Bisexual Behaviors, HIV Knowledge, and Stigmatizing/Discriminatory Attitudes among Men Who Have Sex with Men

Objective To assess the correlates for bisexual behaviors, HIV knowledge, and HIV/AIDS-related stigmatizing/discriminatory attitudes among men who have sex with men (MSM). Methods A cross-sectional survey among MSM was conducted in 2011 to provide demographics, sexual behaviors, HIV knowledge, HIV/AIDS-related stigmatizing/discriminatory attitudes, and services in Jinan, Qingdao, and Yantai of Shandong Province of China. Results Of 1230 participants, 82.8% were single, 85.7% aged <35 years, and 47.2% received college or higher education. There were 28.6% MSM who reported to be married or cohabitating or ever had sex with woman in the past 6 months (P6M). 74.5% had ≥6 HIV-related knowledge score. The average total score of stigmatizing/discriminatory attitude was 37.4±4.4(standard deviation). Bisexual behavior was independently associated with higher levels of HIV/AIDS-related stigma/discrimination(AOR = 1.1, 95% CI:1.0–1.1), older age(AOR = 1.2, 95%CI:1.1–1.2), and lower HIV-related knowledge score(AOR = 1.6, 95%CI:1.2–2.2). HIV knowledge score ≥6 was independently associated with lower levels of HIV/AIDS-related stigma/discrimination(AOR = 1.3, 95%CI:1.2–1.3), less bisexual behaviors(AOR = 0.6, 95%CI:0.5–0.9), ever received a test for HIV in the past 12 months (P12M)(AOR = 3.2, 95%CI:2.3–4.5), college or higher level education(AOR = 1.9, 95%CI:1.4–2.6), consistent condom use with men in P6M(AOR=6.9, 95%CI:4.6–10.6), recruited from internet or HIV testing sites(AOR = 11.2, 95%CI:8.0–16.1) and bars, night clubs, or tea houses(AOR = 2.5, 95%CI:1.7–4.8). Expressing higher levels of HIV/AIDS-related stigmatizing/discriminatory attitudes was independently associated with bisexual behaviors(Aβ = 0.9, 95%CI:0.4–1.4), lower HIV-related knowledge score(Aβ = 3.6, 95%CI:3.0–4.1), the number of male sex partners in the past week ≥2(Aβ = 1.4, 95%CI:1.0–1.9), unprotected male anal sex in P6M(Aβ = 1.0, 95%CI:0.5–1.6), and inversely associated with ever received HIV test(Aβ = 1.4, 95%CI:0.8–2.0) and peer education in P12M(Aβ = 1.4, 95%CI:0.9–1.9). Conclusion HIV/AIDS-related stigmatizing/discriminatory attitudes were associated with bisexual behaviors, low HIV testing rate, lower HIV-related knowledge and risk behaviors. This study called for innovative programs that would reduce HIV/AIDS-related stigmatizing/discriminatory attitudes and bisexual behaviors and improve the uptake of prevention service among MSM.


Introduction
In China, sexual transmission has become the major route of transmission for HIV. Men who have sex with men (MSM) have been contributing to increasing proportions of total and new infections. MSM comprise 29.4% of China's new HIV cases according to the 2011 national estimates [1][2][3][4][5][6][7][8][9][10]. Shandong Province, situated along the east coast of China, is the hometown of Confucius and the second most populous province in China. Through the end of 2014, homosexual transmission accounted for nearly half of all reported HIV cases in Shandong [10]. It is estimated that about 2.2% of Chinese adult males have had sex with another male [11,12]. Hence, the emerging HIV epidemic among the MSM population poses a major public health challenge in China. The prevalence of bisexual behaviors among MSM varies according to a country's culture and its social acceptance of homosexuality [13]. MSM behaviors are highly stigmatized in China under the strong influence of Confucianism and collectivism [12]. Such social environments may lead MSM to hide their sexual orientation by unwillingly engaging in heterosexual relationships [12,14]. Our previous study [15] showed that, of 2996 MSM participants, 39.5% acknowledge bisexual behaviors, of which were more likely than MSM-only to be drug users and participate in other HIV-related risk behaviors. Bisexual behaviors may play a critical bridging role in spreading the HIV epidemic [16,17].
Stigma and discrimination have been identified as major obstacles to effective responses to HIV since the beginning of the HIV/AIDS epidemic [18][19][20]. People who hold stigmatizing and discriminatory attitudes are less likely to have preventive behaviors, and more likely to have multiple sex partners, a commercial sex partner, and some other HIV-related high risk behaviors [19][20][21][22]. Li's study conducted in Beijing, the capital and first-tier city in China, showed HIV/AIDS-related stigmatizing and discriminatory attitudes were inversely associated with recent HIV testing [19]. So far, no study has investigated bisexual behaviors and HIV/ AIDS-related stigmatizing and discriminatory attitudes among MSM in China, which is critical information in controlling the growing epidemic. This study was conducted in three secondtier cities to further describe the associations between HIV/AIDS-related stigmatizing and discriminatory attitudes, bisexual behaviors and HIV-related knowledge among MSM in Shandong Province, China.

