Generating Political Priority for Neonatal Mortality Reduction in Bangladesh

Abstract

The low priority that most low-income countries give to neonatal mortality, which now constitutes more than 40% of deaths to children younger than 5 years, is a stumbling block to the world achieving the child survival Millennium Development Goal. Bangladesh is an exception to this inattention. Between 2000 and 2011, newborn survival emerged from obscurity to relative prominence on the government’s health policy agenda. Drawing on a public policy framework, we analyzed how this attention emerged. Critical factors included national advocacy, government commitment to the Millennium Development Goals, and donor resources. The emergence of policy attention involved interactions between global and national factors rather than either alone. The case offers guidance on generating priority for neglected health problems in low-income countries.

In 2001 United Nations member states agreed to 8 Millennium Development Goals (MDGs), poverty alleviation objectives to achieve by 2015. Goal 4 concerns child survival: “Reduce by two thirds, between 1990 and 2015, the under-five mortality rate.”¹ Although analysts expect the world to achieve many of the 8 goals,² only 31 of 137 developing countries are predicted to reach MDG 4.³

A stumbling block is the slow decline in deaths to newborn babies—those aged 28 days and younger. Early neonatal mortality (0–6 days) has declined at a rate of only 1.7% per year since 1990, slower than the pace for children older than 28 days.³ In consequence, newborns now constitute more than 40% of all deaths to children younger than 5 years.

This slow decline may be partly because of the low priority most low-income countries afford neonatal mortality reduction. Although many governments have addressed child survival, few have focused on the neonatal period.⁴ Bangladesh is an exception. Between 2000 and 2011, newborn survival rose from near obscurity to a prominent place on the government’s health policy agenda. This change is surprising, as there was no swift spread of a pathogen harming neonates or sudden rise in the number of newborn deaths; on the contrary, newborn death rates have been declining in the country, if slowly.⁵

We analyzed how newborn survival emerged as a health priority in Bangladesh. We drew on a public policy framework previously developed to explain political attention for maternal mortality reduction in 5 developing countries.⁶ The appearance of an issue on a national policy agenda is only 1 of multiple factors that stand behind policy effectiveness and is hardly enough to ensure that the political system will carry out plans or that these plans will be successful in reducing neonatal mortality. However, reaching the policy agenda facilitates policy effectiveness and is therefore useful to study.

THE FRAMEWORK

Political scientists have a long-standing interest in how and why some issues come to attract the attention and resources of policymakers.^7–11 Social scientists who investigate health policymaking in low-income countries also have considered agenda-setting processes.^12–16 This research suggests that although randomness plays a role, there are systematic elements. Drawing on this scholarship, we developed a framework that we used to analyze variance in policy attention for maternal survival across 5 developing countries (Table 1). The framework consists of 9 factors in 3 categories: transnational influence, domestic advocacy, and national political environment (a more thorough discussion of the framework can be found in Shiffman⁶).

TABLE 1—

Factors Influencing the Degree to Which Maternal Mortality Reduction Appeared on National Policy Agendas in 5 Developing Countries: early 1990s to mid-2000s

Factor	Category	Description
1. Norm promotion	Transnational influence	International agencies’ efforts to establish a global norm for the unacceptability of maternal death
2. Resource provision	Transnational Influence	International agencies’ offer of financial and technical resources to address maternal mortality
3. Policy community cohesion	Domestic advocacy	The degree to which national safe motherhood promoters coalesced as a political force pushing the government to act
4. Political entrepreneurship	Domestic advocacy	The presence of respected and capable national political champions willing to promote the cause
5. Credible indicators	Domestic advocacy	The availability and strategic deployment of evidence to demonstrate the presence of a maternal mortality problem
6. Focusing events	Domestic advocacy	The organization of forums to generate national attention for the cause
7. Clear policy alternatives	Domestic advocacy	The availability of clear policy alternatives to demonstrate to political leaders that the problem is surmountable
8. Political transitions	National political environment	Political changes, such as democratization, that positively or adversely affected prospects for safe motherhood promotion
9. Competing health priorities	National political environment	Priority for other health causes that diverted policymaker attention from maternal mortality reduction

Note. The 5 developing countries from which these data were derived are Guatemala, Honduras, India, Indonesia, and Nigeria.

