Shaping the Health Policy Agenda: The Case of Safe Motherhood Policy in Vietnam

Background: Maternal health remains a central policy concern in Vietnam. With a commitment to achieving the Millennium Development Goal (MDG) 5 target of maternal mortality rate (MMR) of 70/100 000 by 2015, the Ministry of Health (MoH) issued the National Plan for Safe Motherhood (NPSM) 2003-2010. In 2008, reproductive health, including safe motherhood (SM) became a national health target program with annual government funding. Methods: A case study of how SM emerged as a political priority in Vietnam over the period 2001-2008, drawing on Kingdon’s theory of agenda-setting was conducted. A mixed method was adopted for this study of the NPSM. Results: Three related streams contributed to SM priority in Vietnam: (1) the problem of high MMR was officially recognized from high-quality research, (2) the strong roles of policy champion from MoH in advocating for the needs to reducing MMR as well as support from government and donors, and (3) the national and international events, providing favorable context for this issue to emerge on policy agenda. Conclusion: This paper draws on the theory of agenda-setting to analyze the Vietnam experience and to develop guidance for SM a political priority in other high maternal mortality communities.


Implications for policy makers
The available evidence will help to justify the severity of health problem • The perceived quality of evidence is not always meet all criteria (credibility, generalisability, reliability, objectivity, and rootedness) • Advocacy skills are important for getting attention from policy-makers to health problem.

Implications for public
High maternal mortality remains a central policy concern in Vietnam. With great efforts from Ministry of Health (MoH) and other international development partners, safe motherhood (SM) policy aiming to reduce maternal mortality, was received attention from government and became a national health target program with annual government funding since 2008.

Key Messages
Background Maternal health is a central policy concern in Vietnam, as reflected in the National Strategy on Reproductive Health 2001-2010. 1 In 2001, the national maternal mortality rate (MMR) was 165/100 000 live births and much higher in remote and mountainous regions 411/100 000. 2,3 With a commitment to achieve the MDG 5 target to reach MMR of 58.3/100 000 by 2015, the Ministry of Health (MoH) issued the National Plan for Safe Motherhood (NPSM) 2003-2010, which aimed to ensure safe motherhood (SM) services available to every woman and newborn. 4 In 2012, MoH renewed the NPSM and Newborn Care 2011-2015 with particular attention to difficult regions and vulnerable groups. 5 In the last 15 years, SM still high profile in the policy agenda in Vietnam. This paper explores how the political priority for SM emerged and was sustained during period 2001-2008 in Vietnam. The results of this study should be of interest to policy-makers, researchers and other policy actors interested in improving their understanding of processes for initiating health policies in Vietnam and similar contexts. We use the Kingdon's theory of agenda-setting as a framework for analysis. 6 Agenda-setting is the stage of the policy process during which the issues rises to prominence. It is the first stage in policy processes and precedes the others: development and implementation. 7 Political priority refers to the degree to which political leaders consider an issue to be worthy of sustained attention and back this up with provision of financial, human, and technical resources. 8 According to Kingdon, an issue emerges on national agendas when three streams -problems, policies, and politics -flow together and eventually converge. 6 The problem stream refers to broad issues faced by societies. The policy stream refers to the proposed interventions to address the problems. The politics stream refers to global or national political events that are associated with the problem. According to Kingdon, convergence of these 3 streams can create favourable environment to open a window of opportunity to place a particular issue on a policy agenda.

Methods
This paper is developed from a larger HEPVIC project -"Health Policy Processes in Vietnam, India, and China" -a multi-partner 3-year research which explored maternal health policy processes in the 2 Asian countries. 9,10 The research analyzed all stages of policy processes: policy agenda-setting, development and implementation. 9 In Vietnam, the NPSM 2003-2010 was selected to study the policy on skilled birth attendance (SBA). 10 The data collection was undertaken between December 2006 and July 2008, using in-depth interviews and focus group discussions with key policy actors and document review, as we set out next.
Purposive sampling was used to develop the initial list of respondents for interviews on the basis of their roles in health policy processes, which was followed by snowballing technique to identify further respondents. There were 13 indepth interviews and 1 focus group discussion with policymakers, managers/planners, researchers, civil societies, international development partners/donors and politicians. All interviews and a focus group discussion were conducted using a semi-structured guide which included questions exploring respondents' understanding of policy processes, their role in these processes as well as their perceptions of strengths and weaknesses of these processes. Following informed consent from all respondents, all interviews and focus group discussion were audio-recorded and transcribed for analysis. A total of 37 documents related to SM were reviewed. 10 These documents included government policies (5), minutes of meeting (1), annual reports (11), newspapers (4), websites (2), research reports (12) and guidelines by government (2). These documents were identified through the discussions with key informants and searching the MoH, government and programme websites. The research team at the Hanoi School of Public Health, Hà Nội, Vietnam supported by European partners, carried out the data collection and analysis. The data was analysed using a framework approach 11 comprising 5 stages (familiarisation with the data, identifying thematic framework, indexing, charting, and mapping and interpretation), with the support of qualitative data analysis software (NVivo 7.0). As mentioned earlier, the analysis of data reported in this paper was guided by Kingdon's multiple streams theory of agenda-setting 12the only theory we found which helped us to understand the complexity of agenda-setting processes. The multiple streams theory helped to categorise the data for analysis (ie, provided thematic framework and informed indexing and charting of data) and structure reporting of the results. Draft findings were discussed with, and validated by, researcher partners and other stakeholders.

