Mortality, loss to follow-up and transfer-out in paediatric ART programmes in Southern Africa

Background: Paediatric HIV infections in sub-Saharan Africa considerably increase mortality in African children. The aim of the present study was to describe the patient characteristics and to analyze mortality, loss to follow-up (LTFUP) and transfer-out (TO) among HIV-infected children on antiretroviral therapy (ART) in the IeDEA-Southern Africa collaboration, and to evaluate the effect on mortality assuming different mortality rates among LTFUP. Methods: Treatment-naïve children who started ART before the age of 16 years were included. A child was judged to be LTFUP if the last visit was more than 6 months before the closing date of the database. Weibull models were used to calculate crude mortality rates at one year after start of ART and assuming a mortality rate of 45\% among LTFUP. Results: A total of 9,460 HIV-infected children from 11 different cohorts in South Africa (77.2\%), Malawi, Mozambique and Zimbabwe were included. Overall, 5,228 (55.3\%) were younger than five years (range among cohorts 18.9\% to 78.7\%). Crude mortality at one year after start of ART ranged from 0.7\% to 17.7\% in children aged 5 years and from 0\% to 8.9\% in children ≥5 years. Corresponding figures for LTFUP were 1.9\% to 33.7\% (overall 13.3\%), and 0\% to 45.5\% for TO (overall 12.5\%). Assuming 45\% mortality among children LTFUP, mortality estimates at one year ranged from 1.5\% to 19.8\% in children aged 5 years and 0\% to 9.8\% in children ≥5 years. Conclusion: In children in Southern Africa, mortality at 1 year after starting ART varies widely and LTFUP may lead to underestimation of true mortality. The observed variation does not appear to be related to different rates of LTFUP suggesting the importance of programme- and individual-level factors. Therefore, dedicated sampling- or linkage-based field studies among children LTFUP or TO are needed.