LAUSR

Objective: To analyze the trajectories of HIV testing self-efficacy among men who have sex with men (MSM) based on latent class growth model. Methods: From August 2019 to May 2021, 404 MSM were recruited in Shandong Province and subjected to a 1-year follow-up study with individual intervention (pushing intervention pictures and videos in WeChat and follow-up questionnaires) and community intervention (forwarding to friends and sharing and discussing HIV testing-related information in WeChat groups). The level of HIV testing self-efficacy among MSM was measured. The long-term trend of HIV testing self-efficacy was analyzed using the latent class growth model (LCGM), and the influencing factors of the trend were analyzed. Results: A total of 404 MSM were (28.25±8.95) years old, with the oldest being 58 and the youngest being 18. The scores of HIV testing self-efficacy M(Q1, Q3) at baseline and 4 follow-ups were 18.00 (17.00, 21.00), 19.00 (18.00, 22.00), 19.00 (18.00, 22.00), 19.00 (18.00, 22.00) and 19.00 (18.00, 22.00). The results of the freely estimated two-category LCGM model showed that the trend of HIV testing self-efficacy among MSM could be divided into two classes, "intervention response group" [255(63.1%)] and "intervention non-response group" [149(36.9%)]. The former had a higher level of HIV testing self-efficacy which tended to increase at first and then decrease over time, while the latter had a lower and more stable level. The results of the multifactorial logistic regression analysis showed that the OR values of MSM in business or service and jobless or unemployed were 0.261 (95%CI: 0.108-0.633) and 0.186 (95%CI: 0.057-0.610), respectively, using the students as the reference group. Conclusion: There is a group heterogeneity in the trend of HIV testing self-efficacy in the intervention conditions among MSM, and occupation may be an influencing factor.