Health Education Research, Vol. 14, No. 2, 177-183,
April 1999
© 1999 Oxford University Press
Transgender HIV prevention: implementation and evaluation of a workshop
Program in Human Sexuality, Department of Family Practice and Community Health, Medical School, University of Minnesota, 1300 South Second Street, Suite 180, Minneapolis, MN 55454-1015, USA
 |
Abstract |
|---|
 Top Abstract IntroductionInterventionEvaluationResultsDiscussionReferences |
|---|
Virtually no HIV prevention education has specifically targeted the transgender community. To fill this void, a transgender HIV prevention workshop was developed, implemented and evaluated. A 4 h workshop, grounded in the Health Belief Model and the Eroticizing Safer Sex approach, combined lectures, videos, a panel, discussion, roleplay and exercises. Evaluation using a pre-, post- and follow-up test design showed an increase in knowledge and an initial increase in positive attitudes that diminished over time. Due to the small sample size (N = 59) and limited frequency of risk behavior, a significant decrease in unsafe sexual or needle practices could not be demonstrated. However, findings suggested an increase in safer sexual behaviors such as (mutual) masturbation. Peer support improved significantly. Future prevention education should make special efforts to target the more difficult-to-reach, high-risk subgroups of the transgender population.
|
Introduction |
|---|
|
TopAbstractIntroductionInterventionEvaluationResultsDiscussionReferences |
|---|
The transgender community comprises a diverse group of individuals—including cross-dressers or transvestites, transexuals, bi-gender persons, drag queens and kings, and (fe)male impersonators—who cross or transcend culturally defined categories of gender. The prevention of HIV among this population is considered critical because: (1) prevalence indicators suggest high levels of HIV infection, (2) behavioral studies identify the two highest risk behaviors for HIV—unprotected receptive anal intercourse and needle sharing—to be disproportionately common and (3) transgender people remain a marginalized group with little access to prevention education.
Reporting statistics for HIV and AIDS do not distinguish transgender people as a risk group. As such, the prevalence of HIV infection among the transgender population as a whole has not been quantified. However, studies of transgender sex workers have revealed alarmingly high rates of HIV/AIDS, higher than among non-transgender male and female sex workers. For example, Elifson et al. (Elifson et al., 1993
) found that 68% of 53 transvestite sex workers in Atlanta were HIV-positive, compared to 27% of 152 non-transvestite male sex workers (Elifson et al., 1989). Similarly, Gattari et al. (Gattari, et al., 1991) found that 86% of 22 drug-using transgender sex workers in Rome were HIV-positive compared to 32% of 396 non-transgender injecting drug users. A few studies show that HIV/AIDS has affected other subgroups of the transgender population as well. For example, a female-to-male transexual was diagnosed with AIDS (Coleman and Bockting, 1988), two (2%) of 91 transexuals at the Rosenberg Clinic, Galveston, Texas tested HIV-positive (Avery et al., 1995) and three (8%) of 39 male-to-female transexuals at the Free University Hospital, Amsterdam, the Netherlands, died of AIDS (Kesteren et al., 1997).
Despite the high prevalence of HIV infection among transgender sex workers, their awareness of AIDS was found to be low (Ratnam, 1986). Identified risk behaviors include multiple partners, frequent receptive anal sex, irregular condom use, both injecting and non-injecting drug use, and needle sharing. For example, among 27 transgender sex workers in Italy, Tirelli et al. (Tirelli et al., 1991) found a median number of 260 clients (range 10–1000) per month. In Singapore, Kok et al. (Kok et al., 1990) compared 23 transexual sex workers with non-transgender controls matched for race, age and sex, and found that 16 (70%) compared to one (4%) had anal sex with clients, during which only five (22%) used condoms; in Atlanta, 46 (87%) of 53 transgender sex workers reported engaging in receptive anal sex (Elifson et al., 1993). Finally, in Milan, seven (29%) of 24 Italian transexual sex workers used intravenous drugs (Galli et al., 1991). Only one study has assessed knowledge, attitudes and HIV risk behavior among a sample not limited to sex workers (Avery et al., 1995). In a clinical population of 99 transexuals, basic knowledge was high and number of sexual partners was low. Twenty-four percent of sexually active respondents engaged in receptive anal sex, 19% reported condom use and 4% reported intravenous drug use.
