Sexual attitudes, sexual behaviors, and use of HIV prevention services among male undergraduate students in Hunan, China: a cross-sectional survey

This cross-sectional study was conducted in Hunan, a Central China province. Home to 70 million people, Hunan has approximately 20,000 HIV-infected patients [19]. The 109 universities and colleges in Hunan provide higher education with a variety of disciplines. According to the document on HIV prevention jointly issued by the Ministry of Education and Ministry of Health, all colleges and universities are required to provide HIV/AIDS prevention education at the very beginning of university life [8]. To fully mobilise students’ enthusiasm in participating in AIDS prevention, various youth-friendly services such as lectures, billboards, videos, and school media publicity are encouraged in the universities. All universities/colleges in Hunan offer health education on AIDS prevention, but HIV testing is not available on university campus. Students can access HIV testing at local CDCs, nearby health facilities, or community-based organisations for HIV counselling and testing.

The target population of this study was male undergraduate students. Approximately 60,000 male undergraduate students from three comprehensive universities in Hunan were purposefully sampled to ensure an adequate representation of students from various majors and different grades. With the permission of the Student Union of the three universities and departments, we asked for recruitment assistance from each grade. The participants were recruited via peer referral in the selected cohorts. Via WeChat (a Chinese cell/web app for messaging, social media, communications – including answering questionnaires – and other purposes), we successfully sent a unique two-dimensional code linking the electronic questionnaire to potential participants who worked for the student unions of various grades in the universities. In all three universities, these participants then disseminated the two-dimensional code via WeChat group to undergraduates of different grades and various majors. Male undergraduate students who were 18 years and above and used WeChat were eligible for this study. The sample size was calculated based on the proportion of HIV testing. The rate of HIV testing reported in previous studies in Chinese undergraduate students varied from 2.8–47.1% [20, 21]. Based on the detection table of the sample size for proportions in Hulley et al. [22], assuming the prevalence of HIV testing was 25%, and for a confidence level of 95% with the width of confidence interval of 0.10, the optimum estimate of 288 students was required. As sexual experience is a precondition of HIV testing, the total sample size based on the proportion of sexual behavior occurrence among Chinese male undergraduate students (21.1%) [23] was 1365 by assumption in this study.

Demographic data regarding the students’ age, grade, ethnicity, area of residence, with or without a sexual partner, etc. were collected. Behaviours such as smoking, alcohol drinking, and sexual orientation were also assessed.

HIV-related knowledge was assessed with a 13-item questionnaire. The questionnaire was developed on the basis of the unified National AIDS Sentinel Surveillance Questionnaire designed specifically for university students [24] and the existing literatures [25]. Each item was measured via a yes/no/uncertain format, and only the correct response scored 1 point. Mastering 85% or more of the knowledge in the questionnaire was considered as having sufficient knowledge, and a total score of 11 or higher was required to demonstrate ‘sufficient knowledge’; otherwise, respondents were regarded as having ‘insufficient knowledge’. The internal reliability (KR-20) for this questionnaire was 0.64.

In this study, the Sexual Attitude Scale was used to measure the undergraduates’ sexual attitudes [26]. The 7-item scale assessed the attitudes towards premarital sex, extramarital sex, commercial sex, homosexual behaviours, one-night stand, and multiple sex partners. Participants rated items on a Likert scale ranging from 1 (completely unacceptable) to 5 (completely acceptable). For each item, scores lower than 3 were recoded as ‘unacceptable’, and scores of 3 and higher were recoded as ‘acceptable’. A composite score was obtained by summing up the scores of each item, with higher scores indicating more conservative sexual attitudes. The original 6-item scale assessed commercial sex behavious without distinguishing between ‘paying for sex’ and ‘selling sex’. In the current study, participants were asked to respond to both situations. The Cronbach’s alpha for this scale was 0.84.

Sexual behaviours were assessed by asking students about their sexual experiences, heterosexual/ homosexual behaviours, commercial sexual behaviours, and condom use in the past year [24]. Participants who had sexual intercourse in the last year were asked to rate the frequency of condom use on the basis of four response options (1 = never, 2 = sometimes, 3 = often, 4 = always). HIV prevention services were operationally defined in this study as HIV educational programs, counselling services, and testing services. Considering the differences between the HIV prevention services provided by colleges and universities, public health sectors, and community-based organisations, the following five questions were asked: (1) Did you receive any HIV prevention education organised by the university in the last year? (2) What types of educational activities did you participate in at the university? (3) Did you take an HIV test in the last year? (4) Are you willing to take an HIV test? and (5) If you are not willing to take an HIV test, what are your reasons for refusing to do so? These questions were designed by the investigators after collecting information from people who organized campus HIV prevention programs and after conducting literature reviews [27, 28].