Source. Shiffman.⁶

International organizations and advocates use several mechanisms to influence national political systems to embrace the causes that concern them. One is norm promotion (factor 1): they try to convince governments that it is appropriate, for ethical reasons, to address a particular issue.^10,11 For instance, the Joint United Nations Programme on HIV and AIDS and other organizations present HIV/AIDS as an exceptional disease and encourage governments to set up institutions dedicated to AIDS prevention and control. Another mechanism is resource provision (factor 2): the enticement of financial and technical assistance to governments if they agree to adopt particular priorities and policies.¹⁷

Although in some cases international actors may bring issues to global and national agendas, more often than not domestic factors are equally crucial. One such factor is policy community cohesion (factor 3). Policy communities are networks of actors from different types of organizations—government agencies, legislatures, nongovernmental organizations (NGOs), and others—committed to common causes. Among the factors that shape their degree of influence are their levels of moral authority, knowledge, and coherence.¹⁸ Political entrepreneurs (factor 4) also shape policy priority⁷—politically influential and particularly capable individuals willing to exert effort to advance a cause. Former United Nations Children’s Fund director James Grant, who did much to advance the cause of child survival globally, is a prototype. Credible indicators (factor 5) also matter.^7,19 These make a difference because they have the uniquely powerful effect of giving visibility to that which has remained hidden, serving as catalysts that may provoke political elites to act. Focusing events (factor 6)—large-scale happenings such as crises, conferences, and discoveries that attract notice from wide audiences—like indicators, bring visibility to hidden issues.²⁰ In addition, clear policy alternatives (factor 7) are influential: policymakers are more likely to act on an issue if they are presented with clear proposals that convince them a problem is surmountable.^7,9

The political and social environments in which international and domestic advocates work also shape policy attention. Many such factors may be influential, including cultural barriers, the ethnic composition of societies, civil strife, weak administrative infrastructures, and endemic corruption. Two factors, however, may be particularly critical in health agenda setting. Political transitions (factor 8) are major political changes such as democratization and public sector decentralization that alter public priorities by giving new actors agenda-setting power and by changing the processes by which public policies are made and implemented.^21,22 Competing health priorities (factor 9) may make it more difficult for new issues to gain attention because most health sectors in low-income countries lack resources, and health causes must vie against one another for this scarce funding.

STUDY DESIGN

We sought to piece together the history of newborn survival efforts in Bangladesh to assess the level of attention to the issue and the factors and processes behind the emergence of attention. We chose a case study approach, which is better suited to achieve these objectives than are other research methodologies, such as structured surveys or statistical analyses of health service utilization.²³ This is true because the defining feature of the case study is that it considers a phenomenon in its real-life context, thereby giving it the capacity to reveal underlying processes. In the language of case study methodology, our inquiry was holistic in nature and selected on the basis of its revelatory and unique characteristics.²³ That is to say, we analyzed the nation-state of Bangladesh holistically as a unit rather than any of its subregions; we sought to make use of our access to policymakers and other officials to reveal insights that may not have been available otherwise; and we justified selection of Bangladesh for analysis because of its uniqueness in being one of the few low-income countries where newborn survival has apparently come to receive significant policy attention.

We used 5 types of sources to conduct this study, triangulating among these to minimize bias: key informant interviews, government reports and documents, donor and NGO reports, published research on newborn survival in Bangladesh, and observations of the sites of several newborn survival projects. In 2009 and 2010 we conducted 26 interviews, lasting on average 60 minutes, with 3 groups of individuals: those centrally involved in efforts to address newborn survival, those in a position to observe and offer authoritative information about the effectiveness of these efforts, and those critical of these efforts. We identified these individuals through publicly available documents, commentaries, and consultation with individuals working on the issue in Bangladesh—a key informant rather than a sampling selection strategy.

We interviewed individuals from the following organizations: the government sector (the Ministry of Health and Family Welfare—Office of Joint Secretary, Ministry of Health and Family Welfare—Directorate General of Health Services, Ministry of Health and Family Welfare—Directorate General of Family Planning), the civil society sector (Bangladesh Perinatal Society, Bangladesh Neonatal Forum, Bangladesh Rural Advancement Committee, International Centre for Diarrheal Disease Research-Bangladesh, Obstetrics and Gynecology Society of Bangladesh), and development partners (Gates Foundation, Japan International Cooperation Agency, Pathfinder International, Maternal and Child Health Integrated Program, Save the Children USA, the Saving Newborn Lives [SNL] program, United Nations Children’s Fund, United States Agency for International Development [USAID]).

We informed interviewees that they would not be identified in the text. We did not record interviews because a number of interviewees felt uncomfortable with that practice, but took detailed notes on each. Rather than follow a set of structured questions, we sought through open-ended questions to elicit the unique knowledge that each informant held about efforts to address newborn survival.

