Factors associated with transmasculine adults recently engaging in sexual behavior with partners of unknown STI and HIV status

Abstract

The sexual partnerships of transmasculine (TM) adults—those assigned female at birth who identify as transgender men or a masculine spectrum gender identity—and characteristics associated with STI/HIV risk behavior remains understudied. Participants in the current study were TM adults (n = 141) receiving care at a community health center in Boston, Massachusetts between March 2015 and September 2016. Using generalized estimating equations, we examined individual- and partnership-level factors associated with TM adults’ odds of engaging in sexual behavior with a sexual partner of unknown STI/HIV status in the past 12 months. TM adults with casual sexual partnerships (vs. monogamous partnerships) and those in partnerships with cisgender men, other TM individuals, or transfeminine partners (vs. cisgender women) had statistically significantly higher odds of engaging in sexual behavior with a partner of unknown STI/HIV status in the past 12 months. Findings may inform future efforts to improve sexual health communication and STI/HIV disclosure between TM adults and their sexual partners.

Introduction

Currently incurable sexually transmitted infection (STIs) – including HIV, human papillomavirus (HPV), and herpes simplex virus type 1 (HSV-1) and type 2 (HSV-2) – give rise to substantial morbidity and mortality in the United States. For instance, high-risk strains of human papillomavirus (hr-HPV) are linked to 99.7% of all cervical cancers,[1] and infection with HSV-2 can cause genital lesions (i.e., genital herpes) which significantly increase an individual’s risk for acquiring HIV during sexual activity.[2–4] A synergistic relationship also exists between HSV-2 and HIV virology, making comorbid infection with these STIs particularly problematic.[3, 5, 6] Guidelines from the Centers for Disease Control and Prevention (CDC) list methods for mitigating STI/HIV-related risk, including: abstention, having fewer sexual partners, pre-exposure vaccination (e.g., HPV vaccine), pre-exposure and post-exposure prophylaxis (PrEP, PEP) for HIV-risk reduction, and using protective barriers during sex (e.g., external/internal condoms).[7] While the CDC recommends notifying recent sexual partners about a positive STI/HIV diagnosis, few guidelines help sexual partners safely disclose their STI/HIV status and initiate a discussion about strategies for STI/HIV risk reduction. Disclosure may also be complicated by state-level laws criminalizing STI/HIV exposure.[8] Consequently, prospective guidelines should help sexual partners navigate both interpersonal and legal barriers to STI/HIV disclosure.

The sexual health and partnerships of transmasculine (TM) adults – who were assigned female at birth and identify as transgender men or a masculine spectrum gender identity – are understudied. Historically, medical professionals frequently presumed that TM adults desire sexual partnerships with cisgender (non-transgender) women.[9] This presumption subsequently fueled a misconception that TM adults are at low risk for STIs and HIV.[9] As a consequence, how TM adults negotiate and navigate STI/HIV-risk reduction strategies with sexual partners of diverse gender identities has received little attention in public health research. One prior study with TM men who have sex with men (TMMSM) found that unequal power dynamics, low self-esteem, and a desire for gender affirmation impeded the ability of TMMSM to negotiate condom use with sexual partners.[10] When another study asked TMMSM to rank their top health concerns, the majority of participants (87.5%; n = 16) did not include sexual health in their top three health concerns.[11] Instead, TMMSM prioritized access to gender affirming medical care (e.g., hormones, surgery).[11] Given the psychological importance of gender affirmation, TM adults may be at greater risk for engaging in sexual behavior with a partner of unknown STI/HIV status when that partner’s gender identity serves to further affirm their own gender identity.[11] Additionally, TM adults are at increased risk for intimate partner violence,[12, 13] relative to their cisgender peers, and may be unable to safely inquire about their sexual partner’s STI/HIV status or negotiate STI/HIV-risk reduction strategies before sex. Research is needed to assess whether characteristics of TM adults’ diverse sexual partnerships – including those of the individual, the sexual partner, and the sexual partnership dyad – are associated with having engaged in sexual behavior with a partner of unknown STI/HIV status.

