Stillbirth and newborn data quality and use and related input and process factors: findings of the IMPULSE study in Uganda.

BACKGROUND

The reduction of preventable newborn deaths in low- and middle-income countries is a global priority. The availability of high-quality newborn and stillbirth data is essential for shaping action-oriented policies and interventions towards resolving these challenges. Using a mixed-methods approach, we evaluated the input and process factors hindering data quality and use in Uganda.

METHODS

We conducted a cross-sectional study from November 2022 to September 2023 across three regions and one city administration in Uganda (51 sites, 30 facilities, 20 district health offices (DHOs), and a Ministry of Health). We collected data primarily through direct observation based on standardised Every Newborn - Measurement Improvement for Newborn & Stillbirth Indicators tools and analysed them using the Performance of Routine Information System Management framework. We synthesised data quality, data use, and their technical, organisational, and behavioural determinants using sub-domain level indicators designed to provide a novel approach for policymakers.

RESULTS

Newborn data availability and completeness were high, with denominator elements exceeding 90% at all levels and numerator elements at facility level ranging from 86% to 100%. In contrast, data accuracy was consistently low (range = 26-61%), and data use for performance review remained limited, particularly at facility level (21-69%) compared to DHOs (33-81%). Among underlying factors, key strengths included that most sites had staff to compile and analyse data (93-100%), used data visualisations (93-95%), and showed strong technical and behavioural performance at the DHO level. Conversely, promotion of evidence-based decision-making was low (58-64%), and critical resource and capacity gaps persisted, including limited availability of minimum item bundles (22-24%), functional internet (60-83%), staff development plans (55%), user skills at facility level (20-61%), and data analysis and feedback mechanisms (50-74%). Overall, 350 (74%) end users reported a need for improvement, with no significant differences across site levels.

CONCLUSIONS

These findings provide actionable guidance for policymakers to improve the quality and use of newborn and stillbirth data. Emerging priority actions include strengthening the data verification processes, building analytical capacity at facility level, and institutionalising regular data review with focus on decision-making.