Sex Education and HIV Testing Among Young Men Who Have Sex With Men: Findings From the 2006–2010 and 2011–2015 National Survey of Family Growth

Abstract

Background: 

Young men who have sex with men (YMSM) are disproportionally affected by HIV and exhibit low levels of HIV testing. The purpose of this study was to examine the relationship between sex education and subsequent HIV testing among YMSM.

Methods: 

Data from sexually active men aged 15–24 years at interview in the 2006–2010 or 2011–2015 National Survey of Family Growth. Sex education included 3 contexts [formal institutions (eg, schools), parents, and health care providers] and specific topics. Outcome variables were ever- and recent-HIV testing. Multivariate models were adjusted for sociodemographics, and data were weighted to account for the complex survey sampling design.

Results: 

A total of 323 sexually active YMSM met inclusion criteria. Overall, 42.4% had ever-tested for HIV and 16.9% tested in the previous 12 months. YMSM were more likely to have ever-tested for HIV if they talked with a parent/guardian about how to prevent HIV/AIDS [adjusted prevalence ratio (aPR) = 1.48; 95% confidence interval (CI): 1.07 to 2.06], talked with a health care provider about how HIV/AIDs is transmitted (aPR = 1.64; 95% CI: 1.13 to 2.38), sexually transmitted diseases (aPR = 1.49; 95% CI: 1.02 to 2.19), condom use (aPR = 1.61; 95% CI: 1.13 to 2.30), and the importance of HIV testing (aPR = 1.83; 95% CI: 1.22 to 2.73).

Conclusions: 

Tailored sex education by parent(s) and health care providers related to HIV/AIDS seems to significantly increase the likelihood of HIV testing among YMSM.

The HIV epidemic continues to receive attention from a variety of national directives, including Healthy People 2020.¹ Nevertheless, adolescent and young adult populations continue to be severely affected by HIV. In 2015, 22% of the 39,500 new HIV diagnoses in the United States occurred among those aged 13–24² years. Approximately 51% of the 60,900 HIV-seropositive adolescents and young adults aged 13–24 years were living undiagnosed in 2013, thus unaware of their serostatus and ability to transmit HIV.²

Among youth and young adult populations, HIV is more prevalent among gay and bisexual men.^2,3 In 2015, 81% of the 8807 adolescents and young adults aged 15–24 years diagnosed with HIV were gay and bisexual men.² Although self-identification of sexual orientation is 1 way to categorize high-risk populations, HIV transmission is not dependent on sexual orientation but rather on the high-risk behaviors such as unprotected same-sex intercourse (ie, without a condom).⁴ As such, national HIV surveillance began to collect data in the early 1990s using survey questions specifically designed to capture information about high-risk populations, such as men who have sex with men (MSM) and more recently young MSM (YMSM), regardless of self-identified sexual orientation.⁴

The Centers for Disease Control and Prevention (CDC) recommends annual HIV testing for sexually active MSM,⁵ and HIV testing every 3 to 6 months for persons with multiple sex partners and YMSM.^5,6 Furthermore, the CDC recommends that HIV tests be a part of routine health care for all adolescents and be used in all health care settings, unless the patient opts-out of screening.^5,6 Despite national testing recommendations, research suggests that HIV testing among YMSM remains suboptimal.^1,3,7,8 To improve HIV testing behaviors among YMSM, Healthy People 2020 now includes the objectives of increasing the proportion of adolescents and adults who have ever been tested for HIV from 66.9% to 73.6%, and increasing the proportion of MSM who report having been tested for HIV in the previous 12 months from 62.2% to 68.4%.¹

Sex education in a variety of settings may increase youth engagement in HIV testing.^9–12 Formal sex education is instruction that takes place in a school, youth center, church, or other community setting¹³ and has been shown to influence adolescent sexual health behaviors, such as condom use.^9,14 A 2014 Youth Risk Behavioral Surveillance System (YRBSS) study found that adolescents who received HIV/AIDS education were 1.5 times more likely to have ever-tested for HIV.¹¹ Other work examined school-based sex education and HIV testing and found that sexual minority men were significantly less likely to report having received HIV/AIDS education than their nonminority peers but were almost 2 times more likely to report HIV testing.¹⁵ Similar findings have been reported by other studies among both sexes.^9,11 As a result, it is still unclear whether formal sex education focused on broad sexual health topics (eg, condom use) and HIV-/AIDS-related health messages impacts HIV testing among YMSM the same as their peers.

