Division of Adolescent and School Health Why the Disparities: The First National Look at HIV-Related Risk Behaviors Among Sexual Minority Male High School Students, United States, 2015 Laura Kann PhD, Emily O’Malley Olsen MSPH, Tim McManus MS, and Stephanie Zaza MD AIDS 2016 July 20, 2016 Thank you. Before I begin - I would like to thank my co-authors that you see listed on this slide. All of us are from the Division of Adolescent and School Health at the CDC in Atlanta.

Background 22% of all new HIV diagnoses in the US in 2014 occurred among 13-24 year olds Most of these diagnoses occurred among males who have sex with males (MSM) Insufficient information exists about The number of young sexual minority males nationwide How HIV-related risk behaviors vary between young sexual minority males and non-sexual minority males In the United States, an estimated 22% of all new HIV diagnoses occur among 13-24 year olds and most of these diagnoses occur among males who have sex with males. This makes young males – an important focus for HIV prevention efforts. Key to these prevention efforts is a clear and accurate understanding of the number of young sexual minority males nationwide and an understanding of how HIV-related risk behaviors vary (or not) between young sexual minority and non-sexual minority males. To date, insufficient information has existed about both of these issues in the US.

Purpose To use 2015 National Youth Risk Behavior Survey (YRBS) data to describe The number of sexual minority males in US high schools How HIV-related risk behaviors vary between sexual minority and non-sexual minority males Sexual behaviors directly related to HIV infection Drug use directly related to and associated with HIV infection Consequently, the purpose of this presentation is to describe (1) the number of sexual minority males in US high schools and (2) how HIV-related risk behaviors vary between sexual minority and non-sexual minority males.

Turning to our Methods…

Sample 3-stage cluster sample design Schools selected with probability proportional to size Classes selected randomly All students in sampled classes included Nationally representative of public and private school students in grades 9-12 in the US The 2015 National YRBS used a 3-stage cluster sample design with schools selected with probability proportional to enrollment in grades 9-12 followed by random selection of classes. All students in sampled classes were included in the survey. The 2015 National YRBS data are representative of public and private school students in grades 9-12 in the US.

Sampling Outcome Response rates School – 69% Student – 86% Overall – 60% Sample size - 15,624 This slide shows the school, student, and overall response rates from the 2015 national YRBS. More than 15,000 students participated.

Survey Administration Anonymous and voluntary Parental permission following local procedures Self-administered, computer-scannable questionnaire with 99 multiple choice questions Completed in one class period The National YRBS is an anonymous and voluntary survey. Parental permission was obtained prior to survey administration following local procedures. The students were provided with a self-administered, computer scannable questionnaire with 99 multiple choice questions that they completed on their own during one regular class period. IRB approval for the survey was obtained at CDC.

Questionnaire Content Behaviors that contribute to unintentional injuries and violence Tobacco use Sexual behaviors Alcohol and other drug use Unhealthy dietary behaviors Inadequate physical activity The National YRBS has always measured the 6 categories of behaviors that you see on this slide.

Questionnaire Content Behaviors that contribute to unintentional injuries and violence Tobacco use Sexual behaviors Alcohol and other drug use Unhealthy dietary behaviors Inadequate physical activity This presentation will focus on selected sexual behaviors and alcohol and other drug use behaviors that are most closely linked with HIV infection.

Questionnaire Content: Measurement of Sexual Minority Status Sex of sexual contacts Sexual identity In addition, starting with the 2015 national YRBS questionnaire 2 questions were added for the first time to measure sexual minority status. The first focused on sex of sexual contacts. The second question focused on sexual identity regardless of whether or not the student had had sexual contact. Both constructs were deemed important for identifying sexual minority students and for gaining the most complete understanding of how HIV-related risk behaviors vary between sexual minority and non-sexual minority students. Because adolescence is a time when many people develop and explore their sexuality it is possible for dissonance to occur between sexual identity and sex of sexual contacts.

