STRUCTURAL EQUATION MODELLING OF PROXIMATE DETERMINANTS OF FERTILITY AMONG WOMEN LIVING WITH HIV IN IBADAN, SOUTH WEST NIGERIA.

Abstract

Proximate determinants of fertility, behavioural and biological factors through which socioeconomic variables influence the fertility, plays a crucial role in influencing the fertility. Understanding of these variables is essential for the proper integration of HIV care and reproductive health services. Thus, this study employs structural equation modelling in the assessment of interrelationship among the proximate determinants of fertility among women living with HIV in Ibadan, Nigeria. A data from a cross-sectional study on childbearing progression among 933 women living with HIV and receiving care at University College Hospital, Ibadan, Nigeria was extracted and analyzed. ANOVA was used for association and comparison of mean of children ever born for the different categories of explanatory variable(s). Structural Equation Modeling was employed to assess the inter relationship among the various proximate determinants of fertility at 5% significance level.The average age of the women is 38.05 (± SD = 6.08) and most of the women had their first sex at mean age of 20.41 (± SD = 3.68) with 24.27(± SD = 4.57) as their age at first birth. Almost all the women, 812 (87.7%) ever used contraceptive with the condom, 753 (80.7%), as the most preferred method of the contraceptives. Most of the women had 3 or more children before the HIV diagnosis. ANOVA result showed that women who had their first marriage and started child bearing before the age of 24 years had significantly higher number of children ever born compared to women who married and started child bearing at age 25 and above and significant at p < 0.005. Women who had at least secondary education had significantly lower number of children ever born than those with Primary or no education at P < 0.05. Structural Equation Model provided a good fit to the data with each indicator contributing significantly to their respective constructs at P < 0.05. The model showed that there is inter relationship between the proximate variables. Marriage (β= -0.10,p=0.305) had direct negative relationship with the contraceptives while contraceptives (β= -0.072,p=0.036) had direct negative relationship with postpartum infecundability. Increase in the component of marriage will cause a corresponding decrease in the use of contraceptives. The model showed that marriage (β= -0.227,p=0.000) and postpartum infecundability (β= -0.032,p=0.953) had a direct negative relationship with the children ever born while contraceptives had direct positive relationship with the children ever born. Women who married and started childbearing late, will have fewer number of children ever born compared to woman who married and started childbearing at early stage. The model had showed that education, age, age at first sex, age at first birth, and duration of postpartum were associated with fertility. Based on this result, intervention targeting the integration of HIV care and reproductive health service should be given adequate attention.