for GBM.^11,12,15,16

A common barrier to service utilization by both providers and patients is stigma.¹⁷ Stigma can include explicit acts of discrimination (e.g., by health workers), anticipation of the future reoccurrence of discrimination, and internalization of one's devaluation by others,¹⁸ all of which interfere with a range of health services among GBM.^19–24 In qualitative studies, GBM report that stigma functions to some extent as a barrier to care for anal infections and cancer.^25–27 Relatedly, providers cite their own discomfort with anal examination and perceptions of patient reluctance as barriers to screening,^28,29 even though GBM appear willing to participate when educated about the need.^30–32 In addition, after people experience anal health problems, they tend to not seek care and to conceal their symptoms during care,^33,34 and this includes GBM.^7,35

These findings suggest that stigma and subsequent psychological discomfort during health care interactions, on the part of providers as well as patients, may play a role in low receipt of anal health screening among GBM.

We examined whether stigma and psychological discomfort are associated with anal health screening among adult cisgender GBM living in the United States. We hypothesized that men's elevated endorsement of anal sex stigma would be associated with a lower likelihood of having ever received either an anorectal examination or swabbing, and that this association would be mediated by a lower level of comfort discussing anal sex with health care providers.

Methods

Participants

In 2017, we conducted a cross-sectional online survey with adult cisgender GBM. The Human Subjects Division of the University of Washington and the Institutional Review Board of the New York State Psychiatric Institute approved procedures. We screened, collected consent from, and surveyed participants through an online survey platform.³⁶ Participants were eligible to participate if they: (1) identified as currently male and were assigned male at birth; (2) reported at least one male anal sex partner in the past year; (3) lived in the United States; (4) reported being able to read, write, and speak in English; and (5) were at least 18 years of age.

We defined “anal sex” broadly, including penile, oral, and manual stimulation. Participants who engaged in the parent project's earlier interview and pilot studies aimed at developing and validating measures of stigma toward anal sexuality were deemed ineligible. The parent project's component studies are described elsewhere.^27,37,38

Of the 1936 respondents who consented by continuing to respond beyond the information statement, we flagged potentially fraudulent responses using an algorithm of key variables,³⁹ ultimately excluding 17 responses. Of the 1919 remaining, 1513 (78%) answered questions relevant to the current analyses. Among those who completed demographic questions, respondents who ended the survey before answering questions about anal sex stigma were more likely to report younger age, being out to “Nobody,” Black/African American racial identification, and “I'm not sure” about their sexual position preferences.

Procedures

To recruit participants to complete the survey, we engaged in targeted sampling, including advertisement on social media and men-seeking-men platforms, electronic announcements to affinity groups within each U.S. state as well as Puerto Rico, and snowball sampling via e-mails to interviewees from a formative stage of the parent project to encourage their recruitment of other eligible GBM.²⁷

In our online survey, we reserved space for racial and ethnic minority men by including a quota trigger that precluded enrollment of additional non-Latino White men once they comprised one-third of the initial targeted sample size. After additional advertising and outreach to men of color, recruitment ended after reaching a revised goal of nearly half of the sample identifying as men of color, with subgroups proportional to their presence in the national population. Respondents who completed the survey could enter a raffle to win one of twelve $50 gift certificates to an online retailer. We collected survey responses over 10 weeks, between July and September 2017. Procedures are described in greater detail elsewhere.³⁸

Measures

Primary explanatory variable

Stigma toward anal sexuality was measured by the 17-item Anal Sex Stigma Scales,³⁷ comprising manifestations of experienced, anticipated, and internalized stigma within subscales for provider stigma (Cronbach's α = 0.79; e.g., “Health workers would treat me badly if they knew the ways I have anal sex”), self-stigma (Cronbach's α = 0.72; e.g., “I may never let go of the shame I feel about anal sex”), and silence (Cronbach's α = 0.79; e.g., “Even if someone brought it up, most guys would hide their true feelings about anal sex”).³⁷

Each subscale represents a theme derived from in-depth interviews with cisgender GBM and key informants,²⁷ which were then operationalized as scale items and validated within a large online sample.^37,38 Silence stigma refers to the omission of information about anal pleasure and health, for example, from a lack of sex education, limited research on the subject, and others' constrained social disclosure of sexual concerns and experiences. Further description of the development and content of the scales is published elsewhere.³⁷ Likert response categories ranged from 0 (Disagree strongly) to 3 (Agree strongly). Higher scores indicate higher levels of stigma.

