Version classiqueVersion mobile

Non-Communicable Disease Prevention

Wanrudee Isaranuwatchai
Rachel A. Archer
Yot Teerawattananon
et al.

8. Cross-Sectoral Policies to Address Non-Communicable Diseases

Melitta Jakab et Peter C. Smith

Texte intégral

8.1 Introduction

  • 1 Melita Jakab et al., ‘Health Systems Respond to Non-communicable Diseases: Time for Ambition’, Heal (...)
  • 2 World Health Organization, Closing the Gap in a Generation: Health Equity through Action on the Soc (...)
  • 3 Michael Marmot and Ruth Bell, ‘Social Determinants and Non-Communicable Diseases: Time for Integrat (...)

1It is well-established that many — if not the majority — of the determinants of health lie outside the immediate control of the health system.1 The WHO Commission on the Social Determinants of Health2 collected a vast body of evidence showing that the risk factors associated with poor health arise overwhelmingly from behavioral and social circumstances that cannot be addressed by the health system alone. This insight has led to movements such as ‘Health in All Policies’, which seek to ensure that health outcomes are given full consideration in all policy areas, including education, housing, transport, environment and fiscal policy. The link between social determinants and non-communicable diseases (NCDs) is especially strong and well-documented.3

  • 4 Public Health Agency of Canada and World Health Organization, Health Equity Through Intersectoral A (...)

2The importance of other sectors for health-related outcomes has led to a growing interest in the development of cross-sectoral policies to address health objectives. We define the concept of ‘sector’ broadly, to include both governmental and non-governmental parts of the economy. The only requirement for cross-sectoral working should be that the non-health sector is capable of developing and implementing policies in pursuit of its own sectoral objectives and is prepared to enter into a dialogue with the health sector on matters of mutual interest. Examples include joint working between health and education ministries to improve child health and educational progress, or public-private partnerships to improve the health and productivity of the workforce. It is noteworthy that many reported experiments with cross-sectoral programs seek to target disadvantaged groups and specifically address health inequalities, for which underlying social determinants might be especially important.4

3The WHO defines intersectoral action as ‘actions affecting health outcomes undertaken by sectors outside the health sector, possibly, but not necessarily, in collaboration with the health sector’.5 Using this definition, responsibility for implementing the actions lies outside the health sector, although of course the health sector may be the driving force behind the program and may finance part or all of it. In this chapter we adopt a broader view of collaboration between sectors, which might include but is not limited to the WHO concept of intersectoral actions. Specifically, our definition also includes actions led by the health sector that either have benefits for other sectors beyond health improvement, or where collaboration with another sector is essential for success. An example might be an occupational health intervention that is undertaken by a health system, partly with the aim of improving health per se, but also offering potential benefits for employers and the broader economy. We therefore use the term cross-sectoral actions in this chapter to describe joint working, whether or not implementation is led by the health sector. This definition of multisectoral actions captures the active collaboration of two or more sectors that deliberately seeks to promote some of the objectives of the health sector.

4In many cases, cross-sectoral policies are intended to promote the objectives of all the sectors involved and not just those of the health system. For example, an educational policy to promote healthy diets amongst schoolchildren may have an immediate objective for the education sector of improving attendance and performance at school, but may have the additional objective of improving health (and reducing health inequalities) amongst young people. Such effects are sometimes referred to as ‘spillovers’ of the educational policy. However, the use of this term suggests an incidental (or accidental) benefit for the health sector, with an implication that the education sector would have implemented the program regardless of its effects on health-system objectives. In contrast, in this chapter we are mainly concerned with purposefully designed programs offering joint benefits that might not be implemented without active cross-sectoral collaboration. In such circumstances, the fact that one particular sector may have ultimate responsibility for implementing a program should not disguise its essential cross-sectoral nature.

  • 6 Jeffrey D. Stanaway et al., ‘Global, Regional, and National Comparative Risk Assessment of 84 Behav (...)
  • 7 Frank Pega et al., ‘The Need to Monitor Actions on the Social Determinants of Health’, Bulletin of (...)

