# Quantifying the Contributions of Training Data and Algorithm Logic to the Performance of Automated Cause-assignment Algorithms for Verbal Autopsy

## TL;DR

This paper investigates how the quality of symptom-cause information and the algorithm logic influence the accuracy of automated cause-of-death assignment in verbal autopsy surveys, highlighting the importance of data quality.

## Contribution

It provides a comparative analysis of different VA algorithms, emphasizing the impact of symptom-cause information quality on performance over algorithmic differences.

## Key findings

- Symptom-cause information quality significantly affects algorithm performance.
- Algorithm logic has less impact than data quality on accuracy.
- Performance varies with different input combinations and evaluation metrics.

## Abstract

A verbal autopsy (VA) consists of a survey with a relative or close contact of a person who has recently died. VA surveys are commonly used to infer likely causes of death for individuals when deaths happen outside of hospitals or healthcare facilities. Several statistical and algorithmic methods are available to assign cause of death using VA surveys. Each of these methods require as inputs some information about the joint distribution of symptoms and causes. In this note, we examine the generalizability of this symptom-cause information by comparing different automated coding methods using various combinations of inputs and evaluation data. VA algorithm performance is affected by both the specific SCI themselves and the logic of a given algorithm. Using a variety of performance metrics for all existing VA algorithms, we demonstrate that in general the adequacy of the information about the joint distribution between symptoms and cause affects performance at least as much or more than algorithm logic.