Study Participants and Settings
A cross-sectional study was conducted from April to June 2011 in the three second-tier cities of Jinan, Qingdao, and Yantai of Shandong Province. (The tiers of cities in China usually refer to key characteristics of the city, including its economic development, provincial Gross Domestic Product, advanced transportation systems and infrastructure, and historical and cultural significance. China's first-tier cities usually refer to Beijing, Shanghai, Guangzhou, and Shenzhen.) Prior to the recruitment of the participants, we conducted in-depth interviews to collect background information and the venues to access them. Participants were recruited and interviewed by trained health professionals through multiple methods, such as community outreach, venue-based recruitment, HIV testing areas, peer referrals and Internet advertisement. Trained health professionals conducted the structured questionnaire-based interview with assistance from trained MSM peers. The enrollment criteria included being male, 18 years of age or older, having self-reported having had sex with another male in the past 12 months, and willing to complete the study. All potential participants were invited for eligibility assessments. Voluntary participation, anonymity, and confidentiality were ensured for all participants. China is still a relatively conservative country, and the traditional Chinese culture does not openly endorse MSM behaviors. Many MSM face strong social pressure to hide their identity and therefore become very apprehensive when signing the informed consent for fear of exposing their MSM identity. In order to reduce the refusal rate, verbal consent was obtained from those who refused to sign the written consent. Participants who did not agree to provide written or verbal consent were excluded in interviews. The written or verbal consent process was performed by two trained health professionals. This study and the consent procedure were approved by the Institutional Review Board of Shandong Center for Disease Control and Prevention.

Measures
We collected data on demographics, sex and drug use behaviors, HIV knowledge, HIV testing, HIV-related prevention services, and stigma and discriminatory attitudes towards people living with HIV/AIDS (PLWHA). Bisexual behaviors were defined as participants who were married or cohabitating with a woman or who reported having had sex with a woman in the past 6 months (P6M). Eight HIV-related questions including HIV transmission modes and prevention as well as misperceptions were used to assess HIV knowledge level. Questions included "A person infected with HIV can be recognized by appearance?"; "Mosquito bites can spread HIV/AIDS?"; "Eating together can spread HIV/AIDS?" et al. All questions were weighted equally. Each correct and incorrect answer was combined into an overall score with a range of 0-8. Two groups were stratified according to whether they got a six or greater HIV-related knowledge score. Individual attitudes towards PLWHA were measured by asking participants about their agreement and disagreement (1 = yes, 2 = no) with 22 questions, details of which are described in a previous study conducted in Thailand and Zimbabwe by Genberg BL [23] and in China by Li X [19]. This scale measured 3 dimensions of HIV-related stigma and discrimination including shame, blame and social isolation. Questions included "People living with HIV/AIDS should be ashamed"; "People with AIDS should be isolated from other people"; "People who have HIV/AIDS are cursed"; "A person with HIV/AIDS should be allowed to work with other people" et al. Questions were summed to create total scale scores with a range of 22-44, where a higher score indicated a lower level of HIV/AIDS-related stigma and discrimination. Serum samples were screened for HIV-1 antibodies by enzyme-linked immunosorbent assay and confirmed as positive by Western Blot test. Syphilis screening was performed by rapid plasma reagin and confirmed by the Treponema pallidum particle agglutination assay.

