Reducing intrapartum-related deaths and disability: Can the health system deliver?

Article Outline

• Abstract
• 1. Introduction
  • 1.1. Objective
• 2. Intrapartum-related outcomes: Evidence and integration into health system
  • 2.1. Interventions and strategies: Overview of the evidence
    • 2.1.1. Obstetric care
    • 2.1.2. Neonatal resuscitation and post-resuscitation management
    • 2.1.3. Linking community and facility
    • 2.1.4. Delivering care in community settings
    • 2.1.5. Perinatal mortality audit
    • 2.1.6. Evidence summary
  • 2.2. Integration of intrapartum-related interventions into health system packages in the continuum of care
  • 2.3. Delivering interventions: Who can do what?
• 3. Health system performance and context
  • 3.1. Health system performance categories
• 4. Prioritization, phasing, and potential lives saved
  • 4.1. Principles for data-based prioritization for planning
  • 4.2. Identify and address missed opportunities within existing programs
  • 4.3. Potential for lives saved through addressing missed opportunities during contact points
  • 4.4. Invest to reduce major gaps in coverage for care at birth
  • 4.5. Potential for lives saved at high (90%) coverage
  • 4.6. Priorities to reduce intrapartum-related burden, by the 5 Categories
• 5. Considerations for policy, planning, and implementation
  • 5.1. National policy
  • 5.2. Financing
• 6. Innovation to improve tools, evidence, and data
  • 6.1. Innovation for tools and technologies
  • 6.2. Improving the data for decision making
  • 6.3. Implementation research
• 7. Conclusions
• 8. Conflict of interest
• 9. Funding
• Acknowledgments
• Appendix A. Supplementary data
• References
• Further reading
• Copyright

Abstract 

Background

Each year 1.02 million intrapartum stillbirths and 904000 intrapartum-related neonatal deaths (formerly called “birth asphyxia”) occur, closely linked to 536000 maternal deaths, an estimated 42% of which are intrapartum-related.

Objective

To summarize the results of a systematic evidence review, and synthesize actions required to strengthen healthcare delivery systems and home care to reduce intrapartum-related deaths.

Methods

For this series, systematic searches were undertaken, data synthesized, and meta-analyses carried out for various aspects of intrapartum care, including: obstetric care, neonatal resuscitation, strategies to link communities with facility-based care, care within communities for 60 million non-facility births, and perinatal audit. We used the Lives Saved Tool (LiST) to estimate neonatal deaths prevented with relevant interventions under 2 scenarios: (1) to address missed opportunities for facility and home births; and (2) assuming full coverage of comprehensive emergency obstetric care and emergency newborn care. Countries were first grouped into 5 Categories according to level of neonatal mortality rate and examined, and then priorities were suggested to reduce intrapartum-related deaths for each Category based on health performance and possible lives saved.

Results

There is moderate GRADE evidence of effectiveness for the reduction of intrapartum-related mortality through facility-based neonatal resuscitation, perinatal audit, integrated community health worker packages, and community mobilization. The quality of evidence for obstetric care is low, requiring further evaluation for effect on perinatal outcomes, but is expected to be high impact. Over three-quarters of intrapartum-related deaths occur in settings with weak health systems marked by low coverage of skilled birth attendance (<50%), low density of skilled human resources (<0.9 per 1000 population) and low per capita spending on health (<US $20 per year). By providing comprehensive emergency obstetric care and emergency newborn care for births already occurring in facilities, 327200 intrapartum-related neonatal deaths could be averted globally, and with full (90%) coverage, 613000 intrapartum-related neonatal deaths could be saved, primarily in high mortality settings.

Conclusion

Even in high-performance settings, there is scope to improve intrapartum care and especially reduce impairment and disability. Addressing missed opportunities for births already occurring in facilities could avert 36% of intrapartum-related deaths. Improved quality of care through drills and audit are promising strategies. However, the majority of deaths occur in poorly performing health systems requiring urgent strategic planning and investment to scale up effective care at birth, neonatal resuscitation, and community mobilization as well as to develop, adapt, and introduce tools, technologies, and task shifting to reach the poorest.

Keywords: Birth asphyxia/asphyxia neonatorum, Health systems, Intervention, Intrapartum-related neonatal mortality, Lives saved, Neonatal mortality, Prevention, Stillbirth, Systematic review

Back to Article Outline

1. Introduction 

This is the final paper in a series of 7 reviews summarizing the size of the burden of intrapartum-related stillbirths and intrapartum-related neonatal deaths, and the evidence to reduce these, especially in low-resource settings where most of these deaths occur. The first paper mapped the staggering size of the problem [1]—almost 2 million deaths, comprised of an estimated 904000 (uncertainty range, 650000–1.17 million) intrapartum-related neonatal deaths, formerly referred to as “birth asphyxia,” and 1.02 million (uncertainty range, 660000–1.48 million) intrapartum stillbirths [2], as well as a poorly measured burden of long-term impairment and disability. Global Burden of Disease, Millennium Development Goals, and Countdown to 2015 metrics do not currently include stillbirths [3], illustrating the legacy of invisibility of this massive loss of life [4]. Perinatal deaths are intimately linked with the annual global toll of approximately half a million maternal deaths [5].

This burden is not just a disease metric, but represents immeasurable loss for families just at a point when new life is expected. Globally, intrapartum-related conditions are implicated in 23% of neonatal deaths, 32% of stillbirths, and 42% of maternal deaths [1]. Yet the time when the risk of these deaths is highest—during childbirth and the immediate postnatal period—is also the time in the continuum of care when women and newborns in high mortality settings are least likely to have contact with health care, particularly skilled care during childbirth and follow-up care in the early postnatal period [6], [7]. While two-thirds of women in high mortality settings attend prenatal care, only one-third access skilled care at birth, and a much lower number receives early postnatal care. In addition, more than half the world's births occur in high mortality countries with a neonatal mortality rate (NMR) greater than 30 per 1000 and coverage of skilled attendance at birth of approximately 50% (Fig. 1). These high mortality settings, mainly in South Asia and Sub-Saharan Africa, account for approximately three-fourths (77%) of neonatal deaths and a similar proportion of maternal deaths [8]. Hence, where the burden is highest, the probability of effective care is lowest [9]. In addition, within countries, it is the poorest families, often in rural areas and urban slums, who continue to have the highest risk, and yet the greatest delays in accessing obstetric and early postnatal care [10], [11]. These delays may not be due to distance, or even financial barriers alone, but also gender, ethnicity, fatalism, or different perceptions of complications at birth [1].

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 1. 

  Countries (193) organized according to 5 Categories of neonatal mortality rate as a marker of health system performance, showing the associated variation in maternal mortality, as well as intrapartum mortality outcomes for stillbirths and neonatal deaths, and an inverse association with skilled attendance at birth. Abbreviation: NMR, neonatal mortality rate. Note that the association between skilled neonatal mortality and skilled attendance is at ecological level and cannot be assumed to be causal based on this data. Country groupings by category of NMR level are adapted from The Lancet Neonatal Series 2005 [74]. Updated for 2009 births and mortality rates.

• Sources: Maternal mortality ratio data from Hill et al. [5]; intrapartum stillbirth data from Lawn et al. [2]; neonatal mortality rate data WHO (UNICEF [75]); intrapartum-related neonatal deaths (“birth asphyxia”) from Countdown 2008 [26] based on methods from Lawn et al. [76].

This burden is large, yet, as summarized in this Supplement, interventions exist even if the GRADE quality of evidence is often moderate or low. The fact that there is a 24-fold difference in intrapartum-related neonatal mortality between high-income and low-income countries [1] indicates that solutions are possible. It has been convincingly argued that the greatest test of health system function is the ability to provide timely care at birth [12]. Delays can be important for many of the critical conditions that cause neonatal and maternal mortality and stillbirth, but delays of even a few hours can be the difference between life and death for a woman with obstetric hemorrhage, or a delay even of minutes can contribute to the death of a baby not breathing at birth. A health system that can provide timely cesarean delivery and resuscitate a non-breathing baby is likely to be able to respond to other acute and chronic conditions; hence, we propose that effective care at the time of birth is a litmus test of health systems performance.

The huge challenge remains how to implement these solutions—especially high-quality obstetric care and immediate newborn care—in low-income settings and particularly how to reach 60 million home births a year. Strategies to better link communities to effective obstetric and immediate newborn care are also critical and yet often not systematically implemented [11]. Can interventions be prioritized for varying levels of health system performance, and are there some interventions that can also be adapted for effective delivery now in lower levels of the health system or even at community level? [13].

1.1. Objective 

In this paper, the concluding paper in a series entitled “Intrapartum-related neonatal deaths: Evidence for action,” we summarize the evidence for interventions to reduce intrapartum stillbirths and intrapartum-related neonatal deaths based on a systematic review of almost 30000 article titles or abstracts as detailed in the preceding 6 papers. The level of evidence was assessed using the GRADE system criteria, to evaluate the quality of the evidence (high, moderate, low, or very low) and make recommendations (strong, conditional, weak) based on standard criteria [1], [14]. Here we place the problem and solutions in a health systems context, integrating effective interventions and strategies with existing health system packages, and synthesizing evidence and experience regarding delivery strategies across the continuum of maternal, neonatal, and child health (MNCH) programs. Statistical modeling based on the Lives Saved tool (LiST) is used to estimate the lives saved through immediate and medium/long-term program priorities for 5 different health system settings categorized by level of NMR as an outcome marker of health system performance. Finally, we highlight evidence gaps and priorities for innovation and further research with a focus on reaching under-served populations, and on current experience with large-scale programs.

Back to Article Outline

2. Intrapartum-related outcomes: Evidence and integration into health system 

2.1. Interventions and strategies: Overview of the evidence 

In this systematic literature review of strategies to reduce intrapartum-related mortality, we screened approximately 30000 article titles or abstracts, but identified fewer than 100 trials that reported an effect on neonatal mortality rate (NMR), stillbirth rate (SBR), or perinatal mortality (PMR), and even fewer that reported intrapartum-related outcomes, specifically intrapartum-related neonatal deaths or intrapartum stillbirths. Furthermore, many of the reports identified were primarily set in high-income settings, thus, the results or comparisons may not be readily generalized to low-income settings. In many of the studies of maternal health interventions, only maternal outcomes were reported, missing an opportunity to also evaluate intervention effect on perinatal outcomes. The converse may also be true—with trials reporting only on neonatal outcomes. It was disappointing to identify how many trials missed the opportunity for integrated evaluations of relevant outcomes for stillbirth, neonate, and mother, as noted in another recent large systematic review process for stillbirths [3]. Interpretation is also simpler if stillbirth and neonatal outcomes are reported separately, rather than always combined as perinatal.

We did not identify any interventions that had high GRADE level of evidence on intrapartum-related mortality (Table 1). This lack of high-quality evidence from randomized controlled trials (RCTs) may also reflect the ethical complexity of undertaking such trials, particularly for interventions that are already considered standard practice, such as cesarean delivery. Nevertheless, there are major knowledge gaps that could be addressed and require a more systematic research agenda, prioritized based on likely impact [15].

Table 1. Evidence of effect of interventions and strategies on perinatal, neonatal, and intrapartum-related mortality outcomes.