Proposed mediator

We conceptualized comfort discussing anal sex practices as a mediator between stigma and health services engagement, based on our analyses of in-depth interviews from formative work with key informants and GBM, who described that stigma contributed to men's discomfort talking about anal sex during clinical encounters.²⁷ Comfort was measured by a question about a respondent's comfort talking with medical providers about their “specific anal sex practices.” A 5-point Likert response scale ranged from 0 (Not at all comfortable) to 4 (Extremely comfortable), with higher scores indicating greater comfort.

Primary dependent variable

Lifetime receipt of anal health screening comprised anal examination and anal swabbing, modeled as a formative latent factor in response to the prompt “Has a medical provider ever done the following with you?” Participants could endorse “Yes,” “No,” or “Not sure” to two activities: “Swabbed your anus (inside your butt) with a cotton swab” and “Conducted an anal examination.” Those who endorsed “Not sure” were recategorized as “No.”

Covariates

During the formative phase for the larger parent project to develop new measures of anal sex stigma,²⁷ interview respondents described that sexual racism^40,41 and social support related to anal sex^42,43 exacerbated and mitigated, respectively, the effects of stigma on health seeking behavior. We therefore empirically examined the potential of these constructs to function as statistical confounders as well as additional potential confounding by other sociodemographic variables.

Discrimination, including endorsement of sexual racism, was measured with the abbreviated version of the Everyday Discrimination Scale (EDS),⁴⁴ a validated 5-item scale of the perceived frequency of experienced discrimination (e.g., “You are treated with less respect than others”) on a 6-point Likert response scale ranging from 0 (Never) to 5 (Almost every day). The EDS allows respondents to select multiple sources of discrimination, including discrimination based on one's perceived race and sexual orientation. Reliability was high in our sample (Cronbach's α = 0.87). Higher scores indicate higher endorsement of discrimination.

Social support specific to anal sex was adapted from the 8-item emotional and informational subscale of the Medical Outcomes Study Social Support Scale.⁴⁵ Participants endorsed the availability of support (e.g., “Someone to share my most private worries and fears with about anal sex” and “Someone who's advice about anal sex I really want”) with Likert response categories ranging from 1 (None of the time) to 5 (All of the time). The adaptation yielded high reliability in our sample (Cronbach's α = 0.97). Higher scores indicate greater social support.

Analytic plan

We conducted structural equation modeling using Mplus in a two-step procedure: first, confirmatory factor analysis to assess the fit of the measurement model, and then, path analysis with model specification and the opportunity to modify paths based on modification indices.⁴⁶ Given the number of ordinal variables in our model, we chose a weighted least-square estimator using a diagonal weight matrix, with theta parameterization.⁴⁷ To assess model goodness of fit, we examined the comparative fit index (CFI), Tucker–Lewis index (TLI), and the root mean square error of approximation (RMSEA), using their widely accepted cutoff scores (CFI and TLI >0.9; and RMSEA <0.05).^48,49

We requested modification indices >50 to assess options to increase model fit that might be congruent with theory. To select measures for inclusion as covariates in model testing, we examined bivariate statistics, using Spearman's rho (r_s), point biserial correlations for dichotomous components of lifetime anal health screening, and chi-square (χ²) and t-test analyses. To conclude mediation, we used the joint significance test, where both paths have to be significant.⁵⁰

Mplus conducts full information maximum likelihood estimation by default,⁴⁷ thereby including participants with missing endogenous items. We conducted Little's missing completely at random test⁵¹ and the results were significant [χ²(df) = 665.15(493), p < 0.001], indicating that data were not missing completely at random. However, missing data were minimal across most variables, less than 0.7% at the item level across respondents. Missing data were higher for social support (7.7%) and everyday discrimination (5.8%).

Results

Sample and descriptive findings

The sample reflected a broad distribution across ages, incomes, racial/ethnic group identification, and geography (Table 1). Most reported a frequency of engaging in anal sex from twice a week to monthly. The top quartile reported 13 or more anal sex partners in the past year and most expressed a preference for versatile sexual positioning, with exclusively insertive anal intercourse the least preferred. Most men reported that their last HIV test was seronegative, and over one-third of these men reported currently using pre-exposure prophylaxis.

Table 1.