5Many of the NCD interventions discussed elsewhere in this book can be implemented successfully only with the involvement of other (non-health) sectors. It will often be the case that — from a health perspective — such cross-sectoral policies address the risk factors associated with ill-health, rather than specific NCDs. Broad areas of concern include nutrition, sanitation and water quality, air quality, alcohol and drugs, exercise and smoking.6 The diversity of important risk factors associated with NCDs is an indication of the wide variety of potential cross-sectoral collaborations that might be considered, often addressing the social determinants of health. Note that cross-sectoral work is considered central to the achievement of the United Nations Sustainable Development Goals.7

6Yet, notwithstanding the manifest importance of cross-sectoral projects for controlling the rise of NCDs, the health sector in many countries has found it difficult to initiate and sustain such working. As we shall discuss, this is in part because of the administrative complexity of managing cross-sectoral projects. But the difficulties are also due in part to limitations in the traditional approach towards evaluating projects that rely on cross-sectoral working. In short, it will usually be the case that cross-sectoral projects need to take account of the objectives of the partner sectors as well as the health sector. We argue that this is not in principle difficult, but does require a reorientation of the cost-effectiveness analysis traditionally applied in the health sector.

7The purpose of this chapter is therefore to offer a framework for thinking about the implementation and evaluation of cross-sectoral work to address NCDs. The next section examines the reasons why cross-sectoral work has in many circumstances proved challenging. We then offer a simple analytic framework for assessing the cost-effectiveness of cross-sectoral projects. The fourth section examines the institutional requirements for managing cross-sectoral work and we then briefly present two successful case studies. We conclude by underlining the need for progress in this area if the rise of NCDs is to be successfully moderated.

8.2 Why Are Cross-Sectoral Policies So Challenging?

  • 8 Kumanan Rasanathan et al., ‘Governing Multisectoral Action for Health in Low-and Middle-Income Coun (...)

8There is widespread evidence that countries are not exploiting all the opportunities that exist for effective cross-sectoral action to promote health-system objectives.8 There are many reasons for this. First, it is often extremely difficult to formulate persuasive policies relating to cross-sectoral working. Successful design requires knowledge of all the sectors involved, often requiring novel methods of policy development and knowledge sharing. The various sectors will have different objectives, different budgetary, legal and other constraints and different metrics of success. Reconciling these differences and creating a unified policy is likely to be more challenging than remaining in the ‘comfort zone’ of single-sector programs.

9Second, the institutions of public administration often militate against successful cross-sectoral working. Within government, ministries are usually given discrete budgets, sometimes further constrained by ‘budget lines’ dedicated to specific services or functions. It is often extremely difficult, or even impossible, to introduce a degree of flexibility into how the budgets are spent. There may in any case be a reluctance to cede some part of a ministry’s budget to another sector, as it may be seen to be a signal that the current budget allocation was unnecessarily generous. By spending on cross-sectoral projects, the ministry may fear that in future years its current level of finance will come under threat.

10Third, in the same vein, a ministry will usually be judged according to an accountability system that focuses on a narrow set of objectives specific to its own sector. Pursuit of cross-sectoral projects may appear to be diluting its focus on those objectives. Furthermore, if a ministry transfers some of its budget to cross-sectoral activities, it may feel that it loses some degree of control over how the money is spent and the outcomes to be pursued. Existing monitoring systems may be illsuited to tracking the use of resources and outcomes. Even if good results can be demonstrated, the health ministry may find it difficult to argue that those results are attributable to its own efforts. In short, if cross-sectoral projects appear to sacrifice some degree of control over resources and performance, there may be a reluctance to pursue them. The accountability problem becomes particularly acute when the goals of the partner sector are in direct conflict with those of the health sector — for example, a trade ministry responsible for promoting economic growth may be reluctant to implement taxes on alcohol that could have an adverse impact on (say) the brewing industry.

11Finally, implementation of cross-sectoral projects can be especially challenging. Compared with conventional single-sector projects, which have well-established and simple lines of command, a cross-sectoral project may require commitment of resources and authorization from a variety of sources. There may, moreover, be no arbiter to resolve disagreements or accelerate implementation. A potentially effective cross-sectoral project may therefore languish unimplemented, or be poorly implemented, because there is neither the commitment nor the authority amongst the participating sectors to overcome challenges and see the project through to a successful conclusion. In short, the administrative transaction costs associated with cross-sectoral projects may be very high compared to those associated with more conventional projects.

  • 9 Rasanathan et al.
  • 10 Ibid.
  • 11 Kirk Emerson, ‘Collaborative Governance of Public Health in Low- and Middle-Income Countries: Lesso (...)

