Latent Profile Analysis of Disease Self-Management and Its Associated Factors Among People Living with HIV with Dyslipidemia

Purpose: To identify latent self-management profiles in people living with HIV (PLWH) with dyslipidemia and factors associated with profile membership, thereby facilitating targeted clinical intervention. Methods: A cross-sectional survey was conducted from December 2024 to June 2025 among 333 PLWH with dyslipidemia at Nanjing Second Hospital. Data were collected via sociodemographic/disease-related questionnaire, the HIV Self-Management Scale (HIVSMS), and the Health Literacy Management Scale (HLMS). Latent profile analysis (LPA) was performed in Mplus 8.3, and multinomial logistic regression was used to examine factors associated with profile membership. Results: Fit indices (entropy = 0.993) supported a three-profile solution: low self-management–low social support-seeking (C1, 42.3%), moderate self-management–stable (C2, 37.8%), and high self-management–emotion regulation dominant (C3, 19.8%). Seeking social support was relatively low across profiles. Compared with C1, C2 membership was significantly associated with higher education and income, lipid-lowering medication use (OR 3.735, 95% CI 1.597– 8.736), and CD4 350– 500 cells/μL, and was less likely among participants with VL > 1000 copies/mL or chronic comorbidities (all P 500 cells/μL, and higher HDL-C, and was less likely among those with VL > 1000 copies/mL (OR 0.037, 95% CI 0.004– 0.380) or chronic comorbidities (all P < 0.05). Compared with C2, C3 membership was independently associated with higher health literacy (HL) (OR 1.038 per point, 95% CI 1.012– 1.064) and was less likely among those with LDL-C ≥ 3 mmol/L (P < 0.05). Conclusion: We identified three distinct self-management profiles among PLWH with dyslipidemia. Profile membership was significantly associated with HL and socioeconomic, HIV-related, lipid-related, and comorbidity factors, supporting the need for profile-tailored strategies to improve self-management. Keywords: HIV/AIDS, lipid abnormalities, health-behavior regulation, finite mixture modeling, predictors