Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

After Reading This Chapter, the Reader Will Be Able to

  • Determine some of the features of the appraisal approaches presented in this chapter (i.e. Critical Comparative Literature Review (CLR) and Evidence Based Research (EBR)), to assess some of the evidence available, whether it has enough robustness (validity and generalisability) or not on the impact outcomes of decentralization as either multi-sectoral or sectoral process aimed to health systems/sectors.

  • Understand the kinds of impact outcomes and effects amongst some of the different components of the health systems provided through specific CLR and EBR examples, which are related to the eight main objectives of decentralization impact on health systems.

  • To identify the pros and cons of achieving or not the main eight objectives of decentralization according to some of the CLR and EBR issues found. Likewise, what are the reasons these objectives can become either positive or negative outcomes overall.

Introduction

There are two related conceptual paradigms for the study of decentralization affecting any health system structure and functioning (Asfaw et al. 2007; Jiménez-Rubio 2010). Both are based on how widely or narrowly decentralization is applied amongst institutionalised and organised sectors of society in different low-, middle- and high-income countries worldwide.

In the first paradigm, decentralization is a wide multi-sectoral process or aim, which is implemented to modify, change or transform legal, political, administrative, managerial, economic and fiscal policies and institutions of the government (UNDP 1999; WB 2001). Also, other political, economic and social sectors of society may simultaneously or not be modified, changed or transformed. Decentralization is established progressively by a rationale of conceptual and comprehensive planning to produce the changes in the middle and long run. Within the first paradigm, decentralization efforts also modify or enhance codes, norms, regulations, structures, infrastructure, resources and procedures for their implementation and development seeking to improve the quality, effectiveness and efficiency of services provided (Evans 2002; Erlandsen 2007).

In the second paradigm, decentralization is a wide or narrow focused or specific and mostly sectoral process or aim, which is implemented and concentrated on reforming different parts of the structure and functions of a specific institutional sector structure or social service provision, (e.g. the health sector, its different tiers and components) (Docteur and Oxley 2003; Burau and Blank 2006). Decentralization within this paradigm touches the three infrastructural, financial and organisational levels of the health services, (primary, secondary and tertiary). The second paradigm (sectoral decentralization) is the most important part addressed by this chapter, which, simultaneously acknowledges the interconnection between both paradigms and the fact that the second paradigm is usually part of and depends on the first one. For instance, it usually depends on general strategic policy planning, financial allocation and transfers and labour force pension schemes, etc.

Both kinds of decentralization paradigms are institutionally or organisationally planned and carried out by the Ministry of Health (MOH) and the government sector, in contractual or another kind of combined partnerships between statutory, civil and voluntary and/or private sectors (Berger and Messer 2002; Eggleston 2009). Also, both paradigms are designed, implemented and evaluated using diverse standard approaches, methodologies and conceptual frameworks according to specific ideological, political, conceptual, technical, financial and economic agendas (i.e. techno-bureaucratic systems, market mechanisms and democratic participative approaches) (Cole 1999; Green 2007; de Leon 2010). Hence, the focus of this chapter is:

First, to analyse critically and comparatively some key elements of decentralization. It is mainly done as the second paradigm and within the health sector. Second, to apply two methodologies Critical Comparative Literature Review with 48 synthesised examples (24 positive and 24 negative) and Evidence-Based Research (EBR) with 8 examples summarised presenting the recommendations and conclusions (Tables 8.1–8.8). These examples demonstrate the positive and negative interrelationships and impact outcomes produced by decentralization upon the different components and aspects of health systems in the short, middle and long run worldwide. Third, to discuss some of the most important positive and negative (sometimes undefined, undetermined or neutral) impact outcomes that decentralization has produced amongst different health systems.