Questionnaire Content: Measurement of Sex of Sexual Contacts During your life, with whom have you had sexual contact? I have never had sexual contact Females Males Females and males This is the question used to measure sex of sexual contacts. Sexual contact was not defined but presumably students considered a wide range of sexual activities including, but not limited to, kissing, touching, and sexual intercourse as well as involuntary sexual activities.

Questionnaire Content: Measurement of Sex of Sexual Contacts During your life, with whom have you had sexual contact? I have never had sexual contact Females Males Females and males Sexual minority For this analysis, male students who selected response option C (males) or response option D (females and males) were combined into a single sexual minority subgroup

Questionnaire Content: Measurement of Sex of Sexual Contacts During your life, with whom have you had sexual contact? I have never had sexual contact Females Males Females and males Sexual minority and compared to male students who selected response option A (I have never had sexual contact) and to male students who selected response option B (females).

Questionnaire Content: Measurement of Sexual Identity Which of the following best describes you? Heterosexual (straight) Gay or lesbian Bisexual Not sure This is the question used to measure sexual identity.

Questionnaire Content: Measurement of Sexual Identity Which of the following best describes you? Heterosexual (straight) Gay or lesbian Bisexual Not sure Sexual minority For this analysis, male students who selected response option B (gay) or response option C (bisexual) were combined into a single sexual minority subgroup

Questionnaire Content: Measurement of Sexual Identity Which of the following best describes you? Heterosexual (straight) Gay or lesbian Bisexual Not sure Sexual minority and compared to male students who selected response option A (heterosexual or straight).

Data Analysis Weighting Based on student sex, race/ethnicity, and grade in school Adjusted for non-response and oversampling of black and Hispanic students All females excluded T-tests to detect significant (p<0.05) pairwise differences in HIV-related risk behaviors by sexual identity and by sex of sexual contact subgroups Prior to data analyses - The national YRBS data were weighted and adjusted for non-response and the oversampling of black and Hispanic students. For this analysis, all females were excluded. T-tests were then used to detect significant pairwise differences in HIV-related risk behaviors by sexual identity and by sex of sexual contact subgroups

And now for the results…

Never Had Sexual Contact Distribution of sexual identity and sex of sexual contact subgroups among male high school students in the US, 2015 Sexual Identity Sex of Sexual Contacts Heterosexual Gay Bisexual Unsure Females Only Males Only Both Males and Females Never Had Sexual Contact 93.1% 2.0% 2.4% 2.6% 53.3% 1.3% 1.9% 43.6% 7,479,188 160,670 192,804 208,871 4,281,856 104,436 152,636 3,502,606 This first slide provides the first ever estimates of the number of sexual minority male students in US high schools. 2015 National Youth Risk Behavior Survey

Never Had Sexual Contact Distribution of sexual identity and sex of sexual contact subgroups among male high school students in the US, 2015 Sexual Identity Sex of Sexual Contacts Heterosexual Gay Bisexual Unsure Females Only Males Only Both Males and Females Never Had Sexual Contact 93.1% 2.0% 2.4% 2.6% 53.3% 1.3% 1.9% 43.6% 7,479,188 160,670 192,804 208,871 4,281,856 104,436 152,636 3,502,606 Of the approximately 8 million male high school students in the US, 4.4% identify as gay or bisexual. 2015 National Youth Risk Behavior Survey

Never Had Sexual Contact Distribution of sexual identity and sex of sexual contact subgroups among male high school students in the US, 2015 Sexual Identity Sex of Sexual Contacts Heterosexual Gay Bisexual Unsure Females Only Males Only Both Males and Females Never Had Sexual Contact 93.1% 2.0% 2.4% 2.6% 53.3% 1.3% 1.9% 43.6% 7,479,188 160,670 192,804 208,871 4,281,856 104,436 152,636 3,502,606 And about 3.2% have had sexual contact with only males or with both males and females. 2015 National Youth Risk Behavior Survey

Behaviors related directly and indirectly to HIV infection The next slides show estimates for sexual behaviors that are directly related to HIV infection and drug use behaviors related to HIV infection directly (when injection is the mode of administration) and indirectly (when drug use increases the likelihood of unsafe sexual behaviors). These estimates are shown for male high school students by sexual identity and by sex of sexual contact subgroups. Red arrows indicate where statistically significant differences were identified.

