This article has Open Peer Review reports available.
Adaptation of a probabilistic method (InterVA) of verbal autopsy to improve the interpretation of cause of stillbirth and neonatal death in Malawi, Nepal, and Zimbabwe
© Vergnano et al; licensee BioMed Central Ltd. 2011
Received: 18 November 2010
Accepted: 5 August 2011
Published: 5 August 2011
Verbal autopsy (VA) is a widely used method for analyzing cause of death in absence of vital registration systems. We adapted the InterVA method to extrapolate causes of death for stillbirths and neonatal deaths from verbal autopsy questionnaires, using data from Malawi, Zimbabwe, and Nepal.
We obtained 734 stillbirth and neonatal VAs from recent community studies in rural areas: 169 from Malawi, 385 from Nepal, and 180 from Zimbabwe. Initial refinement of the InterVA model was based on 100 physician-reviewed VAs from Malawi. InterVA indicators and matrix probabilities for cause of death were reviewed for clinical and epidemiological coherence by a pediatrician-researcher and an epidemiologist involved in the development of InterVA. The modified InterVA model was evaluated by comparing population-level cause-specific mortality fractions and individual agreement from two methods of interpretation (physician review and InterVA) for a further 69 VAs from Malawi, 385 from Nepal, and 180 from Zimbabwe.
Case-by-case agreement between InterVA and reviewing physician diagnoses for 69 cases from Malawi, 180 cases from Zimbabwe, and 385 cases from Nepal were 83% (kappa 0.76 (0.75 - 0.80)), 71% (kappa 0.41(0.32-0.51)), and 74% (kappa 0.63 (0.60-0.63)), respectively. The proportion of stillbirths identified as fresh or macerated by the different methods of VA interpretation was similar in all three settings. Comparing across countries, the modified InterVA method found that proportions of preterm births and deaths due to infection were higher in Zimbabwe (44%) than in Malawi (28%) or Nepal (20%).
The modified InterVA method provides plausible results for stillbirths and newborn deaths, broadly comparable to physician review but with the advantage of internal consistency. The method allows standardized cross-country comparisons and eliminates the inconsistencies of physician review in such comparisons.
Cause-specific mortality data on childhood deaths are vital to identify health needs, compare patterns of death across populations, plan and monitor interventions, and inform policy [1–3]. In high-income countries, all births and deaths are enumerated through vital registration systems, and death certification is routine. In low-income settings, most births and deaths occur at home, death certificates are rarely available, and vital registrations are often inadequate or nonexistent [2–4].
Verbal autopsies (VAs) provide an alternative means of identifying probable causes of death through interviews with a close caregiver of the deceased, in which information about the circumstances, signs, and symptoms leading to death are gathered. VAs have limitations: they require recollection of events at the time of death, rely on understanding and reporting of signs and symptoms by interviewees, and may be influenced by interviewer skills. The data must also be interpreted to establish a diagnosis . Conventionally, VA questionnaires are read by two or more physicians separately and one or more causes of death are attributed. A cause of death is established when physicians' opinions correspond; otherwise diagnosis is reconsidered and discussed with or without the input of an additional physician. If no agreement is reached, the cause of death is considered undetermined. Repeatability of this diagnostic process over time and in different settings is problematic, particularly when diagnostic criteria are not standardized amongst different clinicians [6–8]. In some situations, disagreement between physicians is such that a large proportion of causes of death remain indeterminate [7, 9]. Moreover, the method is costly, time-consuming, and requires the involvement of physicians who are an already overstretched resource in low-income countries [6, 10].
Despite these limitations, VAs are useful in estimating cause-specific mortality fractions (CSMFs) in population studies [6, 8, 11]. They have been used extensively in epidemiological studies, household surveys, and sentinel surveillance sites, and have been piloted in subsamples from sample registration systems. There remains a need to refine the technique to make it more comparable, repeatable, easy to apply, and cost-effective.
VA questionnaires devised by the WHO attempt to standardize the interview process, but more standardized approaches to interpreting VA data are needed. Hierarchical algorithms and computer programs based on logistic regression have been used, but they are difficult to standardize across cultures and age groups and can usually only identify single causes of death [12, 13]. InterVA uses a probabilistic method and has been tested in a range of settings for deaths at all ages, across sexes, and for maternal deaths [14, 15].