By contrast, informal sex education refers to instruction that takes place outside of a classroom setting through various interpersonal relationships. Parents/guardians are a crucial informal source of knowledge and support for adolescents and can shape their sexual attitudes and behaviors during the early life course.^16–18 Adolescence and emerging adulthood are crucial life-span development periods in which youth begins to develop and construct personal beliefs and values. Moreover, many adolescents engage in risk-taking behaviors to develop their identity. Parents, families, and family cohesion may mitigate high-risk behaviors during adolescence.¹⁹ For example, parent–adolescent sex communication can positively impact safer sex behaviors (eg, condom use),²⁰ reduce sexual risk-taking behaviors among adolescents (eg, unprotected anal intercourse and drug use before sex),¹⁶ and even predict future visits to a health care provider.¹⁷ However, most parent–adolescent sex communication research has been conducted with heterosexual adolescents,^10,21,22 which may not generalize to all youth. Parent–adolescent dynamics may be strongly influenced by sexual orientation disclosure and, consequently, parental reactions to disclosure.²² Adolescents who receive supportive reactions from parents after sexual orientation disclosure often report overall better health and sexual health decision-making.²³

Patient–health care provider communication, such as provider recommendations for HIV testing,^24,25 sexual orientation disclosure to a provider,²⁶ and discussion about HIV prevention and behaviors in the health care setting, is another form of informal sex education.^7,27 Health care providers are in ideal positions to recommend HIV testing because they have the ability to conduct in-house HIV screenings and discuss sexual health issues.²⁷ Patient–provider sexual health communication has been linked to variety of positive health behaviors, including HIV testing.²⁷ Providers who feel comfortable discussing sexual orientation and sexual behaviors are more likely to recommend HIV tests to YMSM.^25,27 However, many YMSM report that provider conversations lack inclusive language and do not contain the right questions about sexual health issues impacting YMSM.²⁸ Although health care visits in the previous 12 months are associated with HIV testing,²⁹ MSM/YMSM rarely receive recommendations for HIV tests from a provider (<60% of the time).^24,25,27,29 Sexual health topic discussions may mediate the relationships between visits to a health care provider, receiving an HIV testing recommendation, and actual engagement in HIV testing.^7,27 Although these previous studies revealed a connection between discussions about general HIV/AIDS prevention and HIV testing among MSM/YMSM, research on specific discussion topics, such as condom use and the importance being tested for HIV, is scarce.

Little research has examined the relationship between sex education and HIV testing specifically among YMSM populations. The purpose of this study is to examine the association between sex education and HIV testing in a nationally representative sample of sexually active YMSM aged 15–24 years. We hypothesized that YMSM who received formal or informal sex education on how to use a condom, sexually transmitted diseases (STDs), and/or HIV/AIDS will be more likely to have ever been tested for HIV than those who did not receive education, and that YMSM who had discussed HIV/AIDS transmission and/or getting tested for HIV with a health care provider will be more likely to have ever been tested for HIV than those who did not have these discussions with a provider.