12As argued by Rasanathan and colleagues,9 the fundamental difficulty associated with cross-sectoral projects is one of governance. They argue that ‘effective governance is key to the development of shared policy visions and, even more critically, the effective implementation of programs and policies that require coordination across different sectoral agencies and different levels of government’.10 From the health sector perspective, there has often been a failure to learn from the insights of disciplines such as political economy and public administration, which can offer important lessons for how cross-sectoral working can be pursued successfully. In broad terms, the key requirement for successful cross-sectoral working is what has become known as ‘collaborative governance’, relying on characteristics such as mutuality, trust and leadership amongst autonomous partners.11 Such methods are in stark contrast to the conventional ‘command and control’ models adopted within many ministries.

13This chapter is principally concerned with the choice of cross-sectoral interventions to address the prevention of non-communicable diseases. We shall argue that — with minor amendments — cross-sectoral projects can be evaluated using the same cost-effectiveness principles as are customarily used elsewhere. However, it is important to keep in mind the context of governance when considering cross-sectoral projects and to take their feasibility and the costs of implementation fully into account.

8.3 Analytic Framework

14The normative principle underlying this book is that cost-effectiveness analysis should form a central pillar for guiding priorities in the prevention of NCDs. As discussed elsewhere, CEA involves estimating the incremental costs to the health system of a proposed intervention and comparing them to the health benefits that would arise, with adjustments for equity considerations if needed. Health benefits will usually be measured in terms of QALYs or their DALY counterparts. Projects should then be ranked according to the chosen cost-effectiveness criterion and any projects with a cost per QALY that is less than the health system’s cost-effectiveness threshold should be funded. We assume that the health system’s threshold value indicates the maximum the health system is willing to pay for an additional QALY, given its current level of funding.

  • 12 Bengt Jönsson, ‘Ten Arguments for a Societal Perspective in the Economic Evaluation of Medical Inno (...)

15There has been a great deal of debate in the economics literature concerning the appropriate ‘societal’ perspective to adopt for evaluating health projects that have consequences (costs or benefits) beyond the health sector.12 In this chapter we argue that each sector involved in a cross-sectoral project should assess its maximum willingness to pay (WTP) for the project according to its usual evaluation criterion, given the benefits of the project that would accrue to that sector. Then, if the aggregate willingness to pay across the sectors involved exceeds the project costs, the project should go ahead. For the health sector, this means that, when considering contributing to a cross-sectoral project, the same cost-effectiveness principle can be applied to the use of health system funds as is used for conventional single sector projects.

  • 13 Simon Walker et al., ‘Striving for a Societal Perspective: A Framework for Economic Evaluations Whe (...)

16If we know each sector’s WTP for the project, based on its specific outcome measures, then we can add these up to obtain the maximum joint WTP for the cross-sectoral project across all the collaborating sectors. If this exceeds the costs of the project, then it should in principle be implemented. The precise funding contribution of each sector to the project will be determined by bargaining and agreement, but the contribution of each sector should be no more than its maximum WTP. In that way, each sector will be participating in a cross-sectoral project that contributes in a cost-effective way to its own objectives. Of course, the bargaining over the precise magnitude of each sector’s funding contribution will determine what sort of a Buy (Best, Wasted or Contestable) the project turns out to be for the sector. Fuller details are given in the analytical appendix. This approach is consistent with the ‘extended impact inventory’ approach described by Walker et al.,13 which presents the effects of an intervention across a number of sectoral dimensions, and applies societal values to each dimension to see if the intervention is worthwhile.

  • 14 Michael F. Drummond et al., Methods for the Economic Evaluation of Health Care Programmes (Oxford: (...)

17Notwithstanding its conceptual simplicity, the usual challenges associated with undertaking persuasive CEA remain when adopting this approach, principally those associated with modelling and quantifying all the relevant health outcome consequences of the initiative.14 Furthermore, compared with conventional applications of CEA, the benefits of many cross-sectoral NCD initiatives are likely to be distributed across a wide population over a long period, with considerable associated uncertainty. In many cases there is likely to be a need for country-specific epidemiological modelling to identify the impact of NCD initiatives. The need for contextual modelling and the high levels of uncertainty are therefore challenging, However, the principle of using CEA to assess health-sector actions is not altered, even though some of the benefits and costs accrue to other sectors.

