PRB Discuss Online: Why Are Stillbirths an Invisible Loss of Life in Developing Countries?

(November 2008) Each year more than 3 million stillbirths occur, and approximately 1 million of these occur after the onset of labor, but these deaths remain invisible in international and national policies, programs, and investment agendas. Stillbirths are rarely measured in large surveys, are not included in routine data reported to the United Nations, and are not tracked in the Millennium Development Goals (MDGs). Yet MDG’s 4 and 5 (reducing child and maternal mortality, respectively) cannot be reached without improving newborn and obstetric care, which also affect stillbirths. What can be done to ensure that research, adequate monitoring, and program planning take stillbirths into account?

During a PRB Discuss Online, Cindy Stanton, assistant professor at Johns Hopkins Bloomberg School of Public Health, answered participants’ questions about important issues surrounding stillbirths in developing countries.

November 13, 2008 1 PM EST

Transcript of Questions and Answers

Candice York: While I beleive the questions is of concern I am wondering if there is any consideration for the different causes of stillbirths and as such is research going to demarcate the different causes and thereby focus on the ones that may be affected or influenced by human or developmental factors?
Cindy Stanton: As is often the case, where the problem is the greatest is where we have the least information. 98%+ of stillbirths occur in the developing world and we know much much less about causes of stillbirth than we do about the magnitude. Lawn and colleagues undertook a large literature review to estimate the percent of intrapartum stillbirths, that is stillbirths occurring after the onset of labor. Their work suggests that just under 1 in 3 stillbirths occur during the intrapartum period. If ever there was an issue that serves as a bridge between the concerns of the maternal AND the newborn health communities, it is intrapartum stillbirth. Improved care at delivery, particularly for women with complications, is key to progress for both. The fact that we consider “intrapartum” stillbirth a cause (strictly speaking it is timing of death and not biological cause) speaks volumes about the specificity of information that we have on SB cause of death. There are many different stillbirth cause of death classification schemes, many of which are highly sophisticated for use in developed countries and not really applicable in developing countries. However, there are quite a few that are applicable in low resource settings. Some of these schemes are more programmatically relevant than others. However,given the lack of consensus on a classification scheme, comparisons across study sites is very difficult. As part of the Global Alliance for the Prevention of Preterm Birth and Stillbirth (, I am working with colleagues on a set of global estimates of SB cause of death now, which is to be completed next spring. But, again, these are estimates from a statistical exercise, and what is needed is empirical data from improved cause of death registration in countries.

Agatha Onovo: Still births go unnoticed because people want to forget the loss. In addition to the medical, social and economic indices for high rate of still births in Nigeria, the major problem is non-accessibility and non-availability of proper medical equipments. Consequently, care providers are constrained to provide the necessary care even when something could have been done. The question is what can be done to get political leaders to be commited to equiping the health facilities and providing basic amenities?
Cindy Stanton: I agree, but I suspect the need is broader than that. When there are problems around delivery, women need to recognize them and seek care in a timely manner in order to benefit from the services available at facilities (ie, birth preparednes and all the things the safe motherhood community has been promoting for a couple of decades). Appropriate diagnosis and treatment is needed during pregnancy re: maternal infections and other maternal health conditions associated with stillbirth. I think there will always be a question of coordinating supply and demand.

Mary Kent: What are the constraints on getting better estimates of stillbirths in less developed countries? What are some of the benefits of getting more accurate estimates?
Cindy Stanton: Only 3% of stillbirths occur in countries with complete civil registration. What we know about the magnitude of stillbirths globally comes from 2 separate statistical exercises, which surprisingly came to the same global totals (3.2 million – Stanton et al. (uncertainty 2.5-4.1 million) 2006; and 3.3 million stillbirths – WHO 2006). However, despite agreement at the global level, the two series of estimates differ substantially at the country level, where the data are needed most. The concern is that where different sources provide highly varying estimates or estimates with great uncertainty, policy makers and others who could/should use the data, lose faith and find it easy to dismiss these numbers. All of the authors above acknowledge that these estimates are not precise and are a first attempt to prompt discussion/debate and hopefully improved data collection. Given that household-based surveys of women of reproductive age will probably remain the source of data that we all need to rely on in the immediate future (ie, while civil registration systems improve over time), there is a tremendous need to improve methods re: retrospective data collection. Current methods (or current practices at implementing existing methods) tend to seriously under-estimate stillbirths. However, health facilities can be a very important source of routine data, particularly in countries with fairly high institutional delivery. In those settings, one can benefit from a broader array of data re: cause of death, maternal condition and the health care received – all of which can be used to improve outcomes.