The current analysis investigates whether characteristics of TM adults’ sexual partnerships were associated with engaging in sexual behavior with a partner of unknown STI/HIV status in the past 12 months.Findings can inform the development of guidelines to help foster sexual health communication and disclosure between TM adults and their sexual partners, integrate biobehavioral prevention (e.g., PrEP to prevent HIV acquisition), and subsequently reduce their risk for STIs and HIV.

Methods

Participants and procedures

Between March 2015 and September 2016, 150 transmasculine adult patients enrolled in a sexual health study at Fenway Community Health Center in Boston, Massachusetts. Participants were considered eligible for enrollment into the primary study upon meeting the following criteria: (1) were assigned a female sex at birth and identify with a masculine-spectrum gender identity; (2) were between the ages of 21–64 years (inclusive); (3) had a cervix; (4) were sexually active in the past three years; (5) could speak and understand English; (6) were willing to provide informed consent. Additional information regarding the primary study is published elsewhere.[14] All study activities were approved by the Institutional Review Board at Fenway Health.

TM participants self-reported demographic and sexual health data via a cross-sectional survey for themselves and their three most recent sexual partners from the past 12 months.In this analysis, a sexual partnership was defined as having engaged in one of the following sexual behaviors: (1) penetrative sex using fingers or fists; (2) oral-genital sex; (3) oral-anal sex; (4) receptive and/or insertive frontal/vaginal sex; (5) receptive and/or insertive anal sex; or (6) the use of prosthetics/toys. This secondary analysis utilizes data from 141 TM participants with at least one sexual partner in the past 12 months.

Measures

STI/HIV status of a sexual partner.

For each sexual partner, TM participants were asked: “Has this partner ever been diagnosed (by a physician, nurse, or other medical provider) with any of the following STIs?- (1) HIV, (2) human papillomavirus (HPV), or (3) herpes simplex virus (HSV) type 1 or 2.” Response options included “yes,” “no,” “I do not know,” and “prefer not to respond” for each STI. These STIs (HIV, HPV, HSV type 1 or 2) were selected because they are currently incurable and remain transmissible through sexual behavior. Responses were dichotomously recoded as “I do not know” and “yes/no” before modeling.

Participant characteristics.

TM participants’ characteristics included their (1) ages (21–24, 25–29, or 30–50 years); (2) racial identities (Asian, African American/Black, Multiracial, Native Hawaiian or other Pacific Islander, or White); (3) Hispanic/Latinx ethnicities (Yes, No); (4) gender identities (man/male, transgender man, female-to-male [FtM], trans man, man of transgender experience, trans masculine, genderqueer, gender non-conforming, non-binary, agender, bigender, another gender identity); (5) STI/HIV histories (a prior diagnosis with HIV, HPV, Chlamydia, Trichomoniasis, Syphilis, Gonorrhea, genital herpes, hepatitis B, hepatitis C, or another STI); (6) engagement in sexual behavior with a partner of unknown STI/HIV status in the past 12 months (yes/no); and (7) number of sexual partners in the past year. In order to improve cell sizes for the purpose of statistical modeling, participants’ racial identities were crudely recoded as “racial minority/non-White” (Asian, African American/Black, Multiracial, Native Hawaiian or other Pacific Islander) or “White.” Participants’ gender identities were also recoded as “non-binary” (genderqueer, masculine of center, gender non-conforming, non-binary, bigender) or “binary or another gender identity.”

Sexual partnership characteristics.

Sexual partnership-level measures included the partnership’s configuration and the sexual partner’s gender identity. Partnership configurations were coded using the following: (1) casual sex, (2) monogamous relationship, or (3) non-monogamous relationship. Sexual partners’ gender identities were coded as: (1) cisgender female; (2) cisgender male; (3) transfeminine (transgender female [male-to-female/MtF] or nonbinary/genderqueer and assigned-male-at-birth [AMAB]); or (4) transmasculine (transgender male [female-to-male/FtM] or nonbinary/genderqueer and assigned-female-at-birth [AFAB]).

