sae4health: An R Shiny Application for Small Area Estimation in Low- and Middle-Income Countries

2025-05-02

AI Summary

In low- and middle-income countries (LMICs), subnational health indicator estimation is hindered by sparse data and high methodological barriers. To address this, we developed the first R Shiny platform tailored for LMICs, integrating areal and unit-level spatial small-area models implemented via integrated nested Laplace approximation (INLA) for efficient Bayesian inference. The platform incorporates over 150 Demographic and Health Surveys (DHS) health indicators, supports multi-level administrative unit analysis, and features interactive visualization and automated report generation. Its key innovation lies in unifying both spatial small-area modeling paradigms within a lightweight, web-based interface—significantly lowering technical entry barriers. Applied in Nigeria, the platform produced high-resolution maps of childhood stunting, improving subnational estimation accuracy and policy interpretability. It is now actively deployed to support public health decision-making.

Technology Category

• Application Domains: Humanities & Computational Social Science
• Intelligent Robots: State Estimation
• Reasoning under Uncertainty: Applications

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Web data visualization

Abstract

Accurate subnational estimation of health indicators is critical for public health planning, especially in low- and middle-income countries (LMICs), where data and tools are often limited. The sae4health R shiny app, built on the surveyPrev package, provides a user-friendly tool for prevalence mapping using small area estimation (SAE) methods. Both area- and unit-level models with spatial random effects are available, with fast Bayesian inference performed using Integrated Nested Laplace Approximation (INLA). Currently, the app supports analysis of over 150 indicators from Demographic and Health Surveys (DHS) across multiple administrative levels. sae4health simplifies the use of complex prevalence mapping models to support data-driven decision-making. The app provides interactive visualization, summary, and report generation functionalities for a wide range of use cases. This paper outlines the app's statistical framework and demonstrates the workflow through a case study of child stunting in Nigeria. Additional documentation is available on the supporting website (https://sae4health.stat.uw.edu).

Problem

• Provides user-friendly SAE tool for health indicators in LMICs
• Enables prevalence mapping with spatial Bayesian models via INLA
• Supports analysis of 150+ DHS indicators across administrative levels

Innovation

• R Shiny app for small area estimation
• Uses Bayesian inference with INLA
• Supports 150+ DHS indicators