ANACONDA Mortality Data Quality Assessment Tool
The systems that produce mortality and cause of death data in many countries are often poorly developed and fragmented, resulting in poor quality data that are generally not fit-for-purpose, and as a result are grossly under-used or not used at all.
A common concern with any mortality dataset, including data produced from civil registration systems, is their reliability in describing actual mortality patterns in the population to which they refer.
As a first step to improving the policy utility of vital statistics systems in countries, it is very important to have a detailed understanding of problems in the data, particularly with regard to completeness and accuracy. Higher quality data reduces uncertainty about the leading causes of death in a population and how they are changing, thus better meeting policy needs.
Analysis of Causes of (National) Death for Action tool (ANACONDA) meets this need. ANACONDA is an electronic tool that assesses the accuracy and completeness of mortality and cause of death data by checking for potential errors and inconsistencies.
Using ANACONDA helps build analytic capacity in the core epidemiological and demographic concepts that underlie mortality data based on decades of recorded observations in a wide range of countries.
The tool only requires the age and sex structure of the source population (denominator) and the ICD-10 codes by sex and standard age groups (numerator). It then applies over 23 tests of the data, conducting all the calculations needed for a comprehensive data quality review and automatically generates the associated figures and tables from which a report can be written. It reviews the rates of mortality as well as the causes of mortality. The structure of the tool is logical and all the computational steps are automated. It provides a summary composite quality score for each data set analysed.
The tool is particularly useful for those who are responsible for the production of routine mortality data, as it allows them to regularly monitor the quality of their datasets over time and between sub-national levels.