Percentage of male high school students who reported HIV-related sexual behaviors, by sexual identity, 2015 Ever had sexual Had intercourse Current sexual Condom use** intercourse with 4+ persons activity* Starting first with the sexual behaviors - This slides shows the percentage of male high school students by sexual identity who ever had sexual intercourse, had sexual intercourse with 4 or more persons, were currently sexually active, and who used a condom at last sexual intercourse. Sexual intercourse was not defined and could include vaginal, anal, or oral sex. None of these behaviors varied statistically between heterosexual and gay or bisexual male students. *Had sexual intercourse during the 3 months before the survey. **Among students who were currently sexually active. 2015 National Youth Risk Behavior Survey

Percentage of male high school students who reported HIV-related sexual behaviors, by sex of sexual contact, 2015 Ever had sexual Had intercourse Current sexual Condom use** intercourse with 4+ persons activity* Similarly, this slide shows the percentage of male high school students by sex of sexual contact who reported the same 4 sexual behaviors. Again, there were no statistical differences between the 2 sex of sexual contact subgroups. Note: Male students who had no sexual contact were excluded from these analyses. *Had sexual intercourse during the 3 months before the survey. **Among students who were currently sexually active. 2015 National Youth Risk Behavior Survey

Percentage of male high school students who ever used a needle to inject any illegal drug into their body,* by subgroup, 2015 Sexual Identity Sex of Sexual Contacts However, when we look at the drug use behaviors we see a different story. This slide shows the percentage of male high school students who ever used a needle to inject any illegal drug into their body one or more times during their life by sexual identity and by sex of sexual contact subgroups. Note that gay and bisexual male students were significantly more likely than heterosexual male students to have injected drugs and that male students who had sexual contact with only males or with both males and females were significantly more likely than the other 2 sexual contact subgroups to report injected drug use. *One or more times during their life. 2015 National Youth Risk Behavior Survey

Percentage of male high school students who ever used heroin, Percentage of male high school students who ever used heroin,* by subgroup, 2015 This slide shows the percentage of male high school students who ever used heroin. Gay and bisexual male students and male students who had sexual contact with only males or with both males and females were significantly more likely than the other subgroups to report heroin use. Sexual Identity Sex of Sexual Contacts *One or more times during their life. 2015 National Youth Risk Behavior Survey

Percentage of male high school students who ever used cocaine, Percentage of male high school students who ever used cocaine,* by subgroup, 2015 Sexual Identity Sex of Sexual Contacts This slide shows the percentage of male high school students who ever used cocaine. Note the same pattern of statistically significant differences. *Any form of cocaine (e.g., powder, crack, or freebase) one or more times during their life. 2015 National Youth Risk Behavior Survey

Percentage of male high school students who ever used methamphetamines,* by subgroup, 2015 This slide shows the percentage of male high school students who ever used methamphetamines. Again, the same pattern of differences exist between subgroups. Sexual Identity Sex of Sexual Contacts *One or more times during their life. 2015 National Youth Risk Behavior Survey

Percentage of male high school students who ever took prescription drugs without a doctor’s prescription,* by subgroup, 2015 Sexual Identity Sex of Sexual Contacts Finally, this slide shows the percentage of male high school students who ever took prescription drugs without a doctor’s prescription and the statistically significant differences. *Such as Oxycontin, Percocet, Vicodin, codeine, Adderall, Ritalin, or Xanax, one or more times during their life. 2015 National Youth Risk Behavior Survey

