Given the gaps in the literature regarding risk and protective factors for mental health morbidity in TM adults in particular, the aims of this study are to: 1) describe the socio-demographic characteristics, discrimination and violence, and resilience and social support experienced by TM adults; and 2) identify risk and protective factors for adverse mental health status in this traditionally under-researched group. These findings can be used to develop interventions that address individual, interpersonal, and societal risk factors, as well as to leverage protective factors to improve mental health status for TM adults.
This is a secondary analysis of data collected from 150 TM individuals enrolled in a Boston-based biobehavioral sexual health study between March 2015 and September 2016. Main findings from the sexual health study have been reported elsewhere . Participants were recruited for a single visit through convenience sampling methods (recruitment flyers, medical provider and staff referrals, community outreach, social media, community listserv posts, and word of mouth referrals) [32, 33]. Individuals interested in participation completed a brief screening survey in-person or by phone. Individuals were eligible if they met the following criteria: age 21 to 64 years ; assigned a female sex at birth, now with a masculine spectrum gender identity; have a cervix; have been sexually active within the past 3 years (sexual partner(s) of any gender); able to speak and understand English; and willing and able to provide verbal informed consent. Individuals were provided with a $100 for study participation.
The Trans Masculine Sexual Health Collaborative at Fenway Health, a federally-qualified community health center that serves the lesbian, gay, bisexual, and transgender (LGBT) community in Boston, Massachusetts, conducted the study . A Community and Provider Task Force comprised of 10 individuals was convened to provide guidance to the investigative team, specifically advising the team on methods and ensuring that all aspects of the study (e.g., design, instruments, protocols, procedures, recruitment, branding, website, implementation, interpretation and dissemination of findings) were gender-affirming.
First, participants provided verbal consent (to ease paperwork burden) to participate in the study, as approved by the Institutional Review Board (IRB) at Fenway Health. Next, participants completed a self-administered quantitative survey via an electronic tablet which included socio-demographics, resilience and social-support, discrimination and violence, and mental health. Following the survey, participants completed a clinical visit, followed by an exit interview, both of which are described elsewhere . All study activities were approved by the IRB.
Age was assessed continuously in years. Participants were asked to check all that apply regarding their race/ethnicity: White; American Indian/Alaskan Native; Asian; Native Hawaiian or other Pacific Islander; Black/African American; other race/ethnicity. Participants reporting more than one race/ethnicity were coded as such. Participants with a race/ethnicity other than White were coded as being a person of color (POC) to maximize sample size for analysis. Participants were asked whether they were Hispanic or Latino. Additionally, participants were asked to indicate their current gender identity. Participants who identified as a man, male, transgender man, female-to-male (FtM), trans man, man of transgender experience, or trans masculine were coded as having a binary gender identity. Participants who identified as genderqueer, gender non-conforming, non-binary, or another non-binary gender identity were coded as non-binary.
Participants reported the highest level of education they had completed; responses were dichotomized as high school degree or equivalent (i.e., GED, high school diploma, trade school certificate) or some college or more (i.e., some college, Associate’s degree, undergraduate bachelor’s degree, some graduate school, or graduate degree). Participants were asked if they were unemployed versus employed (either full or part time). Annual household income was assessed continuously and coded as low ($32,000 or less, representing 300% of the 2013 federal poverty level) or over $32,000 . Participants also had the option to indicate “don’t know” or “prefer not to answer” when reporting their annual household income. Participants were also asked their current relationship status and could check all apply to the following responses: single; partnered; civil union; married; separated; divorced; widowed; other. Participants who indicated that they were partnered, in a civil union, married, or in another committed relationship, and were not single, were coded as yes, partnered, otherwise no. Unstable housing was assessed by asking participants if they had been homeless or unstably housed in the past 12 months (yes/no). Participants were also asked if they had trouble accessing gender-affirming care in the past 12 months (yes/no).
Resilience, social support, and acceptance
Resilience was assessed using 4 items from the Brief Resilience Scale [29, 38]. This scale assesses ability to recover from challenging life events. Item responses are on a scale from 1 = strongly disagree to 5 = strongly agree and included items such as “I tend to bounce back quickly after hard times.” Appropriate item responses were reverse scored and averaged to derive a single scale score (α = 0.86). Higher scores indicated greater ability to recover from stress (i.e., greater resilience).
Social support was measured using an abbreviated version of the Medical Outcomes Study (MOS) Social Support Survey first developed to assess social support in patients with chronic illness . The scale contained 4 items each representing a domain of social support (α = 0.87). Specifically, participants were asked how often, on a scale from 1: None of the time to 5: All of the time, someone is available: To help with daily chores if you are sick? (domain: tangible support); To understand your problems? (domain: emotional/informational support); To get together with you for relaxation (domain: positive social interaction); and To love you and make you feel wanted? (domain: affectionate support). Scores were summed and transformed to be on a scale from 0 to 100, as is standard practice for this measure, with higher scores indicating greater social support .
