Improving data quality through regular flagging of incoherences in PMTCT services data

In Tanzania, a baseline data assessment identified health facilities with incoherent data points, which were primarily attributed to issues with data collection, reporting, and oversight. The data coherence score, used to measure possible data quality issues, revealed a median score of 26 out of 32. This score indicates that nearly 20% of the data points were incoherent, suggesting a need for improvement in data quality. The incoherent data points were identified through the use of Data.FI support, which flagged issues to Council Health Management Teams (CHMTs) for cleaning and improvement. The causes contributing to data quality issues were multifaceted, including a lack of understanding of data tools, incomplete reporting, insufficient oversight, and inadequate data collection tools. To address these issues, several strategies were implemented, including the creation of data validation rules in iMES to detect errors, refresher training for healthcare providers, and sustained coaching by CHMTs. The implementation of these strategies resulted in a significant reduction in the number of health facilities with incoherent data points, from a median of 26 to 5. However, an uptick in incoherent data points was observed in October and November 2022, which was attributed to ongoing cleaning, incomplete coverage of change idea interventions, staff turnover, and incomplete cascading of knowledge. To sustain these improvements, the change idea implementation will be continuous through engagement in situation room meetings and supportive supervision. This will ensure that data quality issues are addressed in the long run and that the improvements are sustained. The Data for Implementation (Data.FI) project, a five-year cooperative agreement funded by the U.S. President's Emergency Plan for AIDS Relief, has been instrumental in supporting these efforts. The project is implemented by Palladium, in partnership with JSI Research & Training Institute, Johns Hopkins University, Right to Care, Cooper/Smith, DT Global, Jembi Health Systems, and Macro-Eyes, with support from expert local resource partners. The project's goal is to improve data quality and inform decision-making in the fight against HIV/AIDS in Tanzania.