METHODS

The National Survey of Family Growth (NSFG) is designed to collect information related to marriage, divorce, pregnancy, use of reproductive health care/services, and general health of men and women.^13,30 NSFG began in 1973 and is ongoing.³⁰ The 2006–2010 and 2011–2015 NSFG were conducted using a multistage, stratified, clustered sampling design to survey a nationally representative household sample population of noninstitutionalized men and women aged 15–44 years that reside in the United States.^13,30 A total of 10,403 men completed interviews during the 2006–2010 NSFG, resulting in a 75% response rate.¹³ After a 15-month gap in interviewing, 4815 men completed the 2011–2013 NSFG and 4506 men completed the 2013–2015 NSFG, with response rates of 72.1% and 67.1%, respectively.¹³ As recommended by the CDC, we used caution when combining the 2006–2010 and 2011–2015 data files because weighted estimates derived from the combined data file may be misleading if estimates from the separate data files vary significantly.¹³ Additional information related to the NSFG sampling design and methodology is described elsewhere.^13,30

For this study, we included 2006–2010 and 2011–2015 NSFG participants who were male, aged 15–24 years at the time of the interview, and had ever engaged in sexual intercourse with another male (eg, receptive anal sex, insertive anal sex, or oral sex). The primary outcome of interest was ever-tested for HIV, measured by the question, “not counting tests you may have had as part of blood donations, have you ever been tested for HIV (yes/no)?” If the participant responded “yes,” he was then asked whether he was tested for HIV in the previous 12 months (ie, within 1 year of the interview date).

Participants were asked seven yes/no questions related to formal sex education before the age of 18 years on the following seven sex-related topics: (1) how to say no to sex; (2) methods of birth control; (3) where to get birth control; (4) STDs; (5) how to prevent HIV/AIDS; (6) how to use a condom; and (7) waiting until marriage to have sex. Participants were also asked whether they have ever talked to a parent or guardian before the age of 18 years about each of those seven topics. For both formal and parental sex education, we created a summary variable to categorize sex education topics as any (defined as ≥1 topic) vs. none (no topics). Formal sex education topics of where to get birth control, how to use a condom, and abstinence were only assessed in the 2011–2015 NSFG. In addition, the sex education by parents topic of abstinence was only assessed in the 2013–2015 NSFG.

Patient–provider HIV/AIDS communication was assessed by asking participants, “has a doctor or other medical care provider ever talked with you about HIV, the virus that causes AIDS (yes/no).” If “yes,” then participants were asked to identify which of the 10 following topics were covered in past discussions: (1) how HIV/AIDS is transmitted; (2) STDs; (3) the correct use of condoms; (4) needle cleaning/using clean needles; (5) dangers of needle sharing; (6) abstinence from sex; (7) reducing the number of sexual partners; (8) condom use to prevent HIV or STD transmission; (9) safe sex practices; and (10) getting tested and knowing your HIV status. Getting tested and knowing your HIV status was only assessed during the 2011–2015 NSFG.

Key sociodemographic variables included age, race/ethnicity, income as percent of poverty level, health insurance status during past 12 months, and sexual orientation. Sexual orientation was assessed by asking participants to answer the question, “do you think of yourself as…” by choosing one of the following answer choices (1) heterosexual or straight; (2) homosexual or gay; or (3) bisexual. Health behavior included STD testing during the previous 12 months.

Data were managed and analyzed using SAS v9.4 (SAS Institute, Inc., Cary, NC). First, we determined the unweighted sample size and frequency of all variables. The estimates for all variables from the 2 separate survey cycles did not vary significantly (Table 1 and Table 2). As a result, we combined 2006–2010 and 2011–2015 data into a total population and calculated weighted estimates for each variable of interest.¹³ Then, we then compared the distribution of all variables in the total population by HIV testing (ever vs. never) using χ² statistics.

TABLE 1.: Descriptive Characteristics of YMSM Aged 15–24 Years Who Have Ever had Same-Sex Intercourse*

TABLE 2.: Receipt of Formal and Informal Sex Education Among Men Aged 15–24 Years Who Have Ever had Same-Sex Intercourse*

We generated unadjusted and adjusted prevalence ratios (PRs) and corresponding 95% confidence intervals (CIs) for the total sample using PROC GENMOD with Poisson error distributions and log link functions. The GENMOD procedure was selected because logistic regression models and odds ratios are poor approximations of the prevalence ratio when the outcome is not sufficiently rare.³¹ Potential covariates for the adjusted models were first identified using the results of χ² tests (P < 0.05) (Table 1). All models were adjusted for age, race/ethnicity, sexual orientation, total number of sexual partners in entire life, and family income as percent of poverty level. Unless otherwise noted, all analyses were weighted to account for the complex multistage, stratified, clustered sampling of NSFG data.