Additionally, we undertook archival research on the history of Bangladeshi newborn survival efforts, gathering and reviewing 105 documents. Among the documents we reviewed were national health plans, assessments of these plans by independent agencies, donor and government project assessment reports, statistical records and analyses, internal records of development partners, commissioned research on newborn survival in Bangladesh, and published articles on this issue. Also, we visited several implementation sites. Among those we observed were an urban maternal, newborn, and child health project in Dhaka run by the Bangladeshi NGO, Bangladesh Rural Advancement Committee, and the Bangabondhu Sheikh Mujib Medical University. We selected sites where major neonatal survival programs or research were being conducted.

Once all the interviews had been conducted and the documents collected, we input these into NVIVO 8 software (QSR International, Melbourne, Australia), a program that facilitates the analysis of qualitative data. We grouped related material from multiple sources into categories, including major historical developments and the factors in the framework delineated above. We then reconstructed the history of newborn survival efforts and analyzed the evidence on the level of policy attention and causal influence of the framework factors. Thereafter, we considered the adequacy of the framework and its limits in explaining the emergence of attention.

Case studies such as these that rely heavily on interviews with involved actors are susceptible to bias. To minimize this possibility, we employed several techniques recommended by case study methodology experts to address potential error.^23–25 First and foremost we triangulated among sources. Our historical information came not just from interviews but also from published sources and independent reports. Second, we did not rely on individual interviews predominantly to check historical accuracy because these were susceptible to recall bias; instead, when interviewees reported a significant event, we checked published literature or reports for corroboration. We also inquired about these events with multiple respondents. Finally, we received feedback on a draft from 3 individuals familiar with the history of newborn survival efforts in the country. One of these individuals was not directly involved with newborn survival but was in a position to comment because that person held considerable knowledge on the history of child survival initiatives in the country.

The research design imposes limits on internal and external validity. In-depth exploration enables us to develop hypotheses concerning why attention may have emerged for newborn survival in Bangladesh and to suggest general propositions concerning health policy agenda setting in low-income countries. On the other hand, the design creates uncertainty about the conclusions, as they are grounded in consideration of only a single case. Additional comparative research on other countries that considers alternative explanations will be necessary to assess the causal power of the factors we identify. Also, any generalization to other settings must be done with caution, given elements of the sociopolitical and health context that are unique to Bangladesh.

THE CASE OF NEWBORN SURVIVAL IN BANGLADESH

A decade ago government officials and donors were paying little attention to newborn survival.²⁶ The country’s overarching health plan for 1998 through 2003 incorporated essential newborn care but only as 1 of 8 elements of reproductive health. Neonatal mortality was not among the indicators the government used to measure health sector performance. Attention was hampered by the fact that many health officials believed, without strong evidence, that existing interventions were sufficient to address the problem.

In the late 1990s domestic conditions were favorable for the emergence of support. Enacted in 1998, a health and population sector program promised greater health sector coherence after 2 decades of ineffectual government coordination and fragmented donor projects. Also, the Bangladeshi government was already concerned about addressing child survival.²⁷ In addition, by 1999 there were 4 medical associations that shared some concern for newborn health, including 1 focused on newborns exclusively—the Bangladesh Neonatal Forum. Among the leaders of these associations were 4 doctors, who, through overseas medical training and observation of conditions in the country, had come to the conclusion that Bangladesh had a severe problem with neonatal mortality.

Globally, too, conditions in the late 1990s were favorable for the emergence of support. The MDGs had norm-setting influence on national governments.²⁸ Several pieces of evidence indicate that the child survival MDG shaped Bangladeshi government priorities. The government set up a national task force on the child and maternal survival MDGs in August 2007, which still meets regularly,²⁹ and achieving the health MDGs became one of the pillars of the government’s health program for 2003 through 2011.³⁰

In addition, in 2000 a global program with an exclusive focus on neonatal mortality formed—the SNL program of Save the Children USA—with $50 million in funding from the Gates Foundation.³¹ Soon after its establishment SNL sought to formalize a global alliance of organizations with an interest in newborn survival, helping to create and becoming the secretariat for the global Healthy Newborn Partnership. SNL leaders selected Bangladesh as 1 of 6 initial focal countries, a decision influenced by a long-standing Save the Children presence there and by the country’s high child mortality.

