No cry at birth: global estimates of intrapartum stillbirths and intrapartum-related neonatal deaths

Lawn, Joy; Shibuya, Kenji; Stein, Claudia

RESEARCH

No cry at birth: global estimates of intrapartum stillbirths and intrapartum-related neonatal deaths

Enfants mort-nés : estimation de la mortinatalité per-partum et du nombre de décès néonataux liés à des problèmes per-partum dans le monde

Silencio en el parto: estimaciones mundiales de la mortalidad intraparto y de las defunciones neonatales relacionadas con el parto

Joy Lawn^I,1; Kenji Shibuya^II; Claudia Stein^III

^ISaving Newborn Lives/Save the Children, International Perinatal Care Unit, Institute of Child Health, London, England
^IIMeasurement and Health Information Systems, World Health Organization, Geneva, Switzerland
^IIIHuman Resources for Health, World Health Organization, Geneva, Switzerland

ABSTRACT

OBJECTIVE: Fewer than 3% of 4 million annual neonatal deaths occur in countries with reliable vital registration (VR) data. Global estimates for asphyxia-related neonatal deaths vary from 0.7 to 1.2 million. Estimates for intrapartum stillbirths are not available. We aimed to estimate the numbers of intrapartum-related neonatal deaths and intrapartum stillbirths in the year 2000.
METHODS: Sources of data on neonatal death included: vital registration (VR) data on neonatal death from countries with full (> 90%) VR coverage (48 countries, n = 97 297); studies identified through literature searches (> 4000 abstracts) and meeting inclusion criteria (46 populations, 30 countries, n = 12 355). A regression model was fitted to cause-specific proportionate mortality data from VR and the literature. Predicted cause-specific proportions were applied to the number of neonatal deaths by country, and summed to a global total. Intrapartum stillbirths were estimated using median cause-specific mortality rate by country (73 populations, 52 countries, n = 46 779) or the subregional median in the absence of country data.
FINDINGS: Intrapartum-related neonatal deaths were estimated at 0.904 million (uncertainty 0.651.17), equivalent to 23% of the global total of 4 million neonatal deaths. Country-level model predictions compared well with population-based data sets not included in the input data. An estimated 1.02 million intrapartum stillbirths (0.661.48 million) occur annually, comprising 26% of global stillbirths.
CONCLUSION: Intrapartum-related neonatal deaths account for almost 10% of deaths in children aged under 5 years. Intrapartum stillbirths are a huge and invisible problem, but are potentially preventable. Programmatic attention and improved information are required.

Keywords: Infant mortality; Pregnancy outcome; Labor complications/epidemiology; Asphyxia neonatorum/epidemiology; Hypoxia-ischemia, Brain/epidemiology; Cause of death; Infant, Newborn; Infant, Premature; Regression analysis; Forecasting (source: MeSH, NLM).

RÉSUMÉ

OBJECTIF: Moins de 3% des 4 millions de décès néonataux annuels se produisent dans des pays disposant de données d'état civil fiables. Les estimations pour l'ensemble du monde du nombre de décès néonataux liés à une asphyxie vont de 0,7 à 1,2 million. On ne dispose pas d'estimations de la mortinatalité per-partum. L'objectif de l'étude est d'évaluer le nombre de décès per-partum et de décès néonataux liés à des problèmes per-partum survenus en l'an 2000.
MÉTHODES: L'étude a notamment utilisé les sources de données relatives aux décès néonataux suivantes: statistiques d'état civil indiquant des décès néonataux fournies par des pays totalement couverts (couverture > 90 %) par des registres d'état civil (48 pays, n = 97 297), études sélectionnées par des recherches bibliographiques (> 4000 sommaires) et remplissant les critères de prise en compte (46 populations, 30 pays, n = 12 355). Un modèle de régression a été adapté pour établir les proportions par causes des décès signalés par les statistiques d'état civil et par les données de mortalité tirées de la littérature. Les proportions par causes prévues ont été appliquées au nombre de décès néonataux par pays et les résultats de ces opérations ont été ajoutés pour obtenir un total mondial. La mortinatalité per-partum a été estimée à partir des taux de mortalité par cause médians dans chaque pays (73 populations, 52 pays, n = 46 779) ou à partir de la médiane subrégionale en l'absence de données nationales.
RÉSULTATS: Le nombre de décès néonataux liés à des problèmes per-partum a été évalué à 0,904 million (intervalle d'incertitude : 0,65 - 1,17 million), ce qui correspond à 23 % de la mortalité néonatale mondiale totale (4 millions de décès). Les prédictions par pays fournies par le modèle sont en bon accord avec des jeux de données en population non intégrés aux données d'entrée du modèle. La mortinatalité per-partum mondiale annuelle est estimée à 1,02 millions de décès (intervalle d'incertitude: 0,66 - 1,48 million), soit 26 % de la mortinatalité mondiale.
CONCLUSION: Les décès néonataux liés à des problèmes per-partum représentent près de 10 % des décès chez les enfants de moins de 5 ans. La mortinatalité per-partum constitue un problème énorme et peu visible, dont la prévention est cependant possible. Une telle prévention nécessiterait que les responsables de programmes s'intéressent à ce problème et que les décideurs soient mieux informés.

Mots clés: Mortalité nourrisson; Issue grossesse; Accouchement compliqué/épidémiologie; Asphyxie néonatale/épidémiologie; Hypoxie-ischémie cérébrale/épidémiologie; Cause décès; Nouveau-né; Prématuré; Analyse régression; Prévision (source: MeSH, INSERM).

RESUMEN

OBJETIVO: Menos del 3% de los 4 millones de defunciones neonatales anuales se producen en países con datos de registro civil (RC) fiables. Las estimaciones mundiales de las defunciones neonatales relacionadas con problemas de asfixia oscilan entre 0,7 y 1,2 millones. No se dispone de estimaciones sobre la mortalidad intraparto. Decidimos estimar el número de muertes neonatales relacionadas con el parto y el número de nacidos muertos en el año 2000.
MÉTODOS: Las fuentes de los datos sobre las defunciones neonatales fueron las siguientes: datos del registro civil (RC) sobre las muertes neonatales en países con cobertura plena ( 90%) de RC (48 países, n = 97 297); estudios identificados mediante búsquedas en la literatura (> 4000 resúmenes), y cumplimiento de los criterios de inclusión (46 poblaciones, 30 países, n = 12 355). Se estableció un modelo de regresión para los datos de mortalidad proporcional por causas específicas extraídos del RC y de la literatura. Las proporciones proyectadas por causas específicas se aplicaron al número de defunciones neonatales por país, y finalmente éstas se sumaron para obtener un total mundial. La mortalidad intraparto se estimó utilizando la mediana de la tasa de mortalidad por causas por país (73 poblaciones, 52 países, n = 46 779), o la mediana subregional a falta de datos del país.
RESULTADOS: Las defunciones neonatales relacionadas con el parto se estimaron en 0,904 millones (incertidumbre: 0,651,17), lo que equivale al 23% del total mundial de 4 millones de defunciones neonatales. Las predicciones del modelo a nivel de país fueron razonablemente coherentes con conjuntos de datos basados en la población no incluidos en el input utilizado. Cada año se registran 1,02 millones de mortinatos intraparto (0,661,48 millones), lo que supone el 26% de la mortinatalidad mundial.
CONCLUSIÓN: Las defunciones neonatales relacionadas con el parto representan casi el 10% de las defunciones de menores de 5 años. La mortalidad intraparto es un inmenso problema oculto, pero es potencialmente prevenible. Se requiere atención programática y una mejor información.