We describe the refinement and evaluation of InterVA to identify causes of death in the perinatal (stillbirths and neonatal deaths in the first seven days) and neonatal periods, using data from three different settings: Malawi, Zimbabwe, and Nepal.
InterVA indicators and cause of death categories.
Cause of Death Categories
was this an elder 65+ years
any chronic/recurrent diarr (4+w)
was this an adult 50-64 years
any abdominal swelling
was this a female 15-49 years
was this a male 15-49 years
was this a child 5-14 years
any abnormality of urine
was this a child 1-4 years
any urinary retention
was this an infant 4 wks-1 yr
was this a neonate < 4 wks
any swelling of ankles/legs
was she pregnant at death
no bilateral swelling of ankle
did pregnancy end within 6 weeks
any skin lesions/ulcers
did final illness last at least 3 weeks
any rash (non-measles)
Other acute infection
did final illness last < 3 weeks
any herpes zoster
was death very sudden/unexpected
any measles rash
was death during wet season
any excessive night sweats
was death during dry season
any excessive water intake
was s/he in a transport accident
any excessive urination
did s/he drown
any excessive food intake
had s/he fallen recently
any acute fever
Acute respiratory disease (not pnem.)
any poisoning, bite, sting
any persistent fever (> 2 wk)
Chronic respiratory disease
was s/he a known smoker
any enlarged/swollen glands
any obvious recent injury
any facial swelling
was s/he known to drink alcohol
was there a coma > 24 hrs
any suggestion of homicide
any weight loss
any convulsions or fits
any diagnosis of epilepsy
was the fontanelle raised
any delayed/regressed development
was the fontanelle or eyeball sunken
any diagnosis of asthma
Disorders of the digestive system
any diagnosis of diabetes
Diseases of the nervous system
was there paralysis on both sides
any diagnosis of heart disease
Sickle cell anemia
any paralysis/weakness on 1 side
any diagnosis of HIV/AIDS
any stiff neck
any diagnosis of hypertension
any oral candidiasis
been discharged from hospital very ill
any suggestion of suicide
abnormal hair coloring
any surgery just before death
any coughing with blood
any diagnosis of TB
any chest pain
was s/he adequately vaccinated
was there a cough for > 3 wks
any diagnosis of liver disease
was there a cough for up to 3 wks
any diagnosis of cancer
did baby have arched back after 2 days
any productive cough
any diagnosis of stroke
baby stopped sucking after day 3
any rapid breathing
any diagnosis of measles
did the baby die on day 1
any breathlessness on exertion
any diagnosis of kidney disease
did the mother fail to receive tetanus toxoid vaccine
any breathlessness lying flat
any diagnosis of hemoglobinopathy
did convulsions happen on day 1
any chest indrawing
any diagnosis of malaria
was there no cry/move/breath at birth
any difficulty breathing
any delivery complications
was baby's skin puffy/mushy at birth
any breast lump or lesion
any heavy bleeding around delivery
did the baby fail to cry at birth
was there prolonged labor > 24 hrs
were there convulsions during delivery
any abdominal mass
was the baby born early < 34 wks
Additional causes of death
any abdominal pain
was the baby small < 2500 g
any diarrhea with blood
was there difficulty breathing at birth
any vomiting with blood
any congenital malformations
any acute diarrhea (< 2 wks)
was this a multiple birth
any persistent diarrhea (2-4 wks)
any umbilical infection
where p (C|I) indicates the probability of a cause of death (C) given the presence of the indicator (I) and p(I/!C) is the probability of I in the absence of C .
Probabilities of final-cause categories increase or decrease in relation to specific signs and symptoms reported in the VA interview. If symptoms are not reported, the probabilities do not change. The program is available online http://www.InterVA.net. Users can enter the data as single cases or in batches, and the model generates up to three causes of death and their respective likelihoods. Prior to the current study, the probability matrix consisted of 34 cause of death classifications and 104 indicators .
To explore the performance of InterVA in different settings, 734 stillbirth and neonatal VAs were obtained from rural areas of three low-income countries.