Jose Luis Diaz-Rossello: Three facts about fetal deaths in Latin America: In Latin America the Perinatal Information System developed by CLAP/WMR a PAHO technical center has been a good registry of fetal deaths since 1987. Currently Congenital Syphilis fetal death toll remains invisible althoug it may represent near 15% of all fetal deaths. Most of the preventable fetal deaths in developing countries are near term.
Cindy Stanton: Thanks for bringing up CLAP. It is a wealth of information and a great example of what routine data can provide, and how it can be presented in programmatically relevant ways. Realistically, some very low resource settings in sub-Saharan Africa or South Asia would have a hard time collecting the breadth and depth of CLAP data, but certainly adaptations could be made and stillbirths could be added to routine health information systems. It’s been said so many times re: the MDGs: what you count is what you do. If stillbirths are not on anyone’s agenda, they will not be (and are not being) counted or addressed. Many people these days think that we currently have a real opportunity to change that now. Several things support this idea – but particularly the focus (and increased funding) on newborn health/survival and a renewed focus on maternal survival.

Elizabeth Bocaletti: Most of the stillbirths that occur at home (the majority of the cases) are buried with out recording either the birth or the death. So, how [can we] deal with the registration of these cases? Are there any experiences in countries that have succeeded with birth and stillbirth registration?
Cindy Stanton: Hi Elizabeth! As mentioned below, only about 3% of stillbirths occur in countries with complete civil registration (not surprisingly these are high income countries), so yes, this is a huge constraint. A number of other countries do have stillbirth death certificates/registration but have highly incomplete data. However, what you describe is not just a constraint re: registration. Most people feel quite certain that these events (the birth and the stillbirth, and no doubt some deaths to liveborn babies during the first week of life) simply do not get mentioned at all during household surveys when women provide basic data re:their birth or pregnancy histories (ie the section of survey questionnaires that give us the data to calculate SB, neonatal, infant, child mortality rates). Education is needed to recognize these events at all levels (women/families; health care providers and health planners). At the level of women/families/communities, I think we need substantial qualitative research to learn “how” to talk/inquire about these events. At the level of health care providers and planners, education is needed to recognize the burden of stillbirth – there are about 4 million neonatal deaths, of which 3 million occur during the first week of life (the majority on days 1 and 2) and 3 million stillbirths (about 1/3 of which occur after the onset of labor). It just makes so sense to ignore this loss. Anecdotally, I had a conversation with a colleague of mine last week who told me that when she drew attention to the issue of stillbirths in a conversation with a Ministry of Health official, he said “but they are already dead, why do we care?” I’m sure there are many people who think this way, and just do not realize the important proportion of these fetal deaths that could be prevented.

Rahat Bari Tooheen: In developing countries, not all stillbirths may be reported, and due to resource constraints, stillbirths may not be accounted. Is budgetary increase the only solution, or will awareness building among the stakeholders hold a more sustainable solution?
Cindy Stanton: In only partially agree. I’m not convinced that lack of resources is the primary reason why we have so very little data on stillbirths. I think many simply do not recognize the numbers or the extent to which stillbirths can be prevented. Awareness is needed at all levels, from families up to Ministries and to the MDGs and Countdown to 2015. A clarification for this question and questions above: when I say that women/families do not report stillbirths – I am in no way implying that these losses do not have a profound effect on the woman/families. There may be cultural reasons for not reporting these events, and/or it may well be that we just do not know how to ask the questions in a way to elicit the correct response.

Achen Annet Nancy: I appreciate the topic of discussion but suggest that an issue of several lives lost in abotions be included in the discussion.
Cindy Stanton: You are right that this discussion is about stillbirths, so that is what we are sticking to here today. However, when discussing stillbirth, the issue of not recognizing other types of pregnancy loss is often the white elephant in the room. It is unclear how much, if any, avoidance of this discussion and the inevitable debates that would ensue, have contributed to the invisibility of stillbirths on national and international health agendas.

