Synthetic verbal autopsy data

This synthetic dataset was generated using the WHO Verbal Autopsy Instrument (2016 release, version 1.5.2). The data creation process involved deploying the verbal autopsy tool into ODK Central, configuring tablets for field simulation, and conducting mock data collection. A small set of records was initially captured and subsequently expanded through hashing techniques to increase the row count for testing purposes.

The dataset is designed primarily for evaluating computer-coded verbal autopsy algorithms such as InterVA and InSilicoVA. In these use cases, the dataset serves as the input, and the algorithms produce outputs indicating the probable causes of death. This enables validation of system functionality and supports research into deriving population-level estimates of disease burden.