In Malawi (Mchinji District), 169 stillbirth and neonatal VAs were collected from 2004 to 2005, as part of a cluster-randomized study evaluating two community interventions to improve maternal and child health . Although designed for the study, the VA questionnaire was comparable in structure and content with the subsequent WHO questionnaire . Completed questionnaires were interpreted independently by two Malawian pediatricians, who assigned up to three causes of death on the basis of a hierarchical classification and algorithm . They were able to use alternative diagnoses where necessary. Discrepancies were resolved by discussion and, if consensus could not be reached, the cause of death was recorded as indeterminate.
In Nepal (Makwanpur district), 385 VAs were collected from 2001 to 2003 as part of a cluster-randomized study of a community intervention to improve maternal and child health . The questionnaire was again comparable with the subsequent WHO tool. Questionnaires were interpreted independently by two Nepalese pediatricians, who each assigned a single cause of death on the basis of the same algorithm used in Malawi. Discrepancies were resolved after review by a third physician.
Characteristics of the three studies used as data sources.
Neonatal mortality rate
Number of VA questionnaires
Mixed open and closed questions
Mixed open and closed questions
Standard WHO tool incorporating open and closed questions (24)
5 lay Malawian interviewers with secondary education
Lay local field coordinators
45 midwife enumerators
2 experienced local pediatricians
3 causes of death
3 experienced local pediatricians
1 cause of death
3 experienced local physicians
International Classification of Diseases and Related Health Problems (ICD-10)
1 cause of death
Initial refinement of the InterVA model was based on 100 (59%) physician-reviewed VAs from Malawi. The use of these data for refinement was pragmatic in that, at the time of refinement, they were the only data available. Data from the VA questionnaire were entered in the InterVA model, which assigned causes of death and associated likelihoods. The open histories, where the caregiver reported the events leading to death, were coded and also entered in the model. CSMFs obtained using the original InterVA and physician review were compared. CSMFs were calculated from the InterVA output as the sum of the likelihoods computed for each single cause of death category, divided by the sum of the likelihoods for all causes. For the calculation of CSMFs from physician-review data, if more than one cause of death was assigned, each was considered as a proportion of the total death. Therefore, if a single cause of death was assigned by all physicians, or if only one was available, it explained 100% of that death. If more than one cause of death was attributed, each contributed an equal proportion of the total 100%. For example, if both reviewing physicians assigned prematurity as a cause of death and one of them also assigned sepsis, then prematurity contributed 75% and sepsis 25% to the death. In this way, every available physician diagnosis contributed to the cause-specific mortality profile, avoiding a potential loss of information and bias that might have been introduced by using consensus diagnoses alone.
Fifty-four neonatal-death questionnaires were analyzed with the original InterVA model. Stillbirths were initially excluded, as InterVA was not designed to classify them. The results of this first analysis identified the need for greater differentiation in the model among causes of death in the neonatal period. The InterVA indicators and matrix probabilities were therefore reviewed for clinical and epidemiological coherence by a pediatrician-researcher (SV) and an epidemiologist involved in the development of InterVA (EF). Following this initial refinement, InterVA was evaluated by comparing case-by-case diagnoses with physician-assigned diagnoses for the same 100 VA cases, as well the population-level CSMFs. A process of refinement and comparisons with physician review was undertaken until InterVA elicited mortality profiles deemed by the researchers to be plausible and satisfactorily comparable to physician review.
Evaluating the refined InterVA model
The modified InterVA model was evaluated by comparing population-level CSMFs derived from the two methods of interpretation (physician review and InterVA) for a further and hitherto-untouched 69 VAs from Malawi, 385 from Nepal, and 180 from Zimbabwe. A diversity of data sources was chosen to assess the performance of InterVA in a range of settings. Comparisons of population-level CSMFs were considered paramount as InterVA is intended as a public health tool for health monitoring and program evaluation, rather than for use in clinical settings. Nevertheless, individual level, case-by-case comparisons between physician diagnoses and InterVA were also conducted and the kappa statistic for interrater agreement was calculated to further evaluate the InterVA against the only available alternative method in our populations .