18The outcomes for one of the partners may be negative for some cross-sectoral projects. This is particularly the case when the health sector seeks collaboration with another sector to create infrastructure that will improve health outcomes. For example, a public-transport initiative might improve access to healthcare facilities and the associated health outcomes. The principle remains the same — the health sector must be prepared to reimburse the transport sector for the necessary opportunity cost this project would impose. However, if the WTP of the health sector exceeds the opportunity cost to the transport sector, then the project should be viable and it should in principle be possible to calculate a financial transfer between the sectors that satisfies both parties.

  • 15 Michelle Remme et al., ‘Financing Structural Interventions: Going beyond HIV-Only Value for Money A (...)

19Some commentators have argued that cost-benefit analysis may be a more appropriate framework than CEA for assessing cross-sectoral projects.15 Under CBA, the full range of societal benefits and costs arising from a project would be estimated and monetized. This is a legitimate (though analytically demanding) approach that will demonstrate whether or not — in principle — the project should be implemented from a societal perspective. However, CBA ignores the institutional reality that society has organized much of the economy into discrete sectors (often in the form of government ministries), allocated budget constraints to each sector and attached distinct objectives to the use of those budgets. Furthermore, many cross-sectoral projects entail the involvement of the private (for-profit and not-for-profit) sectors, which may have quite different evaluation criteria from those in the government sector. These institutional constraints in themselves create the need for cross-sectoral delivery of certain projects, because the design of society and government is not aligned with the organizational needs of the project. In these circumstances, CEA is not only a useful device — it is the most appropriate tool for assessing cross-sectoral projects, because it takes into account the financial constraints and missions of each separate sector.

8.4 Institutional Requirements

20Once the case for pursuing a cross-sectoral project has been established in principle, an organizational structure for delivering and monitoring the project must be established. As noted above, almost by definition, existing structures of accountability will often be inadequate for this purpose and so some feasible and administratively efficient governance structure must be identified. The design of project governance is mainly beyond the scope of this chapter, but it is important to offer a brief outline of the issues involved in order to give some context to the cross-sectoral case studies that follow.

  • 16 Rasanathan et al.

21There are a number of possible models of collaboration for cross-sectoral projects16. They include:

  • The health sector is the lead actor, but receives support in the form of funding or other resources from an external partner to support the project. The principal governance requirements are proper accountability to the partner for the use of resources and the outcomes achieved.

  • The mirror image organizational structure, in which the external partner is the lead actor, but receives support from the health sector. Here the need is for proper accountability to the health sector for the use of resources and the outcomes achieved.

  • The health sector is a partner with one or more other sectors to implement projects with joint benefits across the sectors, with
    a new delivery entity created under the governance of a joint board of control, representing the interests of all partners.

  • The health sector is not a formal partner. There is no contribution of resources to the implementing sector, but the health system seeks to influence the implementation and performance of the project in order to promote health system goals (in the spirit of ‘health in all policies’).

  • 17 Emerson.

22Such modes of working have become quite widespread in some higher income countries and have led to the development of innovative models of management and control, known as ‘collaborative governance’.17 However, such working is less familiar in many LMICs and may require new models of leadership and accountability. For example, a common failing in cross-sectoral projects is a lack of incentives to prioritize the project and a lack of accountability mechanisms to ensure that it is delivered in line with expectations. Although willing to participate, the individual partners may fail to give the project adequate priority because it falls outside their traditional lines of business. Therefore, whatever approach to collaborative governance is adopted, it is likely that the cross-sectoral project will need sustained leadership, often from a very high level of government, to ensure that momentum is sustained and that the outcomes promised by the project are fully realized.

  • 18 Mcdaid.

23McDaid18 suggests a number of ways in which incentives can be introduced to strengthen the chosen governance and leadership arrangements. For example:

  • the national government (in the form of the finance ministry) can make funds available only if an effective cross-sectoral partnership is put in place;

  • the national government could introduce a competitive process for funding cross-sectoral projects;

  • continued funding of such projects could be conditional on demonstration of successful implementation and evaluation; or

  • government ministries could be required to ‘ring-fence’ part of their budgets for cross-sectoral projects.