Results
Health policy processes are generally government-led, and there is increasing recognition of the involvement of actors outside both the government sector and the health system. [13][14][15] In Vietnam, since 2008, all policies have to follow the Law on the promulgation of legal documents, 16 which stipulated the following stages for public policies: policy proposal, development, implementation, and evaluation. The main factors affecting the policy processes which emerged from our analysis and reported elsewhere are: nature of the policy issue, involvement of different actors and national and international context. 10 In recent years, under the Law on promulgation of legal document, the health policy processes in Vietnam have become more consultative, involving more actors, and taking into account people's values and beliefs. 15,16 Figure 1 summarises the key stages of NPSM policy processes in Vietnam. The NPSM policy processes were: agenda-setting, development and nationwide implementation. 10 The     19,20 In Vietnam, the first national workshop on policy for essential emergency obstetric care was organized in 1999. In this workshop, the MoH policy-makers emphasized the need to upgrade staff knowledge and expertise. The indicator: "deliveries with assistance of health workers" was renamed as "deliveries with assistance of trained health workers. " The concept of "trained health worker" was interpreted as "skilled birth attendant"; specific examples included nurses, midwives and assistant doctors trained to be able to handle a 'clean and safe delivery, ' meaning avoiding complications and deaths. Only in 2014, the core skills of SBA that recommended by WHO was approved by MoH in Vietnam. 21 The Confluence of Streams All 3 streams of Kingdon's agenda-setting (problem, policy, and politics) were evident in the emergence of SM as a policy priority in Vietnam. In 2003, these streams converged which appear to have created a window of opportunity to get this issue onto the policy agenda. At that time, the results of survey showing high MMR were recognized by the key policy actors as credible and the problem was pinpointed. At about the same time, as shown earlier, the government and donors agreed on policy development and intervention plan thus giving coherence to the policy stream. In relation to the political stream, the MDGs in particular contributed to a favorable environment for the development of NPSM. "MDG is the most important goal that MoH should be targeted to in terms of reducing MMR in the country" (Policy-maker). This confluence of 3 streams led national policy-makers to give sustained attention to SM issues, authorized a national target program and resulted in the continuous allocation of a state budget for the program since 2008 until now.

Discussion
SM became one of Vietnam's foremost health priorities, and between 1990 and 2003, the maternal mortality was declined from 230 to 165/100 000 live births. This problem was clearly perceived as the responsibility of the health sector and was coordinated by the MoH RH department. The agenda-setting of NPSM in Vietnam could be seen as quite linear process, and emerged almost simultaneously with confluence of 3 streams and having approval of NPSM. 10 Our findings suggest that different factors influenced the speed of agenda-setting for NPSM in Vietnam. These are existence of clear policy option, cohesion between policy actors, and existence of credible indicators, as we set out next.