Palabras clave: Mortalidad infantil; Resultado del embarazo; Complicaciones del trabajo de parto/epidemiología; Asfixia neonatal/epidemiología; Hipoxia-isquemia del cerebro/epidemiología; Causa de muerte; Recién nacido; Prematuro; Análisis de regresión; Predicción (fuente: DeCS, BIREME).

Background

The Millennium Development Goals (MDGs) and The world health report 2005 are fuelling attention to maternal and child health (1). Improved information is required at the global and programme levels to prioritize use of resources to achieve the maximum effect. Almost 40% of deaths in children aged under 5 years occur in the neonatal period and this proportion is increasing as the numbers of post-neonatal deaths fall more rapidly (2). Yet neonatal deaths receive limited attention partly due to the lack of robust estimates of cause of death (3). To meet MDG-4, which calls for mortality in under-5-year-olds to be reduced by two-thirds, more rapid reduction is required particularly in the risk of early neonatal death (death in the first week of life), which has shown the least decline (4). Birth asphyxia is a major cause of early neonatal deaths. Although the estimated numbers of disability-adjusted life years (DALYs) for birth asphyxia exceed those due to all childhood conditions preventable by immunizations (5), birth asphyxia does not feature on most lists of childhood "killers" and is not a policy or funding priority. Associated stillbirths (late fetal deaths after 28 weeks of gestation) are virtually invisible at policy level.

Tension between the increasing demand for data for decision-making, and the reality of health information gaps has been described as a "gathering storm" (6). Information regarding the cause of death for more than 97% of neonatal deaths is scanty in countries without full coverage of vital registration (VR). The majority of the world's stillbirths and neonatal deaths occur where no information is available to guide programmes. Thus, modelling and other estimation approaches are necessary while working to improve coverage and quality of new data. Systematic global estimates for intrapartum stillbirths babies who die during labour have never been published. Reported global totals of neonatal deaths due to the non-specific condition of birth asphyxia vary from 0.7 million (7) to 1.6 million (8) although the data inputs and methods for obtaining these estimates are not available.

Clarifying the language relating to birth asphyxia is necessary for improved measurement, especially if the deaths counted are to be relevant to programme action. Previous estimates have referred to the nonspecific condition of birth asphyxia, or not breathing at birth, which has several causes, including preterm birth, although historically the term birth asphyxia has implied a causal link with intrapartum hypoxia. There is no gold standard test for birth asphyxia fetal distress, acidaemia, Apgar scores and other clinical markers of the process of potential intrapartum injury have low positive predictive values (9). Furthermore, such measurements are not feasible for many of the 99% of neonatal deaths occurring in low- and middle-income countries because half of these deliveries take place without a skilled attendant and a minority has access to assessment of acidbase status.

Epidemiological measurement of intrapartum injury has moved from process-based (e.g. long labour) and symptom-based (e.g. Apgar score) definitions to multiple indicator outcomes particularly neonatal encephalopathy, which refers to an abnormal neurobehavioural state in the first few days of life and is most commonly related to intrapartum insult (10). Such outcomes are more feasible to measure consistently and have direct programme relevance (11, 12). Recent developments in early cooling therapy for babies with neonatal encephalopathy are potentially applicable more widely and require early and specific identification of babies with acute intrapartum brain injury (13). If babies with extreme preterm birth or congenital malformations continue to be misclassified as having intrapartum asphyxia, aside from the issue of litigation in rich countries, expected population-level programmatic solutions may be based on misinformation, as different interventions are required to prevent deaths due to these other causes. Improved global estimates necessitate tighter case definitions; detailed data inputs with explicit inclusion criteria; methods and assumptions described; and provision of associated uncertainty estimates.

Objective and case definitions

The objective of the present study was to provide estimates for 192 countries around the year 2000 for the following mortality outcomes:

Neonatal deaths related to intrapartum events, including neonatal deaths resulting from neonatal encephalopathy, neonates born at term who could not be resuscitated (or for whom resuscitation was not available) or specific birth trauma. Where possible, other causes such as lethal congenital malformations and extreme preterm birth (less than 34 completed weeks of gestation (or birth weight < 1500 g) should have been excluded.

Stillbirths occurring intrapartum or fresh stillbirths (skin still intact, implying death less than 12 hours before delivery), weighing more than 1000 g or after more than 28 weeks of gestation, but excluding those with severe lethal congenital abnormalities.

Methods and data sources

Fig. 1 provides an overview of the methods applied. For the proportion of neonatal deaths related to intrapartum events if full coverage (> 90%) VR data were available with data on cause of neonatal death, we analysed and used these data. For countries where full coverage VR data were not available, a regression (logit) model based on VR and study data was fitted, and the predicted proportion applied to the WHO estimates of number of neonatal deaths by country. The cause-specific rate of stillbirths occurring intrapartum by country was estimated from the median rate using studies that met the inclusion criteria.

Search Strategy

Systematic searches were performed of the MEDLINE, POPLINE, Latin American and Caribbean Health Sciences (LILACS), BioMed Central, African Index Medicus and WHO Regional Office for the Eastern Mediterranean (EMRO) databases. Searches were conducted, without restrictions regarding language, on publications since 1985 for various terms, including all-cause mortality terms (e.g. neonatal/perinatal mortality, stillbirths and fetal deaths) and cause-specific terms related to acute intrapartum events (e.g. birth asphyxia, hypoxic ischaemic encephalopathy, neonatal encephalopathy, birth trauma, fresh stillbirths and intrapartum stillbirths). Extensive attempts were made to identify unpublished databases. Over 4000 documents of potential relevance were identified through these search techniques for neonatal deaths, and almost 14 000 for stillbirths (Fig. 1). After screening the abstracts, the selected publications were examined in detail to determine whether they met the inclusion criteria using two screening filters (Table 1).

Neonatal deaths related to acute intrapartum events

The case definition is given above.