Very few HIV prevention efforts have specifically targeted the transgender community. In Brazil, transgender sex workers were recruited as health agents to act as a bridge between sex workers and an HIV prevention project (Peterson and Szterenfeld, 1992). In the US, outreach has occurred (Bockting et al., 1998). In Singapore, Ratnam (Ratnam, 1990) evaluated the effects of health education on 71 transexual sex workers. The intervention consisted of a pamphlet with HIV prevention information and at least four individual counseling sessions over a 1 year period. Comparison of pre-intervention and follow-up questionnaires demonstrated a significant increase in AIDS awareness; however, no significant change in condom use was found. The authors concluded that low socioeconomic status and educational level might account for failure to comprehend what constitutes safer sex.
In light of both the documented HIV risk in some transgender groups and the scarcity of targeted prevention education, we developed a transgender HIV prevention workshop and piloted it in the Minneapolis/St Paul metropolitan area.
|
Intervention |
|---|
|
TopAbstractIntroductionInterventionEvaluationResultsDiscussionReferences |
|---|
The intervention consisted of a 4 h prevention–education workshop, with three large-group presentations and two small-group meetings. Presentations were facilitated by the first author and transgender community leaders; small groups by trained peer educators. Over 6 months, three of these workshops were offered on Saturday afternoons at our center. The first presentation, based on the Health Belief Model (Janz and Becker, 1984
), aimed to increase perceived threat by outlining the scope of the epidemic, personalizing HIV/AIDS, and discussing fears, myths and transgender-specific HIV risks. It included a video featuring transgender celebrity Kate Bornstein (Bockting et al., 1992) and a panel of transgender persons with HIV/AIDS. A small-group discussion followed, to process feelings, concerns and experiences related to transgender and HIV/AIDS, and to identify personal risk. The second presentation, based on the Eroticizing Safer Sex approach (Palacios-Jimenez and Shernoff, 1986), provided information about HIV transmission, affirmed safer sex and addressed risk reduction in the use of injection paraphernalia. It included sexually explicit videos with transgender models eroticizing safer sex and condom use (Jaccoma et al., 1990). Small-group exercises and roleplay followed, to practice condom use and sexual negotiation. Building on HIV risks identified in the first small group, participants helped each other develop individualized prevention plans resulting in concrete commitments to be safer. The final presentation provided referral for HIV testing, and fostered personal and community empowerment. |
Evaluation |
|---|
|
TopAbstractIntroductionInterventionEvaluationResultsDiscussionReferences |
|---|
Method
Subjects
Participants were recruited through advertisements in community media, postings on the Internet, mailings to clients of health providers, posters and outreach. Recruitment was challenging because of the considerable number of transgender persons who remain `in the closet' and hidden, as evidenced by phone calls we received from people interested but hesitant to come forward. Seventy-four of 86 workshop participants enrolled in the study, of whom 71 (96%) completed both pre- and post-tests.
Instruments
The pre-workshop questionnaire assessed demographics; AIDS knowledge; attitudes toward AIDS, sex, safer sex and condoms; risk behavior; and community support. The post-workshop questionnaire assessed changes in knowledge and attitudes. A 2 month follow-up questionnaire assessed consolidation of the gained knowledge and changed attitudes, changes in risk behavior and community support.
Knowledge was assessed using six true–false questions (Ross, 1988a). Inter-item correlations were low (0.00–0.31); responses on each item were independent from each other. Attitudes were measured using five-point Likert scales. Five items examined fear of AIDS, homophobia and erotophobia; six items examined condom attitudes (Ross, 1988b). Cronbach's 
of this 11-item attitude scale was 0.73, indicating acceptable inter-item agreement (Nunnally and Bernstein, 1994). In addition, two single items assessed overall attitude toward safer sex and intention to use condoms. Risk behavior was assessed through a 20-item survey of sexual and needle practices, prostitution, and alcohol and drug use. This survey assessed the occurrence, frequency and number of partners with whom these behaviors were practiced in the last 2 months. Community support was assessed through the question: `How much of your social time is spent with people who are transgender?'.
Procedure
The study was approved by the University of Minnesota's Human Subjects Committee. Upon arrival at the workshop, the purpose of the study was explained and consent obtained. Participants completed a registration letter, with the option to provide a false name (for both anonymity and transgender sensitivity) and an address for follow-up. Subsequently, participants completed the pre-workshop questionnaire. Immediately following the workshop, they completed the post-workshop questionnaire. Two months after the workshop, follow-up questionnaires were mailed to all participants. A reminder was sent to initial non-respondents.