Self-Acceptance was assessed with a single item taken from the Rosenberg Self-Esteem Scale by asking participants, on a scale from 1: No acceptance to 10: Complete acceptance, “I take a positive attitude toward myself” [40, 41]. Transgender Acceptance was assessed with a single item by asking participants, on a scale from 1: No acceptance to 10: Complete acceptance, how much they believe they are accepted by other transgender people.
Discrimination & Violence
Intimate partner violence (IPV) was measured using 11 yes/no items assessing whether participants had experienced physical IPV (e.g., whether a partner had ever pressured, physically forced, threatened or blackmailed participants to have sex when they did not want) ; sexual IPV (e.g., whether a partner had ever hit, slapped, pushed, shoved, kicked, beat up, choked, burned, used a weapon, or thrown something at a participant in their lifetime) ; transgender-related IPV (e.g., whether a partner ever threatened to out them, destroyed their gender affirming treatments) in their lifetime and the past 12 months . Participants experiencing one or more form of IPV were coded as experiencing IPV.
Everyday Discrimination was assessed with the 11-item Everyday Discrimination Scale, which assesses the frequency of participants’ experiences of everyday discrimination in the past 12 months on a Likert scale ranging from 0 = never to 4 = very often [23, 44, 45]. Sample items include: “You have been treated with less courtesy than other people;” “You have received poorer service than other people at restaurants or stores” [44, 45]. Items were summed, and scores ranged from 0 to 39, with higher scores indicating higher levels of everyday discrimination experiences (α = 0.93).
PTSD symptoms were assessed with a four-item screening scale designed for primary care settings, the Primary Care – PTSD (PC-PTSD) . Participants were asked: “In your life, have you ever had any experience that was so frightening, horrible, or upsetting that, in the past month you: (1) Have had nightmares about it or thought about it when you did not want to? (2) Tried hard not to think about it or went out of your way to avoid situations that reminded you of it? (3) Were constantly on guard, watchful, or easily startled? (4) Felt numb or detached from others, activities, or your surroundings?” Participants responded to each item using binary (yes/no) responses. The response items were summed (α = 0.87) and dichotomized based on a clinical cutoff score of two or more events, validated to ICD-9 PTSD diagnosis or presence of PTSD treatment visit .
Participants completed the 18-item Brief Symptom Inventory [47, 48]. Participants were asked to indicate, from 0 (not at all) to 4 (extremely), how much they felt distressed by six symptoms in the past seven days including “feeling lonely” or “feelings of worthlessness” (α = 0.83) . The six items were summed and standardized using T scores (mean of 50 and a standard deviation of 10). Scores are interpreted by comparison to age-appropriate norms. Raw scores are converted to T scores using tables provided in the BSI manual, and T scores of 63 or higher indicated a positive case for depression . This tool has been used in transgender samples previously .
Participants completed the 18-item Brief Symptom Inventory. Participants were asked to indicate, from 0 (not at all) to 4 (extremely), how much they felt distressed by six symptoms in the past seven days including “feeling fearful” or “feeling tense of cleaned up” (α = 0.88). The six items were summed and standardized using T scores (mean of 50 and a standard deviation of 10). A score of 63 or higher indicated a positive case for anxiety. This tool has also been used in samples of transgender populations .
Non-suicidal Self-injury (NSSI)
Participants were asked whether they had engaged in self-injurious behavior (e.g., burning, cutting, severe scratching, hitting) without lethal intent in the past 12 months. This item is based on the Self-Injury Questionnaire, which has also been used in transgender samples previously [15,16,17, 49].
Statistical analyses of quantitative survey data were conducted in SAS 9.4. Means and frequencies were calculated to describe participant characteristics and assess missingness. Bivariate and multivariable logistic regression models were fit to examine the association between participant characteristics, risk and protective factors, and four mental health status: 1) PTSD; 2) depression; 3) anxiety; and 4) NSSI. Participants reporting “don’t know” or “prefer not to answer” were treated as missing and excluded. Missing data ranged from 1% (n = 148) for depression and 11% (n = 134) for anxiety. Independent variables significant at p < 0.10 in bivariate analyses were included in the multivariable models. Backward selection was used for the multivariable models with significance determined at the p < 0.05 level. Area under the curve statistics were calculated to determine the overall predictive validity of the multivariable model. AUC statistics were evaluated based on the following criteria: outstanding (> = .90); excellent (.80–.89); or acceptable (.70–.79) predictive validity .