An alternative strategy for perinatal verbal autopsy coding: single versus multiple coders

Authors


Corresponding Author Cyril Engmann, Division of Neonatal-Perinatal Medicine, Department of Pediatrics, University of North Carolina, School of Medicine, CB # 7596, 4th Floor, UNC Hospitals, UNC-Chapel Hill, Chapel Hill, NC 27599-7596, USA. E-mail: cengmann@med.unc.edu

Summary

Objective  To determine the comparability between cause of death (COD) by a single physician coder and a two-physician panel, using verbal autopsy.

Methods  The study was conducted between May 2007 and June 2008. Within a week of a perinatal death in 38 rural remote communities in Guatemala, the Democratic Republic of Congo, Zambia and Pakistan, VA questionnaires were completed. Two independent physicians, unaware of the others decisions, assigned an underlying COD, in accordance with the causes listed in the chapter headings of the International classification diseases and related health problems, 10th revision (ICD-10). Cohen’s kappa statistic was used to assess level of agreement between physician coders.

Results  There were 9461 births during the study period; 252 deaths met study enrolment criteria and underwent verbal autopsy. Physicians assigned the same COD for 75% of stillbirths (SB) (K = 0.69; 95% confidence interval: 0.61–0.78) and 82% early neonatal deaths (END) (K = 0.75; 95% confidence interval: 0.65–0.84). The patterns and proportion of SBs and ENDs determined by the physician coders were very similar compared to causes individually assigned by each physician. Similarly, rank order of the top five causes of SB and END was identical for each physician.

Conclusion  This study raises important questions about the utility of a system of multiple coders that is currently widely accepted and speculates that a single physician coder may be an effective and economical alternative to VA programmes that use traditional two-physician panels to assign COD.

Abstract

Objectif:  Déterminer la comparabilité entre la cause de la mort codée par un seul médecin et celle codée par un comité de deux médecins, utilisant l’autopsie verbale (AV).

Méthodes:  L’étude a été menée entre mai 2007 et juin 2008. Endéans la semaine d’un décès périnatal dans 38 collectivités rurales éloignées au Guatemala, en République Démocratique du Congo, en Zambie et au Pakistan, des questionnaires AV ont été remplis. Deux médecins indépendants, ignorant les décisions de l’un et l’autre, ont attribué une cause sous-jacente du décès, conformément aux causes listées dans les entêtes de chapitre de ‘La Classification Internationale des maladies et problèmes de santé connexes, 10e révision (ICD-10). La statistique Kappa de Cohen a été utilisée pour évaluer le degré de concordance entre les médecins codeurs.

Résultats:  Il y a eu 9461 naissances au cours de la période d’étude; 252 décès répondaient aux critères d’inclusion dans l’étude et ont subi une AV. Les médecins ont attribué le même code pour 75% des mort-nés (K = 0,69; intervalle de confiance 95% [IC95%]: 0,61 à 0,78) et pour 82% des décès néonataux précoces (K = 0,75; IC95%: 0,65 à 0,84). Les profils et la proportion des mort-nés et des décès néonataux précoces déterminés par les deux codeurs médecins étaient très similaires comparés aux causes attribuées individuellement par chacun d’eux. De la même façon, l’ordre de classement des 5 principales causes de mort-nés et de décès néonataux précoces était identique pour chaque médecin.

Conclusion:  Cette étude soulève des questions importantes quant à l’utilité d’un système à plusieurs codeurs qui est actuellement largement acceptée, et spécule qu’un seul codeur médecin pourrait être une alternative efficace et économique pour les programmes d’AV par rapport à l’utilisation traditionnelle de comitéà deux médecins pour attribuer un code de décès.

Abstract

Objetivo:  Determinar el nivel de concordancia entre la causa de muerte determinada mediante autopsia verbal (AV) por un único médico y por un panel de dos médicos.

