Work

Progress but not on track: forecasting under-five mortality and the likelihood of achieving SDG target 3.2 in Southern Africa

Year: 2026

Type: article

Abstract: By 2030, Sustainable Development Goal (SDG) 3.2 aims to reduce the number of under-five deaths per 1,000 live births to less than 25. Since 1990, Southern Africa has significantly contributed to regional reductions; nonetheless, it is unclear if developments are in line with the paths required to meet this goal. This study aimed to assess long-term mortality trends, project future mortality, and estimate the probability that ten Southern African Development Community countries will achieve SDG target 3.2 by 2030. Autoregressive integrated moving average (ARIMA), Prophet, and hybrid ARIMA–Prophet time series modelling techniques were used to examine World Bank annual under-five mortality data (1990–2023). The study further introduces a probabilistic forecasting framework to quantify the likelihood that each country will achieve the SDG 3.2 target by 2030. An 80/20 training-testing split based on root mean square error (RMSE) and mean absolute error (MAE) was used to assess the models’ forecast accuracy. Using forecast distributions, the likelihood of reaching the SDG threshold by 2030 was determined. Under-five mortality declined substantially across all countries, particularly after 2000; however, recent progress has slowed. Forecasts suggest that no country met the predefined probability threshold for classification as l ikely to achieve SDG target 3.2 by 2030 under current mortality trajectories. Malawi had the highest estimated probability (54%) of achieving the target but remained classified as at risk according to the predefined probability thresholds. Across the region, projected 2030 mortality rates remain well above the SDG threshold of 25 deaths per 1,000 live births in most countries. Under current trajectories, Southern African countries are unlikely to achieve SDG target 3.2 by 2030 without enhanced child health interventions. Probabilistic forecasting provides a transparent framework for strengthening child survival monitoring and policy planning.

Source: BMC Public Health

DOI: 10.1186/s12889-026-30212-3

Authors: Welcome J. Dlamini, Thabo Lephoto

Institutions: North-West University, University of KwaZulu-Natal

Language: English

Topic: Global Maternal and Child Health

Subfield: Pediatrics, Perinatology and Child Health

Field: Medicine

Domain: Health Sciences

SDG: Good health and well-being

Open Access status: gold

Estimated APC paid: $3,290