Characteristics of Cisgender Gay and Bisexual Men Reporting Anal Sex in the Past Year (N = 1513)

	n (%)
Age in years, M (SD; range)a	36.1 (11.3; 18–72)
18 through 27 years	401 (26.5)
28 through 34 years	362 (23.9)
35 through 44 years	396 (26.2)
45 through 72 years	354 (23.4)
Geography
Rural	116 (7.7)
Small	181 (12.0)
Medium sized	329 (21.7)
Large	691 (45.7)
Suburb	196 (13.0)
Regiona
Northeast	316 (20.9)
Midwest	269 (17.8)
South	646 (42.7)
West	281 (18.6)
Race and ethnicityb
Latino, Hispanic, or Spanish ethnicity (of any race)	323 (21.3)
American Indian/Alaska Native	11 (0.7)
Asian or Asian American	88 (5.8)
Black or African American	224 (14.8)
White (non-Latino)	802 (53.0)
Native Hawaiian or Pacific Islander	5 (0.3)
Bi- or multiracial	53 (3.5)
Education
≤ High school/general equivalency diploma	120 (7.9)
Some college	347 (22.9)
2-year college	94 (6.2)
4-year college	496 (32.8)
Masters	306 (20.2)
Doctoral/professional	150 (9.9)
Incomec
Below $15,000	201 (13.3)
$15,000 to $29,999	263 (17.4)
$30,000 to $44,999	243 (16.1)
$45,000 to $59,999	211 (13.9)
$60,000 to $74,999	167 (11.0)
$75,000 to $89,999	100 (6.6)
$90,000 or more	318 (21.0)
Housing
Shelter/dormitory/drug treatment/other	28 (1.9)
Someone else's home	200 (13.2)
Rent	843 (55.7)
Own	442 (29.2)
Relationship statusd
Single	757 (50.0)
Casually dating several people	101 (6.7)
Boyfriend or girlfriend	165 (10.9)
Partner or lover	241 (15.9)
Legal, civil, committed partnership	238 (15.7)
Open relationship or not sure if opene	468 (30.9)
Sexual orientation
Gay	1221 (80.7)
Bisexual/heterosexual	146 (9.6)
Queer	72 (4.8)
No label	55 (3.6)
Additional/two spirit	19 (1.3)
Sexual position preference
“Bottoming” (receptive)	468 (30.9)
“Topping” (insertive)	337 (22.3)
“Versatile” (both receptive and insertive)	652 (43.1)
No preference/not sure	56 (3.7)
Outness about sexual attraction to men
Nobody	37 (2.4)
Select friends	209 (13.8)
All friends and select family	262 (17.3)
All friends and all family	116 (7.7)
Almost everyone (friends, family, coworkers, etc.)	889 (58.8)
Never tested/never received HIV result	112 (7.4)
HIV seronegative	1195 (79.0)
Tested within the last year	1030 (86.2)
Current pre-exposure prophylaxis use	385 (32.2)
HIV seropositive	206 (13.6)
Diagnosed within the past 2 years	26 (12.6)
Linked to care at the time of diagnosis	191 (92.7)
Retained in care (most recent visit <6 months ago)	196 (95.1)
Prescribed antiretroviral therapy	201 (97.6)
Undetectable viral load	187 (90.8)
Sexual behavior
13 or more anal sex partners in the past yearf	348 (23.0)
Anal intercourse in the past 3 monthsd	1389 (91.8)
Receptive penile–anal intercourse in the past 3 months	1143 (82.3)
Insertive penile–anal intercourse in the past 3 months	1089 (78.4)
Health care engagement
Medical coverage (no health insurance or PCP)	106 (7.0)
Out to PCP about anal sex with men (among 1092 with a PCP)	931 (85.3)
Thought you had an STI in the past 3 months but did not see a provider	140 (9.3)
Lifetime screening
Anal examination	802 (53.0)
Anal swabbing	740 (48.9)
Either examination or swabbing	970 (64.1)

Percentages are of the whole sample even if values are missing.

^a Due to missingness n = 1512.

^b Due to missingness n = 1506.

^c Due to missingness n = 1503.

^d Due to missingness n = 1502.

^e Due to missingness n = 1511.

^f How many people have you had anal sex with in the past year? (anal sex here includes any sexual contact with the ass, such as touching, licking, or penetration).

M, mean; PCP, primary care provider; SD, standard deviation; STI, sexually transmitted infection.

