The Interrelationship between Population Dynamics and Childhood Mortality in Butajira District, South Central Ethiopia
No Thumbnail Available
Date
2012-06
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Addis Ababa University
Abstract
The population of Ethiopia is still growing by 2.6% per annum mainly due to high fertility of 4.8
children per woman. Though increasing in the current decade, only about a quarter of married
women used family planning methods in 2011. Though early childhood mortality diminished in
recent years, this decline was not statistically significant in infant mortality. Levels and extent of
reductions in the three components of population change and their relationships varied across
different regions and by urban-rural setting in the country.
Previous studies elsewhere showed relationship between population dynamics and health.
Studies also revealed that intervention in one component of population change affects another
component. A body of literature on insurance and replacement fertility response of childhood
mortality were documented in least developing countries. Moreover, selection, disruption,
adaptation, environmental theories had also documented the relationship between migration with
fertility and under-five mortality. Migration might select people with different fertility behavior
and childhood mortality experience compared to those without such behavior. If those with less
number of children migrate, the fertility and child mortality of non-migrants in the place of
origin would be inflated compared to those of the migrants. Among the latter group, the
disruption due to migration might contribute to reduction in fertility or increase in early
childhood mortality. Besides, the group might adapt the fertility behavior of the population in the
area of destination.
In this regard, this thesis aimed at measuring levels of and assessing relationships between
fertility, contraception, under-five mortality and migration in the designated area of the study. It
also tried to identify main proximate and distal factors of each of these components of population
change in the context of the recently introduced village-based health extension program,
reproductive health strategy and population policy in the densely populated Butajira District of
South Central Ethiopia.
Methods
The study was hosted by the Butajira Demographic Surveillance System which is located about
135 kms from Addis Ababa in the southern direction. Qualitative and quantitative methods were
employed in this study. The quantitative research used two data sources. The longitudinal
surveillance database up to the end of 2008 was extracted to recruit study women of reproductive
age for the cross-sectional study which aimed at measuring levels and identifying determinants
of fertility, contraception and the unmet need of family planning. The database was also used to
have a detailed insight into early childhood mortality and out-migration in the district. Standard
data collection instruments of the INDEPTH-Network and Measure DHS were contextually
adapted for the longitudinal database and the cross-sectional survey respectively. A priori focus
group discussions were held to incorporate the community’s terminologies and opinions. A total
of 11,133 women of reproductive age were recruited from the surveillance database and 9,996 of
them responded positively.
Especially trained and experienced field staff collected the data. There was rigorous supervision.
Data sources were managed by softwares having internal consistency checking mechanisms.
Cleaning was done at desk. Serious anomalies were taken back to the field for reconciliation,
while others were rectified by imputing values from logical flows in the questionnaire.
Frequency distributions, cross-tabulations and graphical presentations were done. Event history
analysis was used to calculate person time of exposure, incidence and prevalence rates using
longitudinal data. Odds ratio along with the 95% confidence interval in binary logistic regression
was used to determine association between covariates and the binary outcome of interest. In the
case of fertility, Bongaart’s model to measure the inhibition effects of proximate determinants
and the incidence rate ratio in Poisson regression along with the 95% confidence interval was
used to measure the association between fertility and covariates. Poisson regression was also
used to measure associations of background characteristics with out-migration and under-five
mortality. Assumptions of all the statistical models used in this study were checked.
Results
The total fertility rate of 5.3 children per woman was high and comparable to the rest of Ethiopia
with rural-urban disparity (Highland, TFR=5.7, Lowland, TFR=6.6 and Urban, TFR=3.3).
Postpartum infecundability due to breastfeeding (Ci=0.68) significantly deducted fertility from
its biological maximum. The contribution of contraception (Cc-u=0.57, Cc-e=0.43) and nonmarriage
(Cm-u=0.53, Cm-e=0.41) was important among urbanites and educated women. Abortion
contributed a significant role to reduce fertility among school youth (Ca=0.76). The fertility
incidence rate ratio was 1.38: 95% CI (1.27, 1.49) times higher among those married before their
15 birthday, 1.24: 95% CI (1.10, 1.39) times higher among uneducated, 1.95: 95% CI (1.84,
2.06) times higher among those families with large size, 1.67: 95% CI (1.59, 1.76) times higher
with child death experience and 1.06: 95% CI (1.01, 1.13) times higher among women living in
food-secured households compared to their counterparts. Against other findings, fertility was
1.09: 95% CI (1.04, 1.15) significantly higher among women with no child sex preference.
Besides, migration status of women did not seem to predict their fertility levels (1.02: 95% CI
(0.97, 1.07)).
The contraceptive prevalence rate of 25.4%: 95% CI (24.2, 26.5) in Butajira District was
comparable though unmet need of 52.4%: 95% CI (51.1, 53.7) was very high compared to
national and regional estimates. Full stock out and absence of methods’ mix, religion, complaints
related to providers and methods, assumption of having proper diet, and optimum workload
when using family planning methods were barriers of contraceptive use mentioned by study
women in the area. The odds of contraception was 2.3: 95% CI (1.66, 3.18) times higher among
urbanites, 1.99: 95% CI (1.38, 2.88) times higher among those completed secondary level of
education and 1.5: 95% CI (1.12, 2.01) times higher among women whose partners completed
secondary plus level of education, 1.3: 95% CI (1.13, 1.5) times higher among women with no
experience of child death, 2.21: (1.8, 2.7) times higher among couples who discussed on
contraception and 2.59: 95% CI (2.11, 3.17) times higher among women whose partners’ support
family planning use compared to their counterparts.