Table 8.1 Example of the evidence-based research found on IITE
Full size table
Table 8.2 Example of evidence-based research on IAEE
Full size table
Table 8.3 Example of evidence-based research on ELTLG
Full size table
Table 8.4 Example of evidence-based research on IANHP
Full size table
Table 8.5 Example of evidence-based research on the impact of IHSDHSD
Full size table
Table 8.6 Example of evidence-based research on the impact of EMAHS
Full size table
Table 8.7 Example of evidence-based research on the impact of RQHS
Full size table
Table 8.8 Example of evidence-based research on the impact of IEAHS
Full size table

Health Decentralization: Focus and Scope of Impact Outcomes

In the last 20 years, there have been institutional, political and social national and international fora, technical meetings and workshops, political debates and research studies in favour or against its implementation and foreseen outcomes (Nieves and La Forgia 2000; Gragnolati and Marini 2003). Decentralization is still now one of the most controversial policy issues in international development (Colombo and Tapay 2004; Braga de Macedo et al. 2009). Likewise, according to WHO (2009), in the last 20 years more than 80 % of the countries worldwide had experienced and experimented somehow with either one of both decentralization paradigms explained above (Work and UNDP/BDP 2002; Docteur and Oxley 2003). It has been implemented throughout North America, Latin America, Europe, Asia, Australasia and Africa. Furthermore, according to Campbell (2003) and Braga de Macedo et al. (2009), at the start of the 2000s and after more than 20 years of decentralization, national and local governments/ authorities worldwide have spent between 10 and 50 % of total revenues generated by themselves implementing the modalities of health decentralization, although soft loans have been given by the IFIs and the international aid community worldwide supporting that purpose as well (Campbell 2003; Braga de Macedo et al. 2009).

Nevertheless, the ideological contents, rationale and objectives of health decentralization policies have mostly been oriented towards techno-bureaucratic systems approaches (TBSA) and market mechanism approaches (MMA) (Cole 1999; Green 2007; de Leon 2010). In many cases a critical analysis of the two approaches mentioned above, both central governments and private sectors have simultaneously been the main authorities/responsible to implement and develop health decentralization as well as the main beneficiaries of its implementation, adoption and impact outcomes (Cole 1999; de Leon 2010). Thus, there have been very few examples about the levels of success of democratic participative process approaches (DPA), their grassroots and horizontal organisational structures and the evidence of their participation in decision-making, governance, accountability and so on, as overall impact outcomes on health systems through health decentralization efforts worldwide (Cole 1999; Schou 2003; de Leon 2010).

Decentralization and Health Systems

WHO/HSP (2001) has stated that a health system includes all actors, institutions and resources that undertake health actions. A health action is defined as the primary intent to improve health in a comprehensively manner. Hence, although the defining goal of a health system is to improve the population’s health, there are other intrinsic (necessary, related and complementary) goals for which a health system must be responsive to the population it serves (WHO/HSP 2001; WHO 2007). These intrinsic goals are determined by the way and the environment in which people are treated. Also, by ensuring that the financial burden of paying for health services is fairly distributed across households and most or all of the population sectors.

Policymakers and populations are vitally concerned about their health systems, how they are performing and where, what and how they can be improved (WHO 2007, 2009). Many different types of reforms and policies have been aimed at improving the performance of health systems. They both have been introduced over the last decades worldwide and at all levels of development. Yet, the evidence about what works and what does not is limited and mixed, and the debate about appropriate health systems development is often led more by ideology than by evidence (WHO 2007, 2009; Wendt 2009).

Meanwhile, there are many studies on the decentralization impact on health systems and health status, focused on the aim and the objective outcomes of it (see Tables 8.1–8.8). Those studies applied different methods to identify and establish the evidence on the relationships between health decentralization impact outcomes and their best indicators. Nevertheless, it is not easy to measure the array of impacts that health decentralization has on health systems (e.g. in planning decision-making and implementation of infrastructure, resources and services) and thus the effects on the health status of populations. The main aim of decentralization in the health sector is to modify the design or content, the implementation and the function of policies, planning and financing norms, structures, procedures, performances/roles and results-outcomes. Therefore, all these elements should be transformed by decentralization to a more effectively, efficiently, fairly/equitably and accountably redefined, restructured and redeveloped core content of any health system (Cole 1999; Green 2007; de Leon 2010).

