comparison of sojourn time distribution for modeling of AIDS disease progression

What is it about?

In this paper, an application of semi-Markov models to AIDS disease progression utilized to find out best sojourn time distributions. We obtained data from 370 HIV/AIDS patients who were under a follow-up since September 2008 to August 2015. The study reveals that within the "good" states, the transition probability of moving from a given state to the next worse state has a parabolic pattern increases with time until it gets optimum and then declines over time. As compared to exponential, the conditional probability of staying in a good state before it moves to the next good state grows faster at the beginning then reaches peaks and then declines faster for long period. The probability of staying in the same good state of the disease declines over time, keeping higher value for the healthier state. Moreover, the Weibull distribution under the semi-Markov model leads to dynamic probabilities with the higher rate of decline and smaller deviations. Weibull distribution is flexible in modeling and so it is preferable to use as a waiting time distribution in disease progression.

Featured Image

Why is it important?

In Modelling AIDS Disease progression one should critically choose which Sojourn Time distribution to use. In our Study this effect in detail and showed that if one selects weibull distribution it is preferred than any other parametric exponential families. This is because of the fact that the weibull adds the advantage of correctly quantifying the effect of disease progression by handling the parameter effects.

Perspectives