		Mortality Effect					Morbidity effect	GRADE level of evidence	GRADE recommendation
		SBR	PMR	NMR	ENMR	IPR-NMR
Interventions and packages	Obstetric care: Intrapartum interventions[10]
	- Basic Emergency Obstetric Care					38%2		Low	Strong
	- Comprehensive Emergency Obstetric Care					75%2		Low	Strong
	- Partograph use	38%–46%	36%–40%					Low-Moderate	Strong
	- Planned elective cesarean delivery for breech		71% (14%–90%)1					Moderate	Conditional
	- Routine induction of labor at ≥41 weeks		70% (1%–91%)1					Moderate	Conditional
	- Continuous labor support						9% (1%–17%)3	Low	Strong
	- Amnioinfusion for meconium staining						91% (51%–98%)4	Moderate	Weak
	- Amnioinfusion for umbilical cord compression						68% (30%–85%)5	Moderate	Weak
	Neonatal resuscitation[17]
	- Facility settings [94]				40% (6%–61%)	30% (16%–41%)		Moderate	Strong
	- Community settings	31%–33%	15%–29%			20%6 [47%–70%]		Low	Conditional
	Post-resuscitation management
	- Serum glucose-fluid management							Low	Weak
	- Anticonvulsants							Low	Weak
	- Thermal management							Low	Strong
	- Therapeutic hypothermia					26% (6%–42%)7		Moderate (high income settings)	Conditional
Linking strategies	Increasing community demand for obstetric care[11]
	- Community mobilization	28%–47%8	25% (4%–41%)	5%–76%9	36% (15%–52%)			Moderate	Strong
	- Financial strategies							Very Low	Conditional
	Brining pregnant women closer to formal health system
	- Community referral and transport schemes							Very Low/Low	Conditional
	- Risk screening							Very Low	Conditional
	- Maternity waiting homes	48%–90%	16%–51%					Low	Conditional
	Delivering care in community settings[13]
	- Community midwives and birthing centers	15%10	12% (5%–17%)	32%–40%	13% (3%–21%)	22%–47%		Low	Strong
	- Trained traditional birth attendants	31%11	6% (4%–9%)13	29%–41%	15%16	11% (2%–21%)13		Low/Moderate	Conditional
			30% (18%–41%)14
	- Integrated home-based care packages by CHWs	28%–49%12	28% (16%–38%)	25%–54%15	36% (27%–44%)	42%17		Moderate	Strong
	Perinatal mortality audit[23]
	- Facility-based audit, quality of intrapartum care	34%–61%	30% (21%–38%)		1%–56%			Low/Moderate	Strong
	- Community-based audit							Very low	Weak

BOLD results are from meta-analysis or Cochrane review.

^⁎Interventions without sufficient evidence demonstrating significant effect on perinatal-neonatal or intrapartum mortality (active management of labor, cesarean for failure to progress, controlled fundal pressure, instrumental delivery, symphysiotomy, induction for fetal macrosomia, maneuvers for shoulder dystocia, external cephalic version, amnioinfusion for meconium or umbilical cord compression, antihypertensives for hypertensive disorders of pregnancy, anticonvulsants for hypertensive disorders of pregnancy, ultrasound confirmation of previa, cervical cerclage, rapid delivery for placental abruption, membrane sweeping at or after 41 weeks gestation, antibiotics for chorioamnionitis, continuous electronic fetal heart rate monitoring, electronic fetal electrocardiogram, fetal pulse oximitry.

¹From Cochrane Review, trials primarily in high-income settings.

²Estimate based on Delphi Expert Consensus opinion.

³From Cochrane Review, reduction in cesarean birth.

⁴From Cochrane Review, reduction of neonatal encephalopathy.

⁵Reduction of 5 minute Apgar <7.

⁶Estimate based on Delphi Expert Consensus, [Range] from 3 community-based studies however low quality and varying definition of non-breathing baby.

⁷From Cochrane Review, all studies from high-income settings.

⁸Range of 3 studies (O'Rourke et al. [78], Bhutta et al. [79], Kumar et al. [55]), 2 studies with no significant effect (Manandhar et al. [80], Fullerton et al. [81]).

⁹Range of packages with multiple interventions not limited to community mobilization, including case management of neonatal infections (O'Rourke et al. [78], Manandhar et al. [80], Fullerton et al. [81], Baqui et al. [82], Bhutta et al. [79], Kumar et al. [55]).

¹⁰Significant 15% effect in 2 studies (Ibrahim et al. [83], Ronsmans et al. [84]), 2 studies with no effect (Gloyd et al. [85], PATH [86]).

¹¹Results from 1 RCT (Jokhio et al. [21]) and 1 multi-center before-after study (Carlo et al. [22]).

¹²Significant reduction in 3 studies (Bhutta et al. [79], Bang et al. [19], Kumar et al. [55]).

¹³Based on previous meta-analysis (Sibley et al. [20]).

¹⁴Based on new RCT (Jokhio et al. [21]).

¹⁵Broad range of packages and interventions, several including case management of neonatal infections.

¹⁶One significant result (Greenwood et al. [87]), one study with no effect (Carlo et al. [22]).

¹⁷Intervention to Control area comparison for SEARCH trial during tube-and-mask training period (Bang et al. [88]).

Evidence for the impact of obstetric care packages (Basic and Comprehensive Emergency Obstetric Care) was very low quality based on GRADE [16], consisting primarily of ecological and historical data. Nonetheless, access to emergency obstetric care should be a universal right for all mothers and is strongly recommended. Of the individual obstetric interventions reviewed, none had strong evidence based on specific data regarding intrapartum-related mortality outcomes (Table 1) [10]. Amnioinfusion for meconium staining was associated with lower rates of neonatal encephalopathy, planned cesarean delivery for breech presentation, and post-term induction of labor were associated with lower all-cause perinatal mortality; however, further evaluation and specifically risk-benefit analysis in low-resource settings is needed before these procedures can be routinely recommended in low-resource settings. On the other hand, there are several individual obstetric interventions with a low level of GRADE evidence, yet these are conditionally or strongly recommended, given the favorable risk-benefit assessment and inability to evaluate these interventions in RCTs because of current standards of care; these include use of the partograph, external cephalic version, emergency cesarean delivery for breech presentation, management of shoulder dystocia with therapeutic maneuvers, in utero resuscitation, symphysiotomy, rapid delivery for placental abruption, and cesarean delivery for uterine rupture (Table 1). Promising intrapartum care provision strategies include obstetric drills/rapid response teams, continuous intrapartum support, safety checklists, and task-shifting to non-physician clinicians. The dearth of evidence stresses the urgent need to better evaluate obstetric care programs and packages to build this evidence base, and to also evaluate perinatal outcomes [10].

In a meta-analysis of observational before-after-studies, facility-based training for neonatal resuscitation resulted in a 30% reduction in intrapartum-related neonatal mortality [17]. The results were highly consistent in direction and strength of association across multiple studies, and also directly generalizable to low- and middle-income settings; thus, the GRADE level of evidence was upgraded to moderate [18]. There is moderate level evidence for therapeutic hypothermia in the reduction of neonatal encephalopathy-related morbidity and mortality; however, the evidence thus far has been derived in high-income settings. Clinical trials are underway of therapeutic hypothermia modified for use in low-resource settings, and this intervention is not presently recommended until further data is available and cost-effectiveness can be compared with primary and secondary prevention.

Community mobilization is an effective strategy to link families to facility-based obstetric care associated with a significant increase in facility births and a 36% reduction in ENMR in a meta-analysis of 2 RCTs and 1 quasi-experimental trial of high intensity mobilization [11]. This GRADE level of evidence is moderate and community mobilization is strongly recommended as a strategy to increase demand for skilled childbirth care, and possibly to improve intrapartum outcomes through reducing the risk for other factors, such as maternal infection, that may compound the risk. Additional evaluation is needed, however, to define impact of community mobilization strategies on intrapartum-related outcomes and cost-effectiveness. Furthermore, other potential linking strategies such as financial incentives, community referral/transport schemes, risk screening, and maternity waiting homes need to be further evaluated.

At the community level, there was moderate GRADE level evidence supporting a beneficial effect of integrated community health worker (CHW) packages on both perinatal (28% reduction) and early neonatal mortality (36% reduction). Furthermore, there is evidence that stillbirths may be reduced 28%–49%, and one quasi-experimental study of CHW training in neonatal resuscitation demonstrated a 42% reduction in “asphyxia-specific” mortality [19]. The effect of training TBA in neonatal resuscitation resulted in a smaller reduction of intrapartum-specific mortality of 11% [20]. A recently published large RCT of TBA training in primary prevention demonstrated a significant 30% reduction in PMR [21] and 2 large trials showed a 31% reduction in stillbirth [21], [22]. The evidence for effect of community-based skilled birth attendants is low by GRADE criteria, and a meta-analysis of before-and-after studies of skilled birth attendant training showed a 12% reduction in PMR and 13 % reduction in early NMR (ENMR) [13]. Three low-quality studies reported intrapartum-related mortality reductions ranging from 22%–47%. However, these data on effect of skilled birth attendants must be interpreted with caution, given the low quality before-and-after studies.

Perinatal mortality audit has been used primarily in health facility settings, and there are encouraging signs that this can be scaled up even in middle- and low-income countries; the overall GRADE of evidence is low-moderate. There are also some experiences with community-based audit [23], [24]. Our new meta-analysis of 8 studies suggests a 31% effect on perinatal mortality from effective audit, linked to action [23].

Comparing the various effect sizes, the GRADE level of evidence, and the strength of GRADE recommendations, the largest mortality effects are expected through obstetric care packages, but these also have the weakest levels of evidence (Table 1). The Lancet Neonatal Series estimated that obstetric care reduced all-cause neonatal mortality by 20%–60% [25]. A recent Delphi process suggests 75% (60%–85%) reduction in intrapartum-related neonatal deaths through Comprehensive Emergency Obstetric Care (see Panel 1). The strongest evidence is for facility-based neonatal resuscitation, community mobilization, and integrated CHW packages (effect size ranging from 30%–40% reduction in ENMR), with more modest reductions with the other strategies, such as training community skilled birth attendants and TBAs or therapeutic hypothermia. Caution must be applied in using these data as some of the effect estimates are based on studies with design limitations or with small numbers of subjects, or the effect may have been dependent on local factors not easily replicated [18]. However, the massive size of the burden of intrapartum-related mortality mandates cautious use now of the available evidence to guide new policy and program implementation, and a much more aggressive approach to filling key evidence gaps, especially for facility-based obstetric care.

2.2. Integration of intrapartum-related interventions into health system packages in the continuum of care 

Some interventions are more feasible to deliver vertically and have shown major increases in coverage, particularly those that are primarily commodity-based and are receiving major investments such as insecticide treated bed nets, immunizations, and antiretroviral therapy [26]. However, interventions based on clinical case management have made slower progress; for example, skilled birth attendance coverage has not increased in Africa in the last decade and at current rate of increase will still reach less than half of African women at birth by 2015 [1]. Interventions to avert intrapartum-related deaths are less amenable to a vertical approach than many child health interventions, and will be most effectively and efficiently delivered when integrated into existing health service delivery packages along the continuum of care for mothers, newborns, and children. Fig. 2 shows the continuum of care through the pregnancy life cycle and along the continuum of place of health service delivery, adapted from the framework of 8 service delivery packages for MNCH proposed by Kerber et al. [27]. Of these healthcare service packages, the 4 most relevant to intrapartum-related outcomes are: (1) prenatal care; (2) childbirth care; (3) postnatal care, particularly immediate care and in the first 2 days of life; and (4) care for sick neonates/children.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 2. 

  Evidence-based interventions to reduce intrapartum-related stillbirths and neonatal deaths, with interventions integrated in packages of care for implementation within the continuum of care for mothers and newborns and for service delivery within facilities and communities. Adapted from Kerber et al. [27].

Specific content in each package will vary depending on local epidemiology, for example the level of NMR, the prevalence of HIV/AIDS or malaria, but also with local health system and community capacity, coverage of skilled birth attendance, access to illness management, and human resources. Prenatal care visits are more feasible and can reach higher coverage by being provided close to home in primary care clinics; aspects of prenatal care can even be provided in the home by CHWs. The content of routine prenatal care includes risk screening, birth preparedness counseling, and community mobilization. However, the quality of prenatal care may vary, and data are lacking to track this gap effectively in low-income settings.

The content of childbirth care packages will also vary, ranging from full comprehensive emergency obstetric care at referral-level health facilities to basic emergency obstetric care at first-level health facilities, and skilled community childbirth care at primary care level. The packages of care at birth should include evidence-based interventions for the baby as well as the mother. Where the birth occurs at home, at least simple immediate newborn care and essential newborn care can be provided by family members or other community cadres present. Every skilled birth attendant should be able to provide immediate newborn care—drying, warming, assessment, as well as neonatal resuscitation and stabilization if needed. It is not a rational investment to provide effective obstetric care only to have the baby die for lack of a bag-and-mask device or because someone is not competent in neonatal resuscitation. Yet currently this is the norm in many hospitals, given the fact that at least for the 6 countries for which data are available, the majority of attendants are not trained in neonatal resuscitation and more than half of the facilities do not have bag-and-mask devices (Fig. 3) [17]. Facilities that provide comprehensive emergency obstetric care should also provide emergency newborn care services for ongoing care of babies who are ill with neonatal encephalopathy after intrapartum-related injury or who have complications of preterm birth or neonatal sepsis. The United Nations (UN) has proposed that the 6 core competencies of basic emergency obstetric care be expanded to include neonatal resuscitation, and we support this proposal. A UN manual detailing the indicators for EmOC mentions this proposal, but detailed content and indicators for this have yet to be included [28].

