Reporting Practices and Data Quality in Health Information Systems in Developing Countries: An Exploratory Case Study in Kenya

Abstract

Despite increased attention paid to health information systems and their key role for improving health systems in in Low and/or Middle income countries (LMICs), it is believed that data from majority of the health information systems contribute little to the decision-making processes due to poor data quality. We carried out an exploratory assessment of the health information system in Kenya with the main objective of determining the quality of data in terms of accuracy, timeliness and completeness. The study also considered the reasons for the observed data quality status. Data quality audits were carried out in selected health facilities. Data from the source documents at health facilities were compared to the data in the national health information systems for the same period. Key informant interviews were conducted and focus group discussions conducted during quarterly review meetings at regional levels. The study revealed that the completeness rate for the monthly reports was 86.9 percent while the timeliness of the reports was 78.7 percent. In terms of accuracy of the reports, the study showed that while there was a significant amount of low accuracy in many reports evaluated, there was a surprisingly high accuracy of reports coming from the maternity units of all health facilities visited. The accuracy of the number of deliveries could be associated with the financial incentives provided by the government to health facilities as part of the country’s free maternity care project in which mothers deliver free of charge in health facilities. While most health information systems are plagued with poor data quality, a simple and practical incentive can improve accuracy, timeliness and completeness.