Métodos:  El estudio fue conducido entre Mayo 2007 y Junio 2008. Dentro de la semana siguiente a una muerte perinatal, se completaron cuestionarios de AV en 38 comunidades rurales remotas de Guatemala, la República Democrática del Congo, Zambia y Paquistán. Dos médicos independientes, que desconocían las decisiones de los demás, asignaron la causa de muerte subyacente, de acuerdo con las causas listadas en los títulos de los capítulos de la Clasificación internacional de enfermedades y problemas relacionados con la salud, décima edición (CIE-10). Se utilizó la Kappa de Cohen para evaluar el nivel de concordancia entre los diferentes médicos codificadores.

Resultados:  Se produjeron 9461 nacimientos durante el periodo de estudio; hubo 252 muertes que cumplían los criterios de inclusión y a las que se les realizó una autopsia verbal. Los médicos asignaron la misma causa de muerte (CDM) en un 75% de las muertes de mortinatos (K = 0.69; intervalo de confianza (IC) 95%: 0.61–0.78) y en 82% de las muertes neonatales tempranas (K = 0.75; IC 95%: 0.65–0.84). Los patrones y proporción de mortinatos y muertes neonatales tempranas determinadas por los médicos realizando la codificación eran muy similares al compararlas con las causas asignadas de forma individual por cada médico. De forma similar, el orden de rango de las primeras 5 causas de mortinato y muertes neonatales tempranas eran idénticas para cada uno de los médicos.

Conclusión:  Este estudio plantea preguntas importantes sobre la utilidad de un sistema de múltiples codificadores que actualmente es ampliamente aceptado, y especula que un único médico codificando podría ser efectivo y una alternativa económica a los programas de AV que utilizan el panel tradicional de dos médicos asignando la CDM.

Introduction

Understanding population-based causes of perinatal death (stillbirths and newborn deaths in the first 7 days of life) is critical to the development of an effective perinatal health policy (Lopez & Mathers 2006). Because there will always be competing demands for healthcare resources, a well-established system for identifying all perinatal deaths and assigning a medically determined cause of death (COD) for each death is highly desirable (Engmann et al. 2009b). In many high-income countries, there is complete recording of deaths and for over 90% of these, medical certification is provided (Mathers et al. 2005). By contrast, fewer than 3% of all perinatal deaths in low- and middle-income countries (LMIC) have medical certification of COD (Lawn et al. 2005). Many of these countries have the highest burden of poverty and disease and continue to lack routine, representative and high-quality information on the levels and causes of death (Setel et al. 2007). Part of the explanation for this may be that over half of all births and perinatal deaths occur in the home and are frequently unrecorded in vital registration systems (Lawn et al. 2008).

Increasing numbers of LMIC are using verbal autopsy (VA) as a cost-effective and sustainable alternative to a thorough medical diagnostic evaluation as a source of data to inform mortality surveillance systems (Hill et al. 2007).To determine the cause of foetal or infant mortality, the VA method relies on information obtained from a standardized interview with the primary caregiver (usually the mother) of the deceased. During this process, the symptoms, signs and behaviours during the illness of the deceased, or of the mother in the case of foetal death, are recorded. Trained coders review these data and apply diagnostic algorithms to determine COD. Typically, two or three trained physician coders review the data and independently assign a COD (Soleman et al. 2006). Any discrepancies between the COD assigned by each physician member of the panel are resolved by discussion and review of the VA data, and a final consensus COD is agreed upon by the physician panel (Setel et al. 2005). The use of multiple physician coders in VA has been used to prevent random and systematic errors. Some researchers have urged that physicians should be encouraged to assign more than just a single COD, and that discussion of discrepant cases among a panel of physicians to reach consensus be considered less appropriate than allowing all physician diagnoses to contribute to the COD profile, whether individual physician diagnoses agree (Joshi et al. 2009b). Other authors have suggested methods for simultaneous analysis of COD (King & Lu 2008). Alternatively, COD can be assigned by the use of predetermined criteria/algorithms or computer simulations, a method that does not require the presence of a physician (Soleman et al. 2006).

A recent report from a general population in India suggests that one trained physician determining COD facilitated by a series of algorithms developed for the Sample Registration System may be as effective as a physician panel in coding COD (Joshi et al. 2009a,b). We sought to determine the potential effectiveness of using a single physician coder to assign the cause of perinatal deaths by comparing COD assigned by two members of a physician panel. Each panel was based in rural districts in one of four low-income countries.