National Attention Between 2001 and 2004

SNL officials moved rapidly to establish a Bangladeshi program, the first large-scale organized effort in the country exclusively focused on neonatal mortality reduction. In October 2000, Washington, DC–based SNL leaders conducted a scoping visit to the country. They and the local Save the Children office contracted a Bangladeshi physician on the faculty of the University of Dhaka, who had just completed postgraduate training in London, to take the lead on a situation analysis on the state of newborns in the country. She and a colleague interviewed actors in the health sector and produced a report revealing a lack of attention to the issue. She presented preliminary results at a February 2001 meeting to facilitate the development of a national SNL strategy. In attendance were many of the officials in the health sector that she interviewed. Attendees heard preliminary data from the Bangladesh 1999–2000 Demographic and Health Survey³² revealing a high neonatal mortality rate of 42 per 1000 live births³² and learned that two thirds of deaths to children younger than 1 year were to newborns. Attendees also learned of the biomedical causes of newborn mortality and evidence on the tractability of the problem coming from work from an Indian physician, Abhay Bang, who had demonstrated in a controlled study a 25% decline in neonatal mortality in a treatment area.³³ In September 2001, the report was launched publicly at a meeting in the capital, Dhaka, attracting widespread media attention, including national television coverage and articles in 11 daily newspapers.³⁴

From 2001 on, SNL reached out to secure support for newborn survival among government, donor, and medical association officials. The first move the organization made may have been the most important: it hired the Bangladeshi physician who conducted the situation analysis on newborn survival. Working quietly behind the scenes, over the next 5 years she emerged as the country’s foremost newborn survival champion and was widely regarded among those working in Bangladesh on child survival as highly effective in her efforts. In line with its global strategy and under her guidance, SNL decided against an insular approach that would make itself the public face for the newborn and isolate newborn care from that of mothers and children.³⁵ Instead it cultivated ownership of the issue among multiple organizations and individuals—especially those in government—and promoted the integration of newborn, maternal, and child health care.

SNL’s first effort to expand ownership was public presentation of the preliminary results of the situation report. Its second move was the establishment in June 2001 of a newborn working group that brought together about a dozen individuals concerned with newborn survival.³⁶ The Bangladeshi physician exerted considerable efforts initially to get reluctant officials to come. Over the following 2 years, the group expanded in size and became a regular monthly forum with formal terms of reference.³⁷ Among the most active members were the 4 professors from the medical associations with an interest in newborn survival. Their concern for newborn survival predated the existence of SNL, and they formed the backbone of the working group. However, it was not until after SNL’s establishment that a policy community—a network of individuals and organizations in regular contact with one another and sharing an interest in the issue—began to coalesce. SNL’s existence, the persistence of the Bangladeshi physician, and the creation of the working group facilitated the emergence of this community. As one of these doctors put it, “If there had been no SNL there may have been no priority, as our first priority was as clinicians, and the government did not come forward; nor did we go to them.”

Members of this policy community identified several problems contributing to the slow rate of neonatal mortality decline, including small numbers of nurse midwives, lack of skills among community health workers, and low postnatal care coverage.³⁸ They actively reached out to promote adoption of good newborn care practices. For instance, in 2002, they developed a module on essential newborn care,³⁹ which the government later endorsed and put into its operational plan.⁴⁰ Also, policy community members succeeded in getting newborn health added as 1 of 5 components of the Integrated Management of Childhood Illness, a strategy the Bangladesh government had adopted in 2002.⁴¹

The policy community’s efforts were aided by a focusing event in 2003 that raised the visibility of the issue among government officials and influenced government policies and programs. At the suggestion of the Bangladeshi physician, the Healthy Newborn Partnership convened a meeting in Dhaka. Opened by the Minister of Health and Family Welfare, the meeting brought together 31 donors and nongovernmental organizations. It was held concurrently with the first international Bangladeshi Perinatal Congress, attended by more than 500 physicians and health professionals from Bangladesh and other countries.⁴² The coincidence of these 2 meetings enabled global health professionals to interact with domestic neonatologists, obstetricians, and other physicians.⁴³ At the conclusion of these meetings, the secretary of the Ministry of Health and Family Welfare chaired a policy session that resulted in the “Dhaka Declaration for Global Newborn Health,”⁴³ calling for enhanced national and global attention to newborn survival.³⁴ A month thereafter the Bangladesh Perinatal Society, supported by SNL, convened a workshop involving the secretary on incorporating significant newborn care in the national health plan.