RESULTS

A total of 323 YMSM aged 15–24 years met the inclusion criteria from the 2006–2010 and 2011–2015 NSFG. During 2006–2010 and 2011–2015, 42.39% of YMSM aged 15–24 years had ever-tested for HIV and only 16.89% had tested for HIV in the previous 12 months (Table 1). Over three-fifths of YMSM were non-Hispanic white (61.29%), 41.13% self-identified as heterosexual, and 32.44% self-identified as homosexual.

Almost all had received formal sex education on at least one sex-related topic. Most YMSM received formal sex education on how to say no to sex (76.08%), methods of birth control (66.46%), STDs (91.67%), how to prevent HIV/AIDS (86.88%), and how to use a condom (2011–2015 estimate: 52.39%). By contrast, less than half reported sex education by parents on how to say no to sex (36.20%), methods of birth control (38.36%), STDs (48.89%), how to use a condom (40.06%), and how to prevent HIV/AIDS (42.11%; Table 2).

Patient–health care provider conversations about sexual health was the least common method of sex education, as only 34.70% of YMSM had discussed an HIV-/AIDS-related topic with a provider. The most common patient–provider discussion topics were how HIV/AIDS is transmitted (27.30%), STDs (26.54%), “safer sex practices” (25.46%), and condom use (23.28%; Table 2). Only 23.45% of YMSM during the 2011–2015 NSFG period discussed the topic of getting tested and the importance of knowing your HIV status with a provider.

In the total sample, YMSM who talked with a parent/guardian about how to prevent HIV/AIDS were not significantly more likely to have ever tested for HIV as indicated by the initial unadjusted prevalence ratio (PR = 1.42; 95% CI: 0.97 to 2.07). However, this association became statistically significant after adjusting for age, race/ethnicity, sexual orientation, total number of sexual partners in entire life, and family income [adjusted prevalence ratio (aPR) = 1.48; 95% CI: 1.07 to 2.06; Table 2]. By contrast, formal sex education topics and all other conversations with a parent/guardian did not seem to be significantly associated with HIV testing at the α = 0.05 level (Table 3).

TABLE 3.: Associations Between Sex Education and Ever Being Tested for HIV Among Men Aged 15–24 Years Who Have Ever had Same-Sex Intercourse*

YMSM were significantly more likely to have ever-tested for HIV if they had talked to a health care provider about at least one HIV-/AIDS-related topic (PR = 2.06; 95% CI: 1.33 to 3.20), no matter which topic. The prevalence ratio decreased but remained significant even controlling for covariates (aPR = 1.85; 95% CI: 1.27 to 2.71). Sex education from providers on how HIV/AIDs is transmitted (PR = 1.86; 95% CI: 1.13 to 2.38), STDs (PR = 1.65; 95% CI: 1.04 to 2.61), the correct use of condoms (PR = 1.71; 95% CI: 1.12 to 2.57), abstinence from sex (PR = 1.80; 95% CI: 1.20 to 2.70), condom use to prevent HIV or STD transmission (PR = 1.88; 95% CI: 1.24 to 2.79), and “safer sex” practices (eg, abstinence, condom use, etc) (PR = 2.05; 95% CI: 1.37 to 3.06) were all associated with HIV testing. The prevalence ratios for these topics decreased but remained statistically significant after controlling for covariates: how HIV/AIDs is transmitted (aPR = 1.64; 95% CI: 1.13 to 2.38), STDs (aPR = 1.49; 95% CI: 1.20 to 2.90), the correct use of condoms (aPR = 1.63; 95% CI: 1.14 to 2.34), abstinence from sex (aPR = 1.54; 95% CI: 1.10 to 2.15), condom use to prevent HIV or STD transmission (aPR = 1.61; 95% CI: 1.13 to 2.30), and “safer sex” practices (eg, abstinence, condom use, etc) (aPR = 1.86; 95% CI: 1.33 to 2.61). By contrast, conversations about reducing your number of sexual partners was associated with HIV testing (PR = 1.86; 95% CI: 1.24 to 2.79) but was not significant after controlling for covariates (aPR = 1.08; 95% CI: 0.83 to 1.39). The topic of getting tested and knowing your HIV status seemed to have one of the strongest associations with HIV testing (PR = 2.31; 95% CI: 1.51 to 3.54) and remained statistically significant after controlling for demographic covariates (aPR = 1.83; 95% CI: 1.22 to 2.73).