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 3. 

  Coverage of care for low-resource countries showing the missed opportunity between the interaction of a woman with the health system for a given package and the delivery of relevant, effective interventions. Source: New analysis of data from UNICEF [75], Bryce et al. [26], DHS 2000–2007, and Service Provision Assessment Surveys (2003–2008). ⁎ Data from Service Provision Assessment Surveys in: Egypt, Ghana, Kenya, Rwanda, Tanzania, and Uganda. ⁎⁎ Postnatal care is the median from 12 countries based on analysis for Countdown 2008 [54].

In addition to integration of evidence-based intrapartum care interventions within existing packages, there is a need for strategies to link the different levels of the health system in order to reduce delays in receiving referral-level obstetric care [11]. Community mobilization activities may originate via community groups, such as newborn care stakeholder groups and mass media campaigns, and be facilitated by community-based providers including community midwives, CHWs, and TBAs. Important increases in facility birth utilization are possible through such mobilization efforts. In a meta-analysis, the more intensive community mobilization activities were associated with a doubling of skilled birth attendance in a short time (1 to 3 years) [11].

2.3. Delivering interventions: Who can do what? 

Table 2 highlights different delivery channels for interventions to reduce adverse intrapartum-related outcomes, ranging from a generic media/marketing approach through to the most skilled provider. A systematic approach is required to define the individual roles and responsibilities of all the providers of care, both formal and informal. A key aspect, often missed, is to clarify the interactions between providers and levels of care, as well as with the family and community. Appendices 1a, 1b, and 1c (available in the online version) provide a matrix of shared responsibilities, outlining in more detail the tasks for each actor (the woman, through community leaders, community cadres, facility-based cadres, and policymakers) during pregnancy, childbirth, and the postnatal period.

Table 2. Delivery channels for interventions to reduce intrapartum-related adverse outcomes.

⁎ Promotion ✓ Care provision +/− Situational/more evidence required.

Adapted from Lancet Maternal Survival Series, Campbell and Graham [89], and Bhutta et al. [90].

The single most critical person for effective care at the time of birth is the midwife. Expectations and competencies for the skilled birth attendant have been defined in a joint consensus statement by the International Federation of Gynecology and Obstetrics (FIGO), the International Confederation of Midwives (ICM), and the World Health Organization (WHO) [29]. However, data from national service provision assessments in 6 African countries showed that 72%–93% of such attendants were not trained in resuscitation, and basic resuscitation equipment such as bag-and-mask was missing from 53%–84% of facilities providing care at birth—a major missed opportunity for high-impact care [17]. Program reports of neonatal resuscitation being undertaken by community midwives, CHWs, and TBAs [17] warrants further research for the effectiveness of these cadres of health workers in resuscitating the non-breathing baby.

An ongoing debate is whether it should be standard practice to have more than one provider present at birth—one to look after the mother and delivery of the placenta, and another for the baby. In most high-income countries it is standard practice for a midwife to have an assistant at the time of the actual birth, although in some countries this may be another cadre of worker teaming with the midwife. In reality, this sometimes leads to the replacement of the skilled provider with another worker with lesser skills, especially during night shifts [30]. This principle of having two providers at the birth has also been applied in the most intensive home-based care studies—a CHW who is competent in resuscitation cared for the baby while another attendant looked after the mother [19]. Although a second worker makes intuitive sense since effective care simultaneously for mother and newborn even in a normal birth is challenging, there are no data available to support a measured benefit apart from some process data to support the value of a supportive companion for the mother [10], [31]. However, the counter-argument is that a standard of two attendants is not feasible, given the global gap for even one skilled attendant at 45 million births a year. For very high NMR settings (>45) where the density of skilled birth attendants is only 0.7 per 1000 population, urgent attention is needed to train and retain more midwives and to evaluate work load, roles, and identify opportunities for task shifting with paired workers who may not be midwives.

Simpler tasks such as clean childbirth care and simple immediate newborn care (drying, warming, and putting the baby skin-to-skin with the mother) can potentially be provided by any helper present at birth who is given the appropriate information and skills before the birth—even family members who may be the only attendant at 32% of births in Africa and 14% in South Asia [13].

For certain interventions, the skill level required may be so high that task shifting can only pass from one skill level to the next skill level directly below it, and with built-in supervision. For example, comprehensive emergency obstetric care is typically provided by obstetrician medical officers in referral-level health facilities. Task shifting would expand the capacity of the next level of providers, including nurses or medical assistants, to perform aspects of obstetric care, such as cesarean deliveries [32]. Evaluations at scale in Malawi, Mozambique, Senegal, and Tanzania suggest that this is feasible, safe, and cost-effective [33], [34], [35], [36]. Success for such task shifting depends on effective linkages with higher levels of the health system and effective supervision [10], [37]. In fact, in 28 district hospitals in Malawi, non-physician clinicians carried out 90% of cesarean deliveries in a 3-month study period: 70% of those for emergency indications were with subtotal hysterectomy, 60% of those combined with total hysterectomy, and 89% of those combined with repair of uterine rupture [34]. Similarly, basic emergency obstetric care procedures, such as vacuum extraction, have been performed by midwives in many countries. In conflict and post-conflict settings, there are small-scale examples of community extension workers undertaking such procedures, for example in refugee camps in Burma [10], [38].

Back to Article Outline

3. Health system performance and context 

3.1. Health system performance categories 

There is enormous variation in healthcare systems around the world, particularly for care at the time of childbirth. Classifying the strength and performance of a country health system requires data. The WHO health system framework describes health systems in terms of 6 core building blocks: service delivery; health workforce; information; medical products, vaccines and technologies; financing; and leadership and governance (stewardship) [39]. Collectively these affect access, quality, affordability, and safety of personal and public health services, and subsequently, coverage of high impact interventions, theoretically resulting in improved health outcomes. Whilst this describes the components of the system, it does not capture how they articulate and perform in particular settings to produce the required outcomes. There have been many attempts to measure health systems performance and a detailed review of this is not an aim of this paper. However, we note that most previous attempts have focused on national economic indicators such as Gross National Income per capita (GNI) [40], but there are several outliers, for example low-income countries with high performance based on mortality outcomes, coverage of care, and equity, such as Cuba, Thailand, and Sri Lanka [41]. Conversely, there are also countries with high income and relatively poor performance such as South Africa and the USA [41].

The performance of the health system should be measured primarily by the effect on mortality and, for the purpose of this series, the focus is the reduction of intrapartum-related mortality. The key process target of relevance is timely access to high-quality care for every birth. Possible mortality outcome indicators may include child mortality, maternal mortality, or neonatal mortality. Stillbirth rates would also be a useful marker of mortality outcome, but are frequently unavailable or unreliable at population level [4]. Under-five mortality is an equally important mortality outcome, but it is possible to substantially reduce under-five deaths through interventions that are less dependent on health systems function such as oral rehydration solution, immunization campaigns, and bed nets. Facility-based care may only become more critical as under-five deaths fall to lower levels [41]. Hence, under-five mortality appears to be a less discriminatory outcome measure of more complex health system function necessary for effective care at birth. Maternal mortality is a newly accepted indicator of human rights [42], but is also clearly related to effective care at birth. However, routine measurement of maternal mortality ratio (MMR) is challenging [5] and may not be a simple choice to measure health system performance. As shown in Fig. 1 and described in the first paper in this series [1], there is a close relationship between MMR and NMR. Given that NMR is nearly 10 times more common, and is measured routinely in Demographic and Health Surveys, we propose that NMR is a useful surrogate marker of health systems performance to track mortality outcome for care at childbirth. Intrapartum-related NMR or intrapartum stillbirth rate or a composite of the two may be an even more sensitive mortality indicator for childbirth care [43]. However, these data are less available, especially at population level, and are more complex (with different denominators for the two components) than NMR. We endorse the UNFPA proposal of this composite (intrapartum stillbirth and day 1 neonatal mortality) indicator as a useful marker of quality of facility-based childbirth care, as discussed under innovative data needs later in this paper, although, further evaluation of validity is required [43]. Skilled attendance at birth has also been used as a discriminatory marker of access to care, progress toward a comprehensive health system, and of equity, and is also widely measured and available [41]. NMR, and also MMR, however, is inversely associated with skilled attendance at birth, but this association cannot be considered here to be causal since the data is ecological (Fig. 1).

Hence, using one outcome indicator (NMR), we allocated 193 countries of the world into 5 “Categories” of NMR with an even distribution of countries per band apart from the top band (NMR >45), which has only 18 countries but still includes nearly one-fifth of global annual births (20 million). The same Categories of NMR have been used previously for setting priorities for neonatal mortality reduction in varying contexts [44], [45]. Across these 5 NMR Categories shown in Fig. 1, there are also variations, up to about 90-fold, in maternal mortality ratio. Furthermore the cause-specific neonatal mortality varies. For example, in Category 5 countries (NMR >45 per 1000), around half of neonatal deaths are due to infections and the intrapartum-related NMR is around 12 per 1000, which is 20 times higher than in the Category 1 settings with the lowest mortality [1]. As described previously, these Categories act as a useful marker of variation of causes as well as magnitude of mortality. For policymakers with NMR data but no reliable cause of death data, these Categories may provide a simple surrogate that is better than no data at all [7].

The 5 Categories of NMR provide a sensitive indication of health system functioning during childbirth, including the predominant place of birth, person(s) attending the majority of births and providing essential newborn care, and availability of emergency and neonatal intensive care (Table 3). For example, the median cesarean delivery rate ranges from 17% in Category 1 (NMR ≤5), to 3% in Category 5 (NMR ≥45). Likewise, the density of skilled personnel is over 20-times higher in Category 1 (10.9 per 1000) compared with Category 4 (0.9 per 1000) and Category 5 (0.7 per 1000), the latter two of which are less than half that recommended by WHO of 2 per 1000 (Table 3).

Table 3. Countries (193) organized according to 5 Categories of neonatal mortality as a marker of health system performance, showing the variation in skilled birth attendance, health system capacity, and density of skilled human resources.

Abbreviation: NMR, neonatal mortality rate.

Country groupings by level of NMR adapted from Lawn et al. [7], and updated for 2009. Sources: maternal mortality rate [5]; intrapartum stillbirth rate [2]; NMR [75]; and revised neonatal cause-specific estimates for Countdown 2008 [26] based on methods from Lawn et al. [76]. The skilled birth attendance is based on median, and the range is reported in Lawn et al. [1].

In addition, the quality and quantity of health information varies with these 5 Categories, since the Category 1 settings have full availability of vital registration data, often recently updated, but the Category 5 settings are dependent on intermittent household surveys and verbal autopsy data for cause-of-death.

It should be noted that this approach to define a context for priority setting is not restricted to national analyses. For example, in some countries the urban capital city population may have entirely different outcomes (NMR) and health systems performance than in poor rural areas or in urban slums. Thus, while these 5 Categories were used to categorize countries, we recognize that they oversimplify the variable situations encountered in any given country and can also be applied to sub-national populations if the NMR data are available.

Back to Article Outline

4. Prioritization, phasing, and potential lives saved 

4.1. Principles for data-based prioritization for planning 

Priorities and the menu of feasible solutions will vary across settings. Where data and a systematic decision-making process are lacking, it is more likely that choices will be swayed by the loudest voices, whether these are local or international, rather than by the true problems and priorities of the setting and local communities. A reproducible approach to defining the levels of health system performance by NMR is discriminatory and useful for priority setting and a first step in a transparent and data-driven approach to setting priorities. In all health systems, there are missed opportunities for providing life-saving interventions; thus, we focus first on addressing missed opportunities at facility and at community level. Given the major global gap for care at birth, we also examine priorities to close this gap.