Statistical Analysis
We recorded the questionnaire-based data and biological testing results into the EpiData software (EpiData 6.4 for Windows). The Statistical Program for Social Sciences software (SPSS software, Version 15.0) was utilized for all analyses. Descriptive statistics were expressed as frequencies and proportions for categorical variables, and mean and standard deviations for continuous variables. Bisexual behaviors (yes/no), HIV knowledge level (6 or <6), and HIV/ AIDS-related stigmatizing/discriminatory attitudes are the outcomes of interest. Logistic regression for bisexual behaviors and HIV knowledge level, and linear regression for HIV/ AIDS-related stigmatizing/discriminatory attitudes were performed. Univariate analyses were conducted to estimate the relationship between potential risk factors and bisexual behaviors, HIV knowledge level, and HIV/AIDS-related stigmatizing/discriminatory attitudes. We further conducted stepwise backward sequence analyses to select correlates that are independently associated with these outcomes. All variables in the final multivariable models had a pvalue < 0.05 and were considered significant. Adjusted odds ratios (AOR) and 95% Confidence Intervals (CIs) for having bisexual behaviors and for a higher HIV knowledge level, and adjusted effect estimates and 95% CI for HIV/AIDS-related stigmatizing/discriminatory attitudes were reported for each explanatory variable in the final models.

Sexual Behaviors, Prevalence Rates of HIV and Syphilis
Of all participants, 91.4% had sex with men in the P6M, 54.3% had 2 male sex partners in the past week, 70.8% used a condom during their last anal sex encounter, and 31.3% consistently used condoms in the P6M with male partners; 27.2% had commercial sex with men with 29.8% consistently used condoms in the P6M, 21.5% of participants had ever sold sex to a man with only 31.1% consistently used condoms in the P6M, 23.4% had sex with a female with one third (32.5%) consistently having used condoms in the P6M, only 1.1% ever used drugs, half (50.7%) had received HIV testing in the past year, three-quarters (75.2%) and 41.3% had ever received condom promotion/HIV testing, and counseling and peer education, respectively. Of all participants, 1.6% was HIV-infected and 6.8% syphilis-infected (Table 1).

Correlates for Bisexual Behaviors
Of the 1230 participants, 28.6% were married, cohabitating or had sex with a woman in the P6M. Univariate analysis revealed that MSM who were recruited from Jinan and Qingdao, >35 years of age, married or cohabitating, had a high school or lower education, had lived in

Correlates for HIV Knowledge
Of all participants, 74.5% had a 6 HIV-related knowledge score. Univariate analysis revealed that MSM who were recruited from Jinan, the Internet or a HIV testing site, were 35 years of age, single or separated, ethnicity group, college or higher education, self-identified as non-heterosexual, age of first sexual intercourse being 20, never had sex with a man in the P6M, had

Correlates for Stigmatizing and Discriminatory Attitude
Of all participants, the total score of stigmatizing and discriminatory attitude was 37.4±4.4. Univariate analysis revealed that MSM who were recruited from Qingdao, bathhouses, sauna or outdoor cruising area, were >35 years of age, were married or cohabitating, had a lower level of education, a lower score of HIV knowledge, bisexual behaviors, age of first sexual intercourse being >20, had sex with a man in the P6M, the number of male sex partners was 2 in the past week, had commercial sex with a man in the P6M, no condom use during the last Total N for each subgroup may not add up to the total due to missing data; P6M: in the past 6 months