The vast majority of respondents living with HIV reported engagement with health care and viral suppression. The majority of the sample reported having had a medical provider perform either anal swabbing or examination, with slightly less than half reporting swabbing and slightly more than half reporting examination.

Overall, participants reported a low level of stigma (mean [M] = 1.05, standard deviation [SD] = 0.43), with higher levels of silence (M = 1.62, SD = 0.56) than either provider stigma (M = 0.93, SD = 0.70) or self-stigma (M = 0.58, SD = 0.53). Overall, men reported low to moderate levels of both social support to discuss anal sex with other trusted sources (M = 1.96, SD = 1.30) and comfort discussing anal sex (M = 2.05, SD = 2.00), with much higher variability than for the other measures (Table 2).

Table 2.

Loadings, Means, Standard Deviations, Reliability, Correlations, and Covariances Among All Variables (N = 1513 Cisgender Gay and Bisexual Men)

	Variable	λa	M (SD)	α (95% CI)	1	2	3	4	5	6	7	8	9	10	11	12	13	14
					Provider stigma	Self-stigma	Silence	Comfort	Social support	Everyday discrimination	Age	Race/ethnicity	Educ.	Income	Medical coverage	Outness	Anal swab	Anal examination
	Anal Sex Stigma Scales	—	1.05 (0.43)	0.80 (0.79 to 0.82)
1	Provider stigma	0.47	0.93 (0.70)	0.79 (0.77 to 0.81)	2.44	0.58	0.58	−0.55	0.65	−0.55	−0.02	0.17	−0.10	0.01	−0.03	−0.05	0.01	−0.06
2	Self-stigma	0.72	0.58 (0.53)	0.72 (0.70 to 0.74)	0.27	1.85	0.57	−0.56	−1.23	0.57	−0.25	−0.18	−0.31	−0.18	−0.23	−0.43	−0.43	−0.36
3	Silence	0.79	1.62 (0.56)	0.72 (0.70 to 0.74)	0.47	0.53	0.62	−0.26	−0.77	0.31	−0.08	−0.12	−0.14	−0.12	−0.12	−0.23	−0.23	−0.13
4	Comfort discussing anal sex	—	2.05 (2.00)	—	−0.35	−0.41	−0.32	1.00	1.09	−0.22	0.11	0.02	0.13	0.06	0.08	0.38	0.34	0.32
5	Social support	—	1.96 (1.30)	0.97 (0.97 to 0.98)	−0.13	−0.32	−0.35	0.38	8.04	−0.26	0.02	0.02	0.28	0.22	0.35	0.71	0.75	0.51
6	Everyday discrimination	—	1.61 (1.10)	0.87 (0.86 to 0.88)	0.31	0.31	0.30	−0.16	−0.07	1.83	−0.44	−0.06	−0.35	−0.22	−0.26	−0.03	−0.08	−0.20
	Sociodemographics
7	Age	0.66	—	—	−0.01	−0.18	−0.11	0.11	0.01	−0.32	1.00	0.09	0.53	0.26	0.27	0.10	0.19	0.49
8	Race/ethnicity	0.27	—	—	0.11	−0.14	−0.15	0.02	0.01	−0.04	0.09	1.00	0.23	0.25	0.05	0.39	0.19	0.15
9	Education	0.54	—	—	−0.07	−0.23	−0.18	0.13	0.10	−0.26	0.53	0.23	1.00	0.48	0.46	0.25	0.28	0.34
10	Income	0.80	—	—	0.00	−0.13	−0.16	0.06	0.08	−0.16	0.26	0.25	0.48	1.00	0.36	0.21	0.25	0.23
11	Medical coverage	0.55	—	—	−0.02	−0.17	−0.15	0.08	0.12	−0.19	0.27	0.05	0.46	0.36	1.00	0.10	0.11	0.25
12	Outness	0.55	—	—	−0.03	−0.32	−0.30	0.38	0.25	−0.02	0.10	0.39	0.25	0.21	0.10	1.00	0.40	0.30
	Lifetime screening
13	Anal swab	0.85	—	—	0.01	−0.32	−0.30	0.34	0.26	−0.06	0.19	0.19	0.28	0.25	0.11	0.40	1.00	0.68
14	Anal examination	0.80	—	—	−0.04	−0.26	−0.16	0.32	0.18	−0.15	0.49	0.15	0.34	0.23	0.25	0.30	0.68	1.00

Covariances are in bold; correlations are below the diagonal.

^a Factor loadings are all significant (p < 0.001).

CI, confidence interval; Educ., education.

Bivariate analyses

Bivariate analyses with sociodemographic variables (Table 2) indicated that higher age, higher household income, higher education, greater outness about attraction to men, having medical coverage, and racial group identification other than Black/African American were each negatively associated with stigma and positively associated with anal health screening (all p < 0.05). We therefore constructed a formative latent factor for sociodemographic variables, comprising these six variables, to adjust for potential confounding. This composite variable allowed us to account for intercorrelation and measurement error. As a linear composite of these risk factors, higher values indicate responses associated with better screening outcomes.

Measurement model

Our measurement model, illustrated in Figure 1, comprised five latent factors for primary variables and covariates, and converged normally with acceptable fit: χ²(652) = 2628.96, p < 0.0001, χ²/df = 4.03, RMSEA = 0.045 (95% confidence interval [CI]: 0.043 to 0.047), probability of RMSEA <0.05 = 1.00, CFI = 0.99, and TLI = 0.99. Standardized factor loadings (λ) on each component of lifetime screening were significant (p < 0.001): anal examination (λ = 0.80) and anal swabbing (λ = 0.85). Loadings, means, SDs, correlations, and covariances among study variables are seen in Table 2.

FIG. 1.

Anal health stigma model positing direct effects of high anal sex stigma on lower receipt of anal health screening, mediated through comfort discussing anal sex. A circle denotes a latent factor; a square denotes a manifest (observed) variable.

Structural equation model

The structural paths hypothesized in Figure 1 provided good fit for the data: χ²(688) = 2977.58, p < 0.0001, χ²/df = 4.33; RMSEA = 0.047 (95% CI: 0.045 to 0.049), probability of RMSEA <0.05 = 1.00, CFI = 0.99, and TLI = 0.99. However, modification indices suggested an additional and theoretically congruent association between social support and comfort. This inclusion in the final model (Fig. 2) improved fit [χ²(1) = 44.44, p < 0.0001] and indicated good fit in itself: χ²(687) = 2837.82, p < 0.0001, χ²/df = 4.13, RMSEA = 0.045 (95% CI: 0.044 to 0.047), probability of RMSEA <0.05 = 1.00, CFI = 0.99, and TLI = 0.99. This final model accounted for 48% of the variance in screening, 37% in stigma, and 31% in comfort (all p < 0.001).

FIG. 2.

Final model showing standardized beta coefficients, with dotted lines indicating adjustment for confounding by covariates, *p < 0.05; **p < 0.01; and ***p < 0.001. Anal sex stigmas include three latent subscales (self-stigma, provider stigma, and silence).

As seen in Table 3, the direct effect of anal sex stigma on anal health screening was not statistically significant, but total effects indicate that higher stigma was linked to less likelihood of anal health screening (β = −0.14, 95% CI: −0.25 to −0.04, p = 0.01). As seen in Figure 2, the direct association was wholly mediated by comfort discussing anal sex, with higher stigma linked to less comfort (β = −0.44, 95% CI: −0.50 to −0.38), and more comfort linked to greater likelihood of anal health screening (β = 0.28, 95% CI: 0.19 to 0.37; both p < 0.001).

Table 3.

Standardized Effect Sizes from Structural Equation Modeling to Test Direct, Indirect, and Total Effects of Anal Sex Stigmas on Lifetime Screening Among Cisgender Gay and Bisexual Men Who Reported Anal Sex in the Preceding Year (N = 1513)

Dependent variables	Explanatory variables		Covariates adjusted for confounding
	Anal sex stigmas, β (95% CI)	Comfort discussing anal sex with health workers, β (95% CI)	Social support specific to anal sex, β (95% CI)	Everyday discrimination, β (95% CI)	Sociodemographics,a β (95% CI)
Direct effects
Anal sex stigmas	―		−0.35 (−0.41 to −0.29)***	0.35 (0.28 to 0.41)***	−0.20 (−0.27 to 0.13)**
Comfort discussing	−0.44 (−0.50 to −0.38)***	―	0.21 (0.15 to 0.26)***
Lifetime screening	−0.02 (−0.14 to 0.10)	0.28 (0.19 to 0.37)***	0.07 (−0.01 to 0.15)	0.14 (0.06 to 0.23)**	0.59 (0.51 to 0.66)***
Indirect effects
Anal sex stigmas	―
Comfort discussing		―	0.16 (0.12 to 0.19)***		0.09 (0.05 to 0.12)***
Lifetime screening	−0.12 (−0.16 to −0.08)***		0.11 (0.07 to 0.15)***	−0.05 (−0.09 to −0.01)*	0.03 (0.01 to 0.05)*
Total effects
Anal sex stigmas	―		−0.35 (−0.41 to −0.29)***	0.35 (0.28 to 0.41)***	−0.20 (−0.27 to 0.13)**
Comfort discussing	−0.44 (−0.50 to −0.38)***	―	0.36 (0.32 to 0.41)***		0.09 (0.05 to 0.12)***
Lifetime screening	−0.14 (−0.25 to −0.04)*	0.28 (0.19 to 0.37)***	0.17 (0.11 to 0.24)***	0.09 (0.02 to 0.17)*	0.62 (0.55 to 0.69)***

