Estimation of survival times of HIV-1 infected children for doubly and interval censored data

Abstract

International studies have established the association of antiretroviral (ARV) therapy and survival in HIV-infected patients. Our study was carried out to confirm the influence of ARV therapy on survival in a group of vertically transmitted HIV-1 infected children. We employed Turnbull’s methods based on double censoring (1974) and interval censoring (1976) to estimate the survival time distribution for children receiving and not receiving ARV therapy. It was noted that children undergoing ARV therapy had a longer survival than those not under treatment. In our study group, children under treatment survived for 15 years with a probability of almost 95% while children not under treatment had a survival of 11 years with a probability of 75%. Further, the estimates of survival times as obtained by the interval censoring mechanism were found to be more precise than those obtained by the double censoring mechanism. We even utilized imputation approach which showed that the width of the band of estimates was narrower for interval censored data as compared to double censored data.