Discussion
This study provided the first data to assess the correlates for bisexual behaviors, HIV knowledge, and HIV/AIDS-related stigmatizing and discriminatory attitudes among MSM in three Total N for each subgroup may not add up to the total due to missing data; P6M: in the past 6 months second-tier cities of Shandong Province. Multivariable logistic regression analyses showed bisexual behaviors, HIV knowledge and HIV/AIDS-related stigmatizing and discriminatory attitudes predicted each other. The findings of this study further contribute to the deeper understanding of the role of stigma and discriminatory attitudes among MSM in China. Common negative attitudes towards PLWHA, bisexual and unprotected sex among this group has created an emerging challenge in delivering prevention services to contain the rapidly growing epidemic. Studies revealed that stigma can have significant adverse effects on health and disease transmission by promoting a delay in seeking care and reluctance to follow medical advice [20,[24][25][26]. Confronting the rapid expansion of the HIV/AIDS epidemic among MSM in China [1,2], this study provides more detailed evidence and calls for innovative programs that would reduce HIV-related stigmatizing and discriminatory attitudes and risks among the MSM communities in three second-tier cities of Shandong Province. This study showed that HIV/AIDS-related stigmatizing and discriminatory attitudes were inversely associated with having received a test for HIV in the last year. This finding is consistent with Li's report from Beijing [19]. Studies show that HIV testing and counseling can help to substantially reduce risk behaviors [27,28]. However, only 50.7% of participants in this study had received HIV testing in the last year. This low rate of HIV testing indicates that current HIV testing and intervention programs have not yet been carried out effectively among MSM in these three second-tier cities. Given the increase in innovative technology, e.g., Internet, mobile phone, and social media usage, more innovative avenues to promote HIV testing and linkage to care could be considered and developed within the rapidly expanding HIV prevention services among this group. This study showed that bisexual behaviors were associated with a higher level of HIV/AIDS-related stigma and discrimination. This finding is not consistent with Li's report [19]. The operational definition of bisexual behaviors might be the reason. Li's study defined bisexual behaviors based on a self-reported sexual orientation. However, in this study, bisexual behaviors refer to the participants who were married or cohabitating with a woman, or reported having had sex with a woman in the past six months. The latter definition may be more representative than the one that was applied in Li's study. Another reason could be that Li's study was conducted in Beijing, the capital of China and a first-tier city; which may have differences among MSM compared to that in three second-tier cities in China. The previous study [15] reported that bisexual MSM were more likely to have higher risk behaviors than MSM-only, and it highlighted the importance of bisexual behaviors as a potential epidemiologic bridge. As the home of Confucius, MSM behaviors in Shandong are stigmatized and MSM face strong social pressure under the heavy influence of Confucianism and collectivism [12,15]. Bisexual behaviors may further reinforce the negative attitudes towards PLWHA and exacerbate high-risk behaviors, which will, conversely, result in a higher proportion of bisexual behaviors among the MSM group. Therefore, better targeted and more innovative programs are critical to reduce HIV-related stigmatizing and discriminatory attitudes. This may also aid in decreasing bisexual behaviors and improving the availability and coverage of prevention services as well as their acceptance for unique settings, such as the second-tier cities of Shandong Province, thus preventing transmission via bisexual behaviors from their highrisk male sex partners to their wives [16,17]. This study showed that MSM with a higher HIV knowledge were more likely to express less negative attitudes. This finding is consistent with Dias' report [29]. Those MSM who have a higher education level will have more access to HIV-related knowledge and health information, and will have a lower perceived stigma and discriminatory social pressure. This study also found that peer education has a similar effect. Peer education could alleviate social pressure, persuade their peers to seek HIV testing and counseling, and provide a link to care and medical treatment [20,27,30]. The findings of this study suggest HIV-related knowledge promotion, expanding HIV testing, and peer education could be integral parts of the conventional HIV/ AIDS services to reduce the stigmatization and discrimination among this group. However, this study found that high risk behaviors widely exist; 54.3% of the participants had 2 male sex partners in the past week, and only 31.3% consistently used condoms in the past six months with male partners. This finding suggests that greater efforts should be focused on enhancing individuals' motivation to change their behavioral patterns and teaching behavioral skills to reduce risky behaviors. The findings also suggest HIV-related knowledge promotion strategies should vary from venue to venue.
This study highlights the importance of policy considerations for stigma and discrimination, and its related sexual risks among MSM. In spite of considerate planning, implementation and quality control, this study has its limitations. The data relying on retrospective self-reporting may be subject to recall bias. Sensitivity of sex and drug related questions and expressing stigmatizing and discriminatory attitudes towards PLWHA could lead to reporting bias and social desirability bias. In addition, the non-response information was not collected.
Despite these limitations, this study provides important information for further research and suggests that MSM who experience stigmatizing and discriminatory attitudes towards PLWHA may be at a higher risk for an increased number of sex partners, being bisexual, having a lower level of HIV knowledge, a smaller proportion of having received HIV testing and peer education, and having unprotected sex. The findings of this study call for innovative programs that would reduce HIV/AIDS-related stigmatizing and discriminatory attitudes and bisexual risk behaviors and improve the uptake of prevention services among MSM in the three second-tier cities Jinan, Qingdao, and Yantai of Shandong Province.