A comparative study of the determinants of infant & child mortality in Kenya & Ghana
Abstract
Under-five mortality (USMR) and infant mortality (IMR) have been used as measures of
children's well-being for many years (Hill et al, 1996). The high number of children under
five years of age dying each year in Africa is shocking, particularly compared to the
dramatic progress in the rest of the world. While the number of under-five deaths outside
Africa fell from 18.1to 6-4 million (6S percent decrease) annually between 1960 and 2000,
under-five deaths in Africa increased from 2.3 to 4.S million (96 percent increase). With
the technology to address the principal causes of child death now available at reasonable
cost and being applied successfully in other regions, the situation in Africa represents a
grim picture of the failure of African countries and the global community to address this
basic humanitarian issue (SARA, 200S). The Ghana Demographic and Health Surveys
carried out in 1993, 1998 and 2003 indicated that USMR decreased from 119 to 108 and
leveled off to 111per 1000 live births (GSS and MIl, 1999 and 200S). Results of the 1998
and 2003 Kenya Demographic and Health Surveys showed that USMR underwent an
increase from 112to 114 having been 96 per 1000 live births in 1993 (NCPD, CBS & MIl,
1999 and 2003)·
The main study objective was to carry out a comparative analysis of the determinants of
infant and child mortality in Kenya and Ghana. Data was obtained from the Demographic
and Health Surveys carried out in both countries in 2003. Survival analysis was utilized to
calculate the probability of a child's risk of dying within the first five years of life due to
socioeconomic among other factors. Children born alive were th~ f~us of study and Cox's
proportional hazard model was used to account for censoring' in the estimation of
exposure time since not all children had had the chance to survive to the oldest age under
investigation by the time of the interview ~
Results on the analysis of the factors associated with infant and child deaths in Kenya and
Ghana indicated that in Kenya, maternal education, birth order, preceding birth interval
and ethnicity were strongly associated with risk of infant death. In Ghana on the other
hand, the most crucial determinants of the risk of infant death were maternal level of
1 Censored observations occur whenever the dependent variable of interest represents the time to a
terminal event, and the duration of the study is limited in time. Censored observations may occur
in a number of different areas of research, for example, in the demography we may study the
"survival" of children. By the end of the study period, some children will still be alive and such
subjects represent censored observations.
education, birth order and preceding birth interval. Again in Kenya, risk of childhood
death was robustly associated with maternal education, maternal age at birth and
ethnicity while in Ghana, maternal education, source of drinking water, type of toilet
facilities and marital status were strong determinants of the risk of childhood death.
From the findings, it is quite clear that in both countries infant mortality determinants
were more biological than those of childhood mortality which were more to do with the
mother's socioeconomic status. As hypothesized, maternal education was largely
associated with decreased risk of infant and child death. This was however not the case for
infant mortality in Ghana where infants of women with primary education had increased
risk of death. Higher maternal education leads to better hygiene and sanitation for
children hence increasing their survival chances but low level of education has been found
to be associated with increased risk of death of children in the first two years of life.
Both Ghana and Kenya show an increased risk of infant death due to short preceding birth
intervals. Preceding birth interval is ideally supposed to be at least 24 months so as to give
the mother ample time to recuperate after birth and take good care of the newly born child
as well as limit the number of children in the household. Short birth intervals expose both
mother and child to ill health hence one or the other or both has an elevated risk of death
compared to those with longer birth intervals. Higher birth order also increases chances of
death of both mother and child due to maternal depletion and other stress associated with
many children e.g. exposure to more pathogens and financial strain. Ghana showed a
higher association with risk of infant death from short birth intervals and high birth order
than Kenya.
Ethnicity was strongly associated with infant and child deaths in Kenya which can be
explained by the different environments and socioeconomic situation these people are
exposed to. Elevated risk of child deaths is dominant among those that live in areas that
are prone to Malaria, drought/famine and are less-urbanized e.g. Nyanza, Coast, Eastern
and Western Province. The environmental factors such as drinking water and toilet
facilities became important in childhood since this is the time children get more exposed
to the environment around them. Socioeconomic situation is strongly linked to the
hygiene and sanitation that children are exposed to with better facilities increasing
children's survival status.
The study findings pointed out the measures that could be taken to improve child survival
more so in Kenya and Ghana. First, education exposes women to the knowledge of better
healthcare, hygiene and sanitation i.e. seeking prenatal and post natal health services,
delivering in appropriate health facilities, having their children immunized and ensuring
that the environment under which they raise them is clean. It also allows women to
understand that they have the right and obligation to plan and appropriately space their
births in order to give their children the best living conditions.
Citation
A project submitted in partial fulfillment for the award of the degree of Master of Science in Population Studies at the University of Nairobi.Publisher
Institute of population Studies and research