If the Sustainable Development Goal (SDG) target 5.3 to end child marriage by 2030 is to be met, the annual rate of reduction in the prevalence of child marriage must increase from 1.9% to 23%. Over 30 countries have developed, or are developing, national policies/programmes towards this goal. However, many are struggling to operationalise these policies/programmes, particularly at subnational levels. Thus, Girls Not Brides and the WHO commissioned a review of lessons learnt from national and subnational implementation of multi-sectoral policies/programmes targeting other issues that could be applied to the global movement to end child marriage. This review identified a number of pragmatic lessons learnt. At the national level, countries should identify and engage committed and skilled leadership, build a shared understanding of the target issue and how to address it, and delineate and clarify the roles and responsibilities of relevant stakeholders. At the subnational level, countries should establish coordination mechanisms, build awareness and capacity of staff, use subnational evidence to contextualise and tailor interventions, develop coordinated budgets and cost-sharing mechanisms, and integrate monitoring and evaluation systems. These lessons are remarkably consistent, despite coming from different target issues and contexts. The commonality of these findings reveals that various stakeholders are repeatedly and consistently failing to ensure that these fundamental requirements are in place. It is vital that the global movement to end child marriage learns from and uses these lessons if it is to meet its SDG target.
- health policy
- maternal health
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
A recent review of research priorities on multi-sectoral collaboration for health in low-income and middle-income countries found that the top-ranked questions were predominantly pragmatic questions of how to structure, implement and sustain multi-sectoral coordination.
At the national level, countries should identify and engage committed and skilled leadership, build a shared understanding of the target issue and how to address it, and delineate and clarify the roles and responsibilities of relevant stakeholders.
At the subnational level, countries should establish coordination mechanisms, build awareness and capacity of staff, use subnational evidence to contextualise and tailor interventions, develop coordinated budgets and cost-sharing mechanisms, and integrate monitoring and evaluation systems.
These findings demonstrate that the challenges and lessons learned for multi-sectoral coordination are remarkably similar across target issues and contexts.
These commonalities also indicate that relevant stakeholders are repeatedly and consistently failing to ensure that these fundamental requirements for successful subnational implementation of multi-sectoral policies/programmes are in place.
It is vital that the global movement to end child marriage—as well as movements to address other health issues through multi-sectoral approaches—learns from and uses these lessons if it is to achieve its aims.
Globally, rates of child marriage are on the decline: in the past decade alone, 25 million child marriages have been averted.1 However, progress has been uneven across and within regions, as well as countries. While North Africa and the Middle East have decreased the percentage of girls married by age 18 years by half, and South Asia has reduced a girl’s risk of marrying before age 18 years by more than a third, less change has been observed in Latin America and sub-Saharan Africa.1 2 Additionally, evidence from countries such as India and Ethiopia points to substantial variability in subnational rates of decline.3 4 To meet the Sustainable Development Goal target 5.3 to end child marriage by 2030, the annual rate of reduction in the prevalence of child marriage would need to be increased from 1.9% over the past 10 years to 23%.1 5
In recent years, as the evidence base on effective strategies to end child marriage has grown, over 30 countries have developed or are in the process of developing national strategies and National Action Plans (NAPs) to end child marriage.6–11 (Since 2015, Girls Not Brides: The Global Partnership to End Child Marriage has engaged in a process of documenting and learning lessons from the development and implementation of national strategies to end child marriage, the findings of which were published in consecutive reports in 2015 and 2016.) Given the numerous and complex drivers and consequences of child marriage, all of these policies/programmes are multi-sectoral in nature.12 As such, they create a holistic framework to help the sectors involved in preventing child marriage and supporting already married girls—health, education, child protection, justice, social protection and others—to coordinate and integrate their activities with the allocation of the necessary human and financial investments.13 However, many countries are struggling to implement and operationalise their multi-sectoral policies/programmes, particularly at subnational levels.10 Additionally, a recent review of research priorities on multi-sectoral collaboration for health in low- and middle-income countries found that the top-ranked questions were predominantly pragmatic questions of how to structure, implement and sustain multi-sectoral coordination.14
To respond to this knowledge gap, this analysis sought to identify lessons learnt on strategies for multi-sectoral policy/programme implementation at national and subnational levels. However, as child marriage policy/programme implementation is still in its infancy, little evidence exists on the pragmatic aspects of subnational multi-sectoral child marriage policy/programme implementation. For this reason, the authors conducted a literature review and key informant interviews on other, more established policy/programme areas—such as HIV/AIDS, nutrition, maternal health and gender-based violence (GBV)/violence against women (VAW)—to draw out critical actions/elements that could be applied to efforts within the global movement to end child marriage (online supplementary annex 1).15–22 This review identified a number of pragmatic lessons learnt from eight key case studies for action at national and subnational levels (online supplementary annex 2).