Statistical Analysis

Data were analyzed using SAS v9.4. Descriptive statistics were used to summarize demographic data for the participants and their sexual partners. Missing values were excluded from the calculation of descriptive statistics. Demographic and STI awareness data clustered around the TM participant, who reported information for up to three sexual partnerships. Generalized estimating equations (GEE) with exchangeable-within-subject correlation structures and logistic link functions were used for analysis. Eight bivariate GEEs and one multivariable GEE were used to examine individual- and partnership-level factors associated with participants’ odds of having engaged in sexual behavior in the past 12 months with a sexual partner of unknown STI/HIV status (i.e., unaware of whether their partner had been previously diagnosed with HIV, HPV, or HSV type I or II).

Results

Participant characteristics (Table 1).

Table 1.

Characteristics of transmasculine participants (n = 141 participants)

Characteristics	n = 141 (%)
Unaware of at least one sexual partner’s STI/HIV statusa
Yes	67 (47.5)
No	74 (52.5)
Age, years
21–24	45 (31.9)
25–29	57 (40.4)
30–50	39 (27.7)
Racial identity
Asian	8 (5.7)
Black/African American	4 (2.8)
Multiracial	21 (14.9)
Native Hawaiian or other Pacific Islander	1 (0.7)
White	106 (75.2)
Missing	1 (0.7)
Hispanic/Latinx ethnicity
Yes	14 (9.3)
No	133 (88.7)
Missing	3 (2.0)
Non-binary gender identityb
Yes	27 (19.2)
No	114 (80.9)
Ever diagnosed with an STI/HIVc
Yes	28 (19.9)
No	113 (80.1)
Number of sexual partners (last 12 months), median [IQR]	2 [1–4]

The 141 participants had a mean age of 27.4 years (SD = 5.7, range = 21–50). The sample was predominantly White (75.2%) and non-Hispanic/Latinx (88.7%). A total of 27 participants (19.2%) endorsed a non-binary gender identity. Among participants, the median number of sexual partners in the past 12 months was 2 (IQR = 1–4). Collectively, 19.9% of participants had previously been diagnosed with either HIV, HPV, Chlamydia, trichomoniasis, syphilis, gonorrhea, genital herpes, hepatitis B, hepatitis C, or another STI. One participant reported being HIV-positive. Sixty-seven participants (47.5%) reported having at least one sexual partner in the past 12 months with an unknown STI/HIV status.

1. Defined as having at least one sexual partner in the past 12 months with an unknown STI/HIV status, including whether the partner had ever been diagnosed with HIV, human papillomavirus (HPV), or herpes simplex virus (HSV) type 1 or 2.

2. Includes participant gender identities of genderqueer, masculine of center, gender non-conforming, non-binary, and bigender.

3. Includes any lifetime diagnosis with HIV, HPV, Chlamydia, Trichomoniasis, Syphilis, Gonorrhea, genital herpes, hepatitis B, hepatitis C, or another STI.

Sexual partnership characteristics.

The 141 TM participants provided data for a total of 259 sexual partners (median = 1; IQR = 1–3). Sexual partnerships were primarily with cisgender females (42.2%), followed by cisgender males (31.6%), TM individuals (17.2%), and TF individuals (9%). Sexual partnerships were primarily instances of casual sex (42.9%), followed by monogamous relationships (35.6%) and non-monogamous relationships (21.5%). In a brief sensitivity analysis, partnerships with cisgender women had 4.16 times the odds of being monogamous (vs. non-monogamous/casual) compared to partnerships with cisgender men, nonbinary AFAB, or nonbinary AMAB partners (OR = 4.16, 95% CI = 2.57–6.72, p < 0.001). Among the 259 sexual partners, 2.7% were known to have been previously diagnosed with HPV, 5.8% with HSV type I or II, and none with HIV. Participants were unaware of a sexual partner’s STI/HIV status in 41.3% of reported partnerships (n = 107).

Adjusted odds ratios for engaging in sexual behavior with a partner of unknown STI/HIV status (Table 2).

Table 2.