Analysis
A repeated measures design tested significance of changes between pre-, post- and follow-up tests, using SPSS (SPSS Inc., Chicago, IL). Because all Greenhouse–Geisser statistics were non-significant, univariate F statistics were used, followed by paired t-tests to compare means. Due to the small sample size and small number of pair-wise comparisons, Bonferroni correction was not used. Thus, the experiment-wise may be higher than 0.05. Alpha was set at 0.05 (two-tailed); P < 0.10 was seen as evidence of a trend.
A knowledge score was computed by adding correct responses on the six knowledge questions. Recoding appropriate items, scores on the attitude scale were summed; a higher summary score indicates a more desirable attitude. To assess behavior change, items from the behavior survey were grouped into three categories: (1) safer sexual practices (masturbation alone or mutually, vaginal and anal intercourse with a condom), (2) unsafe sexual and needle practices (vaginal and anal intercourse without a condom, shared needles, shared injecting hormones), and (3) behaviors associated with increased risk (received or gave payment for sex, had sex while drunk or high, used intravenous drugs). For each category, responses to the question whether or not participants had done these behaviors in the last 2 months were added into a summary score.
|
Results |
|---|
|
TopAbstractIntroductionInterventionEvaluationResultsDiscussionReferences |
|---|
Background of participants
Fifty-nine participants self-identified as transgender; the remaining 12 were non-transgender significant others and excluded from analysis. Mean age was 41.76 years (SD = 9.61). Like 93.7% of Minnesota residents, almost all participants (97%) were White. The majority self-identified as male-to-female transexual or cross-dresser and were more attracted to women than to men. About half had been tested for HIV; three reported being HIV-positive (Table I
). Thirty-seven (63%) of these 59 transgender participants completed the follow-up questionnaire. Analysis of differences between completers and non-completers (those lost to attrition) showed no significant differences in age, education, religion and number of sexual partners.
|
Knowledge
AIDS knowledge proved moderately high at pre-test; percentage of correct responses per item ranged from 53 to 100%. Comparison of correct responses per subject at pre-, post- and follow-up tests revealed a significant effect (F[2,66] = 3.79, P = 0.028). Knowledge increased significantly between pre- and follow-up tests (t[35] = 2.88, P = 0.007). A trend toward increased knowledge between post- and follow-up tests was found (t[35] = 1.75, P = 0.090); the increase in knowledge between pre- and post-tests was not significant (Table II
).
|
Attitudes
Generally, attitudes proved positive at pre-test. Comparison of attitude scale summary scores at pre-, post- and follow-up tests revealed a significant effect (F[2,60] = 3.71, P = 0.030). Attitudes were significantly more positive at post-test than at pre-test (t[45] = 3.00, P = 0.004) and significantly less positive at follow-up than at post-test (t[32] = 2.87, P = 0.007); between follow-up and pre-tests, no significant difference was found (Table II
). The effect for the single item assessing attitude toward safer sex was significant (F[2,64] = 3.32, P = 0.042); a trend was found toward a more favorable attitude toward safer sex between pre- and post-tests (t[48] = 1.92, P = 0.061); however, this attitude decreased significantly to pre-test levels between post- and follow-up tests (t[33] = 2.26, P = 0.030). For intention to use condoms, no significant effect was found.
Risk behavior
At pre-test, 24 participants (41%) reported having a primary sexual partner with a median length of 7 years. Of these, two-thirds (16) reported never or seldom practicing safer sex, while one-third (8) reported always or often practicing safer sex in their primary relationship; three reported that their relationship was not monogamous. Four participants (7%) reported having had sex with more than one partner—two (3%) with both male and female partners—and only one participant (2%) admitted having had unsafe sex with more than one partner. Because the majority (74%) reported no sexual partners at all, the frequency of sexual activity of the sample as a whole was low. None reported having shared needles.
At follow-up, number of partners remained low; no significant differences between pre- and follow-up tests were found. One participant reported sex with two male partners and one participant reported unsafe sex with seven female partners. Again, no needles were shared.
