Associations between perceived and actual risk of HIV infection and HIV prevention services uptake among men who have sex with men in Shandong province, China: a cross-sectional study

Abstract

Background Associations between perceived and actual risk of HIV infection and HIV prevention services uptake are inconclusive. This study aimed to evaluate the discrepancy between the perceived and actual HIV risk, and quantify the associations between perceived and actual risk of HIV infection and three HIV prevention services utilization among men who have sex with men (MSM) in Shandong province, China. Methods A cross-sectional study was conducted in Shandong province in June 2021. Participants were eligible if they were born biologically male, aged 18 years or older, had negative or unknown HIV status, and had sex with men in the past year. Participants were recruited online. The discrepancy between their perceived and actual risk of HIV infection was evaluated by calculating the Kappa value. Bayesian model averaging was used to assess the associations between perceived and actual risk of HIV infection and HIV prevention services uptake. Results A total of 1136 MSM were recruited, most of them were 30 years old or younger (59.9%), single (79.5%), with at least college education level (74.7%). Most participants (97.4%) perceived that they had low risk of HIV infection, and 14.1% were assessed with high actual risk. The discrepancy between their perceived and actual risk of HIV infection was evaluated with a Kappa value of 0.076 (P < 0.001). HIV testing uptake had a weak association with perceived high HIV prevalence among social networks (aOR = 1.156, post probability = 0.547). The perceived high HIV prevalence among national MSM was positive related to willingness to use PrEP (aOR = 1.903, post probability = 0.943) and PEP (aOR = 1.737, post probability = 0.829). Perceived personal risk (aOR = 4.486, post probability = 0.994) and perceived HIV prevalence among social networks (aOR = 1.280, post probability = 0.572) were related to history of using PrEP. Perceived personal risk (aOR = 3.144, post probability = 0.952), actual risk (aOR = 1.890, post probability = 0.950), and perceived risk among social networks (aOR = 1.502, post probability = 0.786) were related to history of using PEP. Conclusions There is discordance between perceived and actual personal risk of HIV infection among MSM in China. HIV risk assessment and education on HIV prevalence among MSM should be strengthened to assist high-risk populations aware their risk accurately and hence access HIV prevention services proactively.