Vital registration data

Original analysis was performed on the data for all neonatal deaths from 83 countries that had recorded VR data within the last 10 years, as reported to WHO up to March 2004 (Personal communication, D. Ma Fat, Evidence for Information and Policy Cluster, WHO, June 2003). Data from 48 countries met the inclusion criteria, with the median year being 2000 (Table 1). The proportion of intrapartum-related deaths was derived from analysis using the relevant International Classification of Diseases (ICD)-10 or ICD-9 codes relevant to the case definition, and analysed in Stata version 8 (Stata Corporation, College Station, Texas, USA). For countries with suitable VR data (48 countries) the proportion of intrapartum-related neonatal deaths was used. For countries with less than 500 neonatal deaths per year, a weighted average proportion of the most recent 3 years was used.

Model-based estimates

For the remaining 145 countries, a random effects model was used to predict the proportion of neonatal deaths related to intrapartum events. The dependent variable was the logit of the proportion of intrapartum-related neonatal mortality from VR (48 countries) and from 46 published and unpublished studies that met the inclusion criteria specified (Table 1). A variety of potential independent variables, using national-level data for the year 2000, were tested for fit. These were mortality rate in under-5-year-olds; neonatal mortality rate (NMR); gross domestic product; health expenditures; dummy variable for type of data (VR or others); WHO subregions and mortality levels; and coverage of interventions such as vaccination, skilled birth attendance and antenatal care. These data were obtained from the World Bank, WHO, and the United Nations Children's Fund (UNICEF), and refer to the year 2000. A random effects model was fitted using a parsimonious approach, adding predicators if they reached significance at a level of 5%. The final model was used to predict the proportion of neonatal deaths related to intrapartum events for the 145 countries without VR data. The proportion derived was then applied to the number of neonatal deaths in the country according to WHO estimates to produce an estimate of the number of intrapartum-related neonatal deaths. External validity of the estimates was examined by comparing model predictions to unpublished, population-based data sets.

Intrapartum stillbirths

The case definition applied is detailed above. A "fresh stillbirth" is a baby born dead without signs of skin disintegration or maceration and the death is assumed to have occurred < 12 hours prior to delivery (12, 14). For obstetric classifications such as Aberdeen, acute intrapartum events causing death such as antepartum haemorrhage and obstructed labour were considered equivalent.

Studies meeting the inclusion criteria specified were abstracted (Table 1). If more than one study was included per country, then the median rate was applied. The median was selected in preference to the mean in keeping with approaches used for estimations in the Child Health Epidemiology Reference Group (15). If no data were available for a country then the median cause-specific rate for the subregion was used. The correct denominator for the rate of stillbirths is total number of births. Estimates of rates of stillbirth are not currently available by country and therefore total numbers of births at country level are unknown. Hence the derived cause-specific rates were applied to the estimated live births for each country for the year 2000 (UN Population Division, 2001 revision (16)) and summed to a global total which was then corrected to allow for the missing stillbirths by multiplying by 1.03, based on a global stillbirth rate of 30 per 1000 total births previously estimated by WHO (7).

Uncertainty analysis

In countries with full VR coverage, 95% uncertainty levels were derived from the reported data. For modelled estimates, uncertainty bounds were generated using the standard error of the prediction of the logit and running 10 000 Monte Carlo simulations. For stillbirths, the upper and lower bounds of uncertainty for each subregion were taken as the highest and lowest rates for intrapartum stillbirth in the data entered for that subregion. These methods did not take into account uncertainty in the birth cohort or in the WHO estimates for neonatal deaths by country.

Results

Neonatal deaths related to acute intrapartum events

A total of 46 study populations from 30 countries met the inclusion criteria, with a cumulative sample size of 12 355 neonatal deaths (Table A.2, web version only, available at http//www.who.int/bulletin). VR data from 48 countries were included (97 297 neonatal deaths). The data entered (Fig. 2) suggest that in countries with a lower NMR (< 15 per 1000 live births), the proportion of neonatal deaths related to acute intrapartum events is mainly between 10% and 20%. For NMRs between 15 and 30 per 1000, intrapartum events are reported to cause a higher proportion of neonatal deaths, ranging from 23% to 37%. However at higher NMRs (above 30 per 1000 live births), the cause-specific proportion falls to 1525% of neonatal deaths.

The final regression model applied to predict the proportion of neonatal deaths related to acute intrapartum events was (standard errors in parentheses):

Where lnq5 is the natural logarithm of the national risk of dying between birth and 5 years of age, lnGDP is the natural logarithm of gross domestic product in purchasing power parity, logit (%DPT₃) is the logit of national coverage of immunization with three doses of diphtheria, pertussis and tetanus toxoid immunization, and data type is a dummy variable for data input type (VR or literature). The goodness-of-fit was satisfactory, as reflected by R-square (0.61). There was no systematic deviation among the residuals. National data for coverage of emergency obstetric care were not available. Other covariates such as antenatal care were not found to be statistically significant.

The predicted proportion of neonatal death due to intrapartum events in each country was then applied to the WHO estimations of neonatal deaths by country. The model results, combined with full coverage VR data for 48 countries, predicted a global total of 0.904 million (0.651.17 million) deaths. Table 2 compares model predictions with unpublished population-based data from four countries with high mortality levels. These data sets met the inclusion criteria (Table 1), but were not included in modelling. A paired t-test did not detect a statistically significant difference between observed and predicted proportions (P = 0.76). The average absolute difference between observed and predicted proportionate mortality was only 4%.

Intrapartum stillbirths

Studies meeting inclusion criteria included 73 populations in 52 countries with a cumulative sample size of 46 779 stillbirths (Table A.1, web version only, available at: http//www.who.int/bulletin). Approximately 84% were reported in peer-reviewed papers, four were national reports and two were unpublished data sets. Table 3 summarizes the estimates of rates of intrapartum stillbirth by subregion, summing to a global total of 1.02 million (uncertainty 0.661.48 million). The incidence varied markedly, with a rate of less than 1 per 1000 total births in the richest regions, to more than 15 times higher in the Eastern Mediterranean regions, South Asia and sub-Saharan Africa. The disparity reached 50-fold when comparing the lowest (0.33 per 1000 total births) (17) and highest (17.4 per 1000) (18) reported rates.

Discussion

The figures presented here are the first systematic global estimates quantifying intrapartum stillbirths, and the first estimates of intrapartum-related neonatal deaths using a more specific case definition. We estimated that 0.904 million neonatal deaths were associated with acute intrapartum events, approximately 23% of all neonatal deaths globally, and an additional 1.02 million intrapartum stillbirths accounted for 26% of 3.9 million stillbirths. This total of close to 2 million deaths occurring immediately around the time of birth is largely invisible in terms of the priorities of either safe motherhood or child survival programmes. Improving health systems at the time of childbirth potentially offers a triple benefit: reducing not only neonatal mortality and intrapartum stillbirths, but also averting many of the estimated 0.5 million maternal deaths a year.