The Maimwana study (Malawi) received ethical approval from the Malawi National Health Sciences Research Committee; the MIRA Makwanpur, Nepal, study was approved by the Nepal Health Research Council and the Institute of Child Health and Great Ormond Street Hospital ethics committees; and the Zimbabwe Maternal and Perinatal Mortality Study received ethical approval from the Medical Research Council of Zimbabwe (MRCZ/A/1368).
InterVA was modified to include two extra cause of death categories: fresh stillbirth and macerated stillbirth. To define the stillbirth diagnoses and differentiate among possible causes of stillbirth and neonatal death, nine further indicators were added to the model. The resulting modifications to the specific indicators and cause of death categories included in InterVA are shown in Table 1. As these are extra entities in the model, they run in parallel to the existing indicators and causes without directly affecting them.
Evaluation of the Refined InterVA Model
After refining the model, case-by-case agreement between InterVA and reviewing physician diagnoses, for 69 cases from Malawi, 180 cases from Zimbabwe, and 385 cases from Nepal, was 83% (kappa 0.76 (0.75 - 0.80)), 71% (kappa 0.41(0.32-0.51)), and 74% (kappa 0.63 (0.60-0.63)), respectively.
Comparison of cause-specific mortality fractions according to InterVA and physician review.
Fresh/macerated split of stillbirths from Malawi and Nepal based on interpretation by InterVA and physician review.
Multicountry mortality comparison
The deadline for the Millennium Development Goals (MDGs) is less than five years away and the need to quantify childhood mortality, understand its causes, and assess the effects of proposed interventions are central to MDG4. Neonatal deaths contribute about 40% of under-5 mortality globally . A recent evaluation of the INDEPTH network of Health and Demographic Surveillance Sites  calls for all sites to use InterVA for coding of causes of death, since such approaches represent "the only viable strategy to produce timely and comparable cause of death statistics" . Our study has revised the InterVA method for verbal autopsy to improve its ability to identify causes of stillbirth and newborn death and tested it in three populations.
In this study, physician review was used as a reference standard to compare InterVA. The use of physician review was the only alternative source of cause of death assessment for our study populations. This choice has limitations, however. Physicians are influenced by their experience, perception, and interpretation of local epidemiology [23, 27]. Moreover, they mostly use the open history to reach a decision and may not account consistently for all the indicators. Sensitivity and specificity of physician review compared with hospital diagnosis in neonatal populations varied between 64% and 74% in a recent study  and concerns about inter- and intrarater reliability are well described .
An alternative to physician diagnoses is the use of hospital records. Hospital diagnoses have been used to establish sensitive, specific, and positive predictive values of VA diagnoses [8, 12, 20]. The main pitfall of hospital diagnoses in developing countries, particularly in rural settings, is that the CSMF of deaths occurring in hospitals are likely to be different from the ones in communities . There is therefore the risk of increasing precision of an interpretative method, defined as its ability to reproduce hospital diagnoses in the population where it is tested. This would not necessarily produce results that are correct when used in populations where access to hospitals and health care is limited. Moreover, the ability to recognize, recall, and report signs of illnesses may be different among hospital users and nonhospital users.
The results of InterVA as compared with physician review showed an almost identical ranking of causes of death. However, differences exist. Some of these differences can be explained by the way the model was constructed. Prematurity, for example, was over-diagnosed by InterVA in Zimbabwe and Nepal. This probably resulted from using a dataset where clinicians were allowed more than a single cause of death to refine InterVA. In fact, when multiple causes of death are allowed, prematurity is more likely to be listed as a coexisting cause of death than when a single cause is selected . The model did not include "other" as a cause of death and would have classified such causes of death in one of the available diagnoses.
InterVA over-diagnosed neonatal infections compared with physician review in Zimbabwe, while the opposite happened in Nepal. This inconsistency could be due to the interpretation of signs by different physicians. Alternatively, it could be due to the selection of a priori probabilities. Greater understanding of the way physicians decide to value or ignore signs and symptoms may help in future refinements and evaluations of InterVA.
