| dc.description.abstract | The duration of postpartum amenorrhea plays a dominant role in shaping the fertility behavior of any population. The main objective of this study was to estimate the duration of postpartum amenorrhea using data on duration of breastfeeding. Specifically, the objective of the study was to assess the fit of Bongaarts & Potter (1983) and Yadava & Islam (1994) models to Kenyan data and then use the appropriate model to estimate trends in breastfeeding effects on fertility in Kenya. Also, the study sought to examine the extent of heaping in the breastfeeding data, and then re-distribute the excess numbers on heaped point using a Three-point technique to produce a smooth distribution.
The data used was from four Kenya Demographic and Health Surveys (KDHS) of 1989, 1993, 1998 and 2003. This study was focused on the breastfeeding experience of the mothers to their youngest child who has been ever breastfed. Generally, the results of the distribution of breast feeding data showed marked heaping of breastfeeding duration at 3, 6, 12, 18, and 24 months for all the four KDHS data sets. Srinivansan (1980) method of calculating digit preference indicate that approximately 32% to 35% of respondents preferred reporting duration of breast feeding in 12 months in 1989, 1993, 1998 and 2003 KDHS. The results of the prevalence/incidence mean shows that the mean duration of breastfeeding for the last child in 1989, 1993, 1998, and 2003 KDHS is 23.91, 24.06, 23.95, and 24.46 months respectively. Hence, there was no significant change in the mean duration of breastfeeding.
In the Bongaarts and Potter (1983) equation, the estimated amenorrhea periods were nuch lower than the observed duration of amenorrhea for all the durations of rreastfeeding less than 18 months. On the other hand, for durations of breastfeeding nigher than 18 months, Bongaarts and Potter (1983) equation estimated amenorrhea periods higher than the observed duration of amenorrhea. In case of overall population iverage, Bongaarts and Potter equation gave an estimate of an amenorrhoea period as l6.92, 17.51, 16.96, and 17.39 months compared to the observed averages of 7.66, 8.62, 7.61 and 7.75 months respectively for 1989, 1993, 1998 and 2003 KDHS. In the Yadava & Islam (1994) model the estimates for durations of amenorrhea period were much closer to the observed durations of amenorrhea periods except for breastfeeding durations <2 and 2-4 months.
In case of overall population average, Yadava & Islam model gave estimates of the average amenorrhoea period as 13.22, 13.41, 13.23, and 13.38 months relative to observed averages of 7.66, 8.62, 7.61 and 7.75 months respectively for 1989, 1993, 1998 and 2003 KDHS. For the impact of the models, results indicate that Bongaarts and Potter (1983) equation over estimates the fertility inhibiting effect of breastfeeding by almost 24 percent for the four data sets, whereas Yadava and Islam overestimates by only 3 percent. The study recommends that further testing be done to show the relationship among different status and frequency of breastfeeding, age, level of education and parity on postpartum amenorrhea. | en_US |