At the national level
Identify and engage committed and skilled leadership
The importance of high-level commitment from key decision-makers in government was paramount across all eight case studies. ‘Housing’ the target issue at the highest institutional levels (ie, within the office of the prime minister or another prominent figure) can facilitate sustained prioritisation of the issue on the government agenda and budget allocation. Where this was not possible, engaging high-level political figures and/or committed champions in the policy/programme development and implementation processes similarly helped stakeholders to advocate for political commitment, prioritisation and budget allocation. It can also support concrete action through institutional arrangements (box 1).
Illustrative examples: Ensuring there are committed decision-makers
In Senegal, the Committee for the Fight against Malnutrition, a multi-sectoral coordinating body for nutrition, was housed within the Prime Minister's office, with monitoring of the framework of results at his level.15 As a result, the Committee became a centralised political and administrative hub that coordinated sustained attention to nutrition across all sectors and led to the launch of the 15-years Programme de Rènforcement de la Nutrition (PRN—Nutrition Enhancement Programme).
In India, the state of Maharashtra’s Chief Minister personally committed himself and his office to tackling the issue of malnutrition.16 This prioritisation was prompted by advocacy and pressure from civil society and media about the unacceptability of child deaths in his electoral district, which caused some political embarrassment. The Chief Minister continued to chair the State Malnutrition Monitoring Missions’ bimonthly meetings and sustained prioritisation and commitment of the relevant ministries under his jurisdiction.
In Malaysia, a female doctor from the Kuala Lumpur General Hospital spearheaded the development of an interdisciplinary model of healthcare for women who experience abuse and piloted the first One Stop Crisis Centre in 1994.17 However, the lack of commitment among key decision-makers, reflected in the lack of clear policy guidance and operational details from the Department of Health, left the implementation of services to the subnational level, with consequences for quality.
Build a shared understanding and vision of the target issue and how to address it
Different actors may frame and conceptualise the target issue from individual positions and mandates. Developing a common understanding of the target issue at the outset and basing it on rigorous evidence can ensure a shared vision and sense of purpose among different sectors and stakeholders of the target issue and how to address it. This process may be done by developing a common theory of change and clarifying how multi-sectoral action is linked to the broad set of preventative and responsive measures needed to address the issue (box 2).
Illustrative example: Building a shared vision for the multi-sectoral approach
In Ethiopia, the health, economic and social sectors of the government were intimately involved in developing and launching the 2013 National Nutrition Plan and its vision.18 The plan was widely considered legitimate because of the broad consultation and engagement of all relevant sectors that provided technical input into framing nutrition as both a food security/agricultural development concern as well as a health and behaviour change issue, as well as the appropriate strategies to fight malnutrition through a multi-sectoral approach. Additionally, effort was given to understanding the major drivers of nutrition problems in each of the four subnational regions, which supported improved tailoring of the interventions.
Delineate and clarify the roles and responsibilities of multi-sectoral stakeholders
The most common challenge identified across all eight case studies was the lack of clear roles and responsibilities of the stakeholders involved, including government departments/ministries, civil society, the private sector and donors. At the outset, it is important to set out the different roles and responsibilities for these stakeholders (as described below), as well as how their complementary contributions will come together (box 3).
