Home People Bibliography Verbal Autopsy
Software Info/Help Refresh
Software
Last updated 2025-07-15

We have streamlined the organization of our software so that it is available through only two channels: the comprehensive R archive network (CRAN) for R packages, and GitHub for everything, including the source code for the R packages.

At this time we do not have the resources to actively maintain all of the software. We are actively maintaining research-oriented software that supports our publications, but we are not able to support user-facing packages, modules, or applications. Below, links are organized into sections marked 'Actively Maintained' and 'Archived' - NOT maintained.

Open Source R Packages on CRAN

Actively Maintained

  • openVA: The openVA package implements multiple existing open-source algorithms for coding cause of death from verbal autopsies. It also provides tools for data manipulation tasks commonly used in Verbal Autopsy analysis and implements easy graphical visualization of individual and population level statistics.
  • InSilicoVA (2012, 2016): An R package for using the InSilicoVA algorithm for coding cause of death from verbal autopsies collected using the 2012 and 2016 WHO VA instruments. It also provides simple graphical representation of individual and population level statistics.
  • InterVA5 (2016): This is an R package replicating InterVA-5 software (http://www.byass.uk/interva) for coding cause of death from verbal autopsies collected using the 2016 WHO VA instrument. It also provides simple graphical representation of individual and population level statistics.
  • InterVA4 (2012): An R package replicating InterVA-4 software for coding cause of death from verbal autopsies collected using the 2012 WHO VA instrument. It also provides simple graphical representation of individual and population level statistics.
  • Tariff: An R implement the Tariff algorithm for coding cause-of-death from verbal autopsies. The Tariff method was originally proposed in James et al (2011) and later refined as Tariff 2.0 in Serina, et al. (2015). Note that this package was not developed by authors affiliated with the Institute for Health Metrics and Evaluation and thus unintentional discrepancies may exist between the this implementation and the implementation available from IHME.

Archived

  • CrossVA: This is an R package that transforms data collected using the 2016 WHO VA instruments into the format expected by the InSilicoVA and InterVA5 algorithms - 2016 and 2022.

Software and Source Code Available on GitHub

Maintained

Archived

  • CrossVA: see above.
  • vacheck: A Python implementation of the data consistency checks performed by the InterVA5 algorithm.
  • pyInterVA: A Python package replicating InterVA-5 software (http://www.byass.uk/interva) for coding cause of death from verbal autopsies collected using the 2016 WHO VA instrument. It also provides simple graphical representation of individual and population level statistics.
  • pyInSilicoVA: The Python implementation of the InSilicoVA R package, but with fewer features.
  • pyOpenVA (openVA with graphical user interface): A graphical user interface for coding cause of death from verbal autopsy data collected using the 2016 WHO VA instrument with the InSilicoVA and InterVA5 algorithms.
  • openVA Pipeline: A Python package implementing the workflow of (1) downloading verbal autopsy records from an ODK Central server, (2) assigning causes of death using either the InSilicoVA or InterVA5 algorithms, (3) storing results in a local database, and (4) posting the results to a DHIS2 server with the VA Program (optional).
  • DHIS2 VA Program: A metadata repository for DHIS2 to enable the storage of VA results, as produced by the openVA Pipeline.

Experimental Prerelease Software

Warning!

Experimental prerelease software runs but has not been tested, verified, or validated. We do not suggest or guarantee that results from experimental prerelease software will be valid, accurate, or useful in any way whatsoever. Use at your own risk.

The algorithms that work with the 2022 version of the WHO instrument - InSilicoVA2022 and InterVA2022 - both use a recently updated version of 'probbase'. Probbase is a set of values that relate the VA indicators to VA causes. The new, updated probbase has not been tested or validated, and because of that, it is impossible to know how well the new algorithms work. A validation study will be conducted, but it will take some time, and before that is complete, these algorithms will remain experimental.

Prerelease Software

  • InSilicoVA for WHO 2022 instrument: An R package for using the InSilicoVA algorithm for coding cause of death from verbal autopsies collected using the 2022 WHO VA instrument. It also provides simple graphical representation of individual and population level statistics.
  • InterVA for WHO 2022 instrument: This is an R package implementing the InterVA algorithm for coding cause of death from verbal autopsies collected using the 2022 WHO VA instrument. It also provides simple graphical representation of individual and population level statistics.
  • Probbase for 2022 VA instrument: An symptom-cause-information (SCI) matrix that contains the conditional probabilities of observing a symptom/indicator for causes of death included in the WHO VA cause list. This SCI is using by the InSilicoVA 2022 and InterVA 2022 algorithms.

Reporting Issues

You will encounter challenges when using the prerelease software. Although we cannot support these tools right now, it is useful to accumulate descriptions of those issues. To provide feedback, please use GitHub's issue tracking feature. When you follow the links above, you will land on the corresponding GitHub repository. In the upper left corner of the window for each repository, there is a button labeled 'Issues', example below. Push that button and follow the instructions to submit your issue: 'New issue' button, fill out form, and then 'Create' button.


Issues button