EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Bayesian Spatial Conditional Logistic Regression Modelling of Overweight/Obesity Prevalence and Determinants Among Women in Ghana, Accounting for Contraceptive Use

Killian Asampana Asosega iD, Eric N Aidoo iD, Atinuke Olusola Adebanji iD, Ellis Owusu-Dabo iD

Global Health, Epidemiology and Genomics · 2026-01 · 2026 (1)

DOI: 10.1155/ghe3/3074676 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Global Health, Epidemiology and Genomics, online erschienen am 29.07.2026. Verfasst von Killian Asampana Asosega und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Statistics and Actuarial Science, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana, knust.edu.gh. Hauptthemen laut MEDLINE: Overweight, Obesity, Contraception Behavior. Die Arbeit nennt 86 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Overweight/obesity (OW/OB) poses a major public health challenge among reproductive‐aged women across most LMICs, including Ghana. The continuous rise in maternal average BMI coupled with fragile healthcare system makes surveillance and interventions to reverse the trend in Ghana as well as most sub‐Saharan African (SSA) countries, a pertinent issue.

Objective

This study identifies risk factors associated with OW/OB among reproductive‐aged women in Ghana, accounting for spatial autocorrelation and the potential confounding effect of modern contraceptive use (MCU). It also provides updated information on the distribution of OW/OB risk. MCU was used as a stratification variable to control for potential confounding related to contraceptive‐induced metabolic or behavioural influences.

Methods

The recent cross‐sectional Ghana Demographic and Health Survey (2022 GDHS) data, made up of 7054 women aged 15–49 years, was used in this study. The Bayesian spatial conditional logistic regression model was developed to determine significant factors.

Results

About 36.9% of the women were overweight/obese (OW = 22.5% and OB = 14.4%). A significantly high maternal OW/OB rate (> 50) was observed in the Ashanti and Greater Accra regions and the least prevalence (< 20%) in the North‐East region. Regional disparity rates indicate that middle and southern Ghana are associated with higher burden compared to northern Ghana. The Bayesian conditional logistic model with convoluted CAR prior emerged as the best model for studying maternal OW/OB risk while accounting for MCU with the minimum WAIC and DIC values. Significantly higher risk of OW/OB burden is linked to high education (primary [aOR = 1.87, 95% CI:2.27–6.50]; secondary [aOR = 2.02, 95% CI: 2.42–7.60]; and higher [aOR = 2.05, 95% CI: 1.59–2.64]), increase in age (20–29 [aOR = 3.34, 95% CI: 2.67–4.17]; 30–39 [aOR = 6.07, 95% CI: 4.66–7.89]; and 40–49 [aOR = 7.22, 95% CI: 5.40–9.65]), being married (aOR = 1.81, 95% CI: 1.49–2.19), urban residency (aOR = 1.41, 95% CI: 1.23–1.61) and the number of children (aOR = 1.06, 95% CI: 1.03–1.11).

Conclusion

The spatial heterogeneity in OW/OB risk provided evidence for regional targeted preventive strategies, particularly in the Ashanti and Ahafo regions, which are significant hotspots for high OW/OB risk, to help reduce the public health burden of maternal OW/OB in Ghana. This could contribute to ensuring the quality health of women (SDG 3).

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

OverweightObesityContraception BehaviorHumansFemaleGhanaAdultAdolescentPrevalenceBayes TheoremMiddle AgedLogistic ModelsYoung AdultCross-Sectional StudiesRisk FactorsBody Mass Index

Stichwörter der Autor:innen

Bayesian modelSustainable Development Goal 3conditional logistic regressionoverweight and obesityspatial model

Autor:innen und Institutionen

  • Department of Statistics and Actuarial Science, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana, knust.edu.gh.
  • Department of Mathematics and Statistics, University of Energy and Natural Resources, Sunyani, Ghana, uenr.edu.gh.
  • School of Mathematical and Computer Sciences, Heriot-Watt University, Dubai, UAE, hw.ac.uk.
  • Department of Statistics, Purdue University, Indianapolis, Indiana, USA, purdue.edu.
  • Department of Global and International Health, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana, knust.edu.gh.

ORCID-Profile: Killian Asampana Asosega, Eric N Aidoo, Atinuke Olusola Adebanji, Ellis Owusu-Dabo

Zitieren

APA

Asosega, K. A., Aidoo, E. N., Adebanji, A. O., & Owusu-Dabo, E. (2026). Bayesian Spatial Conditional Logistic Regression Modelling of Overweight/Obesity Prevalence and Determinants Among Women in Ghana, Accounting for Contraceptive Use. Global Health, Epidemiology and Genomics, 2026(1). https://doi.org/10.1155/ghe3/3074676

Vancouver

Asosega KA, Aidoo EN, Adebanji AO, Owusu-Dabo E. Bayesian Spatial Conditional Logistic Regression Modelling of Overweight/Obesity Prevalence and Determinants Among Women in Ghana, Accounting for Contraceptive Use. Global Health, Epidemiology and Genomics. 2026;2026(1). doi:10.1155/ghe3/3074676

BibTeX
@article{Asosega2026,
  author = {Asosega, Killian Asampana and Aidoo, Eric N and Adebanji, Atinuke Olusola and Owusu-Dabo, Ellis},
  title = {Bayesian Spatial Conditional Logistic Regression Modelling of Overweight/Obesity Prevalence and Determinants Among Women in Ghana, Accounting for Contraceptive Use},
  journal = {Global Health, Epidemiology and Genomics},
  year = {2026},
  volume = {2026},
  number = {1},
  doi = {10.1155/ghe3/3074676}
}
Kostenlos starten →