Extensive efforts were made to systematically identify and use the best information available. Comparison of the model predictions with unpublished data sets (Table 2) showed reasonable agreement in four countries. However, only 48 countries had suitable VR data. Stillbirth data are not routinely collected by WHO. Given the numerous approaches to classification of perinatal deaths and variable case definitions, the data are intrinsically uncertain, even within Europe (19). Improving the quantity, quality and comparability of data collected in the future is the only way to reduce this uncertainty. Trend analysis with previous global estimates of neonatal deaths due to "birth asphyxia" should be made with caution, because the case definitions here are more specific (9).

Neonatal deaths related to acute intrapartum events

There are several potential sources of bias in the proportion of neonatal deaths related to acute intrapartum events. Early neonatal deaths may be misregistered as stillbirths (20). Misclassification of cause of death occurs, and is more likely with simpler verbal autopsy tools using unspecified algorithms, particularly in distinguishing between intrapartum causes and prematurity, or undetected congenital abnormalities. Furthermore, comorbidity may be complex; for example, infection and acute intrapartum events are synergistic and the direct cause of death may be debatable (21). In litigious societies, there may be systematic avoidance of the diagnosis of "birth asphyxia" on death certificates. Despite higher NMRs, the subregional average percentage of neonatal deaths related to intrapartum events is lower in North America (12%) than in Australia and Europe (14% and 15%, respectively) (Table 2). Industrialized countries tend to restrict the definition of intrapartum-related neonatal deaths to deaths subsequent to neonatal encephalopathy with multi-organ dysfunction, documented acidosis and proven intrapartum insults (9). The use of this definition is not feasible in developing countries and omits an important group, namely, those infants who die immediately after birth if resuscitation is unavailable or fails. Relevant and practical case definitions require validation especially within verbal autopsy tools, as this is the major source of cause-specific mortality information on most of the world's neonatal deaths.

There are several plausible explanations for the variation in the proportion of intrapartum-related neonatal deaths with NMR (Fig. 2). In countries with high NMRs, the lower proportion of neonatal deaths may be due to a higher proportion of deaths being caused by infections and tetanus. Additionally, in the settings with the highest mortality rates, access to obstetric care is limited, and acute intrapartum events more frequently result in stillbirth than in neonatal death. It is also possible that the measurement tools used in the settings where mortality is highest, particularly the simplest verbal autopsy tools, underestimate intrapartum-related neonatal deaths, misclassifying these as stillbirths or as due to other causes of death such as infection. The higher proportion of intrapartum-related deaths reported in transitional countries may be a real result, perhaps because infectious causes are less frequent in the countries shown (China and Latin America), leaving intrapartum causes as leading to a high proportion of deaths or there may be systematic misclassification of cause of neonatal death, for example misclassifying preterm birth as intrapartum-related death if Apgar-based definitions are still used. Population-based studies using validated measurement tools are necessary to improve understanding of this apparent variation, and to address any avoidable factors.

In the final model, the strength of association of the outcome with coverage of vaccination with triple doses of diphtheria, pertussis and tetanus, which is considered a good marker for health system functioning, may suggest that improving coverage by skilled birth attendants alone is insufficient to address intrapartum complications. Coverage of emergency obstetric care could not be tested for fit due to lack of comparable national data. Models able to simultaneously predict a number of causes of neonatal death would be an alternative approach (2, 16).

Intrapartum stillbirths

Stillbirths are invisible in many societies, except for the estimated 4 million women who lose their baby in the last 3 months of pregnancy (7). Intrapartum stillbirths are more frequent than intrapartum-related neonatal deaths, especially in settings with limited emergency obstetric care (22). A mature fetus dying during childbirth is usually defined as a preventable death (2325). Hospital-based studies suggest that 2562% of intrapartum stillbirths could be avoided with better obstetric care and more rapid responses to intrapartum complications, including reducing delays in recognition of complications at home, and reducing delays in transportation to hospital (2527).

Fresh stillbirth is a proxy for stillbirth due to acute intrapartum insult, but probably underestimates the true rate of stillbirths due to acute intrapartum events because a duration of labour in excess of 12 hours usually results in a macerated stillbirth and in settings with poor access to obstetric care, labour lasting more than 24 hours is not infrequent (28). Conversely, a small proportion of fresh stillbirths may be due to non-hypoxic causes (undetected congenital abnormalities or severe infection). The extent of such misclassification biases cannot be quantified without further study.

Conclusion

Intrapartum stillbirths and intrapartum-related neonatal deaths are a huge and potentially preventable burden, yet receive limited policy and programmatic attention. Intrapartum-related neonatal deaths account for 9% of all deaths in children aged under 5 years, a proportion comparable to the estimated number of children who die from malaria. The risks of intrapartum-related stillbirths and neonatal deaths are much higher in poor countries, yet the coverage of care is lower, a modern example of the inverse care law (2, 29). Innovative approaches are required to increase information for decision-making and improve care in settings where far too many babies do not cry at birth.

Acknowledgements

We are grateful to Gary Darmstadt for input on the scope of the study and for reviewing the manuscript, and Anne Tinker (Save the Children Federation, USA) for support. We thank Simon Cousens for assistance in analysing the VR data and the Child Health Epidemiology Reference Group based in Child and Adolescent Health at WHO for critical review of the methods. We appreciate the help given by Kate Wilczynska-Ketende with literature searches. We thank Doris Ma Fat, Colin Mathers and Mie Inoue for providing data related to the WHO mortality database. We acknowledge the following for supplying unpublished data sets: Adult Morbidity and Mortality Project, Ministry of Health, United Republic of Tanzania, particularly Philip Setel, Dave Whiting and Yusuf Hemed; Ana Maria Aguilar (BASICS II, Bolivia); Zulfiqar Bhutta (Aga Khan University, Karachi, Pakistan); Fariyal Fikree (Population Council, New York, USA); Professor Henry Perry (Future Generations, Franklin West, Virginia, USA, formerly of Hopital Albert Schweitzer, Haiti); Gils Walraven (Aga Khan Foundation, Paris, formerly of Farafenni Field Site, Gambia); and David Woods (University of Cape Town, South Africa).

Funding: Joy Lawn was supported by the Bill & Melinda Gates Foundation through a grant to Save the Children/USA for the Saving Newborn Lives initiative.

Competing interests: none declared.

References

1. World Health Organization. The world health report 2005. Making every mother and child count. Geneva: World Health Organization; 2005.

2. Lawn JE, Cousens S, Zupan J. Four million neonatal deaths: Where? When? Why? Neonatal Survival Series Paper 1. Lancet 2005;365:891-900. Available at URL: http://image.thelancet.com/extras/05art1073web.pdf

3. Black RE, Morris SS, Bryce J. Where and why are 10 million children dying every year? Lancet 2003;36:2226-34.

4. Darmstadt GL, Lawn JE, Costello A. Advancing the state of the world's newborns. Bulletin of the World Health Organization 2003;81:224-5.

5. World Health Organization. The World Health Report 2003. Shaping the future. Geneva: World Health Organization; 2003. Available at: URL: http://www.who.int/whr/en/

6. Evans T, Stansfield S. Health information in the new millennium: a gathering storm? Bulletin of the World Health Organization 2003;81:856.

