Population attributable fractions for modifiable risk factors of neonatal, infant, and under-five mortality in 48 low- and middle-income countries

Kedir Y Ahmed; Subash Thapa; Getiye D Kibret; Habtamu M Bizuayehu; Jing Sun; M Mamun Huda; Abel F Dadi; Felix A Ogbo; Shakeel Mahmood; Muhammad J A Shiddiky; Fentaw T Berhe; Setognal B Aychiluhm; Anayochukwu E Anyasodor; Allen G Ross

doi:10.7189/jogh.15.04015

Population attributable fractions for modifiable risk factors of neonatal, infant, and under-five mortality in 48 low- and middle-income countries

J Glob Health. 2025 Jan 17:15:04015. doi: 10.7189/jogh.15.04015.

Authors

Kedir Y Ahmed¹, Subash Thapa¹, Getiye D Kibret², Habtamu M Bizuayehu³, Jing Sun^{1

4

5}, M Mamun Huda¹, Abel F Dadi^{6

7}, Felix A Ogbo^{8

9}, Shakeel Mahmood¹, Muhammad J A Shiddiky¹, Fentaw T Berhe^{10

11}, Setognal B Aychiluhm¹, Anayochukwu E Anyasodor¹, Allen G Ross¹

Affiliations

¹ Rural Health Research Institute, Charles Sturt University, Orange, New South Wales, Australia.
² Faculty of Medicine, Health and Human Sciences, Macquarie University, Macquarie Park, New South Wales, Australia.
³ First Nations Cancer and Wellbeing (FNCW) Research Program, School of Public Health, The University of Queensland, St Lucia, Queensland, Australia.
⁴ Data Science Institute, University of Technology Sydney, Sydney, New South Wales, Australia.
⁵ School of Health Sciences and Social Work, Griffith University, Gold Coast, Queensland, Australia.
⁶ Menzies School of Health Research, Charles Darwin University, Casuarina, Northern Territory, Australia.
⁷ Addis Continental Institute of Public Health, Addis Ababa, Ethiopia.
⁸ Riverland Academy of Clinical Excellence (RACE), Riverland Mallee Coorong Local Health Network, SA Health Government of South Australia, Berri, South Australia, Australia.
⁹ Translational Health Research Institute, Western Sydney University, Campbelltown, New South Wales, Australia.
¹⁰ Department of Epidemiology and Biostatistics, School of Public Health, College of Medicine and Health Sciences, Wollo University, Dessie, Ethiopia.
¹¹ School of Medicine and Dentistry, Griffith University, Gold Coast, Queensland, Australia.

PMID: 39820022
DOI: 10.7189/jogh.15.04015

Abstract

Background: Identifying the modifiable risk factors for childhood mortality using population-attributable fractions (PAFs) estimates can inform public health planning and resource allocation in low- and middle-income countries (LMICs). We estimated PAFs for key population-level modifiable risk factors of neonatal, infant, and under-five mortality in LMICs.

Methods: We used the most recent Demographic and Health Survey data sets (2010-22) from 48 LMICs, encompassing 35 sub-Saharan African countries and 13 countries from South and Southeast Asia (n = 506 989). We used generalised linear latent mixed models to compute odds ratios (ORs), and we calculated the PAFs adjusted for commonality using ORs and prevalence estimates for key modifiable risk factors.

Results: The highest PAFs of neonatal mortality were attributed to delayed initiation of breastfeeding (>1 hour of birth) (PAF = 23.9; 95% confidence interval (CI) = 23.1, 24.8), uncleaned cooking fuel (PAF = 6.2; 95% CI = 6.4, 7.8), infrequent antenatal care (ANC) visits (PAF = 4.3; 95% CI = 3.3, 5.9), maternal lack of formal education (PAF = 3.9; 95% CI = 2.7, 5.3), and mother's lacking two doses of tetanus injections (PAF = 3.0; 95% CI = 1.9, 3.9). These five modifiable risk factors contributed to 41.4% (95% CI = 35.6, 47.0) of neonatal deaths in the 48 LMICs. Similarly, a combination of these five risk factors contributed to 40.5% of infant deaths. Further, delayed initiation of breastfeeding (PAF = 15.8; 95% CI = 15.2, 16.2), unclean cooking fuel (PAF = 9.6; 95% CI = 8.4, 10.7), mothers lacking formal education (PAF = 7.9; 95% CI = 7.0, 8.9), infrequent ANC visits (PAF = 4.0; 95% CI = 3.3, 4.7), and poor toilet facilities (PAF = 3.4; 95% CI = 2.6, 4.3) were attributed to 40.8% (95% CI = 36.4, 45.2) of under-five deaths.

Conclusions: Given the current global economic climate, policymakers should prioritise these modifiable risk factors. Key recommendations include ensuring that women enter pregnancy in optimal health, prioritising the presence of skilled newborn attendants for timely and proper breastfeeding initiation, and enhancing home-based care during the postnatal period and beyond.

MeSH terms

Adult
Africa South of the Sahara / epidemiology
Breast Feeding / statistics & numerical data
Child Mortality* / trends
Child, Preschool
Developing Countries*
Female
Humans
Infant
Infant Mortality* / trends
Infant, Newborn
Male
Prenatal Care / statistics & numerical data
Risk Factors