These meetings had concrete effects. Most significantly, the Ministry of Health and Family Welfare for the first time added a newborn survival target in its national health plan: to reduce neonatal mortality from 42 to 32 per 1000 live births by mid-2006.⁴⁴ In addition, the government allocated money to train health workers in essential newborn care, and in the mid-2000s this care constituted an estimated 11% of the government’s primary health care training budget.⁴⁵

Evidence on the problem’s tractability facilitated the policy community’s newborn survival advocacy, helping to shift perceptions of policymakers concerning the potential for reducing mortality through low-cost interventions. Abhay Bang’s work was influential.³³ Also influential was a study in the Sylhet district, an area with poor access to health care. In 2002, investigators from the Bangladeshi research institution International Centre for Diarrhoeal Disease Research, Bangladesh and from Johns Hopkins University initiated the Projahnmo project, with funding from USAID and SNL. Women community health workers identified pregnant women, provided pre- and postnatal home visits, and referred or treated sick newborns, resulting in dramatic neonatal mortality decline: 34% in the last 6 months of the study in the treatment area.⁴⁶ Rather than work in isolation, designers of the project reached out to government decision-makers from its inception to ensure that they would embrace its results, establishing a secretariat that included government oversight of the project and disseminating results throughout the study period. These results convinced USAID to begin a $15 million neonatal survival program in Bangladesh in 2006, forming a cornerstone for the National Neonatal Health Strategy, which was enacted in 2009.⁴⁷

Credible indicators demonstrating the severity of the problem also facilitated advocacy. Data on high neonatal mortality from the 1999–2000 Bangladesh Demographic and Health Survey presented at the SNL 2001 strategic planning meeting alerted health sector officials about the severity of the problem. In 2003 the SNL Bangladeshi physician convinced USAID to include 6 essential newborn care indicators in the 2004 Bangladesh Demographic and Health Survey,⁴⁸ the first global instance of the inclusion of such a broad set of newborn care indicators in this survey. The 2004 DHS included a neonatal mortality measure as it had in the past, revealing the persistence of high neonatal mortality (41 per 1000), and the 2007 DHS indicated its growing share of overall child mortality (57%).⁴⁹ The 2007 DHS also showed low levels of practice of essential newborn care behaviors; for instance, fewer than 20% of newborns had their first bath delayed until at least 72 hours after birth. These DHS data became reference points for health policymakers to track progress on the issue and reinforced their perceptions that the country’s prospects for achieving MDG 4 might be hampered if they did not address newborn survival.

Expanding Initiatives From 2005 to 2011

Influenced by these policy community activities, focusing events, policy alternatives, and indicators, from 2005 on the government stepped up its involvement in neonatal mortality reduction. In 2006 it added a newborn position in the Integrated Management of Childhood Illness unit of the Ministry of Health and Family Welfare. Also in that year, it began a maternal health voucher scheme to encourage the use of pre-, intra-, and postnatal care for pregnant women.⁴¹

In addition, in the second half of the decade 5 large programs involving donors, government organizations, and NGOs were initiated that had significant newborn survival components. One was the 2006 USAID program, prompted by the Projahnmo results.⁵⁰ In 2007, the domestic NGO Bangladesh Rural Advancement Committee began a maternal, neonatal, and child health project covering a population of 8 million in urban slums, with $25.0 million in funding from the Gates Foundation.⁵¹ In addition, Australia, the United Kingdom, and the European Commission provided $71.5 million to fund 3 maternal, newborn, and child survival programs involving government in 15 of Bangladesh’s 64 districts.⁵²

The emergence of these programs was not only because of national advocacy but also because of transnational influences. Over the course of the decade, organizations involved in global health had become increasingly willing to devote resources at the country level to newborn survival. Many had come to recognize that global progress on MDG 4 required reduction in neonatal mortality, the slowest declining component of child mortality.³¹ Their awareness came in part from the advocacy of a global informal network of newborn survival champions that had formed in the second half the 2000s and that, in concert with SNL, had been exercising global leadership on the issue.³¹

In 2009, Bangladesh became one of the few governments of low-income countries to endorse an official strategy exclusively focused on neonatal health.⁴⁷ The detailed document, which the Ministry of Health and Family Welfare published, outlined strategies for addressing newborn survival and reiterated a call from a report on the national health plan to reduce neonatal mortality to 22 per 1000 live births by 2015.³⁰

The document appeared unexpectedly. A discussion began in 2006 after a regional meeting in Myanmar that involved the United Nations Children’s Fund and the World Health Organization. In 2007 SNL organized a trip to Nepal for several officials, including a joint secretary from the Ministry of Health and Family Welfare, to investigate newborn sepsis treatment practices. The secretary learned that Nepal had its own National Neonatal Health Strategy and asked delegation members why Bangladesh did not have the same. Sensing an opportunity, the new leader of SNL Bangladesh and other neonatal survival champions initiated a process to draft such a strategy. They convened approximately 80 individuals, including officials from the government, donor agencies, NGOs, and medical associations. After extensive consultations and negotiations—some difficult, as agencies advanced different agendas—a draft was produced in 2008 that gained the official endorsement of the government in 2009.

