Intersectional inequalities in demand for family planning satisfied with modern contraceptive methods in Ethiopia: a cross-sectional MAIHDA analysis of the 2024–25 Ethiopia Demographic and Health Survey
Year: 2026
Type: preprint
Abstract: Background National family-planning averages may conceal inequalities across combinations of reproductive age and socioeconomic position. We examined intersectional inequalities in demand for family planning satisfied with modern contraceptive methods (mDFPS) in Ethiopia and assessed how much between-stratum heterogeneity was accounted for by additive social gradients versus residual intersectional departures. Methods This cross-sectional study used data from the 2024–25 Ethiopia Demographic and Health Survey Individual Recode dataset. The analytic sample comprised 6,611 women aged 15–49 years with family-planning demand. Survey-weighted mDFPS was estimated by age, wealth, education and residence. The primary MAIHDA specification cross-classified four age groups, five wealth quintiles, three education categories and two residence categories, yielding 120 theoretical intersectional strata. Null and additive multilevel logistic models estimated between-stratum variance, the variance partition coefficient (VPC) and proportional change in variance (PCV). Sensitivity analyses used a more granular 210-strata specification and an unweighted primary model. Results National survey-weighted mDFPS was 67.3%. mDFPS was 61.9% (95% CI 53.4–70.4) at ages 15–19 years, 74.8% (70.2–79.4) at ages 20–24, 72.6% (69.6–75.6) at ages 25–34 and 56.9% (53.2–60.7) at ages 35–49. Coverage increased from 43.8% (38.5–49.0) in the poorest wealth quintile to 77.3% (73.1–81.6) in the richest, and from 53.7% (49.7–57.6) among women with no education to 78.3% (75.4–81.2) among those with secondary or higher education. In the primary MAIHDA model, between-stratum variance declined from 0.858 (VPC 20.7%) in the null model to 0.175 (VPC 5.1%) after inclusion of the additive social-position variables, corresponding to a PCV of 79.6%. Among strata with at least 20 women, the lowest model-predicted mDFPS was 19.6% (approximate 95% interval 9.6–36.1) among rural women aged 15–19 years in the poorest quintile with no education. The 210-strata sensitivity specification produced a similar reduction in between-stratum variance (PCV 77.0%). Conclusions mDFPS in Ethiopia is strongly patterned by age, household wealth and education. Most between-stratum heterogeneity was accounted for by additive socioeconomic and life-course gradients, while a smaller residual intersectional component remained. Equity-oriented monitoring and programming should address these broad structural gradients while retaining intersectional disaggregation to identify combinations of women with particularly low coverage.
Source: Research Square
Author: Geteneh Moges Assefa
Institutions: Amref Health Africa, Amref Health Africa, Amref Health Africa
Language: English
Field: Medicine
Domain: Health Sciences
Keywords: intersectional inequality, family planning demand, modern contraceptive methods, contraceptive satisfaction, women of reproductive age, social gradient, variance partition coefficient, wealth, residences, Ethiopia, reproductive health equity, adolescent contraceptive use, urban-rural differences, maternal and child health financing
Open Access status: green
OpenAlex ID: W7225889905
