Assessing the quality of data for selected reproductive health indicators in designated public health facilities in Bangladesh.

Abstract

Background: An effective health management information system plays a pivotal role in evidence-based decision-making and strengthening health service delivery in a country. The Directorate General of Health Services and the Directorate General of Family Planning of Bangladesh have adopted digital health management information system platforms named district health information system and management information system, respectively. Despite its significance, health management information system data has numerous issues, such as missing values, inaccuracies, lack of internal consistency, and the presence of outliers. This study aims to assess the data quality of reproductive health indicators in the health management information system of the Directorate General of Health Services and the Directorate General of Family Planning.

Methods: The study examined two aspects of data quality: a) completeness of data, subdivided into completeness of facility reporting (report submission rate) and completeness of indicator data (presence of missing values); b) internal consistency of reported data, subdivided into presence of outliers, inter-indicator consistency, and consistency between reported data and original records (accuracy rate). The study utilised retrospective monthly data gathered from July 2021 to June 2022, covering 21 reproductive health indicators. Multi-stage cluster sampling was employed to select 112 health facilities for data collection, including 48 facilities from Directorate General of Health Services and 64 from Directorate General of Family Planning, representing various administrative levels across the country.

Result: The report submission rate for Directorate General of Health Services facilities was 98%, while for the Directorate General of Family Planning facilities, it was 86%. However, 35% of data points were missing in the district health information system server of Directorate General of Health Services, whereas no missing values were observed in the management information system server of Directorate General of Family Planning. Less than 3% of outliers were detected in the server data of both directorates. Inter-indicator consistency was maintained at a high rate of 98% in health facilities under both directorates. The accuracy of reported data varied across indicators and facility types: Directorate General of Health Services facilities showed accuracy rates ranging between 75 and 92%, with an aggregated rate of 86%. Different tiers of the Directorate General of Family Planning facilities had accuracy rates ranging from 92 to 96%.

Conclusion: This research emphasises the significance of rectifying missing values, ensuring consistency, and improving reporting systems, with a particular focus on lower-tier health facilities, to enhance the validity and reliability of reproductive health data in Bangladesh.