This community innovation has been accepted at the 2026 DHIS2 Annual Conference and will be in abstract track/lightning talk.
Data quality check and analytics for action
Routine health facility data in DHIS2 are critical for monitoring service delivery but are often underused due to the time and technical expertise required for extraction, cleaning, analysis, and visualization. These challenges limit timely identification of data quality issues and reduce country ownership of routine data use. To address this gap, a new application was developed to automate monthly DHIS2 facility data processing and generation of standardized analytical outputs. The application automates four core functions. First, it extracts monthly health facility level data directly from DHIS2 and standardizes indicator definitions, geographic hierarchies, and time periods. Second, it applies a structured, multi-step data quality assessment methodology covering outlier detection, missing data, and internal consistency. Outliers are identified using five complementary statistical methods, while missing data and implausible dose relationships are systematically flagged for review and correction. These steps are designed to support country-led validation and decision-making rather than fully automated replacement. Third, the application produces ready-to-use PowerPoint presentations at national, regional, and district levels. These outputs clearly visualize data quality issues at specific facilities and months, enabling concrete problem identification and targeted follow-up. The current focus is immunization data, but the application is adaptable to other health indicators captured in DHIS2. Early implementation across multiple countries demonstrates its potential to improve the efficiency, consistency, and transparency of routine data quality improvement processes. By reducing technical barriers and strengthening local ownership, the application enhances the practical use of DHIS2 data for monitoring, planning, and improving health service delivery.
Primary Author: Yoshito Kawakatsu
Keywords:
DHIS2, data quality, automation, immunization, routine health data, facility-level analysis, subnational analytics, outlier detection, data completeness, internal consistency, visualization
1 Like