Dr. Khaled Shamsul Islam: It may be serious and urgent for developing countries like bangladesh where do not tracked in the Millennium Development Goals (MDGs). Yet MDGs 4 and 5 (reducing child and maternal mortality, respectively) cannot be reached without improving newborn and obstetric care, which also affect stillbirths. Can we start to ensure through research, adequate monitoring, and program planning take stillbirths into account? Now my question is – for this perspective how we can start a Regional alliance for working together to achieve the MDGs goal as well as our regional well being?
Cindy Stanton: Certainly there are many areas of research that are needed, but routine data sources (health facility registers, adding stillbirths to existing community registers, for example) could be an immediate (relatively immediate) source of data on stillbirths. Of course, collecting data serves no purpose unless it is compiled, reviewed and acted upon and there is no guarantee that that would happen. Implementing such a system would require political will and leadership. The magnitude of the problem, however, justifies such effort. RE: starting a regional alliance, my suggestion would be to pursue this effort through existing organizations. For example, the Partnership for Maternal, Newborn and Child Health, the International Stillbirth Alliance, Saving Newborn Lives/Save the Children, the White Ribbon Alliance and/or your regional professional organizations (Pediatrics, OB/GYN, midwifery association,etc.)

Lopamudra Paul: Maternal Mortality ratio and Infant (specially neonatal) mortality rates are high in south asia. moreover, institutional delivery is very low in this region. it is also noted that still births are common phenomena in many countries in this region. do we reduce the occurance of still birth with incease in institutional deliveries? further, will improved Ante Natal Care reduce the occurance of still births? Do medical assistance at home deliveries also revese the situtaion? in addition, do complications during pregnancy and delivery lead to still births?
Cindy Stanton: Two decades of safe motherhood (and common sense) has certainly made clear that institutional delivery or antenatal care will not lead to improvements in maternal or newborn health if the quality of care (skills, use of evidence-based practices) provided is low and if women/newborns are not able to access care in a timely manner. There are also additional risk factors such as inter-generational nutritional issues that lead to low birth weight that influence pregnancy outcome. So, as with maternal mortality,there is a lot of scatter in the relationship between stillbirth and institutional delivery. RE: medical assistance at home-based delivery, there are not that many places in the developing world where this is common (Indonesia stands out with their bidan di desa program. A recent analysis using DHS data suggested that a home-based professional was not associated with lower risk of first day or first week deaths. Data were not available re:stillbirths). And, yes,stillbirths are associated with maternal complications – if you write me,I can send you 2 articles that might be of interest to you.

Dr. Khaled Shamsul Islam: In Bangladesh only 14% are in institutional delivery and almost 90% [of] data are not available for delivery perspective–so it is really difficult to identify the correct figure of stillbirth. How we can proceed?
Cindy Stanton: Interestingly enough, it looks like Bangladesh seems to have had greater success at measuring stillbirths in large population-based surveys than any other country. The stillbirth rates from the DHS (ranging from around 27-33 per 1000) are quite close to high quality data from the Matlab Demographic Surveillance Site of ICDDR,B where a sophisticated, prospective data collection system is in place. (And, yes, one should not always compare Matlab data to the whole country, but I will here given the lack of other data sources). So, I think we all have much to learn from Bangladesh interviewers or respondents or both. In many of the other countries where this type of data has been added to surveys, the estimates appear implausibly low, particularly relative to early neonatal mortality (which may also be under-estimated in survey-based data). I have wondered if the seeming success in Bangladesh at identifying pregnancy losses in surveys was due to the great interest that exists re: menstrual regulation – leading to very careful completion of data on pregnancy, outcomes, gestational age, etc, and that improved stillbirth rates were just a lucky by-product of those practices. This is just speculation on my part. I do think we need to learn more from Bangladesh.

DR. JAMES AKPABLIE: I think we the experts and health workers know what can be done to prevent or at least reduce still births; why is it difficult to put a birth or late pregnancy monitoring scheme/strategy in place to prevent still birth?
Cindy Stanton: See responses to similar questions in this discussion. Among other things, the lack of prioritization may be due in part to a lack of knowledge re: the magnitude of the problem (if it is never counted, who would know?) and recognition that interventions to address many of causes of SB are readily available.