24Each of these approaches has shortcomings and risks and cannot succeed without appropriate governance and leadership. However, they might serve to underline the importance of cross-sectoral collaboration and emphasize the commitment of the government to such working.

25To support the chosen model of governance, there will be a need for information and analysis, in order to monitor implementation and to check that expected outcomes are being secured. This is often challenging because it may be necessary to integrate information systems and reporting requirements from the different sectors. Moreover, it can often be analytically complex to identify the incremental impact of cross-sectoral interventions on expected outcomes. A specific concern in many low-income countries is the large range of often incompatible reporting requirements required by different donor organizations and the preference of such organizations to work in independent ‘silos’ rather than collaboratively.

  • 19 Peter C. Smith, ‘Performance Budgeting in England: Public Service Agreements’, in Performance Budge (...)
  • 20 Audit Commission / Healthcare Commission, ‘Tackling Child Obesity —First Steps’, 2006, https://publ (...)
  • 21 Nicoline Tamsma et al., Centre for Healthy Living in The Netherlands: Building Sustainable Capacity (...)
  • 22 Tapani Melkas, ‘Health in All Policies as a Priority in Finnish Health Policy: A Case Study on Nati (...)

26Although there have been examples of successful intersectoral projects, few countries have succeeded in institutionalizing cross-sectoral working as a routine undertaking. The UK government experimented with a range of cross-sectoral ‘public-service agreements’ as a basis for setting ministerial targets and monitoring progress.19 Under Tony Blair’s leadership, a Prime Minister’s Delivery Unit was established to drive forward cross-sectoral programs such as childhood obesity reduction.20 However, this cross-sectoral approach generally failed to take account of its inherent institutional complexity and it lost momentum under subsequent prime ministers. In contrast, the Netherlands has established a Centre for Healthy Living that seeks to promote health by adopting a systematic approach to the evaluation of cross-sectoral policies. An evaluation concluded that the Centre’s approach had been ‘instrumental in advancing intersectoral health promotion policy and practice across the country’.21 Finland has an especially successful and long-standing tradition of cross-sectoral health promotion, using instruments such as legislation and administrative reforms at both the national and local level.22 Even there, however, it has at times proved difficult to nurture a sustained commitment to collecting the evidence necessary to design and evaluate cross-sectoral projects.

8.5 Types of Cross-Sectoral Policies

27Whilst it is rare to find cross-sectoral working institutionalized, there are a number of examples of successful cross-sectoral policies in countries at all levels of development. The types of initiatives designed — at least in part — to address NCDs might include, but are not limited to:

  • commissioning of non-health infrastructure (e.g., public transport);

  • adaptation of non-health programs (e.g., changes to school curriculum);

  • sharing delivery platforms (e.g., health sector use of a postal delivery network);

  • legislation/regulation affecting non-health sectors (e.g., food labelling);

  • taxation or subsidy incentives (e.g., alcohol taxes);

  • integrated cross-sectoral programs for specific population groups (e.g., child development programs).

  • 23 Public Health Agency of Canada and the World Health Organization.

28A report of eighteen case studies by the Public Health Agency of Canada and the WHO, albeit focusing on health equity rather than NCDs explicitly, illustrates the wide scope of possible cross-sectoral working and the range of possible institutional arrangements.23 In this section we present two additional case studies from Hungary and Croatia that entailed legally binding commitments to promote the longevity and effectiveness of the cross-sectoral program.

Case Study 8.5.1. The public catering decree in Hungary: Intersectoral public-health action to improve nutrition and address social inequalities with a binding legal instrument24


29Addressing obesity, particularly among children, has been a major public-health concern in Hungary to reduce premature NCD mortality and morbidity. Having recognized that voluntary actions alone have not been successful to change unfavorable nutritional outcomes, a complex set of mandatory legal actions have been launched by the Hungarian Government. School catering policies have become the target of action. Because children spend most of their daytime in preschools and schools and 35–65 % of their daily energy consumption takes place there, schools have a central role in providing access to healthy nutrition and shaping children’s health behavior.


30A binding legal instrument in the form of a Ministerial Decree was used to increase vegetable/fruit intake and to reduce fat, salt and sugar consumption among school children. The decree came into force on 1 January 2015. Its scope covered pre-schools, primary and secondary schools and other educational settings, inpatient care facilities and certain types of services providing social care and child protection.