Methods

Setting, subjects and study design

This prospective observational study was nested within an ongoing, cluster randomized, controlled trial, the FIRST BREATH Trial, conducted by the Global Network McClure et al. 2007). The FIRST BREATH trial investigated the effects of implementing a package of newborn care practices, using the WHO Essential Newborn Care (ENC) programme, and a neonatal resuscitation training programme, a simplified version of the American Academy of Pediatrics Neonatal Resuscitation Program, in community settings. As part of this study, birth attendants were trained to collect basic maternal, foetal and neonatal outcomes data, which included demographics, mode of delivery, birthweight, gestational age, receipt of resuscitation and adverse events. All birth attendants were trained to check for foetal and neonatal vital signs on every baby by auscultating the abdomen of every pregnant woman before delivery, and after delivery by feeling the umbilical cord of the neonate for a pulse, auscultating lungs for breath sounds and assessing for any movement (Engmann et al. 2009a,b). Birth weights were measured within 48 h of delivery using UNICEF spring Salter Scales (UNICEF model 145555) provided for the study.

This study included sites in Guatemala (Chimaltenango province), the Democratic Republic of Congo (Equateur province), Zambia (Kafue district) and Pakistan (Thatta district). Within these sites, 38 communities participated in this study. Each community comprised a cluster of villages with approximately 300 deliveries per year. Data describing births were collected by birth attendants and reviewed by trained nurses or health workers assigned to each community and designated as Community Coordinators.

Within 1 week of an early neonatal death (END) or stillbirth (SB), birth attendants notified Community Coordinators who then visited the family, determined eligibility for the study and requested consent from eligible mothers. Perinatal deaths were excluded if they occurred in a hospital, if a birth attendant was absent at delivery, if the mother was unavailable for any reason (including peripartum death) or attempts to enrol the mother did not occur within 7 days of death. A 7-day window within a perinatal death was chosen to reduce the variability in the quality of reporting introduced by recall bias (Soleman et al. 2006; Lee et al. 2008;Fottrell & Byass 2010). Because the conventional perinatal verbal autopsy respondents are mothers, we elected to enrol only those subjects whose mothers were available for interview. Informed consent was obtained from mothers in a private and confidential setting. The consent form was read to all mothers who then provided their signatures or, if they were illiterate, thumbprints.

Training and VA methodology

All Community Coordinators and physicians participating in this study received standardized training in VA methodology (Engmann et al. 2009a). Community Coordinators were trained to interview mothers using the VA questionnaire. To assign COD, physicians were trained in ICD-10 classification, rules and guidelines (WHO 2005).

Uniform data describing the circumstances surrounding a perinatal death were collected from each mother using a standardized VA questionnaire developed specifically for this study from a validated VA tool (Engmann 2009a; Mswia et al. 2006. See Appendices 1 & 2). The questionnaire was administered by the Community Coordinators who then sent these data separately to two local physicians who independently assigned a COD. All physicians were provided with demographic and other descriptive data collected as part of the FIRST BREATH Trial. Each physician assigned one underlying COD, a final COD and contributing causes of death. Underlying COD was defined as the single most important disease or condition that initiated the train of morbid events leading directly to foetal or neonatal death. Underlying COD was assigned in a non-hierarchical manner by physicians familiar with prevailing local diseases and health conditions/patterns (Thatte et al. 2009). After the COD was assigned and entered independently, any discrepancy in assignment of COD between physicians was discussed and a consensus underlying COD assigned. In all cases, the two physicians were able to reach consensus after discussion.

Data collection and analysis

Data were collected between May 2007 and June 2008. Data were entered and transmitted electronically to the data coordinating centre (RTI: Research Triangle Institute International, Research Triangle Park, NC, USA) where data edits, including inter- and intra form consistency checks, were performed. The study was reviewed and approved by the institutional ethics review committees of the Research Triangle International, the University of North Carolina at Chapel Hill and local institutional review boards.

