Inequalities in child mortality in ten major African cities

Background: The existence of socio-economic inequalities in child
mortality is well documented. African cities grow faster than
cities in most other regions of the world; and inequalities in
African cities are thought to be particularly large. Revealing
health-related inequalities is essential in order for governments
to be able to act against them. This study aimed to systematically
compare inequalities in child mortality across 10 major African
cities (Cairo, Lagos, Kinshasa, Luanda, Abidjan, Dar es Salaam,
Nairobi, Dakar, Addis Ababa, Accra), and to investigate trends in
such inequalities over time. Methods: Data from two rounds of
demographic and health surveys (DHS) were used for this study (if
available): one from around the year 2000 and one from between 2007
and 2011. Child mortality rates within cities were calculated by
population wealth quintiles. Inequality in child mortality was
assessed by computing two measures of relative inequality (the rate
ratio and the concentration index) and two measures of absolute
inequality (the difference and the Erreyger's index). Results: Mean
child mortality rates ranged from about 39 deaths per 1,000 live
births in Cairo (2008) to about 107 deaths per 1,000 live births in
Dar es Salaam (2010). Significant inequalities were found in
Kinshasa, Luanda, Abidjan, and Addis Ababa in the most recent
survey. The difference between the poorest quintile and the richest
quintile was as much as 108 deaths per 1,000 live births (95%
confidence interval 55 to 166) in Abidjan in 2011-2012. When
comparing inequalities across cities or over time, confidence
intervals of all measures almost always overlap. Nevertheless,
inequalities appear to have increased in Abidjan, while they appear
to have decreased in Cairo, Lagos, Dar es Salaam, Nairobi and
Dakar. Conclusions: Considerable inequalities exist in almost all
cities but the level of inequalities and their development over
time appear to differ across cities. This implies that inequalities
are amenable to policy interventions and that it is worth
investigating why inequalities are higher in one city than in
another. However, larger samples are needed in order to improve the
certainty of our results. Currently available data samples from DHS
are too small to reliably quantify the level of inequalities within
cities.