# Multilevel analysis of factors associated with abortion among adolescents in Uganda insights from UDHS 2022 dataset

**Authors:** Stephen Mungau, Joan Nanteza, Genevieve Dupuis, Niguss Cherie Bekele, Niguss Cherie Bekele

PMC · DOI: 10.1371/journal.pgph.0005822 · PLOS Global Public Health · 2026-01-12

## TL;DR

This study explores factors linked to abortion among Ugandan adolescents using 2022 survey data, finding that early sexual activity, education, and marital status significantly influence abortion rates.

## Contribution

The study provides new insights into regional and individual-level factors affecting abortion among Ugandan adolescents using the UDHS 2022 dataset.

## Key findings

- 11.0% of Ugandan female adolescents reported having an abortion.
- Age at first sex, education level, and marital status were key determinants of abortion.
- Regional factors accounted for 4.9% of the variation in abortion rates.

## Abstract

Unsafe abortion is a major reproductive health challenge, causing 7.9% of global maternal deaths and 9.6% in East Africa. In Uganda, about 8% of maternal deaths result from unsafe abortions. Early sexual activity, poor access to sex education, restrictive laws and stigma push adolescents into unsafe practices. Limited safe services force many to use dangerous methods leading to severe complications and high maternal mortality. This study examined determinants and prevalence of abortion among Ugandan female adolescents using the 2022 Uganda Demographic and Health Survey dataset of 5,125 females aged 15–24 who had ever engaged in sexual activity. The dependent variable was binary (1 for ever terminated, 0 for never). Weighted data were analyzed using descriptive statistics, ordinary and mixed effect logistic regression models to explore individual- and cluster-level influences. Intra-class correlation and likelihood ratio tests assessed cluster variation. Findings showed 562 adolescents had ever aborted. Those whose first sex was before age 15 were 3.44 times more likely to abort compared to those aged 20–24 while those aged 15–17 were 2.24 times more likely. Married adolescents had twice the odds compared to never married, and cohabiting adolescents were 2.44 times more likely. Compared to those with education beyond secondary, adolescents with no education, primary and secondary schooling were 5.8, 2.99 and 3.01 times more likely to abort. Regional variations accounted for 16.8% of variance, with intra-class correlation of 4.9%. Overall, 11.0% of Ugandan female adolescents reported abortion. Key determinants included age at first sex, marital status, education, contraceptive use and internet use. Region-level factors contributed 4.9% of variation highlighting the need for cluster-level interventions alongside individual approaches.

## Full-text entities

- **Diseases:** sexual (MESH:D050035), chronic pelvic pain (MESH:D011472), pregnancy (MESH:D011254), miscarriages (MESH:D000022), DHS (MESH:C566369), death (MESH:D003643), infertility (MESH:D007246), maternal (MESH:D000079262), Abortion (MESH:D000026), infection (MESH:D007239)
- **Chemicals:** PGPH-D-25-01751 (-), NO (MESH:D009614)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12795349/full.md

## Figures

2 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12795349/full.md

## References

54 references — full list in the complete paper: https://tomesphere.com/paper/PMC12795349/full.md

---
Source: https://tomesphere.com/paper/PMC12795349