7. Save the Children. WHO estimates published in Saving Newborn Lives. State of the world's newborns. Washington DC: Save the Children; 2001; pp.1-44. Available at: URL: http://www.savethechildren.org/publications/newborns_report.pdf

8. World Health Organization. Global burden of disease, 2000. version 2. Geneva: World Health Organization; 2003. Available at: URL: http://www3.who.int/whosis/menu.cfm? path=evidence,burden,burden_estimates,burden_estimates_2000V2,burden_estimates_ 2000V2_subregion&language=English

9. Anon. Use and abuse of the Apgar score. Committee on Fetus and Newborn, American Academy of Pediatrics, and Committee on Obstetric Practice, American College of Obstetricians and Gynecologists. Pediatrics 1996;98:141-2.

10. Cowan F, Rutherford M, Groenendaal F, Eken P, Mercuri E, Bydder GM, et al. Origin and timing of brain lesions in term infants with neonatal encephalopathy. Lancet 2003;361:736-42.

11. Hey EN, Lloyd DJ, Wigglesworth JS. Classifying perinatal death: fetal and neonatal factors. British Journal of Obstetrics and Gynaecology 1986;93:1213-23.

12. Wigglesworth JS. Monitoring perinatal mortality a pathophysiological approach. Lancet 1980;2:684-6.

13. Gluckman PD, Wyatt JS, Azzopardi D, Ballard R, Edwards AD, Ferriero DM, et al. Selective head cooling with mild systemic hypothermia after neonatal encephalopathy: multicentre randomised trial. Lancet 2005;365:663-70.

14. Amar HS, Maimunah AH, Wong SL. Use of Wigglesworth pathophysiological classification for perinatal mortality in Malaysia. Archives of Diseases in Childhood Fetal Neonatal Edition 1996;74:F56-9.

15. Bryce J, Boschi-Pinto C, Shibuya K, Black RE. WHO Child Health Epidemiology Reference Group. WHO estimates of the causes of death in children. Lancet 2005;365:1147-52.

16. United Nations Population Division. World population prospects the 2002 revision. New York; United Nations: 2003.

17. Holt J, Vold IN, Odland JO, Forde OH. Perinatal deaths in a Norwegian county 1986-96 classified by the Nordic-Baltic perinatal classification: geographical contrasts as a basis for quality assessment. Acta Obstetricia et Gynecologica Scandinavica 2000;79:107-12.

18. Kilonzo A, Kouletio M, Whitehead SJ, Curtis KM, McCarthy BJ. Improving surveillance for maternal and perinatal health in 2 districts of rural Tanzania. American Journal of Public Health 2001;91:1636-40.

19. Buitendijk S, Zeitlin J, Cuttini M, Langhoff-Roos J, Bottu J. Indicators of fetal and infant health outcomes. European Journal of Obstetrics & Gynecology and Reproductive Biology 2003;111 Suppl 1:S66-77.

20. Kramer MS, Liu S, Luo Z, Yuan H, Platt RW, Joseph KS. Analysis of perinatal mortality and its components: time for a change? American Journal of Epidemiology 2002;156:493-7.

21. Peebles DM, Wyatt JS. Synergy between antenatal exposure to infection and intrapartum events in causation of perinatal brain injury at term. British Journal of Obstetrics and Gynaecology 2002;109:737-9.

22. Ellis M, Manandhar DS, Manandhar N, Wyatt J, Bolam AJ, Costello AM. Stillbirths and neonatal encephalopathy in Kathmandu, Nepal: an estimate of the contribution of birth asphyxia to perinatal mortality in a low-income urban population. Paediatric Perinatal Epidemiology 2000;14:39-52.

23. Maternal and Child health Research Consortium L. Confidential enquiry into stillbirths and deaths in infancy (CESDI). Eighth Annual report. 2000. Available at: URL: http://www.cemach.org.uk/publications.htm

24. Bugalho A, Bergstrom S. Value of perinatal audit in obstetric care in the developing world: a ten-year experience of the Maputo model. Gynecologic Obstetric Investigation 1993;36:239-43.

25. Wilkinson D. Avoidable perinatal deaths in a rural hospital: strategies to improve quality of care. Tropical Doctor 1995;25:16-20.

26. De Muylder X. Perinatal mortality audit in a Zimbabwean district. Paediatric Perinatal Epidemiology 1989;3:284-93.

27. Buchmann EJ, Pattinson RC, Nyathikazi N. Intrapartum-related birth asphyxia in South Africa lessons from the first national perinatal care survey. South African Medical Journal 2002;92:897-901.

28. Kusiako T, Ronsmans C, Van der Paal L. Perinatal mortality attributable to complications of childbirth in Matlab, Bangladesh. Bulletin of the World Health Organization 2000;78:621-7.

29. Hart JT. The inverse care law. Lancet 1971;1:405-12.

(Submitted: 4 May 2004 Final revised version received: 31 March 2005 Accepted: 4 April 2005)

1 Correspondence should be sent to this author at 11 South Way, Pinelands, Cape Town 7405, South Africa (email: joylawn@yahoo.co.uk).

Appendix

References

1. Wessel H, Cnattingius S, Dupret A, Reitmaier P, Bergstrom S. Risk factors for perinatal death in Cape Verde. Paediatric Perinatal Epidemiology 1998; 12:25-36.

2. Dujardin B, De Schampheleire I, Sene H, Ndiaye F. Value of the alert and action lines on the partogram. Lancet 1992;339:1336-8.

3. Chalumeau M, Salanave B, Bouvier-Colle MH, De Bernis L, Prual A, Breart G. Risk factors for perinatal mortality in West Africa: a population-based study of 20 326 pregnancies. MOMA group. Acta Paediatrica 2000; 89:1115-21.

4. Chalumeau M, Bouvier-Colle MH, Breart G. Can clinical risk factors for late stillbirth in West Africa be detected during antenatal care or only during labour? International of Journal Epidemiology 2002;31:661-8.

5. Temmerman M, Plummer FA, Mirza NB, Ndinya-Achola JO, Wamola IA, Nagelkerke N, et al. Infection with HIV as a risk factor for adverse obstetrical outcome. AIDS 1990;4:1087-93.

6. Kulmala T, Vaahtera M, Ndekha M, Koivisto AM, Cullinan T, Salin ML, et al. The importance of preterm births for peri- and neonatal mortality in rural Malawi. Paediatric and Perinatal Epidemiology 2000;14:219-26.

7. Kulmala T, Vaahtera M, Rannikko J, Ndekha M, Cullinan T, Salin ML, et al. The relationship between antenatal risk characteristics, place of delivery and adverse delivery outcome in rural Malawi. Acta Obstetricia et Gynecologica Scandinavica 2000;79:984-90.

8. Axemo P, Liljestrand J, Bergstrom S, Gebre-Medhin M. Aetiology of late fetal death in Maputo. Gynecologic and Obstetric Investigation 1995;39:103-9.

