Evaluating an intervention to improve the quality and use of sub-national routine immunization data for improved immunization decision-making in Nigeria: a qualitative implementation research

Abstract Background: Studies evaluating data quality are needed to better identify what interventions might improve the quality of routine immunization (RI) data and use in decision-making. This study evaluated the outcomes of interventions that were implemented to improve the quality and use of RI data for decision-making in Enugu state, Nigeria. Methods: The study was conducted in six local government areas (LGAs) of Enugu State, Nigeria using a cross-sectional, qualitative design. Data was collected through semi-structured in-depth phenomenological interviews. Twenty-one interviews were conducted with purposively selected RI stakeholders including: State-level programme managers (n=3), primary healthcare coordinators (n=6), LGA RI officers (n=6) and facility RI focal persons (n=6). Acceptability, adoption, and feasibility dimensions of the Proctor et al framework was used to assess the interventions, which were 1). data review meetings; and 2). data-focused supervision. The self-selected changes in data management practices including ‘tally as you vaccinate’, ‘adherence to reporting timeline’, ‘use of demographic information to set target’, ‘use of local alternatives to archive paper records’, ‘updating monitoring chart’, and ‘data use in health facilities’ were also evaluated. Data were coded and analysed thematically using NVivo software version 11. Findings: Most participants accepted and adopted the interventions and self-selected data management practices. However, their feasibility depended on context-specific issues. Unpredictable funding undermined the feasibility of implementing the interventions. Resistance among health workers constrained ‘tally as you vaccinate’ due to staff shortage, long waiting time and staff unwillingness to use Open Data Kit mobile data application. Adherence to reporting timeline was constrained by irregular funding, inadequate staffing, and lack of computers. Use of inappropriate denominators constrained target setting by health facilities. Health workers used baskets in place of cupboards and shelves to archive paper-based records. Delayed supplies limited updating monitoring of charts, while mentoring enhanced it. Self-reported use of data for decision-making was facilitated by demand by health facility committees but constrained by inadequate staffing and lack of funds. Conclusion: The intervention improved acceptability and adoption of self-selected changes in immunisation data management practices . Nevertheless, the feasibility of sustaining the practices required decision-makers to address contextual limitations identified in this study.