Disease progression and its influencing factors among HIV/AIDS patients on antiretroviral therapy in Yili, Xinjiang Uygur Autonomous Region: a multi-state Markov model analysis on follow-up data

Objective To analyze disease progression and its influencing factors in human immunodeficiency virus/acquired immune deficiency syndrome (HIV/AIDS) patients on antiretroviral therapy (ART) with Markov modeling.

Methods We recruited a total of 2 000 HIV/AIDS patients (aged ≥ 16 years) receiving free ART at a public hospital in Yining city, Xinjiang Uygur Autonomous Region (Xinjiang) from 2019 through October 2020 for a 18-month follow-up study. CD4⁺T lymphocyte count of the patients were detected at the enrollment and subsequent follow-ups and four disease progression states of the patients were defined based on the count: namely S1, S2, S3, and S4 for the patients with the CD4⁺T lymphocyte count of ≥ 500/μL, 350 – 499/μL, 200 – 349/μL, and < 200/μL and the mortality was defined as S5. A continuous-time discrete-state Markov model was constructed based on the CD4⁺T lymphocyte count to analyze disease progression patterns of the patients.

Results The analysis finally included 1 491 HIV/AIDS patients on ART, with a total follow-up of 2 224.5 person-years. For all the patients, S1 showed the longest total persistent duration of 201.34 months and S4 was of the highest probability of mortality (0.17) compared to other states. The results of multifactorial analysis revealed following factors associated with the intensity of disease state transition: (1) increased intensity of transition from S1 to S2 for patients aged ≥ 50 years compared to those aged < 30 years; (2) higher intensity of transition from S1 to S2 (adjusted hazard ratio aHR = 1.66, 95% confidence interval 95%CI: 1.37 – 2.01) and from S2 to S3 (aHR = 1.65, 95%CI: 1.21 – 2.25) but lower intensity of transition from S2 to S1 (aHR = 0.80, 95%CI: 0.68 – 0.94) and from S3 to S2 (aHR = 0.73, 95%CI: 0.58 – 0.92) for the patients infected heterosexually compared to those infected via drug injection; (3) higher intensity of transition from S1 to S2 but lower intensity of transition from S2 to S1 for the patients with the baseline CD4⁺T lymphocyte count of 350 – 499/μL, 200 – 349/μL, and < 200/μL in comparison with the patients with the baseline CD4⁺T lymphocyte count of ≥ 500/μL; (4) stronger intensity of transition from S2 to S3 for the patients with the baseline CD4⁺T lymphocyte count of < 200/μL compared to the patients with the baseline CD4⁺T lymphocyte count of ≥ 500/μL; and (5) higher intensity of transition from S2 to S3 for the patients at World Health Organization HIV/AIDS stage II compared to those at the stage I.

Conclusion Better transition of disease state was observed in HIV/AIDS patients on ART in Yili city of Xinjiang ant the patients aged 50 years and above, infected via drug injection, and with lower baseline CD4⁺T lymphocyte count were at increased risk of deteriorative transition of disease state.