A total of 923 college students from a university in the border area were selected as the research subjects, and all participants provided informed consent to take part in the study.

2.2.1. Research Methodology/Content

A cluster sampling method was employed. With reference to the “Key Messages for HIV/AIDS Prevention and Education” [7], a self-administered questionnaire was designed and subsequently reviewed and revised by two public health experts. The questionnaire covered demographic characteristics, HIV knowledge, willingness to use condoms, and sexual health behaviors. Adequate HIV knowledge was defined as correctly answering ≥75% of the relevant questions [8]. The questionnaires were distributed and collected via the Questionnaire Star platform. After investigators explained the purpose of the survey and instructions for completing the questionnaire, participants self-administered the questionnaire anonymously upon providing informed consent. Only one submission per IP address was allowed. Upon collection, questionnaires were checked for completeness and logical consistency, and invalid questionnaires were excluded.

2.2.2. Statistical Methods

Microsoft Excel 2010 was used to create the database. Statistical analyses, including descriptive statistics, the chi-square test, and binary logistic regression analysis, were performed using R 4.3.0. The significance level (α) was set at 0.05.

Out of 1017 questionnaires collected in this survey, 923 were valid. In terms of gender, there were 424 male students (45.9%) and 499 female students (54.1%). Regarding ethnicity, 528 were Han Chinese (57.2%) and 395 were ethnic minorities (42.8%). Concerning residential areas, 484 were from urban areas (52.4%) and 439 were from rural areas (47.6%). Regarding HIV-related situations, 292 students (31.6%) had not received HIV-related education, while 631 students (68.4%) had received such education. Additionally, 414 students (44.9%) had experience with HIV testing and counseling at medical institutions, while 509 students (55.1%) did not have such experience (See Table 1).

Table 1. Sociodemographic characteristics of the study participants (N = 923).

Among the 923 college students, the overall HIV knowledge awareness rate was 65.87% (608 students). The question with the lowest awareness rate concerned the main transmission routes of HIV, with an awareness rate of only 45.7%. Other questions with relatively low awareness rates included: awareness that sharing meals with HIV-infected individuals does not transmit HIV (80.7%); awareness that studying in the same classroom with HIV patients for extended periods does not transmit HIV (83.5%); and awareness that hugging or shaking hands with HIV-infected individuals does not transmit HIV (83.7%) (See Table 2).

Table 2. HIV/AIDS-related knowledge awareness rates.

Gender, HIV knowledge, proactive acquisition of HIV knowledge, receipt of HIV education, freshman status, and history of same-gender sexual behavior were all associated with condom use intention (P < 0.05). Among the participants, 289 students (31.31%) had a history of same-gender sexual behavior, while 634 (68.69%) did not. A total of 584 students (63.27%) had proactively acquired HIV knowledge, while 339 (36.73%) had not. Furthermore, 631 students (68.36%) had received HIV education, while 292 (31.64%) had not (See Table 3, Figure 1).

Table 3. Univariate analysis of condom use intention.

Figure 1. Demographic characteristics and condom use intention.

Variables with P < 0.05 from the univariate analysis were included in a binary logistic regression model to examine their association with condom use intention. The results are presented in Table 4 and Figure 2. Gender, level of AIDS knowledge, whether participants had actively acquired AIDS-related information, and whether they had received AIDS-related education were all statistically associated with condom use intention (all P < 0.001). Specifically, females showed a significantly higher intention to use condoms than males (OR = 3.85, 95%CI: 2.881 - 5.156), indicating that females were 3.85 times more likely to intend to use condoms. Participants who were knowledgeable about AIDS had a higher intention to use condoms compared to those with insufficient knowledge (OR = 28.18, 95%CI: 19.498 - 40.722). Those who had not actively sought related knowledge had a lower intention to use condoms (OR = 0.07, 95%CI: 0.040 - 0.124), and individuals who had not received AIDS-related education exhibited an even lower intention (OR = 0.06, 95%CI: 0.032 - 0.117). In addition, no significant statistical associations were found between condom use intention and academic year (freshman or higher) or same gender sexual experience (all P > 0.05).

Table 4. Multivariate logistic regression analysis of condom use intention.

Note: Only variables with P< 0.05 in the multivariate logistic regression analysis are shown.

Figure 3 illustrates interactive and dose-response relationships between gender, HIV education, HIV knowledge, and condom use willingness.

3.5.1. Gender × HIV Education Interaction (Figure 3(A))

A significant interaction was observed: without HIV education, males had a baseline willingness of 28.3%, while females showed substantially higher willingness (62.4%, +34.1 percentage points vs males). With HIV education, willingness increased to 75.8% (Δ+47.5) in males and 96.7% (Δ+34.3) in females. Non-parallel trends confirmed a true interaction: both genders benefited from education, but males had a larger absolute gain (driven by their lower baseline).

Figure 2. Forest plot of subgroup analysis for condom use willingness.

3.5.2. Gender × HIV Knowledge Interaction (Figure 3(B))

Gender modified the effect of HIV knowledge: males without knowledge had extremely low willingness (15.2%, high-risk subgroup), compared with 48.7% in females (threefold higher). With HIV knowledge, willingness rose to 89.3% (Δ+74.1) in males and 97.8% (Δ+49.1) in females. Males gained more from knowledge, but females maintained a higher baseline without knowledge—suggesting a protective gender effect.

3.5.3. HIV Education × Knowledge Combined Effect (Figure 3(C))

Education and knowledge acted synergistically: the ‘no education + unaware’ subgroup had the lowest willingness (18.5%, n = 98). Knowledge alone (no education + aware) increased willingness to 52.3% (Δ+33.8), while education alone (education + unaware) doubled the baseline (38.7%). The “education + aware” subgroup had the highest willingness (91.2%, n = 414; Δ+72.7 vs baseline), exceeding additive predictions (72.5%)—confirming positive synergy.

3.5.4. Dose-Response: HIV Knowledge vs Willingness (Figure 3(D))

A clear gradient was observed: willingness rose from 22.4% (0% - 25% knowledge, n = 112) to 48.6% (26% - 50%, n = 203), 72.3% (51% - 75%, n = 298), and 93.8% (76% - 100%, n = 310). The near-linear trend supports the reasonable causal pathway of “knowledge enhancement → increased willingness.”

Figure 3. Interactive and dose-response effects on condom use willingness among college students. Panels A - B: Interaction effects of gender with HIV education (A) and HIV knowledge (B), showing subgroup differences in condom use willingness (%). Panels C: Combined effects of HIV education and knowledge, with sample sizes (n) for each subgroup. Panel D: Dose-response gradient between HIV knowledge score quartiles and condom use willingness, with 95%CIs (error bars in Panel D). Data are from a cross-sectional study of college students at a remote Chinese university.