Background and objective: Combination antiretroviral therapy (cART) has improved the survival of HIV infected patients significantly. However, in some patients, survival is not guaranteed due to several factors that are either individual-based or cART based. This study presents an HIV, AIDS, Death (HAD) model to analyse the survival of patients on cART.
Materials and methods: Continuous-time Markov models are fitted based on the states occupied for an HIV, AIDS and Death (HAD) model. These states are based on CD4 cell count. Factors that affect the survival of HIV-infected patients on cART are also analyzed. These, among others, include age, gender, routinely collected viral load, time on treatment, non-adherence and peripheral neuropathy.
Results: Patients with higher viral loads than expected are 11.1 times more likely to be at risk of HIV progression to the AIDS state and 1.1 times more likely to be at risk of mortality from a CD4 cell count state above 200 cell/mm3compared to patients with lower viral loads. Non-adherence to treatment increases the risk of transition from CD4 cell count state above 200 cell/mm3 to the AIDS state by 2.2 folds. Patients who were non-adherent to treatment are 3.8 times more likely to transit from the CD4 state above 200 cell/mm3 to death compared to patients who were adherent to treatment. Patients are expected to recover from the AIDS state after one year of treatment.
Conclusions: Recovery from AIDS state by HIV infected patients on cART is likely to occur after one year of cART treatment. However, if the viral load remains higher than expected, this increases risks of immune deterioration even after having achieved normal CD4 cell counts and consequently, mortality risks are increased.
Keywords: HAD model; Kolmogorov�s backward equations; antiretroviral therapy; markov jump process; multistate modelling.