HIV Testing Patterns Suggest Need for Services Specifically Geared to Transgender Individuals

First published online:

| DOI: https://doi.org/10.1363/47e6315

At HIV prevention programs funded by the Centers for Disease Control and Prevention (CDC) in 2009–2011, the proportion of tests that detected new infections was higher among transgender individuals than among males or females.1 The vast majority of HIV tests within the transgender population were conducted among male-to-female individuals, and this group had the highest overall infection rate. Among tests performed on male-to-female transgender individuals, HIV positivity was associated with age, race or ethnicity, and participation in a number of risky behaviors.

While some research has examined how transgender individuals’ HIV risk is linked to behavioral and psychosocial characteristics, little is known about HIV testing patterns in this population. To help fill this gap, analysts used 2009–2011 data from 59 CDC-funded testing sites (health care facilities, community-based organizations, correctional facilities and other types of testing venues) to compare characteristics of HIV testing events across genders and within the transgender population. A testing event was defined as a test or series of tests (e.g., a rapid test followed by a conventional test to confirm a positive result) performed to assess an individual's HIV status; gender was determined by individuals’ responses to a single question.

Of the nearly 9.5 million testing events reported during the study period, more than 9.3 million had all required data and were included in the analysis—roughly 4.8 million among females, 4.5 million among males and 16,000 among transgender individuals (85% of them male-to-female individuals). In all gender groups, the majority of tests were among individuals in their 20s and 30s, and were performed in the Northeast and the South. Tests among males and females were predominantly performed on blacks and whites, and took place primarily in health care facilities. By contrast, those in the transgender population were most often among blacks and Latinos, and were most commonly conducted at HIV counseling and testing sites or community venues. The proportion of testing events that yielded a new HIV diagnosis was highest among transgender individuals—2.4%, compared with 0.9% among males and 0.2% among females. HIV positivity was more common in tests of male-to-female than of female-to-male transgender individuals (2.7% vs. 0.5%).

Reported levels of some risky behaviors were essentially the same in all gender groups. For example, unprotected sex was reported at approximately 60% of testing events for each group, and sex with an injection-drug user at 3–5%. However, several risky behaviors were reported at a higher proportion of testing events among transgender individuals than at others: exchange sex (11% vs. 2% each for males and females), sex with an HIV-positive partner (9% vs. 2–4%) and sex with an anonymous partner (10% vs. 2–5%).

For tests among male-to-female transgender individuals, the analysts used multiple logistic regression to identify correlates of a new HIV diagnosis. (The number of female-to-male transgender individuals was too small to permit similar calculations for that group.) Two background characteristics and three risky behaviors were significant. Testing events among those in their 20s, 30s or 40s were more likely than those among teenagers to yield positive results (odds ratios, 2.5–2.8); infection rates were greater for blacks, Hispanics and other minority groups than for whites (4.6, 2.6 and 1.9, respectively). The odds of a new HIV diagnosis also were elevated in tests performed on individuals who reported unprotected sex (1.9), sexual activity with an HIV-positive partner (1.5) or injection-drug use in the past 12 months (2.0).

The analysts note that the generalizability of their data is limited and that they could not determine the extent to which a testing event represented a single individual. They also acknowledge that certain exclusion criteria may have created biases in the analytic sample and that the question used to determine gender identity may have resulted in some misclassification. While emphasizing the importance of improving the ability of data collection and surveillance systems to identify transgender individuals, they also note that their findings suggest a need to expand HIV prevention services targeted at this high-risk population.

D. Hollander


1. Habarta N et al., HIV testing by transgender status at Centers for Disease Control and Prevention–funded sites in the United States, Puerto Rico, and US Virgin Islands, 2009–2011, American Journal of Public Health, 2015, 105(9):1917–1925.