The top priority should surely be given to interventions with the highest mortality impact that are also affordable and feasible, and improve equity [46]. Hence, we present modeled estimates of lives saved to inform the decision-making process. One of the reasons for ongoing lack of prioritization and investment in intrapartum care may be that advocacy has focused primarily on maternal deaths alone [47]. Linking in the numbers of intrapartum-related stillbirths and intrapartum-related neonatal deaths increases the amount of deaths that are modeled by a quantum, and also allows for a broader analysis of the lives saved using the Lives Saved Tool (LiST) modeling software. LiST is a new module in Spectrum software based on The Lancet [48] Neonatal Survival [25] and Nutrition series modeling of lives saved. While many of the interventions to reduce maternal mortality and stillbirths are already included, however, an important limitation of the current version of LiST is that the output estimates for maternal lives saved and stillbirths averted are not yet available (see Panel 1). Thus, this analysis is based on national modeling using the most recent neonatal mortality rate and cause-of-death estimates, and applying mortality effect estimates in the Lives Saved Tool (LiST) whilst taking into account current national coverage of interventions. The details of the inputs and modeling are described in Panel 1, and the results are displayed in Table 4.

Table 4. Estimates of neonatal deaths averted from intrapartum-related events for countries according to 5 Categories of neonatal mortality.

Source: New analysis using data from UNICEF [75], WHO [91]. Skilled birth attendance is median percentage for each band of countries.

Modeling and inputs and methods described in Panel 1. Upper and lower range of lives saved shown for estimates with single package or intervention, based on 95% CI of meta-analysis or inter-quartile range of Delphi panel responses.

4.2. Identify and address missed opportunities within existing programs 

The most obvious way to increase effective coverage and reduce deaths is to identify a programmatic platform that reaches the target population at the critical times (i.e. during pregnancy, childbirth, and in the immediate postnatal period) and addresses missed opportunities to provide packages of cost-effective interventions at the same contact [25], [46], [49]. To reduce intrapartum stillbirths and intrapartum-related neonatal deaths, some of the key interventions during pregnancy could be provided through prenatal care contacts, yet whilst many women come for one prenatal visit, fewer come for 4 visits and quality of care may be lacking (Fig. 3). It appears that some procedures such as blood pressure monitoring are carried out for almost all women, but it is unknown whether these measurements are taken properly and, even if properly performed, whether this information then leads to the appropriate case management should hypertensive disease of pregnancy be detected. Data for the details of effective care at the time of birth are almost entirely lacking—use of the partograph and tracking fetal heart rate are unknown and may not be amenable to questions in retrospective surveys [50]. While cesarean delivery data are increasingly available [51], it is unknown if the woman who received the procedure had a medical indication or not, or if so, whether the surgical intervention was timely [52] or was associated with complications [53]. The limited data on neonatal resuscitation provision suggest that there is an important missed opportunity whereby few skilled attendants are trained and equipped to provide this core intervention. There is even more limited data for postnatal care, particularly for a visit within 2 days (a global Countdown to 2015 indicator); comparable data are currently available for only 12 countries [54].

4.3. Potential for lives saved through addressing missed opportunities during contact points 

Ensuring that every skilled attendant at birth can resuscitate the non-breathing newborn is a critical and achievable priority [17]. If coverage with neonatal resuscitation were increased to 90% for all current facility births alone (not including births outside facilities), 93700 intrapartum deaths would be averted globally each year. The majority (n=67500) would be prevented in the two highest mortality settings (Categories 4 and 5, NMR >30) where rates of skilled attendance at delivery are about 50%; an additional 21900 deaths would be prevented in Category 3 (88% skilled attendance at delivery) settings. Provision of comprehensive emergency obstetric care is more challenging in weak health systems; however, if achieved for 90% of the deliveries that already occur in facilities, 232500 neonatal lives could be saved each year, or 26% of all intrapartum-related neonatal deaths. The majority of deaths averted (n=175500) are in the two highest mortality settings (Categories 4 and 5, NMR >30), and an additional 48000 deaths would be averted in Category 3 settings. Thus, the greatest mortality effect is observed in the moderate and high NMR Categories, where the burden of deaths is concentrated, and the quality of childbirth care is deficient among babies who are born in health facilities.

There are also missed opportunities at the community level. For the 60 million home births, the person present at birth can at least dry and warm the baby, thereby providing stimulation and thermal control, and put the baby skin-to-skin with the mother. In one program in a poor, rural district in India, within a short time frame the community changed practices of delivering the baby onto the floor, to instead receiving the baby onto clean hands, wrapping the baby and putting the baby skin-to-skin. This change, along with others was associated with a halving of neonatal mortality in 18 months [55]. In settings where there are CHWs or health extension workers at scale, especially for communities with low access to facility care at the time of birth, options for community-based care can be explored, always bearing in mind the goal that these steps should lead toward a more comprehensive health systems solution. Simple immediate newborn care provided at all home births at 90% coverage is estimated to avert 41700 (5%) of intrapartum-related neonatal deaths globally; the bulk of the effect (93%) would occur in Category 4 and 5 settings, where most deliveries occur in the home. Some countries are considering national scale up of community-based neonatal resuscitation. For example, in Nepal where over 80% of 796000 annual births are at home and there is a strong community-based health care system, introduction of community-based neonatal resuscitation could save an estimated 1000 lives a year, based on LiST analysis (see Panel 1).

4.4. Invest to reduce major gaps in coverage for care at birth 

Even if the health system performed to full effectiveness for every interaction that occurs antenatally, intrapartum, and postnatally between pregnant women and newborns with health system providers there remains a huge gap in coverage. Whilst 25% of women in Category 4 and 5 settings do not access prenatal care at all, a staggering 60 million a year give birth outside facilities [1]. Improving quality of care in facilities may draw more women to give birth in hospital; however, countries with high percentages of women giving birth at home urgently need to develop strategies that address both supply and demand factors to increase effective care at birth.

Countries must prioritize and commit to allocating the necessary financial resources to increase the supply of obstetric and newborn care. Several recent analyses have demonstrated the cost-effectiveness of obstetric-neonatal care packages [25], [46], [56]. In The Lancet Neonatal Survival Series, expansion to full global coverage with skilled maternal and immediate neonatal care was estimated to cost $1.8 billion international dollars and avert approximately 100 million DALYS, and met the cost-effectiveness thresholds of the Commission of Macroeconomics and Health [25]. Full coverage with emergency obstetric care packages cost $2.8 billion international dollars and averted 150 million DALYS, also meeting the criteria for cost-effectiveness. When skilled childbirth care, immediate newborn care, and emergency obstetric care are bundled as a package, cost is lowered and cost-effectiveness improved. Full coverage (90%) of emergency obstetric care was estimated to cost between $1.44–2.81 billion international dollars and avert 13%–24% of all neonatal deaths, mainly by reduction in intrapartum-related neonatal mortality [46].

Preliminary evidence suggests that removing financial hindrances and providing financial incentives may substantially increase demand for obstetric care. In Ghana, a rapid increase in facility birth coverage was observed following a Presidential decree removing charges for any care at birth or in the first month of life; institutional deliveries increased by 19% and skilled birth attendance by 14%–17% and likely more since the survey was carried out [57]. In India, the Janani Suraksha Yojna program has paid cash incentives to cover institutional delivery and transport. As a result, the total number of institutional births increased throughout the country, with some large states showing an unprecedented jump (9% in Orissa, 15% in Rajasthan, and 18% in Madhya Pradesh) between 2004 and 2007–2008 [58], [59]. Although the impact on mortality outcomes is unknown, it is critical that quality maternal and newborn care is ensured.

4.5. Potential for lives saved at high (90%) coverage 

Our new analysis for 193 countries suggests that if comprehensive emergency obstetric care was provided to 90% of all women in labor (regardless of place of birth) and was effectively implemented, an estimated 495000 neonates per year could be saved that are currently dying of intrapartum-related causes. If neonatal resuscitation were to reach 90% coverage of all births, around 192000 babies' lives could be saved. If the two packages were combined, with 90% coverage of comprehensive emergency obstetric care and neonatal resuscitation, as well as case management of babies with neonatal encephalopathy, then up to 613000 newborn lives could be saved every year (Table 4). The majority (83%) of neonatal deaths averted would be from Categories 4 and 5, respectively, and 14% would be prevented in Category 3 settings.

An analysis of care packages in Sub-Saharan Africa and South Asia found that both skilled maternal and immediate newborn care as well as emergency obstetric care were highly cost-effective in both settings and highly recommended for universal scale up [56].

4.6. Priorities to reduce intrapartum-related burden, by the 5 Categories 

In Categories 1 and 2, where the burden of intrapartum deaths is low, there is still a substantial burden of intrapartum-related morbidity. Key priorities include improved measurement of chronic disability and impairment, follow-up of long-term outcomes of disabled survivors, and the development and evaluation of early intervention programs and interventions to reduce the impact of chronic disability (Fig. 4). The prevention of risk factors such as adolescent pregnancy, and unhealthy behaviors may also be targeted to reduce future burden. In Category 3 where skilled attendance is relatively high, ensuring timely access to comprehensive emergency obstetric care and neonatal resuscitation, as well as providing post-resuscitation care is also a priority although may be lower impact, including improved recognition and basic management of neonatal encephalopathy with thermal and supportive care and referral to district or referral level facilities with neonatal intensive care facilities.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 4. 

  Priorities to reduce intrapartum-related burden, for countries categorized by 5 Categories of neonatal mortality rate (NMR) as a marker of health system performance, showing linking strategies to increase effective coverage.

• Adapted from Knippenberg et al. [44], Lawn and Kerber [77].

In Categories 4 and 5, programs may first address the missed opportunities for the approximately 50% of deliveries already occurring in facilities, and optimize coverage of prenatal, emergency obstetric, and neonatal care at district or referral-level hospitals. Achieving universal skilled attendance requires specific human resource plans and, simultaneously, increasing demand for skilled birth attendance may be facilitated by community mobilization efforts, financial incentives, communication and transport systems, and community birthing centers [11]. Other potential linking strategies that may deserve further evaluation include maternity waiting homes and targeted risk screening for pregnancy complications as opposed to prevalent maternal prepregnancy risk factors. For the approximately 50% of births that occur in the home, home-based care may be considered by a range of cadres for primary prevention, and even potentially neonatal resuscitation, as discussed in detail in the fifth paper in this series. Priorities must also include the training of skilled personnel and improvement of the quality and capacity of facilities for obstetric and neonatal care.

One of the strengths of the 5 Categories is the emphasis on continual health systems performance improvement, moving from one category to the next by context-specific, strategic strengthening of the health systems. While making the best of the immediate resources and opportunities, the countries must invest and act now to attain the long-term goal of universal coverage of effective care at birth (Fig. 4). High mortality settings (Categories 4 and 5) are working to move toward transitional (Categories 2 and 3) and then to lower mortality (Category 1).

Back to Article Outline

5. Considerations for policy, planning, and implementation 

5.1. National policy 

Program managers and policy makers require more than a global review of effectiveness, or even cost-effectiveness, to decide which strategies and specific interventions will be the most successful in their context. There is no magic “one size fits all” program to address intrapartum-related mortality; the local epidemiology (mortality rates, causes, and risk factors), as well as health system design and performance, financing, and community demand are key factors to consider [44].

More countries now have integrated MNCH strategies, laying out multi-year plans for scaling up key packages and the strategic approach to doing so. This is especially encouraging in large, high-burden countries such as Bangladesh, Nigeria, Tanzania, and Pakistan. The Pakistan MNCH integration and national plan was to a large extent driven by the recognition that perinatal deaths and intrapartum-related neonatal mortality could only be reduced by integrating skilled maternal and newborn care at all levels of the health system [60]. More commonly, countries have a health sector plan that includes some pages on maternal and child health and less commonly on neonatal interventions, given the recognition that to reduce this does require some specific planning and attention [61]. Stillbirths still generally remain invisible and fall outside national health plans.

Few countries have a specific national perinatal strategy, let alone a comprehensive strategy to address intrapartum-related complications. In 2008, the government of Bangladesh identified neonatal health as a critical problem gap and obstacle to meeting Millennium Development Goal 4 and developed a national neonatal health strategy via a collaborative consensus building process between the Ministry of Health, UN agencies, several NGOs, professional organizations, and the private sector. “Birth asphyxia” was identified as a top priority condition and a road map was developed to address intrapartum-related birth complications at all levels of the health system from the community to union-level and district-level hospitals, as well as along the continuum of pregnancy care, from prepregnancy, prenatal, intrapartum, and immediate postnatal care (Panel 2, at the end of the article).

