Assessing data quality of international health indicators

The aim of the work is to propose a conceptual framework for assessing the data quality of health indicators. The WHO Regional Office for the Americas (AMRO) is monitoring the progress of national and international agreed health goals based on data that countries annually report to the Regional Office. Poor data quality undermines M&E and decision-making and the credibility of the institution that shares the indicator. In Latin America and the Caribbean, there is a recognized need to raise the awareness for better and comparable health statistics that will benefit the US in monitoring progress of global health projects.

For this, we used infant mortality rate (IMR) per 1,000 live births and infant deaths, as reported from countries to AMRO in 2014, infant deaths from the AMRO mortality database, and United Nations IMR estimates. We assessed national IMR of countries with good and poor performing health information systems in the Americas.

The framework proposes five quality dimensions: accuracy, consistency, coverage, timeliness, and interpretability. For each dimension, a simple measurement method was defined based on available data and sources. The five dimensions were individually scored and the overall data quality was presented in a radar graph that visualized the strengths and weaknesses of each dimension.

The findings prove a simple method to show the multidimensionality of the concept of data quality. They demonstrated the utility of the concepts, its operational applicability and its usefulness for measuring data quality. Therefore, this approach could serve countries in Latin America and the Caribbean to evidence the quality of their national health indicators.