31The decree gives a definition of nutritious and healthy meals appropriate for age and physiological status and it considers special dietaryneeds. The decree puts special emphasison equity by guaranteeing healthy meals free of charge for children in socially disadvantaged families, with the cost covered by public funds. The regulation obliges caterers to provide adequate information to consumers by displaying the amount of nutrients and presence of allergens. A special chapter is dedicated to the mandatory training of caterers.

Making It Happen

32The preparatory intersectoral work was led by the Ministry of Human Capacities, a supra ministry covering the areas of health, social affairs, education, youth and sport. Having these various government competencies under one roof facilitated more efficient cooperation, more effective alignment of intersectoral cooperation and a strong social and equity focus included in the decree. The decree was widely and thoroughly negotiated with all relevant stakeholders, including governmental bodies, professional and public organizations (such as caterer associations, parental associations, patient associations and local governments) and with the food industry. Robust communication activities ensured good understanding of public health goals.


33Preliminary evaluations show that between 2013–2017, meals in primary schools became healthier: there was increased consumption of milk and/ or dairy products, fruits and vegetables and whole-grain products and cereals and reduced intake of salt and saturated fatty acids. An impact on the food industry was detected in the form of increased willingness to reformulate food with respect to fat and salt content.

Case Study 8.5.2. Employing people with disabilities in Croatia: intersectoral public health action for an inclusive labor market25


34In Croatia, as in many other countries, people with disabilities are an under-represented group in the workforce. This has a significant impact on their welfare, including their health status, and exacerbates social inequalities in society. People with disabilities account for about 12 % of the total population or about half a million people, of which 48 % are in the working age group of 19–64 years.

35Strengthening employment opportunities for people with disabilities has received growing attention as a civil rights issue and as an under-appreciated growth opportunity for businesses and government budgets. For people with disabilities, employment means greater economic self-sufficiency, an opportunity to use their skills and more active participation in community life. Employment in this group is particularly important because having a disability often means being socially isolated, which negatively influences health outcomes over time.


36To address this, Croatia implemented a Law on Vocational Rehabilitation and Employment of Disabled Persons in 2013 with the aim of increasing the number of employed people with disabilities.

37The 2013 Law focused on regulatory mechanisms including i) quotas related to the number of people with disabilities to be employed, ii) incentives for employers, iii) the development of integrative workshops and working centers which seek to match the abilities of people with disabilities to employment opportunities. The Law also regulates reasonable accommodations to be made at the workplace, including the adaptation of physical barriers and provision of working equipment and personal assistance as needed.

38The 2013 law was not an isolated instrument but one component of concerted policy action to support the welfare of people with disabilities, based on prominent regulatory activity developed over 15 years and including more than 250 laws, sub-acts and decisions.

Making It Happen

39The role of the Public Health Institute was essential in the development of the 2013 Law and related intersectoral action. It produced evidence-based briefings on the impact of employment policies on the health of people with disabilities and presented them to various working groups that were established to implement the process. The role of producing and presenting actionable evidence proved critical in catalyzing intersectoral action. The Institute also coordinated preparatory action between the various stakeholders to highlight the importance of civil rights and health issues relating to people with disabilities.


40The 2013 Law serves as an effective incentive for employers to hire, recruit and retain people with disabilities. Around 11,000 people with disabilities have been newly employed since the implementation of the Law.

8.6 Conclusions

41Many NCD interventions rely on cross-sectoral collaboration for implementation. This chapter has shown that cross-sectoral interventions can — in principle — be analyzed from the perspective of the health sector in the same way that conventional health interventions are assessed, by applying CEA to the health benefits and the costs to the health sector associated with the project. However, cross-sectoral interventions are, by their nature, complex. The evidence to support the analysis will often be in short supply, somewhat speculative or of poor quality. Partner sectors are likely to encounter analogous difficulties when assessing the project from their own perspectives. Therefore, cross-sectoral projects will often need to negotiate serious analytic hurdles before they can even be considered. From the health sector perspective, the role of public health institutes might therefore be crucial in assembling and presenting evidence relevant to the development of cross-sectoral NCD policies.

