Multilevel modeling of unintended current pregnancy: In the case of Ethiopian Demographic and Health Survey, 2016.

Abstract

Unintended pregnancy has been a major public health and reproductive health issue imposing a great adverse consequence on the mother and child. However, estimates of unintended pregnancy through the appropriate model are lacking. This study is aimed at modeling and assessing the extent of variation and factors associated with unintended pregnancy among women in Ethiopia. A cross-sectional study was conducted based on 2016 Ethiopian Demographic and Health Survey data related to the reproductive health of 1122 currently pregnant women and a multilevel modeling approach was used. The proportion of unintended current pregnancies was 20.1%. According to random intercept with a fixed slope model, women who had 1 to 3 living children and those who had 4 and above were more likely to be unintended (OR = 3.54, 95% CI: 1.985-6.332) and (OR = 5.47, 95% CI: 2.67-11.227), respectively, compared to women with no living children. Also, married women were less likely to be unintended (OR = 0.14, 95% CI: 0.065-0.304) compared to unmarried women. In addition, women having work were more likely to be unintended (OR = 1.56, 95% CI: 1.079-2.255). Furthermore, women who intend to use contraceptive methods were less likely to be unintended (OR = 0.54, 95% CI: 0.362-0.796) compared to women who do not intend. The number of living children, current marital status, the intention of contraceptive use, and respondents' working status were found to have a significant effect. Giving attention to regional variations and intention of contraceptive use is important to reduce unintended current pregnancies in Ethiopia.