Evaluation of neonatal verbal autopsy using physician review versus algorithm-based cause-of-death assignment in rural Nepal

James V. Freeman, Parul S Christian, Subarna K. Khatry, Ramesh K. Adhikari, Steven C. LeClerq, Joanne Katz, Gary L. Darmstadt

Research output: Contribution to journalArticle

Abstract

Verbal autopsy (VA) is used to ascertain cause-specific neonatal mortality using parental/familial recall. We sought to compare agreement between causes of death obtained from the VA by physician review vs. computer-based algorithms. Data were drawn from a cluster-randomised trial involving 4130 live-born infants and 167 neonatal deaths in the rural Sarlahi District of Nepal. We examined the agreement between causes ascertained by physician review and algorithm assignment by the kappa (K) statistic. We also compared responses to identical questions posed posthumously during neonatal VA interviews with those obtained during maternal interviews and clinical examinations regarding condition of newborns soon after birth. Physician reviewers assigned prematurity or acute lower respiratory infection (ALRI) as causes of 48% of neonatal deaths; 41% were assigned as uncertain. The algorithm approach assigned sepsis (52%), ALRI (31%), birth asphyxia (29%), and prematurity (24%) as the most common causes of neonatal death. Physician review and algorithm assignment of causes of death showed high κ for prematurity (0.73), diarrhoea (0.81) and ALRI (0.68), but was low for congenital malformation (0.44), birth asphyxia (0.17) and sepsis (0.00). Sensitivity and specificity of VA interview questions varied by symptom, with positive predictive values ranging from 50% to 100%, when compared with maternal interviews and examinations of neonates soon after birth. Analysis of the VA data by physician review and computer-based algorithms yielded disparate results for some causes but not for others. We recommend an analysis technique that combines both methods, and further validation studies to improve performance of the VA for assigning causes of neonatal death.

Original languageEnglish (US)
Pages (from-to)323-331
Number of pages9
JournalPaediatric and Perinatal Epidemiology
Volume19
Issue number4
DOIs
StatePublished - Jul 2005

Fingerprint

Nepal
Cause of Death
Autopsy
Physicians
Parturition
Respiratory Tract Infections
Interviews
Asphyxia
Sepsis
Mothers
Newborn Infant
Validation Studies
Infant Mortality
Diarrhea
Sensitivity and Specificity
Perinatal Death

ASJC Scopus subject areas

  • Epidemiology
  • Pediatrics, Perinatology, and Child Health

Cite this

Evaluation of neonatal verbal autopsy using physician review versus algorithm-based cause-of-death assignment in rural Nepal. / Freeman, James V.; Christian, Parul S; Khatry, Subarna K.; Adhikari, Ramesh K.; LeClerq, Steven C.; Katz, Joanne; Darmstadt, Gary L.

In: Paediatric and Perinatal Epidemiology, Vol. 19, No. 4, 07.2005, p. 323-331.

Research output: Contribution to journalArticle

Freeman, James V. ; Christian, Parul S ; Khatry, Subarna K. ; Adhikari, Ramesh K. ; LeClerq, Steven C. ; Katz, Joanne ; Darmstadt, Gary L. / Evaluation of neonatal verbal autopsy using physician review versus algorithm-based cause-of-death assignment in rural Nepal. In: Paediatric and Perinatal Epidemiology. 2005 ; Vol. 19, No. 4. pp. 323-331.
@article{473af3b0e4ed43ad84d281aabd50dd61,
title = "Evaluation of neonatal verbal autopsy using physician review versus algorithm-based cause-of-death assignment in rural Nepal",
abstract = "Verbal autopsy (VA) is used to ascertain cause-specific neonatal mortality using parental/familial recall. We sought to compare agreement between causes of death obtained from the VA by physician review vs. computer-based algorithms. Data were drawn from a cluster-randomised trial involving 4130 live-born infants and 167 neonatal deaths in the rural Sarlahi District of Nepal. We examined the agreement between causes ascertained by physician review and algorithm assignment by the kappa (K) statistic. We also compared responses to identical questions posed posthumously during neonatal VA interviews with those obtained during maternal interviews and clinical examinations regarding condition of newborns soon after birth. Physician reviewers assigned prematurity or acute lower respiratory infection (ALRI) as causes of 48{\%} of neonatal deaths; 41{\%} were assigned as uncertain. The algorithm approach assigned sepsis (52{\%}), ALRI (31{\%}), birth asphyxia (29{\%}), and prematurity (24{\%}) as the most common causes of neonatal death. Physician review and algorithm assignment of causes of death showed high κ for prematurity (0.73), diarrhoea (0.81) and ALRI (0.68), but was low for congenital malformation (0.44), birth asphyxia (0.17) and sepsis (0.00). Sensitivity and specificity of VA interview questions varied by symptom, with positive predictive values ranging from 50{\%} to 100{\%}, when compared with maternal interviews and examinations of neonates soon after birth. Analysis of the VA data by physician review and computer-based algorithms yielded disparate results for some causes but not for others. We recommend an analysis technique that combines both methods, and further validation studies to improve performance of the VA for assigning causes of neonatal death.",
author = "Freeman, {James V.} and Christian, {Parul S} and Khatry, {Subarna K.} and Adhikari, {Ramesh K.} and LeClerq, {Steven C.} and Joanne Katz and Darmstadt, {Gary L.}",
year = "2005",
month = "7",
doi = "10.1111/j.1365-3016.2005.00652.x",
language = "English (US)",
volume = "19",
pages = "323--331",
journal = "Paediatric and Perinatal Epidemiology",
issn = "0269-5022",
publisher = "Wiley-Blackwell",
number = "4",

}

TY - JOUR

T1 - Evaluation of neonatal verbal autopsy using physician review versus algorithm-based cause-of-death assignment in rural Nepal

AU - Freeman, James V.

AU - Christian, Parul S

AU - Khatry, Subarna K.

AU - Adhikari, Ramesh K.

AU - LeClerq, Steven C.

AU - Katz, Joanne

AU - Darmstadt, Gary L.

PY - 2005/7

Y1 - 2005/7

N2 - Verbal autopsy (VA) is used to ascertain cause-specific neonatal mortality using parental/familial recall. We sought to compare agreement between causes of death obtained from the VA by physician review vs. computer-based algorithms. Data were drawn from a cluster-randomised trial involving 4130 live-born infants and 167 neonatal deaths in the rural Sarlahi District of Nepal. We examined the agreement between causes ascertained by physician review and algorithm assignment by the kappa (K) statistic. We also compared responses to identical questions posed posthumously during neonatal VA interviews with those obtained during maternal interviews and clinical examinations regarding condition of newborns soon after birth. Physician reviewers assigned prematurity or acute lower respiratory infection (ALRI) as causes of 48% of neonatal deaths; 41% were assigned as uncertain. The algorithm approach assigned sepsis (52%), ALRI (31%), birth asphyxia (29%), and prematurity (24%) as the most common causes of neonatal death. Physician review and algorithm assignment of causes of death showed high κ for prematurity (0.73), diarrhoea (0.81) and ALRI (0.68), but was low for congenital malformation (0.44), birth asphyxia (0.17) and sepsis (0.00). Sensitivity and specificity of VA interview questions varied by symptom, with positive predictive values ranging from 50% to 100%, when compared with maternal interviews and examinations of neonates soon after birth. Analysis of the VA data by physician review and computer-based algorithms yielded disparate results for some causes but not for others. We recommend an analysis technique that combines both methods, and further validation studies to improve performance of the VA for assigning causes of neonatal death.

AB - Verbal autopsy (VA) is used to ascertain cause-specific neonatal mortality using parental/familial recall. We sought to compare agreement between causes of death obtained from the VA by physician review vs. computer-based algorithms. Data were drawn from a cluster-randomised trial involving 4130 live-born infants and 167 neonatal deaths in the rural Sarlahi District of Nepal. We examined the agreement between causes ascertained by physician review and algorithm assignment by the kappa (K) statistic. We also compared responses to identical questions posed posthumously during neonatal VA interviews with those obtained during maternal interviews and clinical examinations regarding condition of newborns soon after birth. Physician reviewers assigned prematurity or acute lower respiratory infection (ALRI) as causes of 48% of neonatal deaths; 41% were assigned as uncertain. The algorithm approach assigned sepsis (52%), ALRI (31%), birth asphyxia (29%), and prematurity (24%) as the most common causes of neonatal death. Physician review and algorithm assignment of causes of death showed high κ for prematurity (0.73), diarrhoea (0.81) and ALRI (0.68), but was low for congenital malformation (0.44), birth asphyxia (0.17) and sepsis (0.00). Sensitivity and specificity of VA interview questions varied by symptom, with positive predictive values ranging from 50% to 100%, when compared with maternal interviews and examinations of neonates soon after birth. Analysis of the VA data by physician review and computer-based algorithms yielded disparate results for some causes but not for others. We recommend an analysis technique that combines both methods, and further validation studies to improve performance of the VA for assigning causes of neonatal death.

UR - http://www.scopus.com/inward/record.url?scp=21244440590&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=21244440590&partnerID=8YFLogxK

U2 - 10.1111/j.1365-3016.2005.00652.x

DO - 10.1111/j.1365-3016.2005.00652.x

M3 - Article

C2 - 15958155

AN - SCOPUS:21244440590

VL - 19

SP - 323

EP - 331

JO - Paediatric and Perinatal Epidemiology

JF - Paediatric and Perinatal Epidemiology

SN - 0269-5022

IS - 4

ER -