However, a common theme even among those countries who have national strategic plans is a lack of either an implementation plan or a clear process to enable implementation in a decentralized, locally contextualized manner, for example at district level. Countries such as Tanzania and Ghana that have strong district health decision-making tools and processes appear to be making progress in reducing deaths [62].

5.2. Financing 

Financial constraints are a critical barrier to women coming to facilities to give birth and can also be a major determining factor in lack of access or major delays in having a cesarean delivery [63]. The countries with the highest burden of perinatal mortality and the lowest performing health systems have an average GNI of only US $440; these are the poorest countries in the world (Table 5). Yet in these very countries, on average, half the costs for health care are paid out of the pockets of these extremely poor families, compared with less than 20% of the costs on average in the lowest burden/highest health system performance countries [11]. Government spending is only US $20 per capita, compared with the estimated US $34 required for a minimum package of health services [40].

Table 5. Countries (193) organized according to 5 Categories of neonatal mortality, as a marker of health system performance showing the variation in gross national income per capita (GNI) and health system financing from government and donor sources.

Source: New analysis using data from Greco et al. [65], UNICEF [75], WHO [91].

Skilled birth attendance is median percentage for each band of countries.

⁎ The Abuja target calls for African countries to allocate 15% of national budgets to health.

Donor assistance for health has grown dramatically in the last 20 years, increasing from US $5.6 billion in 1990 to US $21.8 billion in 2007 [64]; donor spending has also increased by 66% for maternal and neonatal health and 63% for child health, resulting in nearly a US $1.4 billion increase from 2003 to 2006 [65]. However, both these analyses stress that the major increase has been for “vertical,” disease-specific, commodity-driven interventions such as for HIV/AIDS, malaria, and immunizations. The investment in health systems and especially in care at the time of birth has not been commensurate to burden, or to the potential for lives saved. In addition, our analysis of Overseas Development Assistance (ODA) across the 5 NMR Categories suggests that the Category 5 countries with the highest mortality and lowest health system performance for intrapartum care receive lower ODA per birth for maternal/newborn care and lower ODA per child than the Category 3 and 4 countries (Table 5). This may be related to the countries that are post-conflict or have governance challenges, but does raise the urgent need for more attention to MNCH care for the highest burden settings [10].

One novel approach to donor and government inputs is performance-based funding, which is defined as “transfer of money or goods conditional on taking a measurable action or achieving a predetermined performance target” [66]. To date, use of performance-based financing at scale has mainly been through the Global Fund to Fight TB, HIV/AIDS and malaria, and by the Global Alliance for Vaccines Initiative (GAVI). This strategy has potential to increase health system focus on meeting targets for increased coverage and quality of skilled care at birth. In Rwanda, increased institutional birth coverage was one of the targets set for performance-based funding and rates almost doubled (from 12% to 23%) in pilot provinces, whilst provinces with traditional input-based financing only saw a rise from 7% to 10% [66]. Given major increases in funding, more commitments including the first ever G8 statement on MNCH [67], performance-based funding focused on care at birth may help to ensure specific outputs with the greatest potential effect on MNCH.

In addition to supply-side financing, demand-side financing is crucial to remove economic barriers to care seeking. Experience from India and other countries demonstrates a rapid increase in utilization of public facilities for childbirth with conditional cash transfers [68], [69].

Back to Article Outline

6. Innovation to improve tools, evidence, and data 

6.1. Innovation for tools and technologies 

Creative and low-cost technologies may play a critical role in improving access to life-saving, intrapartum care interventions in low-resource settings. Key development needs have been highlighted in panels in several papers in the series. There is an urgent need to develop affordable and durable equipment, such as portable ultrasound devices, fetal heart rate monitors, hand-held Doptone devices, meconium suction devices, sterilizable bulb suction and self-inflating bag-mask devices. Furthermore, existing technology can be modified or utilization can be increased; for example, through the adaptation of bicycles or motorcycles for use as ambulances, or the promotion of cellular phone usage, supported by affordable and user-friendly software by community birth attendants to communicate with health centers and facilitate referral. These are discussed in more detail in second, third, and fifth papers in this series [10], [13], [17].

6.2. Improving the data for decision making 

In order to guide policy making to monitor intrapartum-related events, there is an urgent need to refine and standardize indicators of intrapartum-related health outcomes, both fatal and non-fatal, as well as process indicators. Table 6 highlights some key considerations in developing improved and feasible indicators for use in low-income settings.

Table 6. Improving the data for decision making for intrapartum care.

Improved measurement of outcome data
Intrapartum stillbirths
• Improved measurement of the numbers/rates of stillbirths, especially in settings where most births occur at home and/or where stillbirths are normally a taboo subject
• Consistent definitions and classification systems to allow comparability of causes of death measurement across low- and high-income settings
• Tools to assess the causes of stillbirths, and to better distinguish intrapartum stillbirths versus antepartum stillbirths versus intrapartum-related neonatal deaths e.g. through verbal autopsy
• Linking to other data collection mechanisms e.g. vital registration, household surveys, demographic surveillance systems
Intrapartum-related neonatal deaths
• Improved measurement of intrapartum-related outcomes (mortality and morbidity)
• Consistent definitions and classification systems to allow comparability of measurement of intrapartum-related neonatal outcomes across low- and high-income settings
• Verbal autopsy tools and hierarchical methods to distinguish intrapartum-related neonatal deaths from other causes of very early death such as early-onset sepsis and preterm birth
Combined marker of intrapartum-related stillbirths and neonatal deaths, and/or intrapartum-related maternal deaths
• Validation of a composite indicator of quality of intrapartum care [43] e.g. intrapartum stillbirths plus first day (or predischarge if earlier) neonatal deaths >2000 g as a surrogate for intrapartum-related neonatal deaths, consider addition of intrapartum-related maternal deaths
• Classification systems to cross tabulate stillbirth and neonatal outcomes with maternal deaths, complications, and risk factors
Impairment and disability
• Feasible case definitions for neonatal encephalopathy in low-income and community settings (for example, surrogate marker proposed is seizures in first 24 hours in neonate with birth weight >2500 g)
• Screening methods (e.g. application of surveillance or screening tool followed by definitive testing of screen positives) for identification of infants at high risk of disability or impairment and who may benefit from early intervention
• Feasible, sustainable instruments to measure disability that are validated at population level to ensure that improved newborn survival (from intrapartum-related or other conditions such as preterm birth or infection) is not contributing to an increase in disability rates
Improvement in measurement of service coverage data
Obstetric care coverage indicators (refinement, consensus, and consistent reporting)
• Attendance at birth
∘ Skilled birth attendance coverage, monitoring of skills, competence, and procedures performed by skilled birth attendants
∘ Place of delivery, and other birth attendants
∘ Cross tabulation by rural/urban and by socioeconomic status
• Emergency Obstetric Care services
∘ Access, utilization, and met need for EmOC services, better determination of baseline marker of ‘need’ in different settings
∘ Consistent definitions of maternal indications, complications, and life-saving interventions
∘ Cesarean deliveries as % of all births: specify those for maternal-fetal indications
• Indicators to track referral systems for obstetric and newborn care from community to facility and between facilities
Neonatal care coverage (refinement, consensus, and consistent reporting)
• Indicators of newborn care at birth: Proportion of facilities with capacity for neonatal resuscitation (training and equipment), proportion of staff competent in neonatal resuscitation, neonates receiving neonatal resuscitation, validation of data collected through facility assessments or through retrospective surveys
• Routine postnatal care: Timing, frequency, cadres, and content of postnatal care visit in facility and at home, validation of data collected through retrospective surveys
• Emergency newborn care: Proportion of facilities with capacity for ongoing care for neonatal encephalopathy (neonatal intensive care, assisted ventilation, nutrition support, and fluid management)

A strong argument can be made to develop a mortality outcome indicator that combines the burden of intrapartum-related injury on the fetus and newborn, given the linked pathophysiology, shared interventions to reduce both intrapartum stillbirths or neonatal deaths, and the frequent misclassification of stillbirths and non-breathing babies. Fauveau [43] suggested a new indicator of the quality of emergency obstetric care, intrapartum case fatality, which is the sum of the late stillbirth rate and first day neonatal deaths and this has recently been added to the UN list of indicators [28]. A more precise indicator of intrapartum-related mortality would be to add the intrapartum stillbirth rate (“fresh” stillbirths >1000 g) to the rate of neonatal deaths among liveborn babies over 2000 g who die in the first 24 hours or prior to hospital discharge. This is likely to be a useful outcome indicator for care at birth that is sensitive enough to reflect quality of intrapartum care, yet common enough to be measurable, and with simple enough data to be feasible. The data in the first paper in this series demonstrates that the vast majority of intrapartum-related deaths in term babies occur within 24 hours of birth. Hence, predischarge data will capture most of these intrapartum-related neonatal deaths and may be feasible even using existing labor ward admission and discharge records. Further validation of this indicator and a standard approach to measuring and reporting the data are required.

Given that the majority of neonatal deaths occur at home, a key challenge is to develop and validate case definitions and classification systems that are feasible to apply consistently in resource-limited settings, and that allow comparability across different settings. Validation of verbal autopsy tools to differentiate stillbirths from livebirths and ascertain cause of death, particularly with hierarchies for categorizing causes of death in the presence of co-morbid conditions is a further challenge. Another critical program question with almost no useable data in low-income settings is regarding long-term disability and impairment after intrapartum-related complications [70]. In order to monitor this outcome, feasible and standard measures for neonatal encephalopathy need to be developed for low-resource, community settings. Furthermore, simple screening methods and tools to measure neurodevelopment and disability need to be developed, standardized, and validated on population-based cohorts in low-income settings.

Improved use of process indicators, including measurement of obstetric and neonatal service coverage and quality, is essential to tracking progress to improved intrapartum health outcomes. The UN process indicators of emergency obstetric care are relatively young and may benefit from refinement and improvement over time, for example the proportion of births in emergency obstetric care facilities or cesarean delivery may be more valuable if they specify the proportion of medically-indicated cases [71], [72]. The indicator, met need for emergency obstetric care, requires clearer guidelines for the definitions of maternal indications and life-saving interventions, as well as clarity in the inclusion of abortion-related care. Consistent information needs to be collected on the quality of the content of interactions between healthcare providers and women during pregnancy, childbirth, and in the postnatal period, in order to quantify missed opportunities for interventions. Certain data, such as the proportion of deliveries that receive intrapartum monitoring or postnatal care, are almost completely lacking. This coverage data may be collected in national Demographic Health Surveys or Service Provision surveys and need to be consistently reported. Furthermore, the measurement of EmOC indicators can and should be linked with neonatal indicators, such as markers of the immediate neonatal care capacity including the proportion of newborns that receive neonatal resuscitation or proportion of skilled birth attendants trained in neonatal resuscitation. Other neonatal indicators may include the capacity of facilities to provide neonatal resuscitation and care for neonatal encephalopathy.

6.3. Implementation research 

A systematic research pipeline has been framed for neonatal health interventions and may be applied for research priorities related to intrapartum-related injury [15]. In this model, research and implementation questions are grouped into 4 major categories:

The pipeline starts with the epidemiologic description and identification of major determinants of intrapartum injury, which is covered in the first paper in this series. Key areas and challenges for future epidemiologic research include the development of consistent case definitions of intrapartum-related neonatal deaths, classification systems, and terminology; comparable cause-specific data across different settings; and identification of reliable data on births and deaths in high-mortality settings. In the discovery phase, mechanisms of disease are investigated to guide the development of new interventions, ranging from elucidating the synergistic inflammatory response of maternal infections and hypoxic brain injury, to neuroprotective mechanisms of xanthine oxidate inhibitors and therapeutic hypothermia. In the development phase, interventions are developed or modified to reduce cost, increase effect or improve deliverability, such as low-cost and durable bag-and-mask resuscitators, wind-up Doptone devices, or CHW training materials. Finally in the delivery phase, existing interventions are implemented in new settings, with appropriate monitoring and evaluation to determine cost-effectiveness and mechanisms for scale up, and to guide the earlier steps of implementation research including the discovery and development science. Delivery research is urgently needed to increase the delivery of evidence-based interventions such as neonatal resuscitation in low-income settings, and to define challenges to effective scale up.

In Table 7, we summarize some of the key research and implementation questions identified in this series of papers. An innovative method to systematically prioritize research questions has been developed by the Child Health and Nutrition Research Initiative (CHNRI) [73], and is being applied to research questions for intrapartum-related neonatal deaths.