Policy Community Cohesion
Policy communities play important roles in shaping national priorities, including SM. 22 A similar finding was reported in this case. The policy community for the NPSM policy included the MoH policy-makers, obstetrician, Vietnam Women Union, CSOs and development partners (RNE) and other implementers at local level. All held commitment to reducing maternal deaths, as shown earlier. Most actors at MoH had medical background and therefore, the issue of SM was considered as a technical one. This explains why the Women Union was not invited, as being perceived by the MoH as having lack of technical knowledge. This perception contrasts with findings from other studies of SM policy development such as in Nigeria or India, where the CSOs were seen differently for example in White Ribbon in India. 23 In our study, the roles of policy champion are prominent in agenda-setting, a finding which is similar to other contexts. In India, the Prime Minister played significant role in shifting the SM from a condition into a problem. In Nigeria, Honduras, Indonesia, and Vietnam, the MoH officials played important roles in developing SM policies. 7,23,24 It can be seen that the policy champion is typically a powerful individual, located at national level and having good connections with different stakeholders including donors. The person can disseminate, advocate and mobilize support and resources. Furthermore, the person can actively facilitate placing problems onto the policy agenda, instead of waiting for windows of opportunity to emerge. However, in Vietnam, it appears that the policy issue and global political agenda have more influence than power from individual policy actors. 10 The development partners can often be important in policy processes. We found that the active participation of RNE had a special effect on the SM agenda. They provided funding to develop, and later implement, the SM policy, which contributed to acceptance of SM as a priority by the government. Similarly, in other countries powerful external partners provided crucial financial and technical support in policy processes: for example, UNICEF, and WB in India; DFID Path in Nigeria; USAID, WB, the Netherlands Embassy in Honduras. 7,23 Although there is no direct evidence from our study, the long history of RNE support to the health sector may have also contributed to their credibility and power in the health sector, as compared to other bilateral and multilateral agencies.

Clear Policy Option
The international norm that advocated for SM such as obstetric care and SBA provided a clear intervention for reducing MMR in the country. 22 The policy option, which was deliveries with assistance of SBA, was non-controversial. However, some modification was introduced for the Vietnamese context: institutional delivery in mainland areas and home delivery by trained health workers in mountainous areas.
We found that existence of a global norm and a clear policy option were an influential force in the country's decisions to address the causes of maternal mortality. This made it easier to bring the topic to the policy table, increased the likelihood of obtaining support from politicians, policy-makers and development partners, and consequently reduced the time for agenda-setting 10 ; a finding which is similar to the results of studies found in India and Indonesia. 8,24 Credible Indicators It is important to have sources of credible evidence to highlight the issues such as high MMR to indicate the severity of problem to the policy-makers. 22 These sources were clearly identified in other countries such as National Family Health Survey in India, 8 maternal mortality survey in Honduras 7 and IDHS survey in Indonesia. 24 A similar phenomenon was reported in our study, with maternal mortality survey Vietnam 2 considered as trustworthy by many policy-makers in making the decision to develop the NPSM. On the other hand, when the reliable data was not available, policy-makers may ignore the issue because they are unaware of problem or unconvinced due to absence of evidence that any problem exists.
Policy Implications for Future Agenda-Setting Initiatives After 10 years, the SM policy remains high on the health policy agenda in Vietnam. The new National Plan of SM was developed for 2011-2020 with a focus to reduce the disparity in MMR rates between the regions. The SM program continues receiving the annual government funding. Three lessons can be shared from our study to inform future efforts on generation of political priority for the public health agenda. First, credible evidence to catalyze action and the strategic deployment of data are important to identify the problem and highlight it to key policy actors. Second, powerful policy champions are crucial in driving the policy process and convincing donors and other policymakers on the needs of developing a particular policy. The MoH officials can often lack effective advocacy skills, which appear essential for such champions. Third, the power of relevant national and international events, informing simple and effective policy proposals cannot be underestimated. In our case, the SM proposal focused on WHO-recommended effective interventions. Such proposals can easily get attention by the policy-makers, compared with more complex ideas which can often have less clearlydefined gains.
In seeking to draw principles from our study to other settings, there is a need for caution. Although not evident in our data, there are some unique aspects of the political context of Vietnam that shaped SM issues, which can also be regarded as preconditions for the agenda-setting in other contexts, in addition to the three streams discussed earlier. Vietnam has a relatively well-functioning health system and good health indicators, compared with countries at similar income levels.
The health system is well-developed, for example reflected in nationwide coverage of essential healthcare services. The government is demonstrating commitment to providing quality healthcare for its people, especially to reduce MMR. The wider reforms allowed the government to have better economic growth with potential further benefits to the society, such as increased resources to achieve better health outcomes.

Conclusion
The generation of priority for SM in Vietnam required combination of three streams: (a) development of reliable data to mark the severity of problem, (b) the persistence and commitment of policy champions leading to support from other policy actors such as international donors and government, and (c) the organization of attention-generating international and national events. The additional unique aspects of the context of Vietnam, such as existence of health reforms, can also be seen as prerequisites for the policy agenda-setting in other settings. Three policy implications for future agenda-setting initiatives are proposed, including the generation of credible evidence, existence of powerful champions and building on related national and international events.

Ethical issues
The ethical approval for this study was obtained from the Institutional Review Board of Hanoi School of Public Health, Hà Nội, Vietnam.
the manuscript. All authors read and approved the final manuscript.