Illustrative examples: Developing clarity on the roles and responsibilities of stakeholders
In Malaysia, the Ministry of Health issued a directive to hospitals to establish comprehensive, multi-sectoral centres for the management of GBV cases.17 However, the directive did not clarify how the centres were to be created, who should direct them, how partnerships with civil society organizations (CSOs) should be structured, how GBV cases should be managed, and how the centres should be funded. Each hospital was therefore left to make its own decisions on how to manage the various agencies involved, including CSOs, the police and religious groups, with major consequences for quality and sustainability of services.
In Zambia, Concern Worldwide played a crucial and well-defined role in the implementation of the national nutrition policy in Mumbwa District.19 They supported the establishment of a District Nutrition Coordinating Committee (DNCC), led an extensive accompaniment programme to study the scale and drivers of malnutrition (including direct consultation with community members to ensure that local perspectives were duly understood and included in local plans), and built the capacity of the DNCC on use of data to develop action plans and on managerial skills to strengthen coordination and joint planning. The success of this partnership is exemplified by the fact that the DNCC is still functional, despite the fact that Concern has since exited the project.
In South Africa, tensions between government and civil society at the time of the establishment of local AIDS Councils, as demonstrated by a lack of trust, shared understanding and joint accountability, affected the effectiveness of AIDS Councils in contributing to the country’s multi-sectoral response to HIV/AIDS.20
Government, as the responsible system for policy/programme developmentand implementation and service delivery, should play the central role in subnational implementation of national policies/programmes if they are to be scaled up and sustained. The ‘government’ is not monolithic but made up of diverse actors and sectors working at many different levels. Therefore, it is vital that the roles and responsibilities of different government departments/ministries are developed through a consultative process with the respective stakeholders so that all stakeholders have a sense of ownership and joint responsibility, and also a sense of accountability for their own specific deliverables in implementation of the multi-sectoral initiative. Importantly, these roles and responsibilities should be delineated at national and all relevant subnational levels and across sectors. Once roles and responsibilities are established, these should be communicated clearly at all levels to the relevant stakeholders, along with the relevant communication channels and reporting requirements and processes.
Civil society supports government by providing complementary technical support in a variety of areas. These may include evidence generation to inform subnational decision-making, advocacy for sustained prioritisation of the issue and accountability to national policies/programme commitments and capacity building of stakeholders at national and subnational levels. They are also well placed to directly engage with local populations in defining an issue or problem, identify the multiplicity and diversity of needs, practices and drivers of specific issues, and to formulate localised solutions to resolve these issues. It is also apparent that civil society plays a vital role in service provision where and while government systems are weak. However, to ensure that civil society contributions are complementary rather than duplicative and that civil society does not permanently assume government functions, the case studies illustrate the importance of developing and communicating the roles and responsibilities of civil society partners and other stakeholders. To do this, government and civil society partners should engage in intensive and sustained consultation to understand civil society partners’ added and complementary value, their geographical area of work, area of expertise and their capacity for sustained engagement. An ongoing challenge identified for civil society is its ability to negotiate the multiple roles it often plays simultaneously when working with government as a policy developer, service provider and advocate.
In many contexts, the private sector is a highly influential player in lobbying for policy/programme development/modification, swaying public opinion and providing services. However, the case studies indicate very little strategic and sustained coordination between government and the private sector during subnational implementation of policies/programmes. Specifically, it was noted that government and civil society stakeholders often have little access to information about the private sector’s plans and operations. Multiple case studies cited the need to proactively engage the private sector and regulate its role in implementation, to ensure services are in line with national policies and complementary to government services.
Donors can play a positive role in giving priority to an issue, pressing for multi-sectoral coordination and sustaining momentum for subnational implementation. The case studies point to two key messages on the roles and responsibilities of donors. First, donors can facilitate national and subnational implementation by providing strategic financial resources (ie, catalytic funding to stimulate strengthened action on specific areas of weakness) and technical support (ie, sharing evidence from international best practice on what works and does not work). Second, it is critical that donors respect principles of aid effectiveness by operating within government-led structures and mechanisms to avoid creation of parallel processes, and coordinating with other donors and with all grantees within the country to ensure harmonisation of investments. However, the case studies noted that power dynamics often restrict governments’ comfort and ability to ensure that donors respect and fulfil their roles and responsibilities.