The level of agreement between physician coders for underlying COD was calculated using Cohen’s kappa statistic (K). Levels of agreement based on ranges of kappa values were defined as follows: 0.81–0.99 almost perfect agreement, 0.61–0.80, substantial agreement, 0.41–0.6 moderate agreement and <0.4 slight to fair agreement. (Viera & Garrett 2005). Data were analysed using sas (SAS/STAT® Software version 9.0). Descriptive statistics were generated for participant demographics and circumstances surrounding the deaths. Relationships between categorical variables were evaluated by examining cross-tabulations. Relationships between continuous variables were evaluated by examining means, standard deviations, medians and ranges.

Results

There were 9461 infants born in the designated communities during the study period (Figure 1). Among these, there were 518 SB and END. The SB, END and perinatal mortality rates were 30/1000 births, 25/1000 live births and 55/1000 births, respectively. Of the 518 deaths, 81 were ineligible for the study because the delivery occurred in a hospital (79) or the birth attendant was absent at the time of delivery (2). Among eligible deaths, 185 were not enrolled because the mother was not available for interview within 7 days after the death (145) or did not provide consent (40). This study includes data describing deliveries of 241 women which resulted in 252 perinatal deaths (134 SBs and 118 ENDs).

Figure 1.

 Proportion of stillbirths assigned an underlying cause of death by each of two physician coders and by physician consensus.

The five major causes of END were attributable to infections (45%), birth asphyxia (26%), prematurity (17%), tetanus (4%) congenital malformations (3%) and other/unknown causes (5%). Major causes of SB were attributable to infections (37%), obstructed/prolonged labour (11%), antepartum haemorrhage (10%), prematurity (7%) and cord complications such as prolapse (6%). For 12% of SBs, a COD could not be determined.

Agreement among coders

Physician coders assigned the same COD for 82% of END and 75% of SB. The kappa statistic for overall inter-coder agreement was 0.75 (0.65, 0.84) for END and 0.69 (0.61, 0.78) for SB.

Early neonatal death

Table 1 compares physician coder 1 and physician coder 2 responses for END. Overall, physicians agreed on the same COD for 97 of 118 (82%) END. Table 2 is a comparison of physician coder 1 vs. physician coder 2 responses for specific causes of END. As an example, physicians agreed 109 times of 118 (92%) on prematurity as a COD. They agreed that prematurity was a COD 13 times, and that prematurity was not the COD 96 times. The kappa value (level of agreement) between physicians was 0.7 (95% CI 0.51–0.88).

Table 1.   Comparison of physician coder 1 and physician coder 2 assigned cause of early neonatal death (n = 118)
Physician 1 responsesPhysician 2 responsesPhysician 1
Total n (%)
Physician consensus
n (%)
PretermInfectionBirth asphyxiaCongenital malformTetanusUnknown/no causeOther
  1. The bold numbers along the diagonal indicate agreement reached independently by the two physicians. Per cent agreement = 97/118 = 82.2%. The percentages in parenthesis, provided along the Physician 1 total column, indicate how often Physician 1 reported the cause of death (COD) out of the total. Similarly, the percentages in parenthesis, provided along the bottom Physician 2 total row, indicate how often Physician 2 reported the COD out of the total.

  2. The percentages provided in parenthesis in the extreme right Physician Consensus column refer to how often the Physician Consensus reported the COD out of the total. Thus, taking infection as an example, Physician 1 and Physician 2 concluded neonatal infection was the COD 52 times. Physician 1 and Physician 2 initially agreed that infection was the underlying COD for 44 of the 52 cases. After discussing the 13 discrepant cases where one but not both attributed the underlying COD to infection, the physicians came to final consensus that 8 of the 13 cases had an underlying COD of neonatal infection. Other causes of early neonatal death were hypothermia, low birth weight and birth trauma.