Limitations Results apply only to males who attend school 3% of all 16-17 year olds not enrolled and not completed high school Sexual minority youth might represent disproportionate percent of dropouts Extent of underreporting or overreporting cannot be determined Some students may not know their sexual identity, may be unwilling to disclose it or label themselves, or may not have understood the question As with all studies, this one has its limitations. Obviously, because this was a school-based survey, the results apply only to males who attend school and not all young males. While most young sexual minority males are in school other research also documents that sexual minority youth might represent a disproportionate percent of school dropouts. Second, as is common with surveys like the YRBS, the extent of under and over reporting cannot be determined although the survey questions demonstrate good test-retest reliability. Third, it is possible that some students may not know their sexual identity, may be unwilling to disclose it or label themselves, or may not have understood the question. However, we do not have evidence that the words used to describe various types of sexual identity are unclear to high school students in the US.

Conclusions Collecting high quality, national population-based data on the number of young sexual minority males is possible The prevalence of injected drug use among young sexual minority males is disturbing Prevalence of sexual behaviors alone that directly cause HIV infection are not causing disparities in rates of HIV diagnoses between young MSM and young males who have sex with females Other factors to be considered Sexual practices Sexual networks Social issues There are three main conclusions from these analyses … First, to support HIV-prevention efforts, it is important to have high quality, national population-based data on the number of sexual minority males in US high schools. It is hard to respond appropriately to a population that has not been counted. We now know that collecting those data is possible. Second, the prevalence of injected drug use among young sexual minority males is disturbing and a cause for great concern because of the efficiency with which injected drug use transmits not only HIV infection, but hepatitis and other diseases. Third, these analyses do not provide evidence that differences in the prevalence of sexual behaviors alone are causing the disparities in HIV diagnoses between young MSM and other young males. This finding is consistent with similar research done among older MSM. This finding also suggests that other factors are placing young MSM at substantially higher risk for HIV infection. For example, we know from other research that the frequency of receptive anal sex and the prevalence of HIV, the amount of undiagnosed and untreated HIV infection, and the prevalence of other STI among the sex partners of young MSM increases risk of HIV infection from any single act of sex. The sexual practices and sexual networks of young MSM simply put young MSM at far greater risk of infection. In addition, many young MSM suffer from social isolation, stress from self-concealment and coming out, discrimination, and even hatred which may occur at home, at school, and within their communities among both their peers and the adults responsible for their care and protection. These factors may lead to harmful coping behaviors (such as drug use as evidenced in this presentation) that can further increase risk for HIV infection.

Recommendations Address the social issues that result in social isolation, stress, and discrimination Conduct new research on injected drug use among young sexual minority males Improve access to HIV testing Improve linkage to and retention in medical care Provide more risk reduction education and HIV prevention strategies Offer more school-centered HIV prevention Continue measurement of health-risk behaviors among young sexual minority males through large, population-based surveys So – it is hard to imagine how even the best intervention technology will be able eliminate the disparities in HIV diagnoses unless these social issues are addressed. Further, new research is needed to better understand injected drug use among young sexual minority males and what can be done to minimize if not eliminate this very high risk behavior that is occurring at alarming rates. In addition, young sexual minority males need improved access to HIV testing. Those who are infected also need to be linked to treatment and retained in medical care. Further, all young sexual minority males need access to the full range of HIV prevention strategies, including condoms, biomedical interventions, as appropriate, and more school-centered HIV prevention. School programs should focus on providing a safe and supportive environment for all youth including reduced bullying and harassment, on access to youth-friendly health care, and on primary prevention of HIV-related sexual and drug use behaviors. Further, it is important to continue to measure health-risk behaviors among young sexual minority males through large, population-based surveys. Reducing HIV infection among young sexual minority males is key to reducing HIV infection in the US. Thank you.

Thank you! Thank you.