At the subnational level
Establish coordination structures and mechanisms to direct and operationalise multi-sectoral action
Coordination structures (ie, committees, councils, working groups, task forces) can be powerful mechanisms to direct and support implementation of multi-sectoral policies/programmes. These groups typically involve representatives from the relevant departments/ministries, and sometimes are broadened to include representatives from civil society and the private sector. Their mandates commonly include operationalising multi-sectoral policies/programmes by defining roles and responsibilities of relevant stakeholders, ensuring coordination and accountability of relevant stakeholders, supporting costing and budgetary allocation, promoting adequate staffing at national and subnational levels, and developing and coordinating a joint monitoring and evaluation system (box 4).
Illustrative example: Limitations of coordination structures
In South Africa, local AIDS councils were established in Mpumalanga Province to support multi-sectoral implementation by mirroring the role of the South African National AIDS Council (SANAC) at the subnational level.20 However, voluntary membership and lack of financial support for participation in the ACs led to poor government and private sector engagement and frequent membership changes. Similarly, lack of decision-making power and funding, coupled within adequate senior political leadership, hindered operationalisation of the ACs’ recommendations.
However, as illustrated by the case studies, coordination structures require a number of facilitating factors to be effective. Specifically, coordination structures required resources to incentivise collaboration, authority to enforce participation, and power and support from leadership at the highest levels to operationalise decisions. Additionally, parallel national and/or subnational coordination structures create additional challenges; for example, individuals may participate in multiple coordination structures; structures may have overlapping objectives, functions and responsibilities; and the transience of coordination structures may undermine their effectiveness and/or sustainability.
Build awareness and capacity of leadership and staff at subnational levels
Skilled leadership at subnational levels is required to carry out the respective roles and responsibilities of the department/ministry, as outlined by the national policy/programme and the coordination structure or mechanism. Importantly, these personnel should be able to drive identification of gaps, weaknesses and bottlenecks in implementation and prompt timely responses. To do this, leadership should negotiate multiple agendas and priorities, including by leveraging agreed-upon roles and responsibilities of each department/ministry (box 5).
Illustrative example: Building awareness and capacity of leadership and staff at subnational levels
In the Indian state of Maharashtra, between 2006 and 2012, the Director General of the Nutrition Mission (an initiative aimed at reducing acute undernutrition in children) provided strong leadership to improve understanding of the multi-dimensional nature of malnutrition and the necessary multi-sectoral coordination between government agencies and civil society organisations to improve nutrition outcomes.16 Within the government, the number of cross-departmental initiatives between the Departments of Health, Women and Child Development, Water Supply and Sanitation, and Tribal Development increased, as did understanding and buy-in from different government departments on the issue. At the grassroots level, the Mission’s Secretariat engaged extensively with field workers and village communities though a learning and monitoring approach of the initiatives underway at the village level to combat malnutrition, which also contributed to increased understanding of the multi-dimensional nature of malnutrition andchild stunting.
Subnational staff—from government officials to frontline workers—should understand the goals of policies/programmes, and be trained and adequately supported to carry out their respective roles and responsibilities. Across all case studies, subnational staffoften lacked awareness of the content and purpose of national policies/programmes, awareness of their roles and responsibilities, and capacity and support to carry out their mandated tasks with quality. This challenge was compounded by frequent turnover of staff. As a result, multiple case studies noted frustration among staff at subnational levels, as staff felt they were required to assume additional roles and responsibilities to ensure multi-sectoral coordination without adequate training, support, and remuneration.