Table 7. Research questions organized by the research pipeline of definition, discovery, development, and service delivery.

Obstetric Care
Definition and description	• Refining consistent, feasible approaches to data collection for intrapartum-related outcomes (perinatal mortality, neonatal encephalopathy, intrapartum stillbirths, intrapartum-related neonatal mortality)
	• Differentiating intrapartum-related events from preterm birth complications, especially in verbal autopsy
	• Developing and testing valid and feasible definitions of EmOC service indicators in different settings
Discovery	• Further research on pathophysiology and synergy between maternal infection and hypoxic brain injury, and mechanisms or medications to intervene
	• Mechanisms linking risk factors with intrapartum stillbirths or neonatal deaths
Development	• Development of low-cost hand held ultrasound device to connect with personal computer, simplified Doppler umbilical artery ultrasound
	• Simple training aids for use of fetal heart rate tracking devices and interpretation of findings (pinard stethoscope, Doppler fetal heart rate monitor)
	• Additional user-friendly, low-cost, durable, accurate alternative-powered fetal heart rate monitors capable of detecting late decelerations without need for complex interpretation
	• Oxygen condenser for adults using alternative power
	• Low-cost, durable training mannequins
	• Solar powered lighting for surgery, refrigerator for blood-banking
Delivery	• Rigorous evaluation of major intrapartum obstetric interventions on intrapartum related morbidity and mortality
	• Evaluation with field trials of widespread distribution of vacuum/forceps, hand held Doppler fetal heart rate monitor and effect on intrapartum-related outcomes
	• Trials with symphysiotomy training and kits/training aids
	• Effectiveness of obstetric drills and emergency obstetric training on intrapartum-related outcomes (fatal and non-fatal)
	• Safe methods of delivery induction in low-resource setting with limited monitoring
	• Effect of obstetric safety or delivery checklists on Intrapartum related morbidity and mortality
Neonatal Resuscitation
Definition and description	• Refining consistent, feasible approaches to data collection for intrapartum-related outcomes (perinatal mortality, neonatal encephalopathy, intrapartum stillbirths, intrapartum-related neonatal mortality)
	• Differentiating intrapartum-related events from preterm birth complications, especially in verbal autopsy
	• Developing and testing valid and feasible definitions of resuscitation services indicators in different settings, including follow-up for long-term morbidity
	• Cohort studies to assess the long-term (2, 5, 10 year) outcome of neonatal encephalopathy at the community level. Do resuscitation programs reduce mortality but increase survivors with chronic disability?
Discovery	• Novel therapeutic interventions for treatment of babies with neonatal encephalopathy
Development	• Development/refining of more robust, simple, low-cost resuscitation equipment for use at the community level
	• Oxygen condenser for neonates using alternative power
	• Low cost, durable, pulse oximeters, with alternative power options
	• Low-cost, durable training mannequins
	• Develop and assess lower cost approaches for therapeutic hypothermia
	• Effect of oropharyngeal suctioning performed routinely for the non-vigorous infant with meconium staining
	• What is the effectiveness of a clean cloth on a finger to clear the mouth at birth?
Delivery	• Validation of an algorithm to recognize and resuscitate a non-breathing baby in the facility and in the community? What are the sensitivity and specificity and effect of these methods?
	• What are key selection criteria for successful providers of neonatal resuscitation? What are the best methods for testing competency?
	• Refining and testing the training process - key content and methods of resuscitation for lower-skill cadres and in the community, especially if they are illiterate, and frequency of re-training and supervision
Linking Families and Facilities
Definition and description	• Developing and validating feasible indicators to monitor the functioning of referral and transport systems, community financing, birth preparedness, levels of community mobilization
	• Testing simple, feasible and more accurate risk screening algorithms with better predictive value focusing on pregnancy complications as opposed to prevalent maternal prepregnancy risk factors for use at the community setting and/or first level facilities
Development	• Assessing the role of new technologies for communication for arranging transport, and/or management of simple conditions in the community (mobile phones, GPS, satellite phones, internet, and options without electricity - solar powered systems, longer lasting batteries etc.)
	• Interactive methods and teaching tools for community mobilization activities, such as participatory women's group and community newborn stakeholder meetings (folk songs, puzzles, games)
	• Feasible means for CHWs or TBAs to identify maternal /fetal/neonatal complications during labor (such as a pictorial partogram). If so, will earlier referral and improved outcomes result?
	• What are strategies to promote community acceptance of and involvement in maternity waiting homes? Should risk-based approaches be used to determine who is referred?
Delivery	• The impact and cost of community mobilization, emergency transport systems, financial strategies, risk screening algorithms, and maternity waiting homes on perinatal-neonatal outcomes, and specifically intrapartum-related mortality (intrapartum stillbirth, intrapartum related neonatal death, or early neonatal mortality)
	• What are effective methods to communicate a list of danger signs for the mother and the newborn? What are effective methods to communicate danger signs with alternate health cadres?
	• Sustainability of community mobilization, - the optimal ratio of ‘mobilizers’: population and at what frequency do contacts need to occur? What are pathways for community mobilization to be taken to national scale?
	• How to sustain costs of fuel and repairs for emergency transport vehicles?
	• If women are identified to be at major risk, will they be able to act on this and access facilities with emergency obstetric and newborn care at the appropriate time? If not, what are the barriers to them doing so (e.g. money, time, other children, social acceptability)?
	• Assess social marketing to promote demand for community birthing centers or maternity waiting homes
Delivering Care at Birth in Community Settings
Definition and description	• Increased understanding of traditional practices at birth and after birth that are commonly performed in different contexts-and may be amenable to change
Development	• Incentives for attracting and retaining skilled attendants to work in isolated birthing centers
	• Content and methods for training TBAs and CHWs in newborn resuscitation should be locally appropriate
	• Content, cost, and sustainability of clean delivery kit (target families) and home birth kit (skilled birth attendance or community cadre)
	• Effectiveness of “picture partographs” in assisting TBAs and CHWs in referring women with complications earlier or other methods to specifically improve TBA recognition and early referral of complications
	• Content and methods for training CHWs and TBAs in newborn resuscitation
	• Action for skilled birth attendant, CHW or TBA to do to stabilize a baby with “asphyxia” and assist in transport/referral
Delivery	• The effect and cost on maternal and fetal-neonatal outcomes of birthing centers, skilled birth attendants, TBAs, CHWs established within the community
	• Assessment of whether BEmOC skills can be safely, effectively, and feasibly done in the home setting vs facility settings
	• What are the outcomes for home-based BEmOC procedures?
	• Does CHW or TBA training with adequate on-going support result in a significant impact on the incidence of asphyxia-related fetal and neonatal mortality?
	• Does provision of community-based resuscitation reduce NMR without increasing disability?
	• What can be achieved at community level for babies with neonatal encephalopathy if referral is not an option?
	• What is the relative contribution of improved management of labor compared with neonatal resuscitation on PMR?
	• What is the social acceptability of community birthing centers, particularly in cultures where childbirth is a private family affair?
	• What are the key selection criteria for “successful” SBAs, TBAs, CHWs (e.g. age, literacy, number of deliveries per year)?
Perinatal Audit
Development	• Development and testing of core perinatal audit module with paper-based and computer database that can be adapted and used in varying settings with comparable outputs with consistent results for cause of death and avoidable factors
Delivery	• Effect, cost and enabling factors for wide scale, sustainable implementation of audit
	• Linking perinatal audit with maternal audit, and/or with child mortality audit – feasibility, practical issues
	• Approaches to mortality audit meeting to overcome challenges of “blame” and “loss of face”
	• Further studies of community audit and models for joint community/facility audit

Back to Article Outline

7. Conclusions 

Care at birth is a sensitive marker of a responsive health system. There is a range of intervention strategies to select from to address intrapartum stillbirths and intrapartum-related neonatal deaths, and almost all of these also benefit maternal survival and health. The call for new global attention to improve and scale up intrapartum care is clear, given the number of deaths, the lack of progress in the last two decades, and the feasibility of action at surprisingly low cost per life saved (Table 8). Global cries are empty, however, without local action. Every family, every facility, and every community member can play a role (see online appendix), but policymakers and clinicians responsible for care at the time of birth have a particular responsibility in leading change. New attention for MNCH health systems strengthening by the G8, even in the midst of global economic crisis, is encouraging [67].

Table 8. Key messages for evidence-based action to reduce intrapartum-related burden.

Paper	Problem	Policy and program actions
1	Intrapartum-related neonatal deaths: 904000	• More visibility of numbers, linking this burden with maternal health advocacy, better use of existing data to set priorities and track progress.
	Intrapartum stillbirths: 1.02 million	• Consistency in terminology and classification systems.
	Neonatal encephalopathy: uncertain	• Improve the data for pregnancy tracking and perinatal outcomes, especially intrapartum-related outcomes, particularly in low-income and community settings, including:
	Intrapartum-related maternal deaths: ~225000	∘ use of NMR Category as marker of health system performance to help set priorities for action;
		∘ new indicators for intrapartum care, including aggregate measure of intrapartum stillbirths and intrapartum-related neonatal deaths;
		∘ measurement of impairment and disability related to intrapartum-related injury.
2	Missed opportunities for effective obstetric care for facility births	Identify and address missed opportunities – quality gap
	Coverage of care at the time of birth is low, progress for scale up of skilled birth attendance is slow, and there is a gap in care for rural and poor populations, especially for cesarean delivery and EmOC.	• Intrapartum care that is focused on the highest impact components, but feasible for low-income settings.
		• Improve quality: training and drills, checklists, audit.
		• Innovative and robust equipment.
		Invest to close the coverage gap
		• Innovative task sharing with good supervision.
		• Strategic attention to solutions for care at birth for 60 million non facility births.
3	Lack of resuscitation and post resuscitation care, even for those born in facilities in low-income countries	Identify and address missed opportunities – quality
		• All births to have someone who can dry and stimulate/rub the newborn.
	Major gap for home births	• Basic resuscitation available for all facility births, especially at primary care level.
		• Post-resuscitation care package at district hospital and above.
		• Innovative and robust equipment.
		Invest to close the coverage gap
		• Strategic attention to solutions for resuscitation for 60 million non-facility births.
4	Delays	Increasing community demand for obstetric care
	1. Decision to seek care	• Community mobilization to increase birth preparedness, recognition of danger signs, and obstetric care seeking.
	2. Transportation to facility	• Financial strategies to reduce barriers to care seeking and provide incentives for obstetric care.
	3. Receiving effective care at facility	Formal healthcare system outreach towards community
		• Community referral and transport schemes to reduce transportation delays.
		• Risk screening to bring high-risk women/babies closer to skilled care, need for new algorithms focusing on obstetric complications at high-risk for intrapartum-related injury (as opposed to old strategies using prepregnancy maternal characteristics).
		• Maternity waiting homes to bring higher risk mothers closer to skilled obstetric care.
5	Gap for service provision	• Training of skilled birth attendants with capacity to provide intrapartum monitoring, BEmOC, and neonatal resuscitation in community settings; creation of community birthing centers.
	−60 million home births	• Training TBAs in primary and potentially secondary prevention and encouraging partnership with the formal health system.
	Lack of demand	• Integrated home-based care packages by CHWs, including community mobilization and pregnancy/delivery care with focus on primary and secondary prevention.
	Policy conflict regarding use of community cadres	• Linking with community cadres as part of process to build stronger, integrated health systems.
6	Missed opportunities, delays lack of accountability	• Audit as a tool to improve quality of care and accountability.
		• Only effective if the data links to action.
7	Weak health systems	• Integration of MNCH – with focus on high impact care at high coverage.
		• Variation of solutions based on context (mortality level, health system capacity).
		• Identify and address missed opportunities in existing health system packages (e.g. adding resuscitation to obstetric care).
		• Invest to reduce major gaps in coverage for care at birth.
		• Innovate – new tools and technologies and new delivery strategies to extend the reach of the health system.
		• Validate and promote an indicator of quality of intrapartum care with a composite indicator of intrapartum stillbirths and intrapartum-related neonatal deaths.
Overall message: The most sensitive test of a health system is provision of effective care at the time of birth and the ability to respond quickly to intrapartum emergencies for mother and/or baby; addressing missed opportunities increases the quality of current facility care, and closing the gap for 60 million non-facility births is critical for accelerating progress toward achieving MDGs 4 and 5.