R² in Lifetime screening = 0.48; RMSEA = 0.045 (95% CI: 0.044 to 0.047, probability of RMSEA <0.05 = 1.00); CFI = 0.99; and TLI = 0.99.

^* p < 0.05, ^**p < 0.01, and ^***p < 0.001.

^a Sociodemographics is a composite variable comprising age, race/ethnicity, education, income, medical coverage, and outness.

CFI, comparative fit index; RMSEA, root mean square error of approximation; TLI, Tucker–Lewis index.

Notably, higher informational and emotional social support specific to anal sex was linked to lower levels of stigma (β = −0.35, 95% CI: −0.41 to −0.29) and greater comfort (β = 0.36, 95% CI: 0.32 to 0.41), in addition to greater screening (β = 0.17, 95% CI: 0.11 to 0.24; all p < 0.001).

Discussion

Our findings indicate that cisgender GBM experience stigma-related barriers to anal health screening. As hypothesized, we detected a significant pathway leading from anal sex stigma to men's discomfort discussing anal sex with health workers and thereby to men's lower self-reported receipt of an anal examination or anal swab. These associations held after adjusting for discrimination, social support specific to anal sex, age, income, education, medical coverage, outness, and ethnic/racial identification.

These findings are consistent with our earlier qualitative work on anal sex stigma²⁷ and conclusions of other researchers that stigma functions as a barrier to anal STI testing and anal cancer screening.^25,26,28 In our formative in-depth interviews, key informants and GBM described how anal sex stigma can delay men's access to services when men themselves choose to conceal their anal sex practices or when providers neglect to offer or conduct procedures, even when GBM candidly request anal health care.²⁷

Men's urges to conceal may, in part, be borne by their accurate detection of providers' explicit discrimination or aversion to performing anal health care^15,27,52,53; in this way, a patient's discomfort could function as a warning system to protect them against disclosing to providers who may mistreat them^54,55—although this may still occur even in the context of supportive providers.⁵⁶ Improving communication about anal health between providers and GBM is, therefore, likely to require intervention on both sides of the clinical encounter.

One implication of discomfort functioning as an impediment to men's participation in anal health screening is to consider alternatives to provider-mediated procedures. Self-sampling for extragenital STIs has demonstrated high acceptability and feasibility,^57–59 including through the use of mail-in collection kits, although these methods have yet to demonstrate accuracy that is comparable with clinic-collected specimens.^52,53 Another approach, also with high acceptability, involves educating GBM and their sexual partners to palpate for anal tumors themselves.⁶⁰

These alternatives might bypass interpersonal discomfort during health care interactions by eliminating men's anticipation of provider stigma. However, these alternatives place the responsibility of performing routine medical procedures on the patient's shoulders. Freeing health systems from modifying the root causes of stigma in their practices may allow stigma to continue in ways that exacerbate other health problems and disparities.⁶¹

Alternatives to clinic-based screening are also unlikely in themselves to mitigate the deterrent effects of stigma on health seeking, given that anal sex stigma comprises elements of self-stigma and silence stigma. Indeed, the relatively higher factor loadings of self-stigma and silence stigma in Figure 2 suggest that these contribute more substantially than provider stigma to discomfort and a lack of anal health screening. This implies that psychological and social forces outside of health systems influence engagement in those systems, and that intervention even in health care settings may not mitigate the deterrent effects of stigma on health seeking behavior in this population.

