Open Access Research

Random forests for verbal autopsy analysis: multisite validation study using clinical diagnostic gold standards

Abraham D Flaxman1*, Alireza Vahdatpour1, Sean Green2, Spencer L James1, Christopher JL Murray1 and the Population Health Metrics Research Consortium (PHMRC)

  • * Corresponding author: Abraham D Flaxman abie@uw.edu

1 Institute for Health Metrics and Evaluation, University of Washington, 2301 Fifth Ave., Suite 600, Seattle, WA 98121, USA

2 Bill & Melinda Gates Foundation, Seattle, USA

For all author emails, please log on.

Population Health Metrics 2011, 9:29 doi:10.1186/1478-7954-9-29

Published: 4 August 2011

Additional files

Additional file 1:

Top 40 items based on Tariff Method for adult causes (Items that are extracted from open text field are marked with *, other HCE variables are marked with +).

Format: XLSX Size: 56KB Download file

Open Data

Additional file 2:

Top 40 items based on Tariff Method for child causes (Items that are extracted from open text field are marked with *, other HCE variables are marked with +).

Format: XLSX Size: 31KB Download file

Open Data

Additional file 3:

Top 40 items based on Tariff Method for neonate causes (Items that are extracted from open text field are marked with *, other HCE variables are marked with +).

Format: XLSX Size: 22KB Download file

Open Data

Additional file 4:

Sensitivity analysis for 54 variants of the RF algorithm applied to 200 splits of the child VA data with HCE.

Format: XLS Size: 44KB Download file

This file can be viewed with: Microsoft Excel Viewer

Open Data

Additional file 5:

Median chance-corrected concordance (%) across 500 splits, by age group and cause with and without HCE.

Format: XLSX Size: 14KB Download file

Open Data

Additional file 6:

Slope, intercept, and RMSE for linear regression of true versus estimated CSMF.

Format: XLSX Size: 18KB Download file

Open Data