9. Osman NB, Challis K, Cotiro M, Nordahl G, Bergstrom S. Perinatal outcome in an obstetric cohort of Mozambican women. Journal of Tropical Pediatrics 2001;47:30-38.

10. Dommisse J. The causes of perinatal deaths in the greater Cape Town area. A 12-month survey. South African Medical Journal 1991;80:270-75.

11. Pattinson RC, editor. Saving Babies 2001: 2nd Perinatal Care Survey of South Africa. Pretoria: The MRC Unit for Maternal and Infant Health Care Strategies, PPIP Users, and the National Department of Health; 2004, pp.1-139. Available at: http://www.hst.org.za/uploads/files/saving_babies.pdf

12. Buchmann EJ, Pattinson RC, Nyathikazi N. Intrapartum-related birth asphyxia in South Africa lessons from the first national perinatal care survey. South African Medical Journal 2002;92:897-901.

13. Hinderaker SG, Olsen BE, Bergsjo PB, Gasheka P, Lie RT, Havnen J, et al. Avoidable stillbirths and neonatal deaths in rural Tanzania. British Journal of Obstetrics and Gynaecology 2003;110:616-23.

14. Hinderaker SG, Olsen BE, Bergsjo PB, Gasheka P, Lie RT, Kvale G. Perinatal mortality in northern rural Tanzania. Journal of Health, Population and Nutrition 2003;21:8-17.

15. Kilonzo A, Kouletio M, Whitehead SJ, Curtis KM, McCarthy BJ. Improving surveillance for maternal and perinatal health in 2 districts of rural Tanzania. American Journal of Public Health 2001;91:1636-40.

16. Aiken CG. The causes of perinatal mortality in Bulawayo, Zimbabwe. Central African Medical Journal 1992;38:263-81.

17. De Muylder X. Perinatal mortality audit in a Zimbabwean district. Paediatric Perinatal Epidemiology 1989;3:284-93.

18. Luginaah IN, Lee KS, Abernathy TJ, Sheehan D, Webster G. Trends and variations in perinatal mortality and low birthweight: the contribution of socio-economic factors. Canadian Journal of Public Health 1999;90:377-81.

19. Chek K, Kerr GR. Factors associated with fetal mortality in the triethnic population in Texas, 1993 through 1995. Texas Medicine 1999;95:78-83.

20. Conde-Agudelo A, Belizan JM, Diaz-Rossello JL. Epidemiology of fetal death in Latin America. Acta Obstetricia et Gynecologica Scandinavica 2000;79:371-8.

21. Gadow EC, Castilla EE, Lopez CJ, Queenan JT. Stillbirth rate and associated risk factors among 869 750 Latin American hospital births 1982-1986. International Journal of Gynecology and Obstetrics 1991;35:209-14.

22. Barros FC, Victora CG, Vaughan JP, Estanislau HJ. Perinatal mortality in southern Brazil: a population-based study of 7392 births. Bulletin World Health Organization 1987; 65:95-104.

23. Menezes AM, Barros FC, Victora CG, Alves C, Rocha C, Albernaz E, et al. [Perinatal mortality in two population-based cohorts from southern Brazil: trends and differentials.] Cadernos De Saude Publica [Reports In Public Health] 1996;12(Suppl 1):33-41.

24. Ferraz EM, Gray RH. A case-control study of stillbirths in northeast Brazil. International Journal of Gynecology and Obstetrics 1991;34:13-19.

25. Ashley D, McCaw-Binns A, Golding J, Keeling J, Escoffery C, Coard K, et al. Perinatal mortality survey in Jamaica: aims and methodology. Paediatric and Perinatal Epidemiology 1994;8(Suppl 1):6-16.

26. Ashley D, Greenwood R, McCaw-Binns A, Thomas P, Golding J. Medical conditions present during pregnancy and risk of perinatal death in Jamaica. Paediatric and Perinatal Epidemiology 1994;8(Suppl 1):66-85.

27. Escoffery C, Greenwood R, Ashley D, Coard K, Keeling J, Golding J. Deaths associated with intrapartum asphyxia in Jamaica. Paediatric and Perinatal Epidemiology 1994;8(Suppl 1):119-42.

28. Bassaw B, Roopnarinesingh S, Sirjusingh A. An audit of perinatal mortality. West Indian Medical Journal 2001;50:42-46.

29. Bocaletti E, Schumacher R, Hortado E, Bailey P, Matute J, McDermott J, et al. Perinatal mortality in Guatelmala: Community study. Arlington, VA: BASICS; 1999.

30. el Shafei AM, Sandhu AK, Dhaliwal JK. Perinatal mortality in Bahrain. Australian and New Zealand Journal of Obstetrics and Gynaecology 1988;28:293-8.

31. Abu-Heija AT. Causes and factors affecting perinatal mortality at Princess Basma Teaching Hospital in North Jordan. Asia-Oceania Journal of Obstetrics and Gynaecology 1994;20:415-18.

32. Abu-Ekteish F, Daud AS, Sunna E, Obeidat, Al-Rimawi HS. Perinatal mortality at Princess Badia teaching hospital, northern Jordan. Annals of Saudi Medicine 1997;17:120-3.

33. Kishan J, Soni AL, Elzouki AY, Mir NA, Magoub MR. Perinatal outcome at Benghazi and implications for perinatal care in developing countries. Indian Journal of Pediatrics 1988;55:611-5.

34. Kishan J, Soni AL, Elzouki AY, Mir NA. Perinatal mortality and neonatal survival in Libya. Journal of Tropical Pediatrics 1988;34:32-3.

35. Bittar Z. Rates of perinatal mortality and low birth weight among 3367 consecutive births in south of Beirut. Lebanese Medical Journal 1998; 46:126-30.

36. el Zibdeh MY, Al Suleiman SA, Al Sibai MH. Perinatal mortality at King Fahd Hospital of the University Al-Khobar, Saudi Arabia. International Journal of Gynecology and Obstetrics 1988;26:399-407.

37. Thomassen PA, Langemark L, Kumar R. Stillbirths and antenatal care at a rural district hospital. Annals of Saudi Medicine 1989;9:186-9.

38. Serenius F, Swailem AR, Edressee AW, Ohlsson A. Causes of perinatal death at a Saudi maternity hospital. Acta Paediatrica Scandinavica (Suppl) 1988;346:70-9.

39. Campbell O, Gipson R, el Mohandes A, Issa AH, Matta N, Mansour E, et al. The Egypt National Perinatal/Neonatal Mortality Study 2000. Journal of Perinatology 2004;24:284-9

40. Khan SR, Jalil F, Zaman S, Lindblad BS, Karlberg J. Early child health in Lahore, Pakistan. X. Mortality. Acta Paediatrica (Suppl) 1993;82(Suppl) 390:109-17.