Mary Kent: How does the rate of stillbirths in a population compare with rates of infant mortality? Are the differentials similar, for example, with regard to education and poverty?
Cindy Stanton: Infant mortality rates vary from around 5 per 1000 live births in industrialized countries to about 110 per 1000 live births (West Africa), with the highest IMR for an individual country at 165 (Afghanistan) and the lowest at 2 (Sweden) (source: Unicef’s State of the World’s Children). Stillbirth rates for industrialized countries are around 5 per 1000 births (live and stillbirths), again with northern European countries as low as about 3 per 1000. The stillbirth rates for sub-Saharan Africa and South Asia according to the global estimates I worked on, as well as the WHO estimates, are both at 32 per 1000 births. At an individual country level, one sees estimates of the (population-based) stillbirth rate as high as 45-60 per 1000 births. I’ve never seen a cross-country comparative study of stillbirth by SES differentials among low income countries. Disparities re: stillbirth have been documented for high income countries (see Goy et al Pediatric and Perinatal Epi 2008 as an example). I just ran perinatal mortality by woman’s education on the DHS Statcompiler. At quick glance, you do not see a strong relationship (in some countries there is a clear gradient – decreasing SBRs with higher education), but not really in the majority. However, this is NOT an authoritative answer to your question. Such surveys do not measure SBRates very well.

Y.S. Sivan: 1. Are there exclusive studies on “Social Determinants of Still Birth”? Has the UN / WHO / UNICEF taken any initiative to encourage nations to include specific questions in the national Census and / or sample surveys, national health surveys to ensure regular flow of data (and integrate into a reliable global data-tracking mechanism)? 2. What is the proportion of research spending on still birth from the point of view of the 10/90 Gap? (Major health and social development search engines may consider including ‘still birth’ as a sub-set for easily tracking down the research priorities and data).
Cindy Stanton: I can answer some of your questions, but all of them. RE: what have various agencies have done to date: 1) WHO developed one of the two existing series of global stillbirth rates, and raised the issue of global perinatal mortality back in the 1980’s. WHO staff along with members of the Child Health Epidemiology Reference Group (CHERG) and the Global Alliance for the Prevention of Prematurity and Stillbirth (GAPPS) are currently working on updating/improving the global estimates for 2005. GAPPS staff is working on generating global stillbirth cause of death estimates and updated estimates for intrapartum stillbirth are in preparation. Dean Jamison authored a chapter in the recent edition of Disease Control Priorities in Developing Countries showing how burden of disease can be calculated for stillbirths. Just last week the International Stillbirth Alliance had a large conference in Norway – from which I am sure there will be proceedings with results from very recent research on the topic. From May 7-10,2009 in Seattle, there will be a Gates-funded Landscape Review of Prematurity and Stillbirth in order to highlight a) prioritized research questions urgently needing attention and b) existing evidence-based interventions which require immediate action re: scale up (see: Findings from a validation study in Ghana of verbal autopsy for stillbirth cause of death was recently published (Edmond, K and colleagues). This is not at all meant as a definitive list of efforts that are underway. These are merely the efforts that immediately came to mind as I respond to these questions. The takehome message is that a great deal of exciting work is underway and it seems that the invisibility of stillbirths may be changing. Just a few notes: RE measuring stillbirths in a census, this is not something that I personally would advise, and I suspect many demographers would agree. The census interview is simply not appropriate for in-depth questioning and very careful formulation of potentially sensitive questions, etc. However, I am a proponent of trying to measure stillbirths using other data collection approaches and feel that improving existing methods in use in household-based surveys is a top priority.

Ngozi Enelamah: What is the medical status of stillbirths? Are they reported or regarded as abortions?What are the major causes of stillbirths? How can this be seen as a public health issue and Could the cause be publicized so that they can be prevented?
Cindy Stanton: This is not straight forward to answer. Birth weight and/or gestational age cut-offs determine when pregnancy losses are considered spontaneous abortions versus late fetal death (actually, stillbirth is a colloquial term, the term used by the International Classification of Diseases, Rev 10, is late fetal death, at least for pregnancy losses at 1000 grams birthwt or 28 weeks gestation or more). If I remember correctly, when my colleagues and I were working on the literature review for the global stillbirth estimates, we identified around 20 different definitions of stillbirth used by statistical agencies in various countries; gestational age cut-offs were as low as 20 weeks; birthwt as low as 500 grams. So, in short, it varies from country to country. RE: cause of death, as mentioned elsewhere, there are many different stillbirth cause of death classification schemes, and they are infrequently comparable. My colleagues and I are close to finishing a systematic review of SB cause of death, and we are abstracting data into the following common categories: Maternal conditions (eclampsia/PIH, other maternal pre-existing conditions); abruption or antepartum hemorrhage; infections (syphilis and other maternal and fetal specific infections); congenital abnormalities; intrapartum (fresh SBs – obstruction, CPD); unexplained intrapartum causes, unexplained antepartum causes and unclassifiable. This does not constitute a “classification scheme” – it was designed to capture data from a variety of different classifications. To note – after reviewing over 36,000 abstracts, applying inclusion/exclusion criteria, etc for the review, we will end with SB cause of data from only approximately 100 papers. In other words, there are not a lot of data out there, particularly for low income countries.

