Data Systems Hero

Data Systems

Countries with the highest neonatal mortality burden often have the least data. Timely, accurate data are essential to improve neonatal care and track progress towards ending preventable newborn deaths.

Having comparable variables at the individual-level and using standard indicators enables hospitals and countries to compare Small and Sick Newborn Care intervention coverage and quality.

We have combined NEST360 and the African Neonatal Network (ANN) to form the largest neonatal network across Africa.

The standardisation of indicators and variables allows multi-hospital QI and multi-country learning, helping us all to go faster together in improving outcomes for newborns and their families. This is even more crucial in a world where household surveys are no longer funded.

What does the data systems workstream do?

The PANSAA data systems workstream focuses on:

  • Data quality and comparability;
  • Data use, especially for neonatal care coverage and quality; and
  • Quantitative support to Early Career Researchers (ECRs), especially in R coding.

We have mapped core variables across the NEST360 Neonatal Inpatient Dataset (NID) and ANN/Vermont-Oxford (VON) datasets, with the results showing:

  • 34 variables matching exactly;
  • 21 nearly matching (minor recording);
  • 8 differing conceptually; and
  • 55 being unique to one of the datasets (28 exclusively within NEST360, 27 exclusively within ANN).

We aim to use this data to answer priority implementation research questions codesigned with parents, clinicians, biomedical engineers, and laboratory personnel. This will directly contribute to the work being undertaken by the Kangaroo Mother Care (KMC) and Neonatal Infections workstreams, while also linking to our government and UN partners.

Our expected data systems outputs

Our three-year plan for action is structured as follows:

  • Year 1: Focus on variable mapping and planning;
  • Year 2: Enable ECR-led analysis on topics of interest to them; and
  • Year 3: Collective learning and paper writing on our findings while exploring opportunities for add-on grants.

We have also identified an important gap for a prioritised, practical neonatal follow-up dataset to better track child development and health outcomes.