DISCUSSION

The purpose of this study was to determine the relationship between sex education and HIV testing in a nationally representative sample of YMSM in the United States. Our findings suggest that the proportion of YMSM aged 15–24 years who have ever-tested for HIV and tested for HIV in the previous 12 months falls well below the Healthy People 2020 objectives of increasing the portion of youth and MSM who have been tested for HIV in the previous 12 months.¹ This finding is a major public health concern because over half of the adolescents and young adults aged 13–24 years living with HIV are living undiagnosed and, as a result, are unaware of their ability to transmit HIV.²

The CDC suggests that educating students about HIV/AIDS and other STDs in a formal setting could increase adolescents' likelihood of being tested for HIV and STDs.³ In our study, we found that most YMSM received formal sex education on methods of birth control, STDs, and how to prevent HIV/AIDS, but there was no association between sex education in formal contexts and HIV testing after controlling for confounding variables. Additional research is needed to examine the quality of formal sex education,³ as quality may mediate and/or strengthen the relationship between receipt of formal sex education on HIV/AIDS prevention and HIV testing. For example, past research has shown that teacher training is significantly associated with the provision of comprehensive sex education and increased fidelity of implementation.^32,33

In a sensitivity analysis, we summed the total of number of sex education topics for both formal and parental sex education separately to determine whether the quantity of topics, conceptualized as a proxy for comprehensiveness, would influence likelihood of HIV testing. We found that almost 90% YMSM had discussed at least 3 topics or more in a formal setting. Sex education by parents was not as common, as one-third had never discussed a sex topic with a parent. We found no association between number of topics and HIV testing for both formal sex education and sex education by parents (results not shown). More work is needed to determine whether the quantity or comprehensiveness of the sex education impacts HIV testing among YMSM.

Families, and in particular parents and guardians, are integral in the delivery of sexual health messages and interventions to adolescents and young adults, including YMSM.^16,18 In previous research, the most common topics discussed with parents among male adolescents were abstinence, STDs, condom use, and HIV/AIDS.^1,12 YMSM who had talked with a parent/guardian about how to prevent HIV/AIDS were approximately 50% more likely to have ever-tested for HIV. These findings suggest that parent-based sex education does play a role in increasing adolescent engagement in HIV testing and that the topics that directly impact the YMSM population, such as HIV/AIDS prevention and condom use, are linked to HIV testing. Additional research is needed to examine specific subtopics of HIV/AIDS prevention to determine which pieces of information are linked to HIV testing. Moreover, additional information is needed to assess strength of parent–adolescent connection and perceptions of parental warmth/care. Strength of parent–child relationship and warmth/care may mediate the relationship between HIV/AIDS discussions and youth engagement in prevention measures, such as HIV testing.^21,34 It is also recommended that additional topics previously linked to HIV testing, such as parent–child discussions about same-sex intercourse behavior,³⁵ also be assessed in national survey efforts. These findings suggest the need for innovative family-based HIV-prevention strategies and interventions.

