Background: Despite a 35% reduction in infant mortality rate (IMR) in Ghana between 1990
and 2012, there were still 38,000 infants in 2012 that died before their first birthday. In order
to address a further reduction of infant mortality (IM) and IMR it is important to gain insight
into the determinants that are associated with IM in Ghana. Results from previous research
implied a protective effect of good housing, but the evidence for the association between
housing characteristics and IM is conflicting.
Aim: To examine the association between housing characteristics and IM in Ghana,
controlling for socio-economic and demographic variables.
Methods: Secondary analyses of data from the birth recode files of the 2003 and 2008
Ghanaian Demographic and Health Surveys were performed, using multivariable logistic
regression for both cross-sectional and longitudinal analyses.
Results: Results showed that it is mostly socio-economic and demographic variables that were
significantly associated with IM in Ghana. Housing characteristics significantly associated with
IM were unimproved drinking water, unfinished flooring and solid fuel use. The use of solid
fuels increased the risk of IM, whereas unimproved drinking water and unfinished flooring
decreased the risk of IM. In addition, the risk of IM decreased across time.
Conclusion: Solid fuel use was associated with an increased risk of IM in Ghana, whereas other
housing characteristics did not seem to be (important) determinants of IM.
Clinical Relevance: Reducing IM in Ghana through housing is expected to have only limited
results. New, longitudinal data on the effect of housing upgrading on IM hopefully provide a
clear insight into the association between housing characteristics and IM.