Data Harmonisation Methodology: Unified Multi-Country Perinatal Dataset for Predictive Modelling in Sub-Saharan Africa (Version 6.1)

Read full text
Authors: Joseph Akuze, Phillip Wanduru, Hannah Blencowe, Eric O. Ohuma, Tanida Chaweewong, Audencio Victor, Miriam Bourke
Cited by: 0

Abstract

This component describes the data harmonisation methodology used to construct a unified individual-level perinatal dataset of 5,996,390 birth records from 66 countries across seven contributing studies: ALERT, EN-INDEPTH, PTBi, PRECISE, WHOMCS, NCOPS, and the Demographic and Health Surveys (DHS) programme. The document covers the five-stage harmonisation pipeline: ethical data acquisition and governance, variable mapping across 13 harmonised domains, value standardisation and recoding, linkage of NASA environmental and climate data, and quality assurance procedures. Completeness profiles are reported by study and design stratum. The master data dictionary (Version 10.17) is referenced throughout. This documentation supports transparency and reproducibility for researchers seeking to understand, evaluate, or replicate the harmonisation approach. Version 6.1, January 2026. Part of the Wellcome Accelerator Award project at LSHTM.

Fields of Study & Subjects

Maternal and Neonatal Healthcare Global Maternal and Child Health Research Data Management Practices

Publication details

Publication date
1 January 2026
Language
EN
Document type
Preprint