Structural and Psychosocial Syndemic Conditions and Condomless Anal Intercourse Among Transgender Women - National HIV Behavioral Surveillance Among Transgender Women, Seven Urban Areas, United States, 2019-2020

Abstract

Psychosocial and structural syndemic conditions, including polydrug use and experiencing homelessness, frequently co-occur and might jointly increase HIV risk. Limited studies have assessed racial and ethnic differences in exposure to syndemic conditions and behaviors associated with HIV transmission among transgender women. This report examines the relation between syndemic conditions and condomless anal intercourse (CAI) among transgender women in seven urban areas in the United States to develop HIV prevention interventions for transgender women. During 2019-2020, transgender women in seven urban areas were recruited using respondent-driven sampling for a biobehavioral survey. Reported syndemic conditions (psychosocial: polydrug use, sexual violence, and psychological distress; structural: homelessness, incarceration, and exchange sex) were summed to create a syndemic score. Using modified Poisson regression to account for RDS, the study assessed whether the strength of the association between syndemic score and CAI differed by race and ethnicity. To assess additive interaction, the relative excess prevalence owing to interaction (REPI) and 95% CIs for selected pairs of syndemic conditions on CAI prevalence stratified by race and ethnicity were estimated. Of 1,348 transgender women (Black = 546, White = 176, and Hispanic = 626), 55% reported CAI; and 24% reported ≥3 syndemic conditions. Reporting additional syndemic conditions was associated with CAI for White, Hispanic, and Black participants. The association was significantly stronger for White than Black and Hispanic participants. Limited significant superadditive interactions were found, although the majority were between structural syndemic conditions. Racial and ethnic differences in REPI estimates were observed. Reporting more syndemic conditions was associated with increased CAI across racial and ethnic groups, demonstrating that HIV prevention efforts for transgender women should address structural and psychosocial syndemic conditions. Results differed by race and ethnicity, indicating that syndemic-focused interventions for transgender women should be tailored to racial and ethnic groups.

FIGURE 1

Conceptual model of analysis showing factors contributing to condomless anal intercourse — National HIV Behavioral Surveillance Among Transgender Women, seven urban areas, United States, 2019–2020. * Atlanta, GA; Los Angeles, CA; New Orleans, LA; New York City, NY; Philadelphia, PA; San Francisco, CA; and Seattle, WA.

FIGURE 2

Estimated condomless anal intercourse as a function of syndemic score and race and ethnicity for Black or African American, White, and Hispanic or Latina transgender women^† — National HIV Behavioral Surveillance Among Transgender Women, seven urban areas,^§ United States, 2019–2020. Abbreviations: Black = Black or African American; Hispanic = Hispanic or Latina. * Persons of Hispanic or Latina (Hispanic) origin might be of any race but are categorized as Hispanic; all racial groups are non-Hispanic. ^† N = 1,309 participants had an HIV-negative or HIV-positive National HIV Behavioral Surveillance HIV test result; identified as Black, White, or Hispanic; and had no missing data. ^§ Atlanta, GA; Los Angeles, CA; New Orleans, LA, New York City, NY; Philadelphia, PA; San Francisco, CA, and Seattle, WA.

FIGURE 3

Relative excess prevalence owing to interaction on condomless anal intercourse estimates between syndemic conditions^,† — National HIV Behavioral Surveillance Among Transgender Women,^§ seven urban areas,^¶ United States, 2019–2023**^,†† Abbreviations: CAI = condomless anal intercourse; REPI = relative excess prevalence owing to interaction. * Models account for respondent-driven sampling methodology by clustering on recruitment chain and adjusting for urban area. Models also control for age, education level, relationship status, health insurance, and National HIV Behavioral Surveillance HIV test result. ^† An REPI estimate >0 indicates superadditivity between syndemic conditions on CAI. A REPI estimate <0 indicates subadditivity effects between syndemic conditions on CAI. ^§ Persons of Hispanic or Latina (Hispanic) origin might be of any race but are categorized as Hispanic; all racial groups are non-Hispanic. ^¶ Atlanta, GA; Los Angeles, CA; New Orleans, LA, New York City, NY; Philadelphia, PA; San Francisco, CA, and Seattle, WA. ** REPI estimates with a 95% CI that does not include zero are marked with a triangle (∆). ^†† N = 1,309 participants had an HIV-negative or HIV-positive NHBS HIV test result; identified as Black, White, or Hispanic; and had no missing data.