Verbal autopsy methods with multiple causes of death

Gary King, Ying Lu

Research output: Contribution to journalArticle

Abstract

Verbal autopsy procedures are widely used for estimating cause-specific mortality in areas without medical death certification. Data on symptoms reported by caregivers along with the cause of death are collected from a medical facility, and the cause-of-death distribution is estimated in the population where only symptom data are available. Current approaches analyze only one cause at a time, involve assumptions judged difficult or impossible to satisfy, and require expensive, time-consuming, or unreliable physician reviews, expert algorithms, or parametric statistical models. By generalizing current approaches to analyze multiple causes, we show how most of the difficult assumptions underlying existing methods can be dropped. These generalizations also make physician review, expert algorithms and parametric statistical assumptions unnecessary. With theoretical results, and empirical analyses in data from China and Tanzania, we illustrate the accuracy of this approach. While no method of analyzing verbal autopsy data, including the more computationally intensive approach offered here, can give accurate estimates in all circumstances, the procedure offered is conceptually simpler, less expensive, more general, as or more replicable, and easier to use in practice than existing approaches. We also show how our focus on estimating aggregate proportions, which are the quantities of primary interest in verbal autopsy studies, may also greatly reduce the assumptions necessary for, and thus improve the performance of, many individual classifiers in this and other areas. As a companion to this paper, we also offer easy-to-use software that implements the methods discussed herein.

Original languageEnglish (US)
Pages (from-to)78-91
Number of pages14
JournalStatistical Science
Volume23
Issue number1
DOIs
StatePublished - 2008

Fingerprint

Certification
Parametric Model
Mortality
Statistical Model
China
Proportion
Classifier
Software
Necessary
Cause of death
Estimate
Review
Physicians
Generalization
Caregivers
Tanzania
Statistical model

Keywords

  • Cause of death
  • Cause-specific mortality
  • Classification
  • Sensitivity
  • Specificity
  • Survey research
  • Verbal autopsy

ASJC Scopus subject areas

  • Mathematics(all)
  • Statistics and Probability

Cite this

Verbal autopsy methods with multiple causes of death. / King, Gary; Lu, Ying.

In: Statistical Science, Vol. 23, No. 1, 2008, p. 78-91.

Research output: Contribution to journalArticle

King, Gary ; Lu, Ying. / Verbal autopsy methods with multiple causes of death. In: Statistical Science. 2008 ; Vol. 23, No. 1. pp. 78-91.
@article{e88cdbde24ed4c11b71116acdf71a845,
title = "Verbal autopsy methods with multiple causes of death",
abstract = "Verbal autopsy procedures are widely used for estimating cause-specific mortality in areas without medical death certification. Data on symptoms reported by caregivers along with the cause of death are collected from a medical facility, and the cause-of-death distribution is estimated in the population where only symptom data are available. Current approaches analyze only one cause at a time, involve assumptions judged difficult or impossible to satisfy, and require expensive, time-consuming, or unreliable physician reviews, expert algorithms, or parametric statistical models. By generalizing current approaches to analyze multiple causes, we show how most of the difficult assumptions underlying existing methods can be dropped. These generalizations also make physician review, expert algorithms and parametric statistical assumptions unnecessary. With theoretical results, and empirical analyses in data from China and Tanzania, we illustrate the accuracy of this approach. While no method of analyzing verbal autopsy data, including the more computationally intensive approach offered here, can give accurate estimates in all circumstances, the procedure offered is conceptually simpler, less expensive, more general, as or more replicable, and easier to use in practice than existing approaches. We also show how our focus on estimating aggregate proportions, which are the quantities of primary interest in verbal autopsy studies, may also greatly reduce the assumptions necessary for, and thus improve the performance of, many individual classifiers in this and other areas. As a companion to this paper, we also offer easy-to-use software that implements the methods discussed herein.",
keywords = "Cause of death, Cause-specific mortality, Classification, Sensitivity, Specificity, Survey research, Verbal autopsy",
author = "Gary King and Ying Lu",
year = "2008",
doi = "10.1214/07-STS247",
language = "English (US)",
volume = "23",
pages = "78--91",
journal = "Statistical Science",
issn = "0883-4237",
publisher = "Institute of Mathematical Statistics",
number = "1",

}

TY - JOUR

T1 - Verbal autopsy methods with multiple causes of death

AU - King, Gary

AU - Lu, Ying

PY - 2008

Y1 - 2008

N2 - Verbal autopsy procedures are widely used for estimating cause-specific mortality in areas without medical death certification. Data on symptoms reported by caregivers along with the cause of death are collected from a medical facility, and the cause-of-death distribution is estimated in the population where only symptom data are available. Current approaches analyze only one cause at a time, involve assumptions judged difficult or impossible to satisfy, and require expensive, time-consuming, or unreliable physician reviews, expert algorithms, or parametric statistical models. By generalizing current approaches to analyze multiple causes, we show how most of the difficult assumptions underlying existing methods can be dropped. These generalizations also make physician review, expert algorithms and parametric statistical assumptions unnecessary. With theoretical results, and empirical analyses in data from China and Tanzania, we illustrate the accuracy of this approach. While no method of analyzing verbal autopsy data, including the more computationally intensive approach offered here, can give accurate estimates in all circumstances, the procedure offered is conceptually simpler, less expensive, more general, as or more replicable, and easier to use in practice than existing approaches. We also show how our focus on estimating aggregate proportions, which are the quantities of primary interest in verbal autopsy studies, may also greatly reduce the assumptions necessary for, and thus improve the performance of, many individual classifiers in this and other areas. As a companion to this paper, we also offer easy-to-use software that implements the methods discussed herein.

AB - Verbal autopsy procedures are widely used for estimating cause-specific mortality in areas without medical death certification. Data on symptoms reported by caregivers along with the cause of death are collected from a medical facility, and the cause-of-death distribution is estimated in the population where only symptom data are available. Current approaches analyze only one cause at a time, involve assumptions judged difficult or impossible to satisfy, and require expensive, time-consuming, or unreliable physician reviews, expert algorithms, or parametric statistical models. By generalizing current approaches to analyze multiple causes, we show how most of the difficult assumptions underlying existing methods can be dropped. These generalizations also make physician review, expert algorithms and parametric statistical assumptions unnecessary. With theoretical results, and empirical analyses in data from China and Tanzania, we illustrate the accuracy of this approach. While no method of analyzing verbal autopsy data, including the more computationally intensive approach offered here, can give accurate estimates in all circumstances, the procedure offered is conceptually simpler, less expensive, more general, as or more replicable, and easier to use in practice than existing approaches. We also show how our focus on estimating aggregate proportions, which are the quantities of primary interest in verbal autopsy studies, may also greatly reduce the assumptions necessary for, and thus improve the performance of, many individual classifiers in this and other areas. As a companion to this paper, we also offer easy-to-use software that implements the methods discussed herein.

KW - Cause of death

KW - Cause-specific mortality

KW - Classification

KW - Sensitivity

KW - Specificity

KW - Survey research

KW - Verbal autopsy

UR - http://www.scopus.com/inward/record.url?scp=49749130876&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=49749130876&partnerID=8YFLogxK

U2 - 10.1214/07-STS247

DO - 10.1214/07-STS247

M3 - Article

AN - SCOPUS:49749130876

VL - 23

SP - 78

EP - 91

JO - Statistical Science

JF - Statistical Science

SN - 0883-4237

IS - 1

ER -