Preterm1311000116 (14)20 (17)
Neonatal infection2444011052 (44)52 (44)
Birth asphyxia3328001035 (30)31 (26)
Congenital malformation00040004 (3)4 (3)
Tetanus00005005 (4)5 (4)
Unknown/no cause00000202 (2)4 (3)
Other11100014 (3)2 (2)
Physician 2
Total n (%)
19 (16)49 (42)34 (28)4 (3)6 (5)4 (3)2 (2)118 
Table 2.   Comparison of physician coder 1 and physician coder 2 responses for specific causes of early neonatal deaths (n = 118)
Underlying cause of deathPhysician response –n (%)Kappa with corresponding 95% CI
Physicians agreedPhysicians disagreed (one physician noted the condition positive and other noted the condition negative)
Condition noted as underlying causeCondition not noted as underlying causeTotal
Preterm1396109 (92%)9 (8%)0.70 (0.51, 0.88)
Neonatal Infection4461105 (89%)13 (11%)0.78 (0.66, 0.89)
Birth asphyxia2877105 (89%)13 (11%)0.73 (0.60, 0.87)
Congenital malformation4114118 (100%)01.00 (1.00, 1.00)
Tetanus5112117 (99%)1 (1%)0.90 (0.72, 1.00)
Unknown/no cause2114116 (98%)2 (2%)0.66 (0.22, 1.00)

Stillbirth

Table 3 compares physician 1 and physician coder 2 responses for SB. Overall, physicians agreed on the same underlying COD at the chapter-heading level of ICD-10 for 101 of 134 (75%) of SBs.

Table 3.   Comparison of physician coder 1 and physician coder 2 responses for cause of stillbirth (n = 134)
Physician 1 responsesPhysician 2 responsesPhysician 1
Total n (%)
Physician consensus n (%)
Antepartum haemorrhageMaternal infectionPretermMaternal accidentProlonged labourCord prolapse/ complication Unknown/ no causeOther
  1. The bold numbers along the diagonal indicate agreement reached independently by the two physicians. Per cent agreement = 101/134 = 75.4%. The percentages provided in parenthesis along the Physician 1 total column indicate how often Physician 1 reported the cause of death (COD) out of the total. Similarly, the percentages provided in parenthesis along the total row indicate how often Physician 2 reported the COD out of the total.

  2. The percentages provided in parenthesis in the extreme right Physician Consensus column refer to how often the Physician Consensus reported the COD out of the total. Other causes of stillbirth were identified as malpresentation, folic acid deficiency, hypertension, post-term delivery, multiple birth, polyhydramnios and multipara.

Antepartum haemorrhage8300000112 (8)13 (10)
Maternal infection04200000143 (32)50 (37)
Preterm1350004013 (10)9 (7)
Maternal accident000610007 (5)7 (5)
Prolonged labour00001110012 (9)15 (11)
Cord prolapse/complication000005308 (6)8 (6)
Unknown/no cause12100011015 (11)16 (12)
Other05103201324 (18)16 (12)
Physician 2
Total n (%)
10 (7)55 (41)7 (5)6 (4)15 (11)8 (6)18 (13)15 (11)134 

Table 4 compares physician coder 1 vs. physician coder 2 responses for specific causes of SB. Using maternal infection as an example, physicians agreed 120 times of 134 (90%). They agreed that maternal infection was the cause of SB 42 times and that maternal infection was not the cause of SB 78 times. Physicians disagreed 14 times on the designation of infection as a cause of SB. The kappa level of agreement was 0.78 (0.67, 0.89).

Table 4.   Comparison of physician coder 1 and physician coder 2 responses for specific causes of stillbirth (n = 134)
Underlying cause of SBPhysician response –n (%)Kappa with corresponding 95% CI
Physicians agreedPhysicians disagreed (one physician noted condition positive and other noted condition negative)
Condition noted as underlying causeCondition not noted as underlying causeTotal
Antepartum haemorrhage8120128 (96%)6 (4%)0.70 (0.48, 0.93)
Maternal infection4278120 (90%)14 (10%)0.78 (0.67, 0.89)
Preterm5119124 (93%)10 (7%)0.46 (0.19, 0.74)
Maternal accident6127133 (99%)1 (1%)0.92 (0.76. 1.00)
Prolonged labour11118129 (96%)5 (4%)0.79 (0.62, 0.97)
Cord prolapse/complication5123128 (96%)6 (4%)0.60 (0.31, 0.89)
Unknown/no cause11112123 (92%)11 (8%)0.62 (0.42, 0.83)

Individual assignment of COD and consensus COD

The proportion of SBs and ENDs determined by the two physicians were very similar, as were the patterns derived from the consensus process, compared to causes individually assigned by each physician. Similarly, the rank order of the top 5 leading causes of SB and END was identical for each physician (Figures 1 and 2).

