Abstract
To evaluate the mortality risk in the HIV-positive population, we conducted an observational cohort study involving routine data collection of HIV-positive patients who presented at HIV clinics and multiple treatment centers throughout Guangxi province, Southern China in 2011. The patients were screened for tuberculosis (TB) and tested for hepatitis B (HBV) and C (HCV) virus infections yearly. Following the registration, the cohort was followed up for a 60-month period till the end-point (December 31, 2015). Univariable and multivariable Cox proportional hazards regression models were used to analyze the hazard ratio (HR) and 95% confidence interval (95% CI) for mortality after adjusting for confounding factors stratified by patients' sociodemographic and behavioral characteristics. HRs were compared within risk-factor levels. With the median follow-up of 3.7-person years for each individual, 5,398 (37.8%) (of 14,293 patients with HIV/AIDS) died; among whom, 78.4% were antiretroviral therapy (ART)-naïve; 43.6% presented late; and 12.2% and 3.3% of patients had Mycobacterium tuberculosis (MTB) and HBV and HCV co-infection, respectively. Of individuals with CD4 counts, those with CD4 count >350 cells/μL formed 14.0% of those who died. Furthermore, gender [multivariable HR (95% CI):1.94 (1.68-2.25)], Han ethnicity [2.15 (1.07-4.32)], illiteracy [3.28 (1.96-5.5)], elementary education [2.91 (1.8-4.72)], late presentation [2.89 (2.46-3.39)], and MTB co-infection [1.28 (1.10-1.49)] strongly increased the all-cause mortality risk of HIV-positive individuals. The HR for ART-based stratification was 0.08 (0.07-0.09); and for HBV and HCV co-infection, HR was 1.02 (0.86-1.21). The findings emphasized that accessibility to HIV testing among high-risk populations and screening for viral hepatitis and TB co-infection are important for the survival of HIV-positive individuals. Initiating early ART, even for individuals with higher CD4 counts, is advisable to help increase the prolongation of lives within the community.