41. Jalil F, Lindblad BS, Hanson LA, Khan SR, Yaqoob M, Karlberg J. Early child health in Lahore, Pakistan. IX. Perinatal events. Acta Paediatrica Suppl 1993;82(Suppl)390:95-107.

42. Mogilevkina I, Bodker B, Orda A, Langhoff-Roos J, Lindmark G. Using the Nordic-Baltic perinatal death classification to assess perinatal care in Ukraine. European Journal of Obstetrics, Gynaecology and Reproductive Biology 2002;100:152-7.

43. Holt J, Vold IN, Odland JO, Forde OH. Perinatal deaths in a Norwegian county 1986-96 classified by the Nordic-Baltic perinatal classification: geographical contrasts as a basis for quality assessment. Acta Obstetricia et Gynecologica Scandinavica 2000;79:107-12.

44. Dahl LB, Berge LN, Dramsdahl H, Vermeer A, Huurnink A, Kaaresen PI, et al. Antenatal, neonatal and post neonatal deaths evaluated by medical audit. A population-based study in northern Norway 1976 to 1997. Acta Obstetricia et Gynecologica Scandinavica 2000;79:1075-82.

45. Petersson K, Bremme K, Bottinga R, Hofsjo A, Hulthen-Varli I, Kublickas M, et al. Diagnostic evaluation of intrauterine fetal deaths in Stockholm 1998-99. Acta Obstetricia et Gynecologica Scandinavica 2002;81:284-92.

46. Maternal and Child Health Research Consortium L. Confidential enquiry into stillbirths and deaths in infancy (CESDI). Eighth annual report. London: Maternal and Child Health Research Consortium L; 2000.

47. National Health Service, Scotland. Scottish perinatal and infant mortality and morbidity report 2000. Edinburgh: Information & Statistics, NHS Scotland; 2001.

48. Stewart JH, Andrews J, Cartlidge PH. Numbers of deaths related to intrapartum asphyxia and timing of birth in all Wales perinatal survey, 1993-5. BMJ 1998;316:657-60.

49. Erdem G. Perinatal mortality in Turkey. Paediatric and Perinatal Epidemiology 2003;17:17-21.

50. Jansone M, Lazdane G. Perinatal problems and quality assurance in Latvia a country in economic transition. Acta Obstetricia et Gynaecologica Scandinavica Suppl 1997;164:31-3.

51. Langhoff-Roos J, Larsen S, Basys V, Lindmark G, Badokynote M. Potentially avoidable perinatal deaths in Denmark, Sweden and Lithuania as classified by the Nordic-Baltic classification. British Journal of Obstetrics and Gynaecology 1998;105:1189-94.

52. Lucas GN, Ediriweera RC. Perinatal deaths at the Castle Street Hospital for Women in 1993. Ceylon Medical Journal 1996;41:10-12.

53. Horpaopan S, Puapondh Y, Ratrisawasdi V, Prasertsom W, Vichitpahanakarn P, Sunakorn P. Perinatal mortality at Children's and Rajvithi Hospitals in 19831987. Journal of the Medical Association of Thailand 1989;72:376-81.

54. Sri-Smith R. Perinatal mortality Chiang Rai Prachanukroh Hospital 1992-1994. Journal of the Medical Association of Thailand 1996;79:16-20.

55. Fauveau V, Wojtyniak B, Mostafa G, Sarder AM, Chakraborty J. Perinatal mortality in Matlab, Bangladesh: a community-based study. International Journal of Epidemiology 1990;19:606-12.

56. Gazi R, Goodburn L, Chowdhury AM. Risk factors for perinatal deaths in rural Bangladesh. Journal of Health and Population in Developing Countries 1999;2:70-77.

57. Kusiako T, Ronsmans C, Van der Paal L. Perinatal mortality attributable to complications of childbirth in Matlab, Bangladesh. Bulletin of the World Health Organization 2000;78:621-7.

58. Shah U, Pratinidhi AK, Bhatlawande PV. Perinatal mortality in rural India: a strategy for reduction through primary care. I. Stillbirths. Journal of Epidemiology and Community Health 1984;38:134-7.

59. Phukan RK, Mahanta J. A study of neonatal deaths in the tea gardens of Dibrugarh district of upper Assam. Journal of the Indian Medical Association 1998;96:333-4; 337.

60. Singh UK, Srivastava SP, Kumar A, Thakur AK, Prasad R, Chakrabarti B. Comparative study of perinatal mortality and morbidity in the community and at Medical College Hospital, Patna. Indian Pediatrics 1996;33:1057-8.

61. Shrivastava SP, Kumar A, Kumar OA. Verbal autopsy determined causes of neonatal deaths. Indian Pediatrics 2001;38:1022-5.

62. Misra PK, Thakur S, Kumar A, Tandon S. Perinatal mortality in rural India with special reference to high risk pregnancies. Journal of Tropical Pediatrics 1993;39:41-44.

63. Kapoor RK, Srivastava AK, Misra PK, Sharma B, Thakur S, Srivastava KI et al. Perinatal mortality in urban slums in Lucknow. Indian Pediatrics 1996;33:19-23.

64. Damodar, Mathur HN, Sharma PN. Some observations on perinatal mortality in rural health centre. Indian Journal of Pediatrics 1983;50:629-33.

65. Geetha T, Chenoy R, Stevens D, Johanson RB. A multicentre study of perinatal mortality in Nepal. Paediatric and Perinatal Epidemiology 1995; 9:74-89.

66. Anon. Australian Institute of Health and Welfare National Perinatal Statistics Unit. Australia's mothers and babies 2000. Canberra: Australian Institute of Health and Welfare National Perinatal Statistics Unit; 2003. Perinatal Statistics Series, no. 12.

67. Anon. Annual report of the year 2000. Incorporating the 39th survey of perinatal deaths in Victoria. Melbourne: Consultative Council on Obstetric and Paediatric Mortality and Morbidity; 2004.

68. Tham WL, Tan KH, Tee CS, Yeo GS. Confidential enquiry of stillbirths in current obstetric practice. International Journal of Gynaecology and Obstetrics 1999;64:287-96.

69. Ben-li L, Dao-zhong Z, Hing-qi T, Pei H. Perinatal mortality rate in 11 Jiangsu cities. Chinese Medical Journal 1985;98:157-60.

70. Zhang J, Cai WW, Chen H. Perinatal mortality in Shanghai: 19861987. International Journal of Epidemiology 1991;20:958-63.

71. Amar HS, Maimunah AH, Wong SL. Use of Wigglesworth pathophysiological classification for perinatal mortality in Malaysia. Archives of Diseases in Childhood. Fetal Neonatal Edition 1996;74:F56-59.

72. Amoa AB, Klufio CA, Moro M, Kariwiga G, Mola G. A case-control study of stillbirths at the Port Moresby General Hospital. Papua New Guinea Medical Journal 1998;41:126-36.