Crude odds ratios (ORs) and adjusted odds ratios (aORs) for having engaged in sexual behavior with a partner of unknown STI/HIV status in the past 12 months (n = 259 sexual partnership ps)^a

Covariates		Bivariate models			Multivariable models
Individual-level	n = 259 (%)b	OR	95% CI	p-value	aORc	95% CI	p-value
Unaware of sexual partner’s STI/HIV statusd
Yes	107 (41.3)	-	-	-	-	-	-
No	152 (58.7)	-	-	-	-	-	-
Age (years)
21–24	85 (32.8)	0.88	0.42–1.85	0.743	0.7	0.30–1.64	0.408
25–29	106 (40.9)	0.57	0.27–1.21	0.142	0.44	0.18–1.09	0.077
30–50	68 (26.3)	Ref	-	-	Ref	-	-
Racial identity
Racial minority/non-Whitee	59 (22.8)	1.55	0.77–3.09	0.217	1.32	0.46–3.80	0.613
White	199 (76.8)	Ref	-	-	Ref	-	-
Hispanic/Latinx ethnicity
Yes	26 (10.2)	1.1	0.37–3.25	0.859	0.95	0.23–3.99	0.949
No	228 (89.8)	Ref	-	-	Ref	-	-
Non-binary gender identityf
Yes	53 (20.5)	1.35	0.63–2.89	0.445	1.87	0.79–4.41	0.1526
No	206 (79.5)	Ref	-	-	Ref	-	-
Previously diagnosed with an STI/HIVg
Yes	52 (20.1)	1.24	0.61–2.53	0.556	1.01	0.43–2.38	0.9731
No	207 (79.9)	Ref	-	-	Ref	-	-
Number of sexual partners (last 12 months)		1.03	0.94–1.12	0.58	0.97	0.90–1.04	0.3785
Sexual partnership-level Sexual partnership configuration
Casual sex	110 (42.5)	6.79	3.47–13.28	< 0.001	6.56	3.17–13.56	< 0.001
Non-monogamous relationship	55 (21.2)	1.77	0.84–3.70	0.133	1.38	0.61–3.13	0.437
Monogamous relationship	91 (35.1)	Ref	-	-	Ref	-	-
Sexual partner’s gender identity
Transmasculineh	44 (17)	2.54	1.29–5.02	0.007	2.83	1.33–6.02	0.007
Transfemininei	23 (8.9)	2.76	1.26–6.02	0.011	3.3	1.28–8.51	0.013
Cisgender male	81 (31.3)	3.12	1.60–6.08	0.001	2.3	1.22–4.32	0.01
Cisgender female	108 (41.7)	Ref	-	-	Ref	-	-

Adjusting for participants’ characteristics (age, racial identity, Hispanic/Latinx ethnicity, non-binary gender identity, prior STI/HIV diagnosis, number of sexual partners in the past 12 months) and sexual partners’ gender identities, TM participants in casual sexual partnerships had 6.56 times the odds (95% CI = 3.17–13.56, p < 0.001) of engaging in sexual behavior with a partner of unknown STI/HIV status in the past 12 months relative to those in monogamous relationships. Compared to participants who had a cisgender female sex partner in the past 12 months, those with sex partners who were transmasculine, transfeminine, or cisgender male had significantly higher odds of engaging in sexual behavior with a partner of unknown STI/HIV status (transmasculine: aOR = 2.83, 95% CI = 1.33–6.02, p = 0.007; transfeminine: aOR = 3.30, 95% CI = 1.28–8.51, p = 0.013; cisgender male: aOR = 2.30, 95% CI = 1.22–4.32, p = 0.010). Notably, none of the individual-level covariates (age, racial identity, non-binary identity, STI/HIV history, number of sexual partners within the past year) were significantly associated with participants’ odds of engaging in sexual behavior with a partner of unknown STI/HIV status in the past 12 months.

Definitions OR = odds ratio, aOR = adjusted odds ratio, 95% CI = 95% Wald confidence interval.

1. Results from generalized estimating equations (GEE) using data from 141 transmasculine adults who reported a total of 259 sexual partnerships.

2. Percentages may not sum to 100% due to missing data (missing data ranged from 1 to 3 observations across variables).

3. aORs reflect adjusted odds of reporting “I do not know” versus “yes/no.”

4. Includes a sexual partner previously diagnosed with either HIV, human papillomavirus (HPV), or herpes simplex virus (HSV) type 1 or 2.

5. Includes Asian, African American/Black, Multiracial, or Native Hawaiian or other Pacific Islander.

6. Includes genderqueer, masculine of center, gender non-conforming, non-binary, and bigender.

7. Includes ever being diagnosed with HIV, HPV, Chlamydia, Trichomoniasis, Syphilis, Gonorrhea, genital herpes, hepatitis B, hepatitis C, or another STI.

