Volume 8 - Article 9 | Pages 261–278
Does biological relatedness affect child survival?
|Date received:||04 Oct 2002|
|Date published:||07 May 2003|
|Keywords:||child survival, fostering, HIV/AIDS, orphans, Uganda|
Objective: We studied child survival in Rakai, Uganda where many children are fostered out or orphaned.
Methods: Biological relatedness is measured as the average of the Wright’s coefficients between each household member and the child. Instrumental variables for fostering include proportion of adult males in household, age and gender of household head. Control variables include SES, religion, polygyny, household size, child age, child birth size, and child HIV status.
Results: Presence of both parents in the household increased the odds of survival by 28%. After controlling for the endogeneity of child placement decisions in a multivariate model we found that lower biological relatedness of a child was associated with statistically significant reductions in child survival. The effects of biological relatedness on child survival tend to be stronger for both HIV- and HIV+ children of HIV+ mothers.
Conclusions: Reductions in the numbers of close relatives caring for children of HIV+ mothers reduce child survival.
David Bishai - Johns Hopkins University, United States of America
El Daw Suliman - Johns Hopkins University, United States of America
Heena Brahmbhatt - Johns Hopkins University, United States of America
Fred Wabwire-Mangen - Makerere University, Uganda
Godfrey Kigozi - Makerere University, Uganda
Nelson Sewankambo - Makerere University, Uganda
David Serwadda - Makerere University, Uganda
Maria Wawer - Johns Hopkins University, United States of America
Ron Gray - Johns Hopkins University, United States of America
Other articles by the same author/authors in Demographic Research
Most recent similar articles in Demographic Research