Patient–provider sexual health communication has been linked to variety of positive health behaviors, including HIV testing.²⁷ We found that less than 40% of respondents had ever talked to a provider about HIV/AIDS prevention, which is consistent with findings from pervious MSM/YMSM studies in which less than 30% had ever discussed getting tested for HIV with a provider.^7,27 In our study, YMSM were significantly more likely to have ever-tested for HIV if they talked to a provider about how HIV/AIDs is transmitted, STDs, the correct use of condoms, dangers of needle sharing, abstinence, condom use to prevent HIV or STD transmission, “safer sex” practices (eg, abstinence, condom use, etc), and getting tested and knowing your HIV status. Our findings suggest that health care providers can influence YMSM engagement in HIV testing behavior by discussing health topics that directly impact YMSM. The results emphasize the need for opt-out HIV testing strategies for adolescents and young adults, especially YMSM.^7,24,26 In addition, our findings support the need for educating providers about how to tailor education for YMSM patients on sexual risk behaviors and creating strategies that will help facilitate the patient–provider discussions about HIV/AIDS prevention, including HIV testing.²⁷

Although our findings suggest that health care providers can influence YMSM engagement in HIV testing, many YMSM in our sample had conversations with their health care provider but had never tested for HIV. In a sensitivity analysis (results not shown), 31% of YMSM participants had ever-tested for HIV and had talked to their health care provider about at least one HIV-/AIDs-related topic. By contrast, 10% of YMSM in our sample had talked to their health care provider about at least one HIV/AIDs-related topic but had never tested for HIV. These findings stress the importance of opt-out HIV testing in the health care setting. Additional research, particularly with inquires designed to understand why YMSM talked with their health care provider about HIV/AIDS but never tested for HIV, and using a larger resample size, is needed.

This study has several limitations. First, data from the 2006–2010 and 2011–2015 NSFG are self-reported and, as a result, are subject to recall bias. Participants who have ever had an HIV test may be more likely to recall sex education experiences. Although NSFG uses age restrictions to limit the threat of recall bias (eg, the questions related to sex education were only asked to participants aged 15–24 years), participants still may have been misclassified based on their own recall of exposure. National surveys that use self-reported data, such as the NSFG, may underestimate the prevalence of adolescent sexual health discussions. In addition, the cross-sectional study design could not adequately assess the temporal relationship between the exposure of interest (ie, sex education/communication) and the outcome of interest (ie, HIV testing). As a result, we cannot determine whether receipt of sex education/communication preceded the HIV testing behavior(s) in question.

Additional unmeasured confounders in the multivariable analysis are also possible because the number of variables in the NSFG is solely based on occurrence/receipt of sex education and communication. We were unable to determine the information source (eg, mother, father, or another guardian) for the sex education by parent(s) variables and the location where formal sex education was received (eg, church, community center, and school-based sex education). Other work has shown that sex education by mothers may impact offspring health and behavior differently than sex education by fathers.^10,34 There is a 15-month gap in interviewing between the 2006–2010 and 2011–2015 survey periods.¹³ The 2013–2015 NSFG guide indicates that the survey weights were not designed or adjusted for the purpose of accounting for the 15-month gap and reporting total findings from the 2 NSFG periods, so interpretations of trends should be made with caution.¹³ In addition, the names for the patient–provider sexual health communication questions in 2006–2010 were only asked to persons who had ever-tested for HIV; however, the corresponding questions were asked to all participants in the 2011–2015 NSFG, regardless of HIV testing. As a result, prevalence ratios for all patient–provider variables were calculated using only 2011–2015 NSFG data.

Our study has important implications for HIV/AIDS prevention in the United States. Sex education on specific topics, such as condom use and HIV testing, both by parents and by health providers increased the likelihood of HIV testing among YMSM. Parents and health care providers should tailor sex education discussions with youth to promote healthy behaviors and HIV testing. Public health professionals should equip parents and health care providers with strategies to initiate HIV-/AIDS-related conversations with YMSM.