JMIR Publications

ABSTRACT

Background:

Zero-dose children - those who receive no routine vaccines - remain highly vulnerable, with global evidence showing they are concentrated in marginalized and hard-to-reach settings. In Bangladesh, stagnant full-immunization coverage, population mobility, and inconsistencies in administrative data complicate accurate estimation of zero-dose and under-immunized children. Gaps in the completeness, timeliness, and accuracy of routine EPI information systems further limit the ability to identify and target these populations.

Objective:

This study aimed to assess the performance of Bangladesh’s immunization data systems and identify strategies for integrating zero-dose tracking into routine EPI information systems.

Methods:

We used a two-step qualitative approach comprising a desk review and targeted stakeholder consultations (February–April 2024). The desk review assessed program documents, grey literature, and published studies to map routine immunization data sources and their relevance for zero-dose (ZD) analysis. Thirteen purposively selected stakeholders, including EPI and DHIS2 managers, were interviewed using semi-structured tools to gather insights on data availability, reporting practices, and system challenges. Information from both steps was synthesized using a structured checklist to evaluate completeness, timeliness, accuracy, and feasibility of ZD tracking across data sources. Data was analyzed using a framework approach developed by Lincoln and Guba. Ethical approval was obtained from the icddr,b Institutional Review Board, and written informed consent was collected from all participants.

Results:

Ten routine immunization data sources were identified, with DHIS2, the RCM tool, and the EPI Coverage Evaluation Survey (CES) emerging as the primary systems for zero-dose (ZD) tracking. While most sources captured child-level data, the use of unique identifiers and household-level variables was inconsistent. Data quality and timeliness varied: DHIS2 faced denominator inaccuracies, CES and BDHS provided high-quality survey estimates, and RCM delivered real-time data from hard-to-reach areas. Multiple systems supported ZD identification, though each had limitations related to data completeness, sampling, geographic granularity, or operational constraints.

Conclusions:

Immunization data systems in Bangladesh are fragmented, with limited zero-dose (ZD) tracking and mainly district-level reporting. DHIS2, RCM, and HDSS are the most promising platforms for routine ZD monitoring, offering nationwide integration, real-time field data, and detailed local analysis. Harmonizing these systems and combining multi-source data can improve targeted interventions, subnational monitoring, and equitable vaccine coverage.