Addressing missed opportunities to improve quality of care for current facility births is critical and immediately feasible—ensuring effective emergency obstetric care and neonatal resuscitation could save 327200 neonates every year. There are also missed opportunities at community level where simple immediate newborn care may save 41700 lives annually, and education on the recognition and rapid care seeking for complications could save many more (Table 4). However, to close the major gap in coverage for 60 million non-facility births will require new and strategic investments in service supply—bringing care closer to home and addressing the global shortfall of 1 million midwives needed, especially in Africa. Closing this coverage gap and providing universal coverage of emergency obstetric and neonatal care would avert an estimated 613000 intrapartum-related neonatal deaths, and likely a similar number of intrapartum stillbirths, each year, as well as maternal deaths. In order to maximize these supply-side investments, more attention to demand strategies, including financial incentives and community mobilization, is also required.

A more aggressive approach to innovation to increase health system performance at the time of birth as well as closing the huge gaps in data and evidence are urgently needed. Unless more priority is given, the world will continue to miss the unheard cry of the 230 babies who die every hour, almost three-quarters of whom could be saved through community mobilization, resuscitation, immediate postnatal care, and well-known obstetric interventions that also save mothers' lives.

Back to Article Outline

8. Conflict of interest 

The authors have no conflicts of interest to declare.

Back to Article Outline

9. Funding 

The publication of this Supplement was supported by Saving Newborn Lives, a special program of Save the Children USA, funded by the Bill & Melinda Gates Foundation.

Back to Article Outline

Acknowledgments 

We acknowledge Rachel Haws for her assistance with the literature searches. We are grateful to Robert Goldenberg, Department of Obstetrics and Gynecology, Drexel University and Anu Shankar of the World Health Organization, for very helpful reviews. We also thank Uzma Syed, Riad Mahmud, and SK Asiruddin of the Saving Newborn Lives/Save the Children office in Bangladesh for help with Panel 2.