Linna Lisette Gröppel: Could health promoters do more in matters of stillbirths in developing countries? Is it possible that health promotion and health communication have an effect in reference to stillbirths and a facility to monitor stillbirths?
Cindy Stanton: Education at the community level is needed, and I do not see why community registers could not track stillbirths, as some do for maternal deaths. One thing that I have not mentioned here yet is the problem of misclassification between stillbirths and early neonatal deaths. THis is a problem when women are asked to report their pregnancy outcomes, as well as a problem at health facilities. In the case of home-based births, a woman may never be shown a fetus born dead or a child that dies very quickly after birth. Thus, she may really not know the status of the fetus/infant at birth. Or, even if she does, there may be cultural or other reasons for claiming the outcome as stillbirth versus death following live birth. The same issue exists in health facilities world wide. Providers’ perception of the viability of the infant (particularly very preterm births) and other reasons may well affect their final call re: its status at birth.

samwel chale: Stillbirth shows that it related directly to poverty. At the same time poverty is problem in most of the developing countries, so how can we advise our government so that they can include into their strategic plans while most of their budgets are donor funded?
Cindy Stanton: The good news is that much of what needs to be done to reduce stillbirths is included in maternal/newborn health care packages now being promoted.

Farid Midhet: The importance of stillbirths is also because the intrapartum deaths are a sensitive indicator of quality of obstetric care; hence the proportion of ‘fresh’ stillbirths (and, until recently, the perinatal mortality rate)could be used as proxy indicators for maternal mortality. Identification of stillbirths and their causes, therefore, is important from EmONC perspective. Your comments?
Cindy Stanton: Hi Farid, I completely agree – see answers to previous questions. One could also argue that it is important to track stillbirths in order to be able to interpret possible changes in early neonatal mortality (as obstetric care improves, it is likely that some fetuses that would have died (ie stillbirths) will survive delivery but may die shortly thereafter. As health care systems and general health improve, both indicators improve.

Dr. Anima Sharma: Stillbirths are indeed a very big issue attributing to the loss of life before the unborn baby even breathes his first. There are several socioeconomic reasons [for] stillborns in the developing countrues like India, which include, Poverty, Early Marriage, Illiteracy, Ignorance, lack of decision making, malnourishmnet/ malnutrition, non-accessibility and non-availability of proper medical facilities, lack of care during pregnancy and these issues [are] further aggrevated by traditions and beliefs. This makes the entire scenario very intriguing. There have been several researches conducted to find out the reasons but I think now the stage has come when we should plan a thoughtful action plan to combat this situation. Do not you think that a multi-disciplinary team consisting of Medical Practitioners, Social Anthropologists, Demographers, Psychologists, Policy makers, bureaucrats etc. should jointly intervene and address the issue unitedly? Are there any such intervention going-on in any of the Developing Countries? If yes, then what is the outcome or the success rate?
Cindy Stanton: I don’t know of any truly multi-disciplinary teams, as you describe. I do agree that anthropologists, sociologists and political scientists could assist these efforts by exploring societie’s reluctance to address or even acknowledge the loss represented by stillbirths.

Rachel Breman: How do you suggest improving education amongst health professionals to ensure that stillbirths are getting recorded in a way that can useful for hopefully preventing them in the future?
Cindy Stanton: Hi Rachel, That is a very good question and not one for which even high income countries have found an answer. (Some argue that differences in infant mortality across developed countries are to some extent due to differences in practices re: defining a live birth.) Personal practices and beliefs come into play, as well as issues regarding things like insurance coverage. I guess the bottom line is that when something is recognized as important, individuals and the systems in which they work, establish expectations to accomplish their goals. I do not have evidence of this, but I strongly suspect that medical and midwifery schools in developing countries pay little heed to the finer points of these definitions. Certainly, few developing countries have stillbirth death certificates, another signal from the government and society as a whole, that a stillbirth is important.

hmal: Appreciating highly the problem of invisibility I would ask about successful examples in developing countries. Are there studies that define stillbirths as “loss of life” in human reproduction models or include them in summary indicators [of] life expectancy or healthy life expectancy, for example?
Cindy Stanton: Very quickly, before we close out today – see Dean Jamison’s chapter on the calculation of disability-adjusted life years (DALY’s) for stillbirth in the recent edition of Disease Control Priorities in Developing Countries.