Still, acting to mitigate the effects of stigma anywhere may reinforce positive changes elsewhere.⁶² For example, GBM are largely interested in anal health care, but lack knowledge²⁷; educating GBM using reliable and credible sources of information, for example, education from a health care provider, could be a catalyst for screening by lessening silence about anal sex and interrupting the generalization of anticipated stigma from past providers to future providers. This aligns with a suggestion we heard in formative work²⁷ and with recommended and theory-based behavior change techniques already applied among GBM.^27,63 This may also help GBM overcome internalized stigma, which might otherwise continue to interfere with swabbing or examination conducted by themselves or a sexual partner.

In addition to perceptions about stigma among GBM in our study, ambiguous policy guidelines likely function as a structural barrier to care. Screening guidelines across the country do not always clearly indicate the need for both examination and swabbing among GBM.⁶⁴ Even clinicians who consider screening to be important and who report confidence performing these procedures still cite a lack of guidelines as a barrier to care^16,65 and, interestingly, providers also anticipate and perceive reluctance and discomfort on the part of GBM patients.^15,29,31 This likely contributes, to some extent, to the low availability and low awareness of screening procedures among GBM that are associated with their low intentions to screen.⁶⁶

We should note, even though our study detected a mediating role for discomfort discussing anal sex with health workers, other studies have found that when GBM are explicitly recruited to receive digital anorectal examination, they report minimal complaints and nearly universal willingness to continue with annual examination and swabbing.^30,67,68 More studies about the acceptability of provider-mediated anal health procedures among GBM may help to dispel concerns among clinicians, influence future policies and guidelines, and help us learn new ways to counter both provider-side and patient-side barriers.^30,69

Limitations

There were several limitations in our study. All measures were assessed through self-report, which is subject to both reporting and recall bias. Reporting bias may have been mitigated through our online data collection, which encouraged anonymity. We attempted to mitigate recall bias by testing associations between stigma and a simple categorical answer to having ever received anal health screening, which may be easier to recall accurately than behavior that falls within a more confined time range. This approach sacrificed the assessment of specificity in time periods and anal health procedures for the sake of increasing the likelihood of a high response rate in an already lengthy online survey. Relatedly, we did not operationalize digital examination specifically, just “an anal examination,” and both measures for examination and swabbing were expansive. Respondents may have had varied interpretations of these questions.

We also restricted recruitment to men who engaged in anal sex within the past year. The long induction period between acquiring HPV and developing anal cancer means that our sample excluded men whose history of anal sex beyond the past year may still make them eligible for examination and swabbing. Relationships between stigma and anal health screenings may differ among these excluded men.

Our cross-sectional survey also meant that our model assessed whether current stigma determined past behavior. Stigma may not be a stable construct over time,⁷⁰ and not being offered anorectal screening in the past could be the cause of current stigma rather than stigma being the cause of lower receipt of anal health screening. Still, the magnitude of our findings, even after adjusting for confounding factors, such as informational and emotional social support specific to anal sex, suggests a strong association that warrants further examination. In contexts where screening is available, a cohort study could more accurately determine whether baseline stigma functions as a determinant of future anal health screening.

Finally, our parent project sought to develop a new measure of anal sex stigma among cisgender GBM; therefore, the current sample excludes trans masculine men who have anal sex with men, who may also experience anal sex stigma as a barrier to anal health screening.

Conclusion

We found that cisgender GBM who endorsed less anal sex stigma reported greater comfort talking with health workers about anal sex and were more likely to have received anal health screening, whether an anal examination or an anal swab, at some point in their lives. Stigma and related discomfort discussing anal sex with health workers are targets for intervention to improve anal health and cancer prevention among cisgender GBM.

Authors' Contributions

B.A.K. and J.M.S. conceived, designed, and acquired data for the study. B.A.K. conducted the analyses. B.A.K., J.M.S., W.D., M.M.G., and T.G.M.S. contributed to the interpretation of data. B.A.K. drafted the article and B.A.K., J.M.S., W.D., M.M.G., and T.G.M.S. reviewed the article for intellectual content, revision, and final approval.

Disclaimer

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This work was supported by the National Institutes of Health under grants K23-MH124569 (Principal Investigator: Bryan A. Kutner, PhD, MPH); T32-AI07140 (Principal Investigator: Sheila A. Lukehart, PhD); T32-MH19139 (Principal Investigator: Theo Sandfort, PhD); and P30-MH43520 (Robert H. Remien, PhD); and by the Bolles Graduate Fellowship through the Department of Psychology at the University of Washington.