Sexual behavioural risk, partnership contexts, and correlates of condomless sexual intercourse among men newly diagnosed with HIV in Shanghai, China

Given low HIV pre-exposure prophylaxis (PrEP) and treatment as prevention uptake, understanding patterns of condomless sexual intercourse is critical for developing HIV prevention strategies in China, where HIV disproportionately affects men through sexual transmission. We quantified condom use and identified correlates of condomless sexual intercourse among men in Shanghai, China. We recruited 151 cisgender men ≥ 18 years within 60 days of their HIV diagnosis from Shanghai’s sole HIV treatment center (05/2021-03/2022). Participants were surveyed about their sexual behavioural risks within six months before survey completion. Condom use was categorized into mutually exclusive groups: low-risk (always used condoms), moderate-risk (condomless with one primary partner only), or high-risk (condomless with ≥1 non-primary partner or >1 primary partner). We used sequential Firth logistic regression to identify correlates of condomless sexual intercourse. Of 151 participants, 4.6% ever used HIV PrEP; at survey completion, 39.7% initiated antiretroviral therapy (ART) and 1.3% achieved viral suppression. Of 129 participants who engaged in sexual intercourse within six months before survey completion, 69.5% only had sex with men and 55.6% reported concurrent sex partners. One-fifth (21.7%) always used condoms (range: male partners: 11.6%-25.9%; female partners: 0%-50%). Among the 129 participants, older age (aOR per 5-year increase = 1.68;95%CI:1.20,2.51;p = 0.0018), lifetime sexually transmitted infections (aOR=3.17;95%CI:1.08-10.02;p = 0.0357), and multiple sex partners (2–3 partners: aOR=10.3;95%CI:3.45-37.48;p < 0.0001; ≥ 4 partners: aOR=17.17;95%CI:5.08-71.94;p < 0.0001) were associated with increased odds of high-risk intercourse (compared with low-/moderate-risk). In contrast, restricting to participants reporting low-/moderate-risk intercourse, ≥ 4 partners (aOR=0.10;95%CI:0.01-0.48;p = 0.0112) reduced the odds of moderate-risk intercourse. Overall, condomless sexual intercourse was common among men newly diagnosed with HIV in Shanghai, who often had multiple and concurrent partners. These findings, alongside suboptimal PrEP uptake, ART initiation, and viral suppression, underscore the need for comprehensive HIV prevention strategies integrating drivers of concurrent partnerships, publicly funded condom distribution, expanded PrEP use, and timely ART initiation among men with large sexual networks.