Study design and setting
The purpose of the Trans*National study was to measure HIV incidence and risk factors for HIV acquisition among trans women in four cities worldwide. Data collection occurred from 2016 to 2018. We analyzed survey data on hate crime victimization from the SFBA site (N=629).
Study sample and recruitment process
Study eligibility criteria were: 1) 18 years of age or older, 2) resident of the SFBA, 3) assigned male sex at birth and identify as a gender other than male, and 4) spoke English or Spanish. Long-chain peer-referral was used for recruitment. Study procedures are discussed in detail elsewhere (37). The initial participants, or “seeds”, came from diverse backgrounds by race/ethnicity, education, and HIV status. Seeds were asked to recruit up to three self-identified trans women from their social network. For every successful recruit enrolled into the study, the participant recruiter received $20 USD. All participants provided informed consent. After consent was provided, interviewers administered surveys with a handheld computer. Participants received $50 USD for completing the survey.
Interviews were administered face-to-face by trained study staff at the study site. Measures were developed in conjunction with community advisors and while conducting formative research with trans women to identify constructs they find most relevant to their lives. Study measures assessed for this study were as follow:
Outcome variables. For the present analysis, there were two outcomes of interest: 1) self-reported experience of transphobic hate crimes (Have you ever been a victim of a transphobic hate crime?), which was reported as “yes/no”, and 2) report of transphobic hate crime incident to the police (Did you file a police report?), which was recorded as “none”, “some”, or “all”. For this analysis, report of transphobic hate crime incident to the police was dichotomized as ever/never. If answering yes to experiences of transphobic hate crimes, a further question was asked about what type, including robbery, physical assault, sexual assault, or battery with weapon specifically motivated by prejudice against trans people.
Sociodemographic and other independent variables. Demographic variables were age, gender identity, sexual orientation (gay, bisexual, or straight), race/ethnicity, marital status, educational level, income, immigration, incarceration history, housing status, and sex work. Gender was categorized as female, transgender female/transwoman, genderqueer, and androgynous. Race/ethnicity was categorized as African American, Latina, White, or Multiracial/Other (which included Asian/Pacific Islander and Native American due to small sample sizes in these groups).
Ever undocumented as an immigrant, childhood and adult homelessness, and history of sex work were dichotomized (yes/no). Participants’ current housing was assessed as own house, renting a house/apartment/room, single room occupancy, homeless/temporary/unstable housing, or other. Incarceration history (number of times incarcerated) was dichotomized as yes if incarcerated one or more times and no if never. For our analysis, we used the following question from the Experiences of Discrimination (EOD) Scale (38,39), “Have you ever experienced discrimination, been prevented from doing something, or been hassled or made to feel inferior because of your gender identity or presentation, or race ethnicity, or color from the police or in the courts?” Participants who indicated “yes” were asked if this experience was related to gender identity or presentation, race/ethnicity, or both.
Gender identity/passing. Gender identity passing (“How well do you feel you pass in society as cis-gender female?”) and gender identity questioning (“How often do you feel you are being clocked (or have your gender identity questioned)?”) were categorized based on a 4-point Likert scale (not at all, a little bit, somewhat, and a lot).
Analyses were conducted using Stata 16 (StatCorp. 2017, College Station: TX). Demographic characteristics were summarized using means for continuous variables, and proportions for categorical variables. Chi-squared analyses tested racial differences with experiences of transphobic hate crimes and police report of the crimes, hate crime type, experiences of discrimination by police/courts, incarceration, gender passing, and other independent variables. Bivariate logistic regression models assessed associations between independent variables (i.e., demographics, gender affirmation, immigration status) and each study outcome (‘ever experienced hate crimes’ and ‘reporting hate crime experiences to police’). Statistical significance was set at P-value <0.05.
The Internal Review Board of the University of California San Francisco reviewed and approved the protocol for the study. Study methods were carried out in accordance with relevant ethical guidelines and regulations.