Use subnational evidence to contextualise and tailor interventions
Collecting and using age and sex disaggregated data at subnational levels can be instrumental to better understand the nature of the issue within the specific region, communicate the relevance of the national policy/programme to subnational stakeholders and tailor interventions to the local context, based on local stakeholders’ needs, experiences and understanding of the target issue. Two of the case studies demonstrated the utility of collecting age and sex disaggregated quantitative data and qualitative data at subnational levels to highlight the specific needs of populations in different regions (and the specific needs of key populations within these different regions), and tailor and target interventions accordingly. Qualitative data, especially, can reflect important information on drivers, including those related to social norms, of the target issue (box 6).
Illustrative example: Using subnational evidence to contextualise and tailor interventions
In Ethiopia, a multi-sectoral nutrition programme highlighted the importance of collecting subnational data to better understand the factors that drive malnutrition at local levels and to design and implement localised solutions.18 At the national level, evidence showed that the main drivers of malnutrition were food insecurity, undernutrition and micronutrient deficiencies. However, at the subnational level, evidence showed that the main drivers of malnutrition were food insecurity, low dietary diversity, low awareness of nutrition, and poor maternal and child feeding practices. Furthermore, the levels of importance of these drivers varied by region; indeed, the drivers of malnutrition were shown to be very much dependent on different contexts. Additionally, it was found that these subnational drivers of malnutrition could not be picked up by quantitative surveys alone. As such, the programme increased efforts to understand and respond to the nutrition problem at local levels using a targeted data collection approach that went beyond classic nutrition indicators, to further explore local practices and perspectives of the nutrition problem.
Develop coordinated budgets and cost-sharing mechanisms
Effective multi-sectoral coordination and implementation requires costings for operationalising policies/programmes, clearly allocated budget lines across departments / ministries, and functional cost-sharing mechanisms. However, the case studies included few details about strategies or lessons learnt in this area. Most insights simply noted that lack of capacity of staff to appropriately cost activities, constrained resources at national and subnational levels, and poor cost-sharing mechanisms were some of the key obstacles to effective subnational implementation. Especially at the local level, key informants indicated a sense that funding for local coordination and implementation should come from national departments/ministries with clear mandates, rather than from pooled local resources, although we also know that the source of funding is very much dependent on whether a country has a centralised or decentralised system of governance. A key message that emerged was that civil society can be requested to support capacity building of relevant stakeholders at subnational levels, and temporarily support implementation of activities where there are resource gaps (box 7).
Illustrative example: Developing coordinated budgets and cost-sharing mechanisms
In Senegal, the Committee for the Fight against Malnutrition lacked authority to enforce collaboration and budgetary commitments across sectors, despite being housed in the Prime Minister’s office.15 As a result, multi-sectoral coherence and cost-sharing was largely left to the willingness of the sectors involved.
In Zambia, implementation of the national nutrition policy faced tremendous financial barriers. Despite creation of dedicated budget items for nutrition in relevant ministries, pledges never translated into actual funding disbursements.19 As a result, nutrition interventions were largely financed by international donors.
Integrate M&E systems
A good quality multi-sectoral M&E system is crucial for effective subnational implementation of multi-sectoral policies/programmes. If of good quality, they can help departments/ministries coordinate their work; monitor progress against specific and time-bound targets; identify gaps, weaknesses, bottlenecks and unintended effects of the policy/programme; adjust programming based on evidence; and hold themselves and others accountable. Additionally, M&E systems should include mechanisms to ensure meaningful participation of civil society and other stakeholders. However, the eight case studies included very little information on design and operationalisation of multi-sectoral M&E systems, including theories of change and other analytical frameworks. What did emerge was a desperate need to increase the ability of sectoral M&E systems to ‘talk’ to each other, and to strengthen capacity of staff, especially at subnational levels, to implement M&E systems and report on results.