Stillbirths were included for practical and public health reasons. Although globally there are about 3.2 million stillbirths per year, reliable statistics are lacking . This information gap has to be addressed. About half of perinatal deaths are accounted for by stillbirths . The refinements including stillbirths in the model eliminate the need to differentiate between live births and stillbirths before processing VA data, making the method more suitable for use in large surveys. The separation between fresh and macerated stillbirths is relevant, as prevention strategies are different. The comparisons between InterVA and physician review in Malawi and Nepal suggest that InterVA can differentiate the two categories, although, as with neonatal deaths, there may be room for further refinement.
Case-by-case agreement was moderate in all datasets, however it was lower for Zimbabwe compared to Nepal and Malawi. The new indicators and matrix probabilities have been chosen and modified on the basis of the personal experience of the researchers, and subsequently tested and modeled on a subset of the Malawi data. There is a risk, therefore, that the tool may be too closely modeled on a sub-Saharan African setting (although the results from Nepal do not support this) or on a particular research setup. In addition, the modifications have so far not been put to a panel of experts and may need to be subject to a wider consensus.
There may be important epidemiological and social explanations for the difference in the CSMF in Malawi, Zimbabwe, and Nepal. However, even if the interpretation of verbal autopsy data by InterVA was consistent, methodological variability in other aspects of VA may have contributed to the observed cause distribution. Indeed, the close comparability of CSMF between Malawi and Nepal may to some degree reflect common data capture processes that differ from those used in Zimbabwe. It is possible that in Nepal and Malawi, the populations were part of research areas and might have been sensitized to recognize, describe, and recall signs of neonatal diseases, while in Zimbabwe the community was part of a government surveillance and may have responded differently. Nevertheless, this is a reality of all VA studies conducted in research settings. Use of lay (in Malawi and Nepal) versus health-professional (in Zimbabwe) interviewers and their gender may also have had an impact on data capture. This highlights the need for further methodological research into the effects of other aspects of VA. It is likely that a number of strategies and international collaborations will be necessary to ensure the success of such investigations.
The modified version of InterVA for stillbirths and neonatal deaths produced plausible results when compared with physicians' opinions but had the advantage of being completely internally consistent, allowing standardized comparisons of data from different countries. Ultimately, standardized methods are essential and their application and evaluation in a wide range of settings is encouraged. Through wider application, the strengths and weakness of InterVA, and VA in general, will become more apparent, thereby better informing the application and public health utility of surrogate methods for measuring mortality in absence of vital registration systems.
Maimwana trial funding in Malawi was provided by Saving Newborn Lives/Save the Children, with additional funding from the UK Department for International Development (DFID), The Wellcome Trust and UNICEF Malawi. We thank the Maimwana office and field staff who made this project happen: Tambosi Phiri, Mikey Rosato, Delia Chikuse, Levie Kamtambe, Queen Sara Soho, Joseph Jaffu, Jeremia Mvula and the Mchinji community without which the study would have not been possible, the Mchinji District Health Management Team and district Executive Committee, and traditional leaders working in the district that supported the project. Prof. Marie-Louise Newell also helped set the study up and advised on its development.
Makwanpur trial funding was provided by the UK Department for International Development, with additional support from the Division of Child and Adolescent Health, WHO, UNICEF, and the UN Fund for Population Activities. We thank the MIRA team in Makwanpur and Kathmandu, project managers Bhim Shrestha and Kirti Tumbahhamphe, Drs. S Manandhar and A Ojha, who read and interpreted the verbal-autopsy questionnaires, the communities in Makwanpur district who allowed the study to take place, and the Makwanpur District Development Committee and its members.
Data from Zimbabwe were available thanks to the support and funding provided by the UK Department for International Development (DfID), the World Health Organization (WHO), the United Nations Fund for Population Activities (UNFPA), and the United Nations Children's Fund (UNICEF). We are particularly grateful to Gwendoline Kandawasvika for assistance in data handling.
