Content area
Abstract
Sterility is defined as the inability of a noncontracepting, sexually active woman to have a live birth. Sterility has proven difficult to measure. A review of previously proposed methods for estimating sterility showed that there was no adequate procedure for estimating age-specific rates of sterility from incomplete birth histories. Also, no demographic method for obtaining individual level estimates of sterility was available. This thesis extends previously proposed group measures of sterility to apply to both complete and incomplete birth histories. In addition, procedures for estimating an individual woman's age at onset of sterility are suggested.
In a simulation study, group measures as well as individual measures of sterility were found to be quite robust to variations among populations in the age pattern of sterility and in other variables which affect the reproductive process in the absence of deliberate fertility control. Sampling variation was demonstrated to have a negligible impact on estimates of sterility when samples include at least 500 women.
On the basis of recent data collected in collaboration with the World Fertility Survey, the methods developed here are applied in a comparative study of the levels and the differentials of sterility in Cameroon, Kenya and Sudan. This analysis showed that sterility is highly prevalent in Cameroon, moderate in Sudan and relatively low in Kenya.
The differentials of sterility in Cameroon, Kenya and Sudan were analyzed in a hazards models analysis. It was found that within country differentials in sterility are substantial, and that a similar patterns prevail across the three countries. The prevalence and incidence of sterility have not changed during recent decades in either Cameroon or Sudan, but have increased slightly in Kenya.
Among other results, it was found that sterility varies considerably across regions and by urban or rural residence. This pattern suggests that sterility is linked to different sexual behaviours.
The present study demonstrates that age-specific rates of sterility and an individual woman's age at onset of sterility can be estimated in noncontracepting populations from information about women's marital histories, ages at last live birth and ages at censoring. The empirical results confirm that World Fertility Survey data can yield useful information about the levels and correlates of sterility. (Abstract shortened with permission of author.)