Improving Institutional and Technical Efficiency

Improved Institutional and Technical Efficiency (IITE) in a health system has been implemented reorganising bureaucracies, their tasks, functions and interactions. Public/private health services have been working in partnerships, established under a single or more decision-making authorities. Stakeholders, beneficiaries and clients/users have been involved in seeking more effectively and efficiently public health interventions through decentralised policies and initiatives (Or 2000; Robalino et al. 2001; Collins and Green 1993)

Positive CLR Evidence on the Impact of IITE

Here in three bullet points, the reader has some examples summarised applying CLR on Improved Institutional and Technical Efficiency (IITE) in the last 20 years:

  • Imposition of greater ‘cost-consciousness policies’ to reduce expenditures have brought up ‘creative ideas’ throughout tier levels of government/authority (Colombo and Tapay 2004).

  • The reconceptualisation of health planning strategies has been focussed on seeking overall effectiveness and efficiency, to implement health packages (Cutler 2002; Robalino et al. 2001).

  • Quality assurance procedures have created environmental, technological and logistic conditions to enhance effectiveness and efficiency levels in health services (Ennis 2006; Hussey et al. 2009).

Negative CLR Evidence on the Impact of IITE

Here in three bullet points, the reader has some negative examples summarised on Improved Institutional and Technical Efficiency (IITE) in the last 20 years:

  • The implementation of health decentralization policies has not clearly been communicated regarding its intensions, aims and objectives (Goetz and Gaventa 2001; WHO 2007).

  • Users have not been provided with clear and efficient access to information to elicit appropriate collaborative responses to healthcare providers (Cornwall et al. 2000; Brinkerhoff 2004).

  • Grants have been managed institutionally and technically in inefficient ways, creating setbacks in health service provision (Litvack et al. 1998; WB 2000).

Increasing Allocative Efficiency and Expenditure

The Increase in Allocative Efficiency and Expenditure has been achieved by consolidating budget allocations and expenditure targets through effective planning and implementation procedures by matching better the health services according to local preferences, by implementing proper assessments on priorities and needs fulfilled throughout timed cycles and by establishing health service packages aligned with programmed expenditure cycles (Goetz and Gaventa 2001; Cantarero and Pascual 2008).

Positive CLR Evidence on the Impact of IAEE

Here in three bullet points, the reader has some positive examples summarised on Increased Allocative Efficiency and Expenditure (IAEE) in the last 20 years:

  • Stakeholders have effectively been involved discussing budget allocation in public health interventions at the local level (WB 2004; Jacobs et al. 2007).

  • Expenditure allocation assessments in market-type relations (e.g. health insurance coverage) have been done through participative agreements with local populations (Gilson and Erasmus 2006; Erlandsen 2007).

  • The funding allocation for innovative health packages have been conceptualised according to local/endemic priorities (Iwami and Petchey 2002; Brinkerhoff 2004).

Negative CLR Evidence on the Impact of IAEE

Here in three bullet points, the reader has some negative examples summarised on Increased Allocative Efficiency and Expenditure (IAEE) in the last 20 years:

  • Public debts have endangered subnational economies, augmenting inequalities between and within regions (Macrae et al. 1996; Medved et al. 2005).

  • Responsibility mismatches have frequently been found regarding revenue-raising/spending decisions between authorities (Craig et al. 2008; Mason and Goddard 2009).

  • Conflicts have lacked enough leverage to elicit solutions for matching funds for underfunded health services (Smith 2008; Wagstaff 2009).

Empowering Lower Tiers of Local Governments (ELTLG)

The Empowerment of Lower Tiers of Local Governments (ELTLG) has been achieved through a more active local users’/customers’ participation in decision-making processes, organised in representational groups or structures. ELTLG has also been achieved through transferring, establishing, enabling and improving the technical skills, administrative and managerial capabilities and good governance criteria to those local authorities and populations involved (PAHO 2004; Stegarescu 2005; Green 2007; de Leon 2010).

Positive CLR Evidence on the Impact of ELTLG

Here in three bullet points, the reader has some positive examples summarised on the Empowerment of Lower Tiers of Local Governments (ELTLG) in the last 20 years:

  • Proper information exchanges have been established between health service providers and consumers in some countries (Belshaw 2000; Habibi et al. 2003; PAHO 2004).

  • Accountability has been integrated beyond the mainstream bureaucratic and human resource realms (Lowensen et al. 2004; Sohani et al. 2005).

  • Devolution has been successful combining capