[Survival analysis since diagnosis of HIV-positive injecting drug users aged 15 years and above in China].

Objective: To understand death's mortality and risk factors among HIV-positive injecting drug users (IDU) aged 15 or above in China and provide further reference to future prevention and treatment policies. Method: Retrospective cohort study was conducted to calculate the mortality rate of HIV-positive IDU based on HIV/AIDS Comprehensive Response Information Management System. Cox proportion hazards regression model was performed to assess the risk factors for deaths. The Excel 2019 and SPSS 22.0 software was used for data cleaning and statistical analysis. Results: Between 2001 and 2020, 119 209 HIV-positive IDU were reported with 59 094 deaths. The all-cause mortality rate was 6.96 per 100 person-years (py), and the AIDS-related mortality rate was 1.91 per 100 py, with a decreasing trend over the years. Multivariate Cox regression indicated for all-cause death risks of HIV-positive IDU, compared with those baseline T ⁺ lymphocyte cells (CD4) counts above 500 cells/μl, the HR (95% CI ) of those CD4 counts untested, between 0-199, 200-349, 350-500 cells/μl was 2.85 (2.78-2.93), 2.47 (2.40-2.54), 1.58 (1.53-1.62) and 1.24 (1.21-1.28) respectively. The HR (95% CI ) of antiretroviral treatment (ART) naïve was 7.13 (6.99-7.27) compared with those under ART. The HR (95% CI ) of methadone maintenance treatment (MMT) naïve was 1.07 (1.04-1.10) compared to those receiving MMT. As for AIDS-related death risks, compared with baseline T+ lymphocyte cell CD4 counts >500 cells/μl, the HR (95% CI ) of those CD4 counts untested, between 0-199, 200-349, 350-500 cells/μl was 3.26 (3.08-3.46), 5.54 (5.24-5.85), 2.35 (2.21-2.50) and 1.41 (1.32-1.50). HR (95% CI ) of ART naïve was 5.96(5.74-6.18) compared to those under ART. Conclusions: Further efforts should be made timely on diagnosis, treatment, and harm reduction programs such as MMT for improvement compliance to reduce mortality risks of HIV-positive IDU.