A cross-sectional analysis of psychosocial and structural barriers and facilitators associated with PrEP use among a sample of transgender women in Chicago, IL

Study sample

Between January 2019 and February 2020, 99 transgender women were enrolled using a convenience sampling approach, in an evaluation of the TransLife Care (TLC) project at Chicago House and Social Service Agency in Chicago, Illinois. The aim of the evaluation was to assess the efficacy of TLC to reduce risk for HIV acquisition. TLC was designed to meet the needs of racially/ethnically diverse urban transgender women by addressing the psychosocial and structural drivers of HIV, including housing, employment, legal aid and health services [12]. Eligibility criteria included: (a) identifying as transgender, transfeminine, and/or female and assigned male sex at birth; (b) ≥ 17 years of age; (c) self-reported history of sex with men in the past 4 months; (d) negative/non-reactive HIV screening test at baseline; (e) able to speak/understand English; (f) willing and able to provide informed consent; (g) intend to reside in the local area throughout the 8-month follow-up period; and (h) had no exposure to any component of the TLC intervention in the prior 4 months. Individuals were excluded if they were unable to provide informed consent due to severe mental or physical illness, or intoxication at the time of interview (those excluded could re-screen if symptoms resolved).

Described herein is a cross-sectional analysis of data collected at the baseline enrollment visit. All study procedures were approved by the Institutional Review Board of Ann & Robert H. Lurie Children’s Hospital of Chicago with a waiver of parental permission for participants aged 17 and written consent obtained for all participants.

Data collection and measures

At the point of enrollment, participants completed a baseline questionnaire via computer-assisted interviewing, which included demographic characteristics, health behaviors, and psychosocial factors. Data for the enrollment visit were collected on-site at the TLC. Participants received $50 token of appreciation for completion of the baseline assessment.

Demographics

Sociodemographic information collected included age (in years), race/ethnicity, sexual orientation/identity, highest level of education, and current income.

Barriers

Structural factors. We assessed recent homelessness with a two-part question: “In your lifetime, have you ever been homeless at all? That is, you slept in a shelter for homeless people, on the streets, at a friend or relative’s house for a few nights or weeks, or another place not intended for sleeping?” (Yes, No); “In the past 4 months, were you homeless at any time?” (Yes, No). History of arrest and incarceration were assessed each with a single question: “Have you ever been arrested by the police?” (Yes, No); “Have you ever been in jail, prison, police lock-up, immigrant detention or juvenile detention?” (Yes, No).

Psychosocial factors. We assessed depression and anxiety symptoms with the 10-item version of the Center for Epidemiological Studies Depression Scale (CESD-10) [13] and the 7-item Generalized Anxiety Disorder (GAD-7) [14], respectively, and exposure to transgender-specific victimization with a 10-item victimization scale [15], adapted for trans women. We assessed substance use with the World Health Organization Alcohol, Smoking, and Substance Involvement Screening Test (ASSIST), which includes assessment of 10 substances and related problems. Following coding guidelines for the CESD-10, GAD-7 and ASSIST, respectively, participants were coded as having high versus low depressive symptoms (CESD-10; ≥10, < 10), minimal, mild, moderate or severe anxiety symptoms (GAD-7; 0–4, 5–9, 10–14, 15–21), and low, medium or high substance use (alcohol: 0–10, 11–26, 27–36; cannabis, stimulants, opioid/sedatives: 0–3, 4–26, 27–36) for each substance.

Facilitators

Structural factors. We assessed health insurance status with a single question (“What kind of insurance do you currently use to pay for health care?” Responses were coded as: any insurance (including Medicaid/Medicare, private, other) versus none. We assessed stable housing with a single item, “Which of the following best describes your current living situation? By current living situation, we mean where have you been staying during the past seven days?” Responses coded to reflect “your own place, a room, apartment, or house that is your home” versus all others (e.g., jail/prison, drug treatment, transitional housing, hotel/motel). We assessed current employment with the item, “Please indicate which of the following is true for you regarding your current work status? The responses were coded with “working for pay at a job or business,” versus all others.

Psychosocial factors. We measured psychological gender affirmation using a 5-item measure of comfort and satisfaction with affirmation on a 5-point scale (e.g., “How comfortable are you with people knowing that you are transgender? Not at all comfortable, slightly comfortable, moderately comfortable, very comfortable, extremely comfortable) [17]. We measured transgender-specific collective self-esteem (CSES) with a 16-item measure of thoughts and feelings related to being part of the transgender community on a 7-point agreement scale (i.e., strongly disagree to strongly agree) with four sub-scales: membership (how “good” or “worthy” they feel as a member of the group), private (how good they feel about their group), public (how others are perceived to view the group) and identity (how important the group is to their self-concept) [18].

PrEP indications

We measured recent (past 30 days) condomless anal/vaginal sex (Yes, No) and recent history of HIV-positive sexual partners with items from the AIDS Risk Behavior Assessment (ARBA), adapted for trans women [16]. We measured exchange sex with two sequential items referencing anal and vaginal sex respectively: “How many partners have you had anal (insertive or receptive)/vaginal sex with in the past month? This includes sex with or without a condom?” “How many did you have anal/vaginal sex with, in exchange for things you needed (like money, drugs, food, shelter, etc.)?” Responses were coded to reflect any exchange sex (anal/vaginal) in the past month. We measured history of sexually transmitted infections (STIs) with a single question: “Have you ever been told by a doctor or nurse that you had a sexually transmitted infection, other than HIV?” (Yes, No).

PrEP use

We measured ever having used PrEP, used PrEP in the last four months, and used PrEP in the last month as well as adherence to PrEP with a series of items used in a prior study of trans women [3]: “Have you ever taken HIV medication before sex because you thought it would lower your chances of getting HIV (also known as PrEP)?” (Yes, No); “Have you taken PrEP in the last 4 months?” “Have you taken PrEP in the last month?” “Please indicate whether or not you have taken PrEP on each day during the past month, beginning with yesterday (timeline follow-back approach with calendar; coded on a scale of 0-100% adherent). For analysis, we created binary variables for PrEP use in the recent four months and in the past one month, respectively, versus none/not recent, which included those who had never used PrEP and those who had used PrEP previously, (i.e., > 4 months or > 1 month ago, but not recently). We also created a 3-category variable including never used, past use (having ever taken PrEP but not in the past four months), and recent use (having taken PrEP in the past four months). We present the results descriptively using all three PrEP recall variables. We used the binary recent four month use for multivariable analysis because the small sample size limited our ability to conduct multinomial analysis. For analysis of predictive accuracy, we used past one month PrEP use as the outcome because it corresponded most closely with the assessment period of the PrEP indications.

Analysis

To assess bivariate associations between barriers and facilitators and recent PrEP use, we calculated chi-square statistics for categorical variables and Kruskal-Wallis tests for continuous variables. We used multivariable Poisson regression with robust error variance to compute univariable and multivariable prevalence ratios for associations between barriers and facilitators and recent 4-month PrEP use, controlling for age, race/ethnicity, insurance, and income. To assess the accuracy of PrEP indications alone and in combination for identifying PrEP users, we computed sensitivity (i.e. the proportion of those on PrEP with any indication for PrEP), specificity (i.e., the proportion of those not on PrEP without an indication for PrEP), positive predictive value (i.e., the proportion taking PrEP among those with PrEP indications), and area under the receiver operating characteristic (ROC) curve (an overall measure of predictive accuracy, where 0.5 indicates no better than chance and higher values indicate higher predictive performance). All analyses were conducted in Stata version 17.0 (Stata Corp, College Station, TX).

留言 (0)

沒有登入
gif