- Murray CJ, Lopez AD, Wibulpolprasert S: Monitoring global health: time for new solutions. BMJ 2004, 329: 1096-1100. 10.1136/bmj.329.7474.1096View ArticlePubMedPubMed CentralGoogle Scholar
- Mathers CD, Fat DM, Inoue M, Rao C, Lopez AD: Counting the dead and what they died from: an assessment of the global status of cause of death data. Bull World Health Organ 2005, 83: 171-177.PubMedPubMed CentralGoogle Scholar
- Setel PW, Macfarlane SB, Szreter S, Mikkelsen L, Jha P, Stout S, et al.: A scandal of invisibility: making everyone count by counting everyone. Lancet 2007.Google Scholar
- Setel PW, Sankoh O, Rao C, Velkoff VA, Mathers C, Gonghuan Y, et al.: Sample registration of vital events with verbal autopsy: a renewed commitment to measuring and monitoring vital statistics. Bull World Health Organ 2005, 83: 611-617.PubMedPubMed CentralGoogle Scholar
- Snow B, Marsh K: How Useful are Verbal Autopsies to Estimate Childhood Causes of Death? Health Policy and Planning 1992, 7: 22-29. 10.1093/heapol/7.1.22View ArticleGoogle Scholar
- Quigley MA, Chandramohan D, Rodrigues LC: Diagnostic accuracy of physician review, expert algorithms and data-derived algorithms in adult verbal autopsies. Int J Epidemiol 1999, 28: 1081-1087. 10.1093/ije/28.6.1081View ArticlePubMedGoogle Scholar
- Soleman N, Chandramohan D, Shibuya K: Verbal autopsy: current practices and challenges. Bull World Health Organ 2006, 84: 239-245. 10.2471/BLT.05.027003View ArticlePubMedPubMed CentralGoogle Scholar
- Chandramohan D, Maude GH, Rodrigues LC, Hayes RJ: Verbal autopsies for adult deaths: their development and validation in a multicentre study. Trop Med Int Health 1998, 3: 436-446. 10.1046/j.1365-3156.1998.00255.xView ArticlePubMedGoogle Scholar
- Freeman JV, Christian P, Khatry SK, Adhikari RK, LeClerq SC, Katz J, et al.: Evaluation of neonatal verbal autopsy using physician review versus algorithm-based cause-of-death assignment in rural Nepal. Paediatr Perinat Epidemiol 2005, 19: 323-331. 10.1111/j.1365-3016.2005.00652.xView ArticlePubMedGoogle Scholar
- Byass P, Huong DL, Minh HV: A probabilistic approach to interpreting verbal autopsies: methodology and preliminary validation in Vietnam. Scand J Public Health Suppl 2003, 62: 32-37.View ArticlePubMedGoogle Scholar
- Quigley MA, rmstrong Schellenberg JR, Snow RW: Algorithms for verbal autopsies: a validation study in Kenyan children. Bull World Health Organ 1996, 74: 147-154.PubMedPubMed CentralGoogle Scholar
- Kalter HD, Hossain M, Burnham G, Khan NZ, Saha SK, Ali MA, et al.: Validation of caregiver interviews to diagnose common causes of severe neonatal illness. Paediatr Perinat Epidemiol 1999, 13: 99-113. 10.1046/j.1365-3016.1999.00151.xView ArticlePubMedGoogle Scholar
- Marsh DR, Sadruddin S, Fikree FF, Krishnan C, Darmstadt GL: Validation of verbal autopsy to determine the cause of 137 neonatal deaths in Karachi, Pakistan. Paediatr Perinat Epidemiol 2003, 17: 132-142. 10.1046/j.1365-3016.2003.00475.xView ArticlePubMedGoogle Scholar
- Fantahun M, Fottrell E, Berhane Y, Wall S, Hogberg U, Byass P: Assessing a new approach to verbal autopsy interpretation in a rural Ethiopian community: the InterVA model. Bull World Health Organ 2006, 84: 204-210. 10.2471/BLT.05.028712View ArticlePubMedPubMed CentralGoogle Scholar
- Fottrell E, Byass P, Ouedraogo TW, Tamini C, Gbangou A, Sombie I, et al.: Revealing the burden of maternal mortality: a probabilistic model for determining pregnancy-related causes of death from verbal autopsies. Popul Health Metr 2007, 5: 1. 10.1186/1478-7954-5-1View ArticlePubMedPubMed CentralGoogle Scholar
- Bayes T: An essay towards solving a problem in the doctrine of chances. 1763. MD Comput 1991, 8: 157-171.PubMedGoogle Scholar