Figure 2.

 Proportion of early neonatal deaths assigned an underlying cause of death by each of two physician coders and by physician consensus.

Discussion

After preparation using a standardized VA training programme, two physicians were in substantial agreement when assigning the major causes of END. There was almost perfect agreement when tetanus and congenital malformations were the causes of END. Any discrepancies noted in the COD assigned to individual cases had little impact on their rank order or the overall pattern of reported mortality. Substantial agreement between physicians was observed in the assignment of the major causes of SB (antepartum haemorrhage, maternal infection and prolonged labour). There was only moderate agreement on the assignment of cord prolapse and prematurity as a cause of SB, while there was almost perfect agreement when maternal accident was assigned as the COD.

Few studies have evaluated the impact of different methods for assigning cause of neonatal death using VA. In a recent paper, Joshi et al. (2009a) compared the assignment of the COD in 45 villages in Southern India by single vs. multiple coders. This was a study of mortality in a general population of all ages, and fewer than 1% of the deaths occurred in children 0–28 days of age. They reported that physician coders agreed on the same diagnosis 94% of the time, with overall kappa values of 0.93 suggesting almost perfect agreement among physician coders. Among deaths in children aged 0–28 days, they reported kappa values of 1.0, although there were only 11 cases. Our study examined the comparability of the assignment of causes by two physicians for perinatal deaths only. For the three most important causes of END (infections, birth asphyxia and prematurity), physicians agreed on the same COD approximately 90% of the time, suggesting substantial agreement. For two other causes of END, congenital malformations and tetanus, physician agreement was nearly 100%. Similar results were reported by Edmond et al. (2008) on levels of agreement among three physicians determining cause of 590 neonatal deaths from verbal autopsies in rural Ghana. There was substantial agreement among three physicians for prematurity, birth asphyxia and infections (kappa values 0.8, 0.77 and 0.72). In contrast to our study, they reported a kappa value of 0.63 for congenital abnormalities as a COD.

In our study, physicians showed substantial agreement for certain causes of SBs (antepartum haemorrhage, maternal infection and prolonged labour), and almost perfect agreement for maternal accidents. There was only moderate agreement for prematurity and cord prolapse. When the diagnostic accuracy of VA as determined by three experienced community paediatricians to determine cause of SBs from rural Ghana was compared to a hospital reference standard, VA performed poorly for causes of SB diagnosis such as congenital abnormalities and maternal haemorrhage, while accuracy was higher for intrapartum obstetric complications and antepartum maternal disease (Edmond et al. 2008).

Even in settings where placental examinations, autopsies, cultures, karyotypes, x-rays, MRIs and other imaging are available, up to 60% of SBs are unexplained, highlighting the inherent difficulties that understanding and obtaining agreement over cause of SB can pose (McClure et al. 2006; Silver et al. 2007). In our study, the low rate of ‘unknown’ COD may be an artefact of the study during which coders may have perceived some pressure to assign a cause of SB. Nonetheless, the rate of concurrence between coders also suggests that VA may be a useful tool in determining population-based causes of SB.

There are economic and resource implications of the results of this study. The cost of programmes using VA to assign COD could be substantially reduced by switching to a system of single coding. Joshi et al. estimate that with deaths coded only once, the cause-of-death assignment costs can be halved and project management costs reduced by one-third. They also suggest that funds currently used for duplicate coding could be reassigned to conduct validation studies that compare COD assignments from single coders against COD derived from reliable medical records, diagnosis by autopsy or physician-diagnosed deaths in the community. Because VA is most typically used within weak health systems that suffer a shortage of physicians, utilizing fewer physicians and provide standardized training to them to code VA and redeploying them to other clinical tasks could be a more appropriate use of scarce human resources.

