Abstract / Summary
Abstract Objective Guided by intersectionality theory, this study examined sociodemographic, disease-related, and psychosocial factors associated with anxiety and depression among HIV-positive MSM and explored their potential pathways using structural equation modeling. Methods A questionnaire survey was conducted among 417 HIV-positive MSM. Sociodemographic characteristics, psychosocial variables, and mental health indicators were collected. Univariate analyses, multiple regression analyses, and structural equation modeling were performed to analyze the data. Results Anxiety and depression were positively correlated with HIV-related stigma (r = 0.571) and internalized homophobia (r = 0.542), and negatively correlated with perceived social support (r = −0.539) and chronic disease self-efficacy (r = −0.586; all p < 0.001). Multiple linear regression analysis showed that HIV-related stigma, internalized homophobia, perceived social support, and chronic disease self-efficacy, together with ethnicity, self-rated health status, monthly income, time since HIV diagnosis, and viral load, were independently associated with the HADS total score (adjusted R² = 0.555). The structural equation model demonstrated acceptable fit (χ²/df = 2.632, CFI = 0.949, TLI = 0.933, RMSEA = 0.063). In the specified model, HIV-related stigma, internalized homophobia, and perceived social support showed indirect associations with psychological distress through chronic disease self-efficacy, while self-efficacy was negatively associated with psychological distress (β = −0.910, p < 0.001). Conclusion Anxiety and depression among HIV-positive MSM were associated with multiple sociodemographic, clinical, and psychosocial factors. The findings suggest that chronic disease self-efficacy may represent an important psychological resource linking stigma-related and social factors to psychological distress. An intersectionality-informed approach to mental health care may therefore benefit from simultaneously addressing stigma-related stress, strengthening social support, and enhancing self-efficacy. Given the cross-sectional design and the presence of suppression-like patterns in some structural paths, the observed indirect associations should not be interpreted as establishing causal mediation.