Data were collected online via the Chinese professional survey website Wenjuanxing (www.​sojump.​com) from November 2017 to January 2018. All potential participants accessed the questionnaires by scanning the two-dimensional code with WeChat, and they were able to submit the questionnaires only when all items were finished. Students were allowed to complete the self-administered questionnaire, which took approximately 20–30 min, in their spare time. Each participant received a bonus of 5 RMB online upon the completion of the survey. To avoid repetition, questionnaires bearing the same IP address as previously submitted questionnaires were not accepted.

Of the 2432 invited male undergraduate students, 254 were excluded by the inclusion criteria (56 were under 18 years old and 198 were females). Of the remaining 2178 students, 747 (34.3%) declined to complete the questionnaire, resulting in a final sample of 1431 completed and valid questionnaires. The participants were 18 to 25 years (mean = 19.5 years; SD = 1.32). Most were of Han nationality (93.6%) from urban districts (56.3%), majoring in nonmedical fields (94.9%), not living with sexual partners (96.7%), and from lower grades (27.9% freshmen and 33.3% sophomores). Roughly one-fifth of the participants self-reported cigarette smoking, and more than half self-reported alcohol drinking. The majority of the students regarded themselves as heterosexual (87.8%), and the remaining students considered themselves as either homosexual, bisexual, or unclear (12.2%) (Table 1).

Overall, 1099 (76.8%; 95% CI: 74.6–79.0%) students scored 11 or above in HIV-related knowledge, and were considered as having sufficient knowledge. Among the 13 HIV knowledge questions, the two items on safe sex were answered correctly mostly: 97.7% of the students believed unprotected sex would increase the probability of HIV infection, and 95.7% thought using condoms during sex would decrease HIV transmission. Four items had a correct response rate below 80%: two items on HIV transmission routes, namely whether one could be infected through mosquito bites and whether one could be infected by kissing HIV-positive people, showing a correct response rate of 75.3 and 69.7% respectively. The other two items related to the situation after HIV infection had the lowest correct response rate of 69.0 and 62.9%, namely whether AIDS was equal to death and whether HIV antibody could be detected within one week after the infection. These answers indicated that students with sufficient HIV knowledge still had misconceptions that needed further clarification.

Our sample demonstrated a highly acceptable attitude towards premarital sex and a neutral attitude towards homosexual sex. For premarital sex, 77.1% of the participants reported acceptable (median = 3, P₂₅ = 3, P₇₅ = 5), and 41.8% reported homosexuality acceptable (median = 2, P₂₅ = 1, P₇₅ = 3). As to other types of sexual behaviours, 27.9% believed one-night stand acceptable (median = 2, P₂₅ = 1, P₇₅ = 3), 22.8% expressed paying for sex acceptable (median = 1, P₂₅ = 1, P₇₅ = 2), 17.3% considered selling sex acceptable (median = 1, P₂₅ = 1, P₇₅ = 2), 15.2% thought having multiple sexual partners acceptable (median = 1, P₂₅ = 1, P₇₅ = 2), and 13.2% stated extramarital sex acceptable (median = 1, P₂₅ = 1, P₇₅ = 2).

Totally 299 out of the 1431 male undergraduate students (20.9%; 95% CI: 18.8–23.0%) reported active sex partners, and 49 out of the 299 (16.4%/) engaged in sexual behaviours with males. Regarding sexual partners in the last year among the 299 sexually actively students, 250 (83.6%) reported having sex with fixed female partners, 81 (27.1%) had casual female partners (81 having both fixed and casual female partners), and 15 (5%) had commercial female partners; whereas 33 (11%) had fixed male partners, 22 (7.4%) had casual male partners (20 having both fixed and casual male partners), and 11 (3.7%) had commercial male partners. Those students who had sex with females (t = 3.865, p <  0.001) or male casual partners (t = 2.257, p = 0.029) scored higher in sexual attitudes with more open sexual attitudes. Frequencies of condom use were significantly different among the groups with different partners ( χ² = 17.016, p = 0.048). The post hoc analysis showed that the variance was mostly attributable to the larger proportion of students who never used condoms when they had sex with casual or fixed male partners and a smaller percentage to those with fixed female partners (p < 0.05) (Table 2).

Of all the male undergraduate students in the sample, 1068 (74.6%; 95% CI: 72.4–76.0%) used HIV prevention education services provided by colleges universities in the last year. More than half of the students had used two forms of the university education services, including publicity materials / billboards and lectures. Other educational services included HIV testing services. Only 105 (7.3%) male students had HIV testing in the last year, but the majority (1110, 77.6%) reported willingness to take the tests, and the reasons of 321 students’ unwillingness to take HIV testing were given (Table 3).

Variables significant at p <  0.1 level were included in the multivariate logistic regression. Students who were older (OR: 0.77, 95% CI: 0.70–0.86), from higher grade (OR: 0.80, 95% CI: 0.70–0.91), and those who drank alcohol (OR: 0.71, 95% CI: 0.55–0.93), and had more open attitudes towards paying for sex (OR: 0.72, 95% CI: 0.54–0.95) were less likely to use HIV prevention education services in colleges and universities (Table 4).