The emergence of this strategy had much to do with the activities of an informal network of 6 individuals, who by the end of the decade had come to constitute the core of the country’s newborn survival policy community. Linking newborn survival proponents in SNL, 2 of the medical associations, a unit in the Ministry of Health and Family Welfare and a Bangladeshi research institution, they initiated the process of drafting the document and took up official positions in the working groups that produced the strategy. This network continued thereafter to function behind the scenes to exercise national guidance over the issue, exerting as much influence as any of the formal organizations working on newborn survival. Whenever any major development took place on newborn survival, these individuals met beforehand to plan strategy.

Influence of Political and Health Environment From 1998 to 2011

Influences in the political environment have also shaped newborn survival promotion. In cultivating government and donor attention, champions have had to confront obstacles connected with national politics and governance of the health sector. From independence in 1971, Bangladeshi politics have been characterized by unstable, semidemocratic rule, punctuated by military-led martial law, political assassinations, and caretaker governments.⁵³ Since 1991, power has alternated between 2 political parties: the secularist Awami League and the center right Bangladesh National Party. As each party has won elections, the other has been reluctant to concede power, organizing national strikes, accusing the other of vote rigging and corruption, and enacting parliamentary walkouts. This instability has created problems for policy continuity. It is never clear whether the new regime will sustain the priorities of its predecessor, a problem with consequences for newborn survival. For instance, newborn survival proponents have had to reeducate new politicians and civil servants on existing policies each time there has been regime change.⁵⁴

The system of health governance, too, has posed difficulties for newborn survival promotion. In the Ministry of Health and Family Welfare, the locus of control over the issue of newborn survival is fragmented, with at least 3 units claiming some authority, hampering the ability of the government to take leadership on the issue of newborn survival. The ministry has had a long-standing problem with governance and fragmentation, in part because of the bifurcation and rivalry between its 2 directorate generals, a problem that has proved politically difficult to resolve despite several reform efforts.⁵⁵ This has resulted in tensions even among grassroots workers, with implications for newborn survival. It is unclear which directorate general’s set of cadres should take the lead on newborn care.

Another health governance problem is weak human resource capacity, although this has affected implementation more than policy attention. Bangladesh has a critical shortage of nurses and midwives and of well-equipped health facilities,^30,55 and only 26.5% of women deliver with skilled attendance and only 23.4% in facilities.⁵⁶ Despite several programs to address this problem, including a government initiative to train community skilled birth attendants, this issue is not likely to be resolved for decades.³⁰

Impact of Newborn Survival Promotional Efforts Through 2011

These obstacles notwithstanding, there is considerable evidence that the issue of newborn survival moved from a position of inattention to a government concern over the course of a decade and that advocacy efforts shaped this change. The creation of the National Neonatal Health Strategy, the inclusion of neonatal mortality reduction as a goal in national health plans, and the initiation of multiple government and donor-led programs with newborn survival aims all indicate the emergence of attention and the impact of advocacy. Moreover, neonatal mortality is prioritized in the government health sector program for 2011 through 2016. The program includes newborn care as a key service component,⁵⁷ with an operational plan solely dedicated to maternal, neonatal, child, and adolescent health services. That plan includes an aspiration to put in place 13 500 community health workers to deliver primary health care at the grassroots, including newborn care.⁴⁰ The estimated budget for this operational plan is $384 million, 14% of the total budget and second highest among the 32 in the health sector program. Explicitly noting slow progress on newborn survival, the program includes a target of reducing neonatal mortality to 21 per 1000 live births by the year 2016.

National uptake of interventions also improved over the course of the decade (Table 2). The percentage of mothers initiating breastfeeding within 1 hour of birth rose from 17.0% to 43.0%, and the percentage receiving a postnatal checkup within 2 days of delivery increased from 10.6% to 22.5%. Also, neonatal mortality may have declined 31.0% over the decade, from 42.7 per 1000 live births in 2000 to 29.5 in 2009 (Figure 1) according to estimates from a recent study, although the study uses a statistical model, so there is uncertainty about the extent of change.⁵ The role of advocacy in these intervention uptake and mortality changes is unclear, as other factors likely contributed, such as improving socioeconomic conditions, health programs not connected to neonatal survival, and private sector health activities.