8. Included transgender female (male-to-female, MtF) or nonbinary/genderqueer and assigned-male-at-birth (AMAB).

9. Includes transgender male (female-to-male, FtM) or nonbinary/genderqueer and assigned-female-at-birth (AFAB).

Discussion

This study found that TM adults in casual sexual partnerships and those in partnerships with cisgender men, TM individuals, or TF sexual partners had significantly higher odds of having engaged in sexual behavior with a partner of unknown STI/HIV status in the past 12 months. Among TM adults, the type of partnership and partner gender - rather than the number of sexual partners - were associated with engaging in sexual behavior with a partner of unknown STI/HIV status. Though a dearth of research surrounds the STI/HIV disclosure practices of TM adults, this study finding does corroborate prior research on HIV disclosure and partnership types in other samples. One prior study with a cisgender heterosexual men and women study population reported that men were 6 times as likely, and women were 2.5 times as likely, to disclose their HIV status to a stable (i.e., main) sex partner relative to a casual partner.[15] One study with men who have sex with men (MSM) found that 28% engaged in sex with a person of unknown HIV status (n = 81).[16] Another study reported that more than 66% of MSM had casual sexual partners and 36% did not know their partner’s STI/HIV status (n = 11,766).[17] Research on the determinants and contexts of disclosure for TM and their sexual partners is needed, including characteristics of the individual being disclosed to and the individual disclosing. Further research is needed to investigate whether TM adults in sexual partnerships with cisgender men, TM individuals, or TF partners - relative to cisgender women - experience interpersonal power dynamics that influence their ability to inquire about a partner’s STI/HIV status. For example, intimate partner violence (IPV) has been associated with non-disclosure of HIV serostatus.[18] When examining the interpersonal barriers TM adults face while ascertaining a sexual partner’s STI/HIV status, investigators may benefit from incorporating a measure of intimate partner violence into their study design and analysis.

Findings from the current study should be interpreted in the context of its limitations. First, with a convenience sample of predominantly young (under 30 years-old) and White participants, study findings are limited in their generalizability. The current analysis did not adjust for education level or socioeconomic status as possible confounders. Second, although we explored factors associated with responding “I do not know” versus a definitive “yes/no” to whether a sexual partner had a prior STI diagnosis, we lacked self-report STI/HIV data from TM adults’ sexual partners.While “yes/no” responses convey a higher level of confidence surrounding the knowledge of a sexual partner’s STI/HIV status, it is unclear whether participants’ perceptions truly reflect their partners’ STI/HIV histories or statuses. Additionally, by not collecting data about how participants knew their partners’ STI/HIV histories (e.g., initiate a conversation, get tested together, assumed), we were unable to ascertain the validity of their beliefs. Future research may benefit from exploring factors associated with TM adults being able to accurately ascertain a sexual partner’s STI/HIV status. Third, due to social desirability bias, participants may have underreported their own STI/HIV histories and the histories of their sexual partners. Fourth, the current analysis did not account for PrEP use, condom/barrier use, or HPV vaccination status, which may factor into a TM adult’s decision to engage in sexual behavior with a partner of unknown STI/ HIV status. Finally, given the low lifetime prevalence of STIs/HIV in the sample (i.e., approximately 20%), findings from the current study may lack generalizability to populations at greater risk for STIs/HIV.

Results from the current study may inform future public health efforts to improve sexual health communication and STI/HIV disclosure between TM adults and their sexual partners.Future guidelines targeted toward reducing TM adults’ STI/HIV-related risk should address ways to safely initiate discussions about STI/HIV status with sexual partners, especially when the sexual partnership is casual and the sexual partner is not a cisgender woman. Such intervention discussion could also integrate information on biobehavioral prevention (e.g., PrEP to prevent HIV) to reduce risk for STIs and HIV in TM and their sexual partners. Findings from the current analysis highlight the importance of evaluating both individual-level and partnership-level factors associated with TM adults’ STI/HIV risk, particularly the configuration of the sexual partnership (e.g., casual sex, monogamous relationship) and the sexual partner’s gender identity.