Comparisons of (1) safer sexual practices, (2) unsafe sexual and needle practices, and (3) behaviors associated with increased risk failed to demonstrate any significant differences across time (Table II). Cross-tabulation analysis using Fisher's exact tests revealed two trends: masturbation alone and mutual masturbation with a partner increased (P = 0.073 and P < 0.001, respectively).
Community support
Social time spent with other transgender people increased significantly between pre- and follow-up tests (t[33] = 2.86, P = 0.007). Sixty-one percent of participants at follow-up compared to 38% at pre-test reported spending most or some of their social time with other transgender people.
|
Discussion |
|---|
|
TopAbstractIntroductionInterventionEvaluationResultsDiscussionReferences |
|---|
We successfully delivered an HIV prevention workshop targeting the transgender community. Evaluation showed that AIDS knowledge was significantly higher at follow-up than at pre-test. It is unclear, however, whether this change can be attributed to the workshop, because analysis failed to show a significant increase between pre- and post-workshop knowledge, and no control group was used. One explanation is that participants did receive information during the workshop, but needed time to digest and assimilate it. Alternatively, the workshop may have raised AIDS awareness, which led participants to find out more about HIV/AIDS following the workshop.
Attitudes improved significantly between pre- and post-tests, yet declined to pre-test levels in the 2 months following the workshop. Thus, unlike knowledge, attitudes appear more immediately malleable but less changeable over time. Theories of attitude change emphasize the need for ongoing support and awareness training to maintain attitude change (Valdiserri et al., 1992). Therefore, continuing education, building support networks, and improving social acceptance of transgender identity and sexuality appear important tasks in sustaining positive attitudes toward HIV/AIDS and safer sex.
At pre-test, most participants reported limited sexual activity. A number reported sex with one partner and a few reported high-risk behaviors with multiple partners. No needles were shared. This does not mean that HIV is not relevant to the target population; clinical experience and focus group research indicated that HIV/AIDS has already significantly affected the local transgender community (Bockting et al., 1998). Rather, the intervention may not have adequately reached those at high-risk; few reported active participation in sex work, few reported injecting drug use and socioeconomic status tended to be high, limiting the generalizability of our findings. It is also possible that participants under-reported risk behavior. Although they were assured of anonymity, some (especially those in therapy at our clinic) may have wanted to present themselves in as favorable a light as possible.
Evaluation failed to show significant risk behavior change, which can be attributed to the overall low frequency of sexual activity and small sample size. Replication of the study with a larger sample from a broader cross-section of the population and the use of a control group are recommended to more adequately evaluate the workshop's effectiveness in changing risk behavior.
Participants' social time spent with transgender peers increased significantly as a result of the workshop. This suggests that for socially isolated individuals and fragmented communities like the transgender population, a community empowerment stage model of HIV prevention is appropriate. Our findings suggest that if you start with HIV prevention education, positive effects of the intervention can be measured, but dissipate in part as individuals begin to discover community. It might be more effective to (1) bring people together around a common topic, (2) develop a sense of community and (3) then promote HIV prevention.
|
Acknowledgments |
|---|
We acknowledge the following transgender community representatives who helped plan this study: Aaron, Celie, Debbie, Dotty/Kevin, Jane, Mira, Sander and Susan. We thank Eli Coleman, Deb Finstad, Anne Marie Moore, Sonia Patten and Anne Marie Weber-Main for their assistance and support. This study was conducted in collaboration with the City of Lakes Crossgender Community, the Minnesota Freedom of Gender Expression, the Aliveness Project and the Minnesota AIDS Project, and funded by the American Foundation for AIDS Research (grant 100108-12-EG).
|
References |
|---|
|
TopAbstractIntroductionInterventionEvaluationResultsDiscussionReferences |
|---|
Avery, E. N., Cole, C. M. and Meyer, W. J. (1995) Transsexuals and HIV/AIDS risk behaviors. Poster presentation at the XIV Harry Benjamin International Gender Dysphoria Symposium, Kloster Irsee, Bavaria, Germany, September 9.
Bockting, W. O. (Producer), Grandell, S. (Editor) and Bornstein, K. (Script) (1992) Gender Defender with AIDS-Be-Gone Infomercial [Video]. (Available from the Program in Human Sexuality, 1300 South Second Street, Suite 180, Minneapolis, MN 55454.)