Evidence shows that single sector interventions have not been effective at ending child marriage, and that coordination between relevant departments/ministries, civil society, and other stakeholders is important to ensure that interventions benefit adolescent girls.23–26
The case studies included in this analysis affirmed that while it may be relatively straightforward to set up the architecture for multi-sectoral coordination at the national level, it is often extremely challenging to ensure national-level coordination structures stimulate and sustain coordinated planning and integrated implementation at subnational levels. This analysis identified a number of lessons learnt from efforts to address HIV/AIDS, nutrition, maternal health and GBV/VAW that can be applied to the implementation of multi-sectoral policies/programmes to end child marriage at the subnational level. Interestingly, the lessons that were identified have very little to do with the specific policy/programme area being addressed, and rather have to do with systems capacity, communication, governance and accountability. At the national level, countries should identify and engage committed and skilled leadership, build a shared understanding and vision of the target issue and how to address it, and delineate and clarify the roles and responsibilities of multi-sectoral stakeholders. At the subnational level, countries should establish coordination structures and mechanisms to direct and operationalise multi-sectoral action, build awareness and capacity of staff, use subnational evidence to contextualise and tailor interventions, develop coordinated budgets and cost-sharing mechanisms, and integrate monitoring and evaluation systems.
The findings of this analysis have important implications for ending child marriage. Given that ending child marriage requires implementing coordinated multi-sectoral responses across health, education, child protection and many other sectors, this analysis shows there is much to be learnt and adapted from implementation of other multi-sectoral policies/programmes. For example, key stakeholders should identify a common conceptual framework around drivers and solutions to end child marriage and use evidence to understand the needs of different groups of adolescents in different contexts to tailor interventions accordingly. Additionally, different departments/ministries and stakeholders should converge, including through NAP linkages, and roll out initiatives in areas with the highest potential, that is, where there is strong departmental/ministerial capacity, clarity of roles and responsibilities, and equitable partnerships across all key stakeholders. Given there is much to learn from multi-sectoral issues already, as a next step it is imperative for the global movement to end child marriage plan and work to operationalise policies/programmes with a commitment to learning and improving on what has gone before.
A key limitation of this analysis is the lack of information available in the published literature on outcomes and/or impact of the multi-sectoral implementation of policies/programmes at subnational levels. Instead, most of the available literature describes, with varying levels of details, the events that led to the development of the policy/programme and how it was implemented, and offered some reflections on what worked or did not work for implementation. Although information on outcomes and/or impact of policies/programmes was attempted to be elicited during key informant interviews, key informants commonly articulated that they had only a partial picture of outcomes and/or impact, and that these would be best investigated through discussions with a broader set ofstakeholders.
The findings of this analysis, while not new or ground-breaking, are striking for two reasons. First, lessons learnt from all of the case studies are remarkably consistent, despite their different target issues and contexts. This consistency indicates that while the review was intended to generate lessons learnt for the global movement to end child marriage, these lessons could very well be applied to the implementation of multi-sectoral policies/programmes to address other health issues as well. Second, the commonality of these findings across target issues clearly reveals that governments, civil society, the private sector and donors are repeatedly and consistently failing to ensure that these fundamental requirements for successful subnational implementation of multi-sectoral policies/programmes are in place. This is potentially due to poor and fragmented planning, inappropriate investments and budgeting, and power dynamics within government departments/ministries and between government and other stakeholders. While child marriage has moved from a taboo issue to one of international concern in a relatively short time, there is still much to do to ensure millions of girls are not married as children. As this analysis has articulated, there are many lessons to be learnt from implementation of national policies/programmes targeting other issues, and it is vital that the movement to end child marriage begins to learn from and use these lessons in its own implementation efforts across different sectors.
CLF and MP are joint first authors.
Handling editor Seye Abimbola
Contributors VC and MB conceived the review. CLF was engaged to conduct the review and prepare a full report. MP led the adaptation of the full report into this manuscript. MP and MB prepared the evidence table. MP prepared the draft manuscript and shared it with the co-authors, collated the feedback and prepared revised drafts. MB and CLF provided feedback on the first and subsequent draft. VC provided feedback on the subsequent draft. All the authors read and approved the final draft.
Funding This work was funded by Girls Not Brides: The Global Partnership to End Child Marriage and the UNDP-UNFPA-UNICEF-WHO-World Bank Special Programme of Research, Development and Research Training in Human Reproduction (HRP), a cosponsored programme executed by the World Health Organization (WHO).
Competing interests None declared.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement All data relevant to the study are included in the article or uploaded as supplementary information.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.