[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
Purchase Options:
[Skip to Content Landing]
December 13, 2000

Antiretroviral Therapy and Mortality Among Children With Perinatal HIV Infection—Reply

Author Affiliations

Stephen J.LurieMD, PhD, Senior EditorIndividualAuthorPhil B.FontanarosaMD, Executive Deputy EditorIndividualAuthor

JAMA. 2000;284(22):2871-2872. doi:10.1001/jama.284.22.2869

In Reply: Mr Teglas and colleagues raise the important issue of selection bias in longitudinal studies to estimate the survival of HIV-infected individuals and suggest that our study's estimated cumulative probability of survival, which is lower than that reported elsewhere, may have been affected by such a bias. Although we agree that enrolling all children at birth would represent an improvement, the ideal study would enroll a very large sample of HIV-infected women and follow up those who became pregnant and, afterward, their children.

Nonetheless, we do not believe that our study's cumulative probability of survival is biased, for the following reasons. First, when restricting our analysis to children followed up since birth, the survival estimate remained unchanged with respect to that found for the overall population. Second, the differences between our study's results and those cited by Teglas et al may have been due to the number of children enrolled, possibly producing confidence intervals wide enough to overlap among the studies. Third, as described in our article, the data from the Italian Register for HIV Infection in Children were cross-linked with the data from the Italian AIDS Registry, to which all AIDS cases must be reported by law. This ensures a high level of completeness of data, reducing the potential for selection bias. Fourth, the way in which death was ascertained may have varied among the studies, resulting in a different capacity to detect deaths: cross-linking the data from the 2 registries ensures the greatest possible completeness in death reporting. Finally, the differences in results between our study and the French study are present even among the children born after 1996, despite the fact that, in our study, nearly all of these children were prospectively followed up from birth.

Differences among studies in the survival of HIV-infected individuals need to be interpreted with caution, because a variety of factors may influence these results. In no way is the message of our study concerning the population effectiveness of combined therapies in children affected.