- Byass P, Fottrell E, Dao LH, Berhane Y, Corrah T, Kahn K, et al.: Refining a probabilistic model for interpreting verbal autopsy data. Scand J Public Health 2006, 34: 26-31. 10.1080/14034940510032202View ArticlePubMedPubMed CentralGoogle Scholar
- Lewycka S, Mwansambo C, Kazembe PN, Phiri T, Mganga A, Rosato M, et al.: A cluster randomised controlled trial of the community effectiveness of two intervantions in rural Malawi to improve health care and to reduce maternal, newborn and infant mortality. Trials 2010., 11: Google Scholar
- Verbal Autopsy Standards: Ascertaining and Attributing Cause of Death. In WHO. France, WHO; 2007. Ref Type: ReportGoogle Scholar
- Edmond KM, Quigley MA, Zandoh C, Danso S, Hurt C, Owusu AS, et al.: Diagnostic accuracy of verbal autopsies in ascertaining the causes of stillbirths and neonatal deaths in rural Ghana. Paediatr Perinat Epidemiol 2008, 22: 417-429. 10.1111/j.1365-3016.2008.00962.xView ArticlePubMedGoogle Scholar
- Manandhar DS, Osrin D, Shrestha BP, Mesko N, Morrison J, Tumbahangphe KM, et al.: Effect of a participatory intervention with women's groups on birth outcomes in Nepal: cluster-randomised controlled trial. Lancet 2004, 364: 970-979. 10.1016/S0140-6736(04)17021-9View ArticlePubMedGoogle Scholar
- Munjanja SP: Ministry of Health and Child Welfare, Zimbabwe: Maternal and perinatal mortality study 2007.2009. [http://www.unicef.org/zimbabwe/ZMPMS_report.pdf]Google Scholar
- Fottrell E, Byass P: Verbal autopsy: methods in transition. Epidemiol Rev 2010, 32: 38-55. 10.1093/epirev/mxq003View ArticlePubMedGoogle Scholar
- Lawn JE, Cousens S, Zupan J: 4 million neonatal deaths: when? Where? Why? Lancet 2005, 365: 891-900. 10.1016/S0140-6736(05)71048-5View ArticlePubMedGoogle Scholar
- Indepth: International network of field sites with continuous demographic evaluation of populations and their health in developing countries.2010. [http://www.indepth-network.org]Google Scholar
- Kinyanjui S, Timaeus IM: The international network for the demographic evaluation of populations and their health (INDEPTH), the importance of core support. SIDA, Stockholm; 2010.Google Scholar
- Coldham C, Ross D, Quigley M, Segura Z, Chandramohan D: Prospective validation of a standardized questionnaire for estimating childhood mortality and morbidity due to pneumonia and diarrhoea. Trop Med Int Health 2000, 5: 134-144. 10.1046/j.1365-3156.2000.00505.xView ArticlePubMedGoogle Scholar
- Lee AC, Mullany LC, Tielsch JM, Katz J, Khatry SK, LeClerq SC, et al.: Verbal autopsy methods to ascertain birth asphyxia deaths in a community-based setting in southern Nepal. Pediatrics 2008, 121: e1372-e1380. 10.1542/peds.2007-2644View ArticlePubMedPubMed CentralGoogle Scholar
- Cousens S, Blencowe H, Stanton C, Chou D, Ahmed S, Steinhardt L, et al.: National, regional, and worldwide estimates of stillbirth rates in 2009 with trends since 1995: a systematic analysis. Lancet 2011,377(9774):1319-1330. 10.1016/S0140-6736(10)62310-0View ArticlePubMedGoogle Scholar
- Mwale MW: Infant and Child Mortality. In Malawi Demographic Health and Survey. Edited by: National Statistic Office M, Macro ORC. Calverton, Maryland: National Statistic Office, Malawi; ORC Macro; 2005:123-132.Google Scholar
- Ministry of Health and Population KN: Infant and Child Mortality. In Nepal Demographic Health Survey, 2006. Katmandu, Nepal: New ERA and Macro Intternational; 2007:123-130.Google Scholar
- Central Statistical Office: Early Childhood Mortality. In Zimbabwe Demographic Health Surevey 2005-2006. Calverton Maryland: CSO and Macro International Inc.; 2007:109-117.Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.