73. Maouris P. Reducing perinatal mortality in Vila Central Hospital, Vanuatu. Papua New Guinea Medical Journal 1994;37:178-80.

74. Leach A, McArdle TF, Banya WA, Krubally O, Greenwood AM, Rands C, et al. Neonatal mortality in a rural area of The Gambia. Annals of Tropical Paediatrics 1999;19:33-43.

75. Schumacher R, Swedberg E, Diallo MO, Keita DR, Kalter H, Pasha O. Mortality study in Guinea: investigating the causes of death for children under 5. Arlington, VA: Save the Children Federation, Inc. and the Basic Support for Institutionalizing Child Survival (BASICS II) Project; 2002.

76. Ekanem EE, Asindi AA, Okoi OU. Community-based surveillance of paediatric deaths in Cross River State, Nigeria. Tropical and Geographical Medicine 1994;46:305-8.

77. Pison G, Trape JF, Lefebvre M, Enel C. Rapid decline in child mortality in a rural area of Senegal. International Journal of Epidemiology 1993;22:72-80.

78. Fantahun M. Patterns of childhood mortality in three districts of north Gondar Administrative Zone. A community based study using the verbal autopsy method. Ethiopian Medical Journal 1998;36:71-81.

79. Woods D. Perinatal Audit System Report for 1st Jan31st Dec 2001. Cape Town: Peninsular Maternal and Neonatal Service Cape Town Metropolitan Area, South Africa; 2001.

80. Crowther CA, Glyn-Jones R, Brown IM. Perinatal mortality in the Greater Harare Obstetric Unit area. South African Medical Journal 1987;72:255-6.

81. Gomes JO, Santo AH. Mortalidade infantil em municipio da regiao Centro-Peste Paulista, Brasil, 1990 a 1992. [Infant mortality in a midwestern municipality of the State of São Paulo, Brazil, 1990 to 1992.] Revista de Saude Publica 1997;31:330-41. [In Portuguese]

82. Samms-Vaughan ME, McCaw-Binns AM, Ashley DC, Foster-Williams K. Neonatal mortality determinants in Jamaica. Journal of Tropical Pediatrics 1990;36:171-5.

83. Mendieta E, Battaglia V, Villalba B. Mortalidad Neonatal en el Paraguay: Analisis de Los Indicadores. [Neonatal Mortality in Paraguay: An analysis of the indicators.] Pediatria 2001;28:8-17. [In Spanish]

84. Aguilar AM, Alvardo R, Cordero D, Kelly P, Zamora A, Salgado R. Mortality survey in Bolivia: the final report. Investigating and identifying the causes of death for children under five. Arlington, VA: Basic Support for Institutionalizing Child Survival (BASICS) Project. Published for USAID: 1998.

85. Aleman J, Brannstrom I, Liljestrand J, Pena R, Persson LA, Steidinger J. Saving more neonates in hospital: an intervention towards a sustainable reduction in neonatal mortality in a Nicaraguan hospital. Tropical Doctor 1998;28:88-92.

86. Ebrahim AH. Perinatal mortality in Ministry of Health Hospitals-Bahrain, 1985 and 1996. Journal of the Bahrain Medical Society 1998;10:95-99.

87. Al Sawan, Rima MZ. Perinatal and neonatal mortality trends over fifteen years in Farwania hospital-Kuwait. Kuwait Medical Journal 1998;30:15-19.

88. Asindi AA, Archibong E, Fatinni Y, Mannan N, Musa H. Perinatal and neonatal deaths. Saudi Medical Journal 1998;19:693-7.

89. Dawodu A, Varady E, Verghese M, al Gazali LI. Neonatal audit in the United Arab Emirates: a country with a rapidly developing economy. East Mediterranean Health Journal 2000;6:55-64.

90. Yassin KM. Indices and sociodemographic determinants of childhood mortality in rural Upper Egypt. Social Science Medicine 2000;51:185-97.

91. Fikree FF, Azam SI, Berendes HW. Time to focus child survival programmes on the newborn: assessment of levels and causes of infant mortality in rural Pakistan. Bulletin of the World Health Organization 2002;80:271-6.

92. Djaja S, Soemantri S. The cause of neonatal death and the attributed health care system in Indonesia: mortality study of household health survey, 2001. Jakarta: National of Health Research and Development, Ministry of Health Indonesia; 2003.

93. Fonseka P, Wijewardene K, Harendra de Silva DG, Goonaratna C, Wijeyasiri WA. Neonatal and post-neonatal mortality in the Galle district. Ceylon Medical Journal 1994;39:82-5.

94. Sivagnanasundram C, Sivarajah N, Wijayaratnam A. Infant deaths in a health unit area of Northern Sri Lanka. Journal of Tropical Medicine and Hygiene 1985;88:401-6.

95. Chongsuvivatwong V, Impat A, Tayakkanonta K. A survey of neonatal tetanus and perinatal mortality in southern Muslim communities in Thailand. Southeast Asian Journal of Tropical Medicine and Public Health 1993;24:654-8.

96. Khanjanasthiti P, Benchakarn V, Saksawad A, Khantanaphar S, Posayanond P. Perinatal problems in rural Thailand. Journal of Tropical Pediatrics 1984;30:72-8.

97. Rahman S, Nessa F. Neo-natal mortality patterns in rural Bangladesh. Journal of Tropical Pediatrics 1989;35:199-202.

98. Bang AT, Bang RA, Baitule SB, Reddy MH, Deshmukh MD. Effect of home-based neonatal care and management of sepsis on neonatal mortality: field trial in rural India. Lancet 1999;354:1955-61.

99. Bang AT, Bang RA, Baitule S, Deshmukh M, Reddy MH. Burden of morbidities and the unmet need for health care in rural neonates a prospective observational study in Gadchiroli, India. Indian Pediatrics 2001;38:952-65.

100. Pratinidhi A, Shah U, Shrotri A, Bodhani N. Risk-approach strategy in neonatal care. Bulletin of the World Health Organization 1986;64:291-7.

101. Shah U, Pratinidhi AK, Bhatlawande PV. Perinatal mortality in rural India: intervention through primary health care. II. Neonatal mortality. Journal of Epidemiology and Community Health 1984;38:138-42.

102. Huang W, Yu H, Wang F, Li G. Infant mortality among various nationalities in the middle part of Guizhou, China. Social Science Medicine 1997;45:1031-40.

103. Ren-Ying Y, McCarthy B, Hui-Fang Y, Chuan-Yan Q, Li Z, Tong-Xiang C, et al. The risk approach in perinatal health, Shunyi County, China. Atlanta, GA: US Department of Health and Human Services, CDC; 1989. DHHS publication no. 890-8412

104. Al-Mohdzar SA, Haque ME, Abdullah WA. Changes of perinatal statistics in a semi urban setup between two time periods in Malaysia. Asia-Oceania Journal of Obstetrics and Gynecology 1993;19:401-5.

Saúde Pública

Saúde Pública