Under-five mortality level of 29 per 1000: 95% CI (27.4, 31.8) in the District recorded over the
22 years of surveillance was low. The difference between infant mortality of 86.6 per 1000: 95%
CI (77.4, 96.9) and child mortality of 19.2 per 1000: 95% CI (17.4, 21.3) was higher. Compared
to their counterparts, the study also showed 0.85: 95% CI (0.79, 0.80) times lower under-five
mortality among female children, 1.14: 95% CI (1.03, 1.25) times higher under five mortality
among Muslim and 15.24: 95% CI (13.75, 16.89) times higher among minority Christian
families, 1.31: 95% CI (1.04, 1.66) and 2.02: 95% CI (1.58, 2.59) times higher among rural
highlanders and rural lowlanders respectively, 1.54: (1.43, 1.67) times higher among families
owning oxen, and 1.92: 95% CI (1.66, 2.22) times higher among families owning houses and
2.4: 95% CI (1.89, 2.06) times higher among those living in rented houses and 2.13: 95% CI
(1.79, 2.53) times higher in children living in houses roofed with thatched grass, and 1.46: 95%
CI (1.26, 1.69) times higher among those living in the neighborhoods located 5-9 kilometers
away from Butajira zonal hospital.
The study also revealed high out-migration of 3.97 per 100 person years (3.93, 4.01) in the
district. The risk of out-migration was 0.94: 95% CI (0.92, 0.96) times lower among females,
1.9: 95% CI (1.85, 1.96), 1.77: 95% CI (1.71, 1.82), 1.55: 95% CI (1.49, 1.62), 1.23: 95% CI
(1.17, 1.29) or 2.82: 95% CI (2.66, 2.98), 1.29: 95% CI (1.26, 1.32), 4.71: 95% CI (4.56, 4.87),
1.18: 95% CI (1.15, 1.22), 1.58: 95% CI (1.52, 1.64) and 2.11: 95% (2.04, 2.18) times higher
among teenagers, the youth, unmarried, primary school completes or above, Orthodox and
minority Christians, urbanites, and those living in rented houses and owned by others compared
to their respective counterparts.
Some relationship between the three components of population change was also observed. There
was statistically significant association between early childhood mortality and fertility (6.07:
95% CI (5.36, 6.87)). However, the association between fertility and migration status was not
statistically significant (1.05: 95% CI (0.92, 1.19)). Neither was the association between underfive
mortality and migration statistically significant (1.04: 95% CI (0.92, 1.19)).
Conclusions and Recommendations
Fertility was still high in the study community with high rural urban disparity. The most effective
proximate determinant to deduct fertility from its biological maximum level was non-marriage
due to disruption of marriage through migration of one of the partners. The contribution of
contraception and non-marriage was also important among urbanites and educated women.
Postpartum infecundability also significantly reduced fertility from its biological maximum in
rural areas and among uneducated women. Abortion had also played an important role in
reducing fertility among in-school youth. Delayed marriage, higher education, smaller family
size, absence of child death in the family, and living in food-secured households were also
significantly associated with small number of children. Besides, fertility was significantly higher
among women with no child sex preference. However, migration status of women was not
statistically significant.
The contraceptive prevalence rate in Butajira District was still low, though unmet need was very
high. Barrier to contraception in the area included, stock out and absence of preferred family
planning methods, religion, complaints related to providers and methods, assumption of having
proper diet, and optimum workload when using family planning methods. Significant predictors
of contraception in the district included urban residence, women’s and their partners’ educational
status, child death experience, couple’s discussion on contraception, and partners’ support.
The magnitude of overall early childhood mortality levels in the district recorded over the 22
years of surveillance, though low compared to the national and regional level was still high.
Infant mortality was higher than child mortality in the district. Under-five mortality was
significantly higher among male children, families confessing Muslim and non-Orthodox
Christian denominations, rural residents, families owning oxen, those having their own houses,
families living in rent-free houses, households living in houses with roofs of thatched grass, and
families living in neighborhoods located between 5-9 kilometers from the zonal hospital as
compared with their counterparts.
A high incidence of out-migration was observed in the district with higher level among males,
teenagers, the youth, primary and secondary education or above completes, those not in marital
unions, Christians, urbanites, and families in rented and owed houses compared to those in
owned ones.
This study had also showed statistically significant association between early childhood
mortality and fertility. The association between fertility and migration was not statistically
significant. Neither was the case between early childhood mortality and migration in the study
area.
We recommend that the ills of fast population growth and its consequences should be intensively
informed to the public. Women must be encouraged to sustain the practice of extended
breastfeeding. Efforts should also be exerted to increase contraceptive use in rural communities.
Besides, in-school youth should be aware of post-abortion complications and youth friendly
family planning methods to reduce fertility, maternal mortality and childhood mortality in the
community. Longer years of women’s education should be scaled up. Health systems in Butajira
District and the capacity of staff should be strengthened. The Government should avail family
planning methods with appropriate method mix and increase competence of providers on
managing temporary side effects. More rigorous child and maternal health education should be
channeled through village-based health extension workers. Household hygiene, antenatal care,
immunization and facility based delivery in the district should be scaled up. More efforts should
also be exerted to improve the quality of residential houses. An insurance scheme to care for the
elderly should be put in place to bring about change in the behavior of families towards small
family size. We suggest that local authorities need to facilitate local employment and housing
opportunities for retaining young and educated people in their own areas, to safeguard the future
well-being of the entire population
Description
Keywords
Childhood Mortality