A major strength of this study is the standardized VA training and tools programme which we have reported on previously. After initial training, a train-the-trainer model was used to spread it in the different countries within the GN. This strategy increases knowledge, promotes ownership, builds capacity and enables sustainability of programmes (Enweronu-Laryea et al. 2009). In contrast to other studies which delay interview, we performed them within 1 week and found mothers eager to discuss their baby’s death. Early interviews may also yield more accurate diagnoses. There are also limitations to our study. It is possible that the duplicate coding process may be a poor method for detecting systematic errors in the assignment of causes of death. Also, poor training of coders could also result in a bias towards a particular COD assignment, which could be repeated by subsequent coders. Each coder was tested after training in the VA programme, making these potential biases less likely. Another potential limitation may be bias towards certain diagnoses resulting from prior knowledge of the coders of disease patterns in their community. Therefore, use of coders from the community in which the deaths occur would be expected to result in a high level of agreement, but with less certainty of the correct assignment of COD. Although the VA tool has been validated previously in hospital settings, its application in a community setting where deaths occur outside of hospitals and the formal health care system has not been validated. Therefore, we cannot be certain of the accurate assignment of COD. However, even if tests of validity discovered incorrect assignment of COD for particular causes, it is unlikely that such a problem would affect agreement between coders.

Although inter-observer and intra-observer variations have been recognized over the years, the impact of these variations has not been studied in detail (Fauveau 2006; Garenne & Fauveau 2006). The findings from our study suggest that a single physician coder may be as effective as two coders in determining cause of SB and END when trained in a standardized VA programme. This study also raises important questions about the utility of a system of multiple coders that is currently widely accepted and speculates that a single physician coder may be an effective and economical alternative to VA programmes that use traditional two-physician panels to assign COD.

Acknowledgements

Funding was provided by grants from the National Institutes of Child Health and Human Development and the Bill and Melinda Gates Foundation.

Appendices

Appendix 1: Verbal autopsy questionnaire

inline image

Appendix 2. Case definitions for maternal and neonatal/foetal underlying, final, and contributing causes of death

Fetal/neonatal causes

Preterm/complications of prematurity:

Infant born before 37 completed weeks of pregnancy (includes maternal subjective assessment of born early).

Fetal/neonatal infection/sepsis/pneumonia:

Congenital or acquired invasion and multiplication of germs (bacteria, fungi or viruses).

Birth asphyxia/intrapartum asphyxia:

Failure to initiate and sustain breathing at birth.

Congenital malformation:

A major structural defect which causes the baby to die.

Birth trauma (during labor):

Injuries to the infant during the process of birth.

Fetal trauma:

Injuries affecting the fetus before it is born.

Neonatal accident: Any injury or trauma affecting the infant after it is born.

Tetanus:

Neonates with a normal ability to suck and cry during the first 2 days of life, and who, between day 3 and 28 cannot suck normally, and become stiff or have fits/convulsions (i.e. jerking of muscles) or both.

Diarrhea:

The passage of loose or liquid stools more frequently than is normal for babies.

Hypothermia:

A neonate who feels cold to touch, or who has a temperature below 36.5 °C.

Low birth weight:

Birthweight <2500 g, or smaller in size than expected for the baby’s gender, genetic heritage and gestational age (includes mother’s opinion of small).

Jaundice:

Yellow coloration of the skin or whites of the eyes.

Maternal causes

Antepartum haemorrhage:

Heavy bleeding in the last week of pregnancy (excludes ‘spotting’)

Maternal infection/sepsis:

Invasion and multiplication of germs (bacteria, fungi and viruses). This definition includes malaria, and excludes minor infections such as colds

Preterm delivery:

Labor and delivery before 37 completed weeks of pregnancy

Maternal accident:

Any injury occurring to the mother after 20 weeks of pregnancy that might affect the fetus

Obstructed/prolonged labor:

Any condition that results in the labor process being obstructed e.g. because the fetus is too large for the birth canal or the presentation of the fetus prevents the mother from pushing the baby out of the birth canal

Multiple delivery:

Birth of two or more babies (even if one of the babies is dead)

Hypertensive disorder/eclampsia: High blood pressure that occurs before or during pregnancy. This may be associated with generalized swelling, headaches or seizures

Cord prolapse/complications: When the baby’s umbilical cord falls into the birth canal ahead of the baby’s head or other parts

Malpresentation: When the baby is in a difficult position for the birth process

Post-term: any baby born after 42 completed weeks of pregnancy

Ancillary