Back to Article Outline

Appendix A. Supplementary data 

Back to Article Outline

References 

1. Lawn JE, Lee AC, Kinney M, Sibley L, Carlo WA, Paul VK, et al. Two million intrapartum stillbirths and neonatal deaths: where, why, and what can we do? Int J Gynecol Obstet 2009:107;S5–S19.
   • View In Article
2. Lawn J, Shibuya K, Stein C. No cry at birth: global estimates of intrapartum stillbirths and intrapartum-related neonatal deaths. Bull World Health Organ. 2005;83(6):409–417
   • View In Article
   • MEDLINE
3. Lawn JE, Yakoob MY, Haws RA, Soomro T, Darmstadt GL, Bhutta ZA. 3.2 million stillbirths: epidemiology and overview of the evidence review. BMC Pregnancy Childbirth. 2009;9(Suppl 1):1–17
   • View In Article
   • CrossRef
4. Stanton C, Lawn JE, Rahman H, Wilczynska-Ketende K, Hill K. Stillbirth rates: delivering estimates in 190 countries. Lancet. 2006;367(9521):1487–1494
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (151 KB)
   • CrossRef
5. Hill K, Thomas K, AbouZahr C, Walker N, Say L, Inoue M, et al. Estimates of maternal mortality worldwide between 1990 and 2005: an assessment of available data. Lancet. 2007;370(9595):1311–1319
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (133 KB)
   • CrossRef
6. Koblinsky M, Matthews Z, Hussein J, Mavalankar DV, Mridha MK, Anwar I, et al. Going to Scale with Professional Skilled Care. Lancet. 2006;368(9544):1377–1386
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (136 KB)
   • CrossRef
7. Lawn JE, Cousens S, Zupan J. 4 million neonatal deaths: when? Where? Why?. Lancet. Mar 2 2005;365(9462):891–900
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (633 KB)
   • CrossRef
8. Ronsmans C, Graham WJ Lancet Maternal Survival Series steering group. Maternal mortality: who, when, where and why. Lancet. 2006;368(9542):1189–1200
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (262 KB)
   • CrossRef
9. Hart JT. The inverse care law. Lancet. 1971;1(7696):405–412
   • View In Article
   • CrossRef
10. Hofmeyr GJ, Haws RA, Bergstrom S, Lee AC, Okong P, Darmstadt GL, et al. Obstetric care in low-resource settings: What, who, and how to overcome challenges to scale up? Int J Gynecol Obstet 2009:107;S21–S45.
    • View In Article
11. Lee AC, Lawn JE, Darmstadt GL, Osrin D, Kumar V, Wall S, et al. Linking families and facilities for care at birth: what works to avert intrapartum-related deaths? Int J Gynecol Obstet 2009;107:S65–S88.
    • View In Article
12. World Health Organization. World Health Report 2005: make every mother and child count. Geneva, Switzerland: WHO; 2005;
    • View In Article
13. Darmstadt GL, Lee AC, Cousens S, Sibley L, Zulfiqar A, Bhutta A, et al. 60 million non-facility births: Who can deliver in community settings to reduce intrapartum-related deaths? Int J Gynecol Obstet 2009: 107;S89--S112.
    • View In Article
14. Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso-Coello P, et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ. Apr 26 2008;336(7650):924–926
    • View In Article
    • CrossRef
15. Lawn JE, Rudan I, Rubens C. Four million newborn deaths: Is the global research agenda evidence-based?. Early Hum Dev. 2008;84(12):809–814
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (330 KB)
    • CrossRef
16. Lee AC, Blencowe H, Hofmeyr J, Moran N, Darmstadt GL, Bhutta S, et al. Care during labour and birth for the prevention of neonatal deaths. Int J Epidemiol in press.
    • View In Article
17. Wall SN, Lee AC, Niermeyer S, English M, Carlo W, Bang A, et al. Neonatal resuscitation in low resource settings: What, who, and how to overcome challenges to scale up? Int J Gynecol Obstet 2009;107:S47–S64.
    • View In Article
18. Walker N, Fischer Walker CL, Bryce J, Bahl R, Simon CE, writing for the CHERG Review Groups on Intervention Effects. Standards for CHERG reviews of intervention effects on child health. Int J Epidemiol in press.
    • View In Article
19. 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(9194):1955–1961
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (112 KB)
    • CrossRef
20. Sibley L, Ann ST. What can a meta-analysis tell us about traditional birth attendant training and pregnancy outcomes?. Midwifery. 2004;20(1):51–60
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (214 KB)
    • CrossRef
21. Jokhio AH, Winter HR, Cheng KK. An intervention involving traditional birth attendants and perinatal and maternal mortality in Pakistan. N Engl J Med. 2005;352(20):2091–2099
    • View In Article
    • CrossRef
22. Carlo W FIRST BREATH study group. Effect of Essential Newborn Care (ENC) Training: A Global Network Study of Site Differences. Honolulu, HI: Pediatric Academic Societies; 2009;
    • View In Article
23. Pattinson R, Kerber K, Waiswa P, Day L, Mussell F, Asiruddin S, et al. Perinatal mortality audit: Counting, accountability, and overcoming challenges in scaling up. Int J Gynecol Obstet 2009:107;S113–23.
    • View In Article
24. Patel Z, Kumar V, Singh P, Singh V, Yadav R, Baqui AH, et al. Feasibility of community neonatal death audits in rural Uttar Pradesh, India. J Perinatol. 2007;27(9):556–564
    • View In Article
    • CrossRef
25. Darmstadt GL, Bhutta ZA, Cousens S, Adam T, Walker N, De Bernis L. Evidence-based, cost-effective interventions: how many newborn babies can we save?. Lancet. 2005;365(9463):977–988
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (146 KB)
    • CrossRef
26. Bryce J, Daelmans B, Dwivedi A, Fauveau V, Lawn JE, Mason E, et al. Countdown to 2015 for maternal, newborn, and child survival: the 2008 report on tracking coverage of interventions. Lancet. Apr 12 2008;371(9620):1247–1258
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (488 KB)
    • CrossRef
27. Kerber KJ, de Graft-Johnson JE, Bhutta ZA, Okong P, Starrs A, Lawn JE. Continuum of care for maternal, newborn, and child health: from slogan to service delivery. Lancet. 2007;370(9595):1358–1369
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (208 KB)
    • CrossRef
28. WHOUNFPAUNICEFAMDD. Monitoring emergency obstetric care: a handbook. Geneva: WHO; 2009;
    • View In Article
29. World Health Organization. Making pregnancy safer: the critical role of the skilled attendant, A joint statement by WHO, ICM and FIGO. Geneva, Switzerland: WHO; 2004;
    • View In Article
30. 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(7132):657–660
    • View In Article
    • MEDLINE
    • CrossRef
31. Hodnett ED, Gates S, Hofmeyr GJ, Sakala C. Continuous support for women during childbirth. Cochrane Database Syst Rev. 2007;(3):CD003766
    • View In Article
32. Vaz F, Bergstrom S, Vaz ML, Langa J, Bugalho A. Training medical assistants for surgery. Bull World Health Organ. 1999;77(8):688–691
    • View In Article
    • MEDLINE
33. Pereira C, Bugalho A, Bergstrom S, Vaz F, Cotiro M. A comparative study of caesarean deliveries by assistant medical officers and obstetricians in Mozambique. Br J Obstet Gynaecol. Jun 1996;103(6):508–512
    • View In Article
    • MEDLINE
34. Chilopora G, Pereira C, Kamwendo F, Chimbiri A, Malunga E, Bergstrom S. Postoperative outcome of caesarean sections and other major emergency obstetric surgery by clinical officers and medical officers in Malawi. Hum Resour Health. 2007;5:17
    • View In Article
35. De Brouwere V, Dieng T, Diadhiou M, Witter S, Denerville E. Task shifting for emergency obstetric surgery in district hospitals in Senegal. Reprod Health Matters. May 2009;17(33):32–44
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (680 KB)
    • CrossRef
36. Mbaruku G, van RJ, Kimondo I, Bilango F, Bergström S. Perinatal audit using the 3-delays model in western Tanzania. Int J Gynecol Obstet. 2009;106(1):85–88
    • View In Article
37. FIGO Safe Motherhood and Newborn Health Committee. Human resources for health in the low-resource world: Collaborative practice and task-shifting in maternal and neonatal care. Int J Gynecol Obstet. 2009;105(1):74–76
    • View In Article
38. Mullany LC, Lee CI, Yone L, Paw P, Oo EK, Maung C, et al. Access to essential maternal health interventions and human rights violations among vulnerable communities in eastern Burma. PLoS Med. Dec 23 2008;5(12):1689–1698
    • View In Article
39. World Health Organization. Everybody's business: Strengthening health systems to improve health outcomes. WHO's framework for action. Geneva: WHO; 2007;
    • View In Article
40. Sachs J. Macroeconomics and health: Investing in health for econic development. Report of the Commission of Macroeconomics and health. Geneva: WHO; 2001;
    • View In Article
41. Rohde J, Cousens S, Chopra M, Tangcharoensathien V, Black R, Bhutta ZA, et al. 30 years after Alma-Ata: has primary health care worked in countries?. Lancet. 2008;372(9642):950–961
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (230 KB)
    • CrossRef
42. United Nations Human Rights Council recognises maternal mortality as human rights concern. The Partnership for Maternal, Newborn & Child Health. Available at: http://www.who.int/pmnch/media/membernews/2009/20090617_humanrightsresolution/en/index.html. Accessed July 15, 2009.
    • View In Article
43. Fauveau V. New indicator of quality of emergency obstetric and newborn care. Lancet. Oct 13 2007;370(9595):1310
    • View In Article
    • Full Text
    • Full-Text PDF (37 KB)
    • CrossRef
44. Knippenberg R, Lawn JE, Darmstadt GL, Begkoyian G, Fogstad H, Walelign N, et al. Systematic scaling up of neonatal care in countries. Lancet. 2005;365(9464):1087–1098
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (301 KB)
    • CrossRef
45. Lawn J, Kerber K, eds. Opportunities for Africa's Newborns: practical data, policy and programmatic support for newborn care in Africa. Cape Town: PMNCH, Save the Children, UNFPA, UNICEF, USAID, WHO; 2006.
    • View In Article
46. Darmstadt GL, Walker N, Lawn JE, Bhutta ZA, Haws RA, Cousens S. Saving newborn lives in Asia and Africa: cost and impact of phased scale-up of interventions within the continuum of care. Health Policy Plan. 2008;23(2):101–117
    • View In Article
    • CrossRef
47. Shiffman J. Generating political priority for maternal mortality reduction in 5 developing countries. Am J Public Health. 2007;97(5):796–803
    • View In Article
    • CrossRef
48. Jones G, Steketee RW, Black RE, Bhutta ZA, Morris SS. How many child deaths can we prevent this year?. Lancet. 2003;362(9377):65–71
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (105 KB)
    • CrossRef
49. Haws RA, Thomas AL, Bhutta ZA, Darmstadt GL. Impact of packaged interventions on neonatal health: a review of the evidence. Health Policy Plan. 2007;22(4):193–215
    • View In Article
    • MEDLINE
    • CrossRef
50. Hussein J, Bell J, Nazzar A, Abbey M, Adjei S, Graham W. The skilled attendance index: proposal for a new measure of skilled attendance at delivery. Reprod Health Matters. 2004;12(24):160–170
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (638 KB)
    • CrossRef
51. Stanton CK, Dubourg D, De BV, Pujades M, Ronsmans C. Reliability of data on caesarean sections in developing countries. Bull World Health Organ. 2005;83(6):449–455
    • View In Article
    • MEDLINE
52. Buchmann EJ, Pattinson RC. Babies who die from labour-related intrapartum hypoxia: a confidential enquiry in South African public hospitals. Trop Doct. 2006;36(1):8–10
    • View In Article
    • MEDLINE
    • CrossRef
53. Guidotti RJ, Kandasamy T, Betran AP, Merialdi M, Hakimi F, Van LP, et al. Monitoring perinatal outcomes in hospitals in Kabul, Afghanistan: The first step of a quality assurance process. J Matern Fetal Neonatal Med. Apr 2009;22(4):285–292
    • View In Article
    • CrossRef
54. Bryce J, Requejo JH. Tracking Progress in Maternal, Newborn and Child Survival: The 2008 Report. New York: UNICEF; 2008;
    • View In Article
55. Kumar V, Mohanty S, Kumar A, Misra RP, Santosham M, Awasthi S, et al. Effect of community-based behaviour change management on neonatal mortality in Shivgarh, Uttar Pradesh, India: a cluster-randomised controlled trial. Lancet. 2008;372(9644):1151–1162
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (185 KB)
    • CrossRef
56. Adam T, Lim SS, Mehta S, Bhutta ZA, Fogstad H, Mathai M, et al. Cost effectiveness analysis of strategies for maternal and neonatal health in developing countries. BMJ. 2005;331(7525):1107
    • View In Article
    • CrossRef
57. Ghana Demographic and Health Survey: Preliminary report 2008. 2009.
    • View In Article
58. International Institute for Population Sciences. District Level Household and Facility Survey III: India Fact Sheet. Mumbai, India: International Institute for Population Sciences; 2009;
    • View In Article
59. Lahariya C. Cash Incentives for Institutional Delivery: Linking with antenatal and post natal care may ensure ‘continuum of care’ in India. Indian J Community Med. Jan 2009;34(1):15–18
    • View In Article
60. Bhutta ZA, Cross A, Raza F, Zaheer Z. Infant and Child Mortality. In: National Institute of Population Studies (NIPS) [Pakistan], Macro International Inc., editors. Pakistan Demographic and Health Survey 2006–07.Islamabad, Pakistan: National Institute of Population Studies and Macro International Inc.; 2008. p. 89–100.
    • View In Article
61. Martines J, Paul VK, Bhutta ZA, Koblinsky M, Soucat A, Walker N, et al. Neonatal survival: a call for action. Lancet. 2005;365(9465):1189–1197
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (112 KB)
    • CrossRef
62. Masanja H, de Savigny D, Smithson P, Schellenberg J, John T, Mbuya C, et al. Child survival gains in Tanzania: analysis of data from demographic and health surveys. Lancet. 2008;371(9620):1276–1283
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (168 KB)
    • CrossRef
63. Borghi J, Hanson K, Acquah CA, Ekanmian G, Filippi V, Ronsmans C, et al. Costs of near-miss obstetric complications for women and their families in Benin and Ghana. Health Policy Plan. 2003;18(4):383–390
    • View In Article
    • MEDLINE
    • CrossRef
64. Ravishankar N, Gubbins P, Cooley RJ, Leach-Kemon K, Michaud CM, Jamison DT, et al. Financing of global health: tracking development assistance for health from 1990 to 2007. Lancet. Jun 20 2009;373(9681):2113–2124
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (869 KB)
    • CrossRef
65. Greco G, Powell-Jackson T, Borghi J, Mills A. Countdown to 2015: assessment of donor assistance to maternal, newborn, and child health between 2003 and 2006. Lancet. 2008;371(9620):1268–1275
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (130 KB)
    • CrossRef
66. Eichler R, Levine R. Performance Incentives for Global Health: Potential and Pitfalls. CGD Brief. Available at: http://www.cgdev.org/files/1422185_file_CGDEV_Brief_PBI_web.pdf. Accessed July 15, 2009.
    • View In Article
67. G8 Leaders Declaration: Responsible Leadership for a Sustainable Future. July 8, 2009. G8 Summit of L'Aquila.
    • View In Article
68. Lagarde M, Haines A, Palmer N. Conditional cash transfers for improving uptake of health interventions in low- and middle-income countries: a systematic review. JAMA. Oct 24 2007;298(16):1900–1910
    • View In Article
    • CrossRef
69. Gertler P. Final Report: The impact of PROGRESA on health. Washington, DC: International Food Policy Research Institute (IFPRI); 2000;
    • View In Article
70. 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(5):e1372–e1380
    • View In Article
    • CrossRef
71. Ronsmans C, Campbell OM, McDermott J, Koblinsky M. Questioning the indicators of need for obstetric care. Bull World Health Organ. 2002;80(4):317–324
    • View In Article
    • MEDLINE
72. Koblinsky MA, Campbell O, Heichelheim J. Organizing delivery care: what works for safe motherhood?. Bull World Health Organ. 1999;77(5):399–406
    • View In Article
    • MEDLINE
73. Rudan I, Gibson JL, Ameratunga S, el Arifeen S, Bhutta ZA, Black M, et al. Setting priorities in global child health research investments: guidelines for implementation of CHNRI method. Croat Med J. 2008;49(6):720–733
    • View In Article
    • CrossRef
74. Lawn JE, Cousens S, Zupan J. 4 million neonatal deaths: when? Where? Why?. Lancet. 2005;365(9462):891–900
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (633 KB)
    • CrossRef
75. Unicef. State of the World's Children 2009. New York: UNICEF; 2009;
    • View In Article
76. Lawn JE, Wilczynska-Ketende K, Cousens SN. Estimating the causes of 4 million neonatal deaths in the year 2000. Int J Epidemiol. 2006;35(3):706–718
    • View In Article
    • MEDLINE
    • CrossRef
77. Lawn JE, Kerber KJ, eds. Opportunities for Africa's Newborns: practical data, policy and programmatic support for newborn care in Africa. Cape Town: PMNCH, Save the Children, UNFPA, UNICEF, USAID, WHO; 2006.
    • View In Article
78. O'Rourke K, Howard-Grabman L, Seoane G. Impact of community organization of women on perinatal outcomes in rural Bolivia. Rev Panam Salud Publica. 1998;3(1):9–14
    • View In Article
    • MEDLINE
79. Bhutta ZA, Memon ZA, Soofi S, Salat MS, Cousens S, Martines J. Implementing community-based perinatal care: results from a pilot study in rural Pakistan. Bull World Health Organ. 2008;86(6):452–459
    • View In Article
80. 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(9438):970–979
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (465 KB)
    • CrossRef
81. Fullerton JT, Killian R, Gass PM. Outcomes of a community-and home-based intervention for safe motherhood and newborn care. Health Care Women Int. 2005;26(7):561–576
    • View In Article
    • MEDLINE
    • CrossRef
82. Baqui AH, El-Arifeen S, Darmstadt GL, Ahmed S, Williams EK, Seraji HR, et al. Effect of community-based newborn-care intervention package implemented through two service-delivery strategies in Sylhet district, Bangladesh: a cluster-randomised controlled trial. Lancet. 2008;371(9628):1936–1944
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (307 KB)
    • CrossRef
83. Ibrahim SA, Omer MI, Amin IK, Babiker AG, Rushwan H. The role of the village midwife in detection of high risk pregnancies and newborns. Int J Gynecol Obstet. 1992;39(2):117–122
    • View In Article
84. Ronsmans C, Chowdhury ME, Alam N, Koblinsky M, El AS. Trends in stillbirths, early and late neonatal mortality in rural Bangladesh: the role of public health interventions. Paediatr Perinat Epidemiol. 2008;22(3):269–279
    • View In Article
    • CrossRef
85. Gloyd S, Floriano F, Seunda M, Chadreque MA, Nyangezi JM, Platas A. Impact of traditional birth attendant training in Mozambique: A controlled study. J Midwifery Womens Health. 2001;46(4):210–216
    • View In Article
    • MEDLINE
    • CrossRef
86. PATH. Reducing Birth Asphyxia Through the Bidan di Desa Program in Cirebon, Indonesia: Final report submitted by Program for Appropriate Technology in Health (PATH) to Save the Children US. Jakarta, Indonesia: PATH; Mar 15 2006;
    • View In Article
87. Greenwood AM, Bradley AK, Byass P, Greenwood BM, Snow RW, Bennett S, et al. Evaluation of a primary health care programme in The Gambia. I. The impact of trained traditional birth attendants on the outcome of pregnancy. J Trop Med Hyg. 1990;93(1):58–66
    • View In Article
    • MEDLINE
88. Bang AT, Bang RA, Baitule SB, Reddy HM, Deshmukh MD. Management of birth asphyxia in home deliveries in rural Gadchiroli: the effect of two types of birth attendants and of resuscitating with mouth-to-mouth, tube-mask or bag-mask. J Perinatol. Mar 2005;25(Suppl 1):S72–S81
    • View In Article
    • CrossRef
89. Campbell O, Graham WJ. Strategies for reducing maternal mortality: getting on with what works. Lancet. 2006;368(9546):1532–1541
    • View In Article
90. Bhutta ZA, Darmstadt GL, Haws RA, Yakoob MY, Lawn JE. Delivering interventions to reduce the global burden of stillbirths: improving service supply and community demand. BMC Pregnancy Childbirth. 2009;9(Suppl 1):S7
    • View In Article
    • CrossRef
91. World Health Organization. World Health Statistics 2008. Geneva: WHO; 2008;
    • View In Article
92. Bhutta ZA, Ahmed T, Black RE, Cousens S, Dewey K, Giugliani E, et al. What works? Interventions for maternal and child undernutrition and survival. Lancet. 2008;371(9610):417–440
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (285 KB)
    • CrossRef
93. Stover J, Johnson P, Zaba B, Zwahlen M, Dabis F, Ekpini RE. The Spectrum projection package: improvements in estimating mortality, ART needs, PMTCT impact and uncertainty bounds. Sex Transm Infect. 2008;84(Suppl 1):i24–i30
    • View In Article
    • CrossRef
94. Lee AC, Wall S, Cousens S, Carlo W, Niermeyer S, Keenan W, et al. Neonatal resuscitation and simple immediate newborn care for the prevention of neonatal deaths. Int J Epidemiol in press.
    • View In Article
95. Chowdhury ME, Botlero R, Koblinsky M, Saha SK, Dieltiens G, Ronsmans C. Determinants of reduction in maternal mortality in Matlab, Bangladesh: a 30-year cohort study. Lancet. 2007;370(9595):1320–1328
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (159 KB)
    • CrossRef
96. Govt. of the People's Republic of Bangladesh. National Neonatal Health Strategy and Guidelines. Bangladesh: IMCI Section, Directorate General of Health Services, Ministry of Health and Family Welfare, UNICEF; 2009;
    • View In Article
97. UNFPA. www.unfpa.org. Accessed July 15, 2009.
    • View In Article
98. Baqui AH, Arifeen SE, Williams EK, Ahmed S, Mannan I, Rahman SM, et al. Effectiveness of Home-Based Management of Newborn Infections by Community Health Workers in Rural Bangladesh. Pediatr Infect Dis J. 2009;28:304–310
    • View In Article
    • CrossRef

Back to Article Outline

Further reading