Analysis of Postpartum Variables in Kenya, Evidence From Kenya Demographic and Health Survey (1989)
Abstract
The objectives of this study were three-fold. First, to estimate the mean duration of
postpartum variables using life table technique and Mosley prevalence method. Second, to examine
whether there exists a relationship between durations of breastfeeding and postpartum amenorrhea
by applying Bongaarts' model. Third, to evaluate the effect of mothers' socioeconomic, sociocultural
and demographic characteristics on postpartum practices. The data used in this study is
the Kenya Demographic and Health Survey of 1989.
The mean duration of breastfeeding, postpartum amenorrhea and postpartum sexual
abstinence were estimated for various variables at the national level in Kenya. This was so
because the sample size used could not enable the analysis to be carried out at the provincial or
district level. These estimates are derived by life table and mosley prevalence methods, after which
the two sets of estimates are compared.
Bongaarts' model was used to estimate the expected durations of amenorrhea from
observed durations of breastfeeding, thereby, comparing the expected durations of amenorrhea
with the observed durations. Simple regression analysis was used to estimate the correlation
coefficient (R2) of the two sets of data.
The variables that were used at the national level include; level of education,' place of
.~ --r residence, work status, parity, age of the m.91ber-.--atchild's birth, sex of the child, region
/'
I
(provinces), religion, ethnicity and marital status.
A multiple regression analysis has been used to evaluate the effect of each variable on
duration breastfeeding, amenorrhea a~d sexual abstinence separately. In regression analysis
contraceptive use was included to the above mentioned variables.
The findings of the study show that durations of all the three postpartum variables are
decreasing. The study asserts that variables which negatively affect durations of postpartum
variables are mothers' level of education, work status and place of residence. However, mothers'
iv
age at the child's birth is found to be positively related to duration of postpartum practices.
Regression analysis shows that all socio-economic and demographic variables considered in the
analysis have a considerable effect on duration of postpartum variable. The most notable finding
is that the duration of amenorrhea is mainly determined by the duration of breastfeeding.
The study recommended that breastfeeding education should be incorporated in school
syllabus at secondary level. Programmes on advantages of prolonging breastfeeding duration be
intensify both on print and mass media to increase public awareness.
Sponsorhip
The University of NairobiPublisher
University of Nairobi population studies and research Institutet