Abstract
Abstract
Background
Quantitative assessments of Routine Health Information System (RHIS) data quality are increasingly common in sub-Saharan Africa, but the lived, procedural, and contextual detail behind how routine data are actually collected, compiled, and verified, and why accuracy, completeness, timeliness, and consistency succeed or fail in practice, is rarely captured with the same rigour. This study explored health workers' and supervisors' own accounts of RHIS tools, methods, and data quality practices in Osun State, Nigeria.
Methods
A qualitative descriptive design was used, comprising 35 key informant interviews: 22 with facility-level health workers (officers-in-charge, health records officers, ART coordinators, and monitoring and evaluation staff) across primary, secondary, and tertiary facilities, and 13 with district-level monitoring and evaluation officers, disease surveillance and notification officers, and Local Government Area (LGA) RHIS focal persons. Interviews were analysed thematically, with themes and subthemes triangulated across facility and district respondents.
Results
Four themes described RHIS tools and methods: routine data collection processes, types of health data handled, data collection tools and forms, and compilation/submission procedures, with universal (22/22) practice of month-end summary preparation but only partial (6/22) reliance on locally adapted forms during stock-outs. Across the four data quality dimensions, timeliness showed the strongest and most consistent adherence (20/22 facility respondents; 10/13 district respondents), while accuracy and completeness were most vulnerable to manual transcription error, staff shortages, and inconsistent supervision. District-level respondents functioned as a second, formal verification layer, most notably through DHIS2's built-in validation rules (12/13) and routine rejection of incomplete submissions (10/13), correcting errors and gaps that facility-level processes alone did not fully resolve.
Conclusion
RHIS data quality in Osun State is underpinned by broadly standardised tools and a disciplined monthly reporting cycle, but accuracy and completeness remain structurally exposed to human-resource pressure at the facility level, a vulnerability substantially, though not completely, mitigated by district-level verification. Strengthening facility-level internal validation capacity, rather than relying on district correction after the fact, is the most direct lever available for improving data quality at its source.