Bockting, W. O., Robinson, B. E. and Rosser, B. R. S. (1998) Transgender HIV prevention: a qualitative needs assessment. AIDS Care, 10, 505–526.[ISI][Medline]
Coleman, E. and Bockting, W. O. (1988) `Heterosexual' prior to sex-reassignment, `homosexual' afterwards: a case-study of a female-to-male transexual. Journal of Psychology and Human Sexuality, (2), 69–82.
Elifson, K. W., Boles, J., Posey, E., Sweat, M., Darrow, W. and Elsea, W. (1993) Male transvestite prostitutes and HIV risk. American Journal of Public Health, 83, 260–262.
Elifson, K., Boles J, Sweat, M., Darrow, W., Elsea, W. and Green, R. (1989) Seroprevalence of human immuno- deficiency virus among male prostitutes. New England Journal of Medicine, 321, 832–833.[ISI][Medline]
Galli, M., Esposito, R. and Antinori, S. (1991) HIV-1 infection, tuberculosis, and syphilis in male transexual prostitutes in Milan, Italy. Journal of Acquired Immune Deficiency Syndromes, 4, 1006–1008.
Gattari, P., Rezza, G., Zaccarelli, M., Valenzi, C. and Tirelli, U. (1991) HIV infection in drug using transvestites and transexuals. European Journal of Epidemiology, 7, 711–712.[ISI][Medline]
Jaccoma, A., Armstrong, J. and Sprinkle, A. (1990) Linda, Les, and Annie: The First Female-to-Male Transsexual Love Story [Video]. (Available from Linda, Les, and Annie, PO Box 1024, LIC, NY 11101.)
Janz, K. and Becker, M. H. (1984) The Health Belief Model: a decade later. Health Education Quarterly, 11, 1–47.[ISI][Medline]
Kesteren, P. J. van, Asscheman, H., Megens, J. A. and Gooren, L. J. (1997) Mortality and morbidity in transsexual subjects treated with cross-sex hormones. Clinical Endocrinology, 47, 337–342.[Medline]
Kok, L. P., Ho, M. L., Heng, B. H. and Ong, Y. W. (1990) A psychosocial study of high risk subjects for AIDS. Singapore Medical Journal, 31, 573–582.[Medline]
Nunnally, J. C. and Bernstein, I. H. (1994) Psychometric Theory, 3rd edn. McGraw-Hill, New York.
Palacios-Jimenez, L. and Shernoff, M. (1986) Facilitators Guide to Eroticizing Safer Sex: A Psychoeducational Workshop Approach to Safer Sex Education. Gay Men's Health Crisis, New York.
Peterson, C. and Szterenfeld, C. (1992) Organizing a project with community-based health agents recruited from prostitutes in Rio de Janeiro. Public Health, 106, 217–223.[Medline]
Ratnam, K. V. (1986) Awareness of AIDS among transexual prostitutes in Singapore. Singapore Medical Journal, 27, 519–521.[Medline]
Ratnam, K. V. (1990) Efficacy of health education programme on awareness of AIDS among transexuals. Singapore Medical Journal, 31, 33–37.[Medline]
Ross, M. W. (1988a) Distribution of knowledge of AIDS: a national study. Social Science and Medicine, 27, 1295–1298.
Ross, M. W. (1988b) Attitudes towards condoms as AIDS prophylaxis in homosexual men: dimensions and measurement. Psychology and Health, 2, 291–299.[Medline]
Tirelli, U., Vaccher, E., Covre, P., Corso, C., Serraino, D. and Rezza, G. (1991) Condom use among transvestites in Italy. Journal of Acquired Immune Deficiency Syndromes, 4, 302–303.
Valdiserri, R. O., West, G. R., Moore, M., Darrow, W. W. and Hinman, A. R. (1992) Structuring HIV prevention service delivery systems on the basis of social science theory. Journal of Community Health, 17, 259–269.[Medline]
Received on December 14, 1996; accepted on May 22, 1998
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Nemoto, D. Operario, J. Keatley, H. Nguyen, and E. Sugano Promoting Health for Transgender Women: Transgender Resources and Neighborhood Space (TRANS) Program in San Francisco Am J Public Health, March 1, 2005; 95(3): 382 - 384. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Nemoto, D. Operario, J. Keatley, L. Han, and T. Soma HIV Risk Behaviors Among Male-to-Female Transgender Persons of Color in San Francisco Am J Public Health, July 1, 2004; 94(7): 1193 - 1199. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||