TABLE 2—

Intervention Uptake Connected to Newborn Survival: Bangladesh, 2000–2011

Indicator	Early 2000s, %	Mid to Late 2000s, %
Mothers initiating breastfeeding within 1 h	17.032	43.049
Mothers receiving postnatal checkup within 2 d of delivery	10.658	22.556
Mothers receiving ≥ 4 antenatal care visits	11.658	23.456
Births delivered by skilled attendants	12.058	26.556
Births delivered in a health facility	9.258	23.456
Mothers receiving ≥ 2 tetanus shots	64.032	83.049

FIGURE 1—

Estimated neonatal mortality rate: Bangladesh, 2000–2011.

Source. Oestergaard, Inoue, Yoshida, et al.⁵

ANALYSIS

In 2000 the government and donors paid little attention to newborn survival. By 2011 the issue had become a concern to both. Factors in the framework help explain this shift (Table 3). Transnational influences contributed: international agencies and governments approved MDG 4 in 2001, which served to advance an existing global norm (factor 1) that states must act to save the lives of children. As evidence spread that newborns constituted more than 40% of child mortality, ensuring newborn survival became an increasingly prominent element of that norm. Financial and technical resources (factor 2) from several international agencies followed, including Save the Children, USAID, and the Gates Foundation. The Bangladeshi state embraced the child survival norm of MDG 4 and was a beneficiary of these resources.

TABLE 3—

Framework Factors Applied to Policy Attention for Newborn Survival: Bangladesh, 2000–2011

Factor	Status Before 2000	Status by 2011
1. Norm promotion	No global norm concerning saving the lives of newborns	MDG 4 helps to advance a norm that governments must act to protect the lives of children, including newborns; Bangladeshi state embraces this norm
2. Resource provision	No major international agencies providing significant resources exclusively for newborn survival	SNL has formed, with an exclusive focus on newborn survival and a strong presence in Bangladesh; Gates Foundation, USAID, UNICEF, and other agencies supporting projects in the country
3. Policy community cohesion	No newborn survival policy community	A cohesive policy community for newborn survival is in place
4. Political entrepreneurship	Some medical professionals have a concern for newborn survival; none consistently pushes the Bangladeshi government	Several health professionals from medical associations act as political champions; a Bangladeshi physician in SNL plays a central role in launching and cultivating attention
5. Credible indicators	Data on newborn survival have not been deployed for advocacy	Policy community and government actively use data on newborn survival, including from BDHS, to promote action and assess progress
6. Focusing events	No focusing event for newborn survival	Several focusing events have been organized; most influential is a 2003 joint meeting of the Healthy Newborn Partnership and International Perinatal Congress
7. Clear policy alternatives	Most medical professionals believe very sick newborns cannot be saved because of the lack of available high technology	Projahnmo project and Abhay Bang’s work convince many Bangladeshi policymakers that much can be done to save newborn lives
8. Political transitions	Ongoing political instability in Bangladesh as power passes back and forth between 2 rival parties	Same political dynamic persists, forcing newborn survival proponents constantly to reeducate government officials
9. Competing health priorities	Newborn survival has not been given priority but unclear whether this is because of policymaker attention to other issues	Unclear whether competing health priorities have hampered attention to newborn survival; however, attention to child survival may have facilitated newborn survival priority

Note. BDHS = Bangladesh Demographic and Health Survey; MDG = Millennium Development Goals; SNL = Saving Newborn Lives; UNICEF = United Nations Children’s Fund; USAID = United States Agency for International Development.

But these transnational influences were insufficient by themselves to get newborn survival on the policy agenda; domestic advocacy also played a role. Domestic political entrepreneurs (factor 4) emerged who carefully cultivated the attention and support of the Bangladeshi state, particularly a Bangladeshi physician in SNL and leaders of several domestic medical associations with long-standing concern for newborn survival. They coalesced into a policy community (factor 3) and eventually formed an informal network exercising leadership on the issue, linking the government, SNL, a UN agency, these medical associations, and a Bangladeshi research institution.