42We have argued that it is difficult to ensure successful implementation of cross-sectoral projects without paying attention to their leadership and governance. To some extent, governance requirements can be met by the suitable design of institutional arrangements, including the specification of the organization responsible for the project, the basis on which it will be held to account and the means of assuring satisfactory performance. Models of collaborative governance are emerging to address such issues, but these are at an early stage of development. There is ample evidence to suggest that any collaborative arrangements must usually be buttressed by a very high level of authority, for example through legislation, or the direct interest of the prime ministerial office.

43Notwithstanding these challenges, the importance of the social determinants of health is so great that — without concerted efforts to engage non-health sectors in health promotion — societies will not be able to address the rising burden of NCDs with any effectiveness. Policy-makers therefore need to put in place arrangements for designing appropriate cross-sectoral interventions, assessing their feasibility and performance from the perspective of all the sectors involved, designing appropriate governance arrangements, monitoring the implementation and performance of the initiatives and holding all relevant parties properly to account. This is a major undertaking, especially for the many countries with little experience of such working. However, the potential gains from carefully targeted policies are likely to be very large and the necessary investment in analytic capacity and policy commitment has the potential to transform a health system’s approach to health improvement.

8.7 Analytical Appendix

44Consider two sectors (say health H and education E) considering a joint project with costs C and joint outputs bH > 0 for health and bE > 0 for education.

45First assume that each sector is concerned only with outputs relevant to its own sector. These can be measured in composite measures relevant to the sector, such as (say) additional QALYs for health and additional quality-adjusted years of schooling for education.

46Then health would implement the project on its own if and only if C/bH ≤ kH, where kH is the cost-effectiveness threshold for the health sector;

47and education would implement the project on its own if and only if C/bE ≤ kE, where kE is the cost-effectiveness threshold for the education sector.

48In either case, the non-implementing sector would ‘free-ride’ on the cost-effective project for the other sector.

49Suppose now that the project, although producing joint benefits, is not cost-effective for either sector on its own. That is, C/bH > kH and C/bE > kE. There might nevertheless still be scope for proceeding if the costs of the project can be shared between the sectors. Given its cost-effectiveness threshold, health should be willing to pay the education sector a side-payment SH of up to bH*kH to implement, given the magnitude of the health-related benefits. Education would in turn be prepared to implement if (C-SH)/bE ≤ kE; that is if the side-payment is adequate to make the project cost-effective from the education perspective.

50Rearranging, this implies SH ≥ C - bE*kE to assure implementation, with equality to ensure that the project is (just) acceptable to education. A similar argument can be advanced to assess the circumstances under which health would implement the project, subject to a side-payment from education.

51Therefore, there is always scope for implementation so long as the project costs C satisfy C ≤ bE*kE + bH*kH, the joint willingness to pay for the project. This requires that health makes a co-funding contribution SH to education satisfying bH*kH ≥ SH ≥ C - bE*kE. Alternatively, the project could be implemented by health if education makes a co-funding contribution SE to health satisfying bE*kE ≥ SE ≥ C - bH*kH. This concept can be extended to multiple sectors, or even the general public, when assessing whether a cross-sectoral project can be a Best Buy. Without extending the analysis beyond the health sector, however, we may misinterpret from a societal perspective whether a cross-sectoral project is a Best Buy, a Wasted Buy or a Contestable Buy. Note that in either case the upper limit of the inequality indicates the maximum payment the co-funder would be prepared to make to secure implementation, whilst the lower limit indicates the minimum payment that the recipient would be prepared to receive in order to proceed with the project. The actual choice of S would be a matter for bargaining between the two sectors.

52Note that there is no scope for joint implementation if project costs C are such that C > bE*kE + bH*kH. This means that this cross-sectoral project is a Wasted Buy, even when a broader societal perspective is adopted.


1 Melita Jakab et al., ‘Health Systems Respond to Non-communicable Diseases: Time for Ambition’, Health Systems Respond to Non-communicable Diseases: Time for Ambition., 2018,

2 World Health Organization, Closing the Gap in a Generation: Health Equity through Action on the Social Determinants of Health (Geneva, 2008),

3 Michael Marmot and Ruth Bell, ‘Social Determinants and Non-Communicable Diseases: Time for Integrated Action’, BMJ (Online), 394 (2019), 1251,

4 Public Health Agency of Canada and World Health Organization, Health Equity Through Intersectoral Action: An Analysis of 18 Country Case Studies, 2008,

5 World Health Organization, Intersectoral action, 2019,

6 Jeffrey D. Stanaway et al., ‘Global, Regional, and National Comparative Risk Assessment of 84 Behavioural, Environmental and Occupational, and Metabolic Risks or Clusters of Risks for 195 Countries and Territories, 1990–2017: A Systematic Analysis for the Global Burden of Disease Study 2017’, The Lancet, 392.10159 (2018), 1923–1994,

7 Frank Pega et al., ‘The Need to Monitor Actions on the Social Determinants of Health’, Bulletin of the World Health Organization, 95.11 (2017), 784–87,

8 Kumanan Rasanathan et al., ‘Governing Multisectoral Action for Health in Low-and Middle-Income Countries’, PLoS Medicine, 14.4 (2017), e1002285,; David Mcdaid, ‘Institutionalising Inter-Sectoral Action: A Time for Leaping and Pole-Vaulting. Eurohealth. 24.1, 13–15.

9 Rasanathan et al.

10 Ibid.

11 Kirk Emerson, ‘Collaborative Governance of Public Health in Low- and Middle-Income Countries: Lessons from Research in Public Administration’, BMJ Global Health, 3. Supplement 4 (2018), e000381

12 Bengt Jönsson, ‘Ten Arguments for a Societal Perspective in the Economic Evaluation of Medical Innovations’, European Journal of Health Economics, 10.4 (2009), 357–59,

13 Simon Walker et al., ‘Striving for a Societal Perspective: A Framework for Economic Evaluations When Costs and Effects Fall on Multiple Sectors and Decision Makers’, Applied Health Economics and Health Policy, 17.5 (2019), 577–90,

14 Michael F. Drummond et al., Methods for the Economic Evaluation of Health Care Programmes (Oxford: Oxford University Press, 2015).

15 Michelle Remme et al., ‘Financing Structural Interventions: Going beyond HIV-Only Value for Money Assessments’, AIDS, 28.3 (2014), 424–34,

16 Rasanathan et al.

17 Emerson.

18 Mcdaid.

19 Peter C. Smith, ‘Performance Budgeting in England: Public Service Agreements’, in Performance Budgeting: Linking Funding and Results, ed. by M Robinson (Washington, DC: , 2007), pp. 211–33,

20 Audit Commission / Healthcare Commission, ‘Tackling Child Obesity —First Steps’, 2006,

21 Nicoline Tamsma et al., Centre for Healthy Living in The Netherlands: Building Sustainable Capacity and Alliances for Effective Health Promotion (Copenhagen, 2018),

22 Tapani Melkas, ‘Health in All Policies as a Priority in Finnish Health Policy: A Case Study on National Health Policy Development’, Scandinavian Journal of Public Health, 41. Supplement 11 (2013), 3–28,

23 Public Health Agency of Canada and the World Health Organization.




M.Sc., Ph.D., is a senior health economist at the WHO Barcelona Office for Health Systems Strengthening. She has twenty years of experience in health system strengthening, health financing, policy analysis and education in global health. Her work includes advising WHO Member States on health financing policy design and implementation, in particular in Moldova, Kazakhstan, Kyrgyzstan, Tajikistan, Turkey, Ukraine and Uzbekistan. She has been leading a multidisciplinary work program on the Health System Response to NCDs. She has been co-director of the Barcelona Courses on Health Systems Strengthening and Health Financing. She is co-editor of Health Systems Respond to NCDs: Time for Ambition (Jakab, Farrington, Borgermans, Mantingh, WHO Regional Office for Europe 2018) and of Implementing Health Financing Reform: Lessons from Countries in Transition (Kutzin, Cashin and Jakab, European Observatory, 2010). She has a PhD from Harvard University and M.Sc. in Health Policy for the Harvard School of Public Health.

Ph.D., is Emeritus Professor of Health Policy at Imperial College London and Honorary Professor of Health Economics, University of York. He is a mathematics graduate from the University of Oxford, with previous appointments at the University of Cambridge and the University of York, where he was Director of the Centre for Health Economics. His main research interest is in the economics of health, and his recent work has focused mainly on the financing and efficiency of health systems in low-and middle-income countries. Peter has published over 150 academic articles and twelve books, and has advised many governments and international agencies, including the World Health Organization, the International Monetary Fund, the Global Fund, the World Bank, the European Commission and the Organization for Economic Cooperation and Development.


Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search