The policy community also organized attention-generating focusing events (factor 6)—most prominently a 2003 joint meeting of the Healthy Newborn Partnership and Bangladesh Perinatal Congress—which served to further advance attention. They also generated credible indicators (factor 5) via the insertion of measures on the coverage of essential newborn care into the Bangladeshi Demographic and Health Surveys and feasible policy alternatives (factor 7) surrounding home-and community-based care emerging especially from Projahnmo research. These provided evidence on the severity and tractability of the problem.

The national political environment also influenced advocacy efforts. Ongoing political instability because of competition between the Awami League and the Bangladesh National Party (roughly, factor 8) constrained advocates in making progress; however, a long-standing government commitment to child survival as a health priority (roughly, factor 9) facilitated advocacy.

The case reveals some need for modification in the framework. Although the first 2 categories—transnational influence and domestic advocacy—identify factors that shaped policy attention in Bangladesh (Table 3; factors 1–7), factors in the third category of national political environment require some modification. The 2 elements in the framework in that category—political transitions and competing health priorities (Table 3; factors 8 and 9)—are applicable to the case but only if their meaning is stretched. There was no fundamental political transition in Bangladesh in this time period. The political system persisted as unstable and semidemocratic as 2 rival political parties alternated in power, each refusing to give up power according to democratic norms. This flux posed problems for advocates, as they were never sure whether a new government would embrace the priorities of its predecessor. Political instability, rather than political transition, was the force at work in this case. Further cross-national research is necessary to identify the full set of characteristics of political systems that are most influential on health agenda setting in low-income countries.

Also, although competing health priorities may have hampered attention to newborn survival (our data did not enable us to assess this), we found evidence of another dynamic facilitating attention: the ongoing priority of child survival, a concern of the Bangladeshi state since the 1970s. Rather than speak of competing health priorities, it may be better to term this factor “existing health priorities,” as these may in some cases hamper attention by claiming scarce resources and in other cases facilitate attention, especially if a new health issue is congruent with existent priorities.

The case may point to the need for an even more fundamental modification to the framework. The framework neatly separates transnational and national influences, as is the custom in research concerning international public policy processes. But are these influences so neatly separable? The SNL Bangladeshi physician was a member of global and national networks of actors concerned with newborn and child survival, facilitating her capacity to influence the Bangladeshi state. It is unclear then whether she is best understood as a transnational actor, a national actor, or a mix. Similarly, most of the core members of the national newborn survival policy community (all Bangladeshi nationals) had transnational links; for instance, some were employees of international agencies, and others were well-recognized international researchers. The same questions may be raised about other causal influences—the MDGs, the indicators in the BDHS, the policy alternatives that were adopted—and even the identity of SNL as an NGO in Bangladesh. All were composed of transnational and national elements. Does the framework, and global health policy analysis more generally, need to dispense with simple demarcations that classify some causal influences as transnational and others as national and with the conventional question concerning which sets of influences are more powerful in setting national policy agendas? Is there a need to consider how actor identities and policy processes are fused and the demarcation of national borders and identities increasingly less relevant for health policy analysis? We have offered no clear answer to these questions but raise them as points for consideration that emerge from analysis of the case.

CONCLUSIONS

Newborn survival in Bangladesh is a case of successful advocacy for the placement of a health issue on the policy agenda of a low-income country. Neglected in 2000, by 2011 neonatal mortality reduction had emerged as a government health priority. The continuity and growth of policy attention for newborn survival may depend on factors identified in the framework, specifically the following:

• capable political entrepreneurs who continue to push for the cause,

• the persistence of a cohesive policy community,

• the ongoing publication of credible data on intervention uptake and mortality levels,

• evidence-based consensus on the set of interventions and policy alternatives needed to make progress on newborn survival,

• global agreements on child survival that put normative pressure on nation-states to act, and

• the availability of resources from international donors to augment any funds the government is willing to provide.

These factors may also be relevant for promoting other health issues in low-income settings. Some factors, such as the emergence of capable political entrepreneurs, are not easily cultivated. Such individuals are uncommon and must come to their own decisions to support a health issue. Other factors, however, are at least in part under the control of communities seeking to advance particular health issues. Where policy communities are divided over strategy, they might take steps to bridge differences to advance their cause. Where credible data and policy alternatives do not exist, policy communities can support efforts to generate these to demonstrate the severity and tractability of the problem. And at the global level, where resources are limited and global agreements do not exist, policy communities can advocate global commitments that place normative pressure on donors to provide such support and on nation-states to act. The case of newborn survival promotion in Bangladesh suggests that although policy communities cannot control all elements of the agenda-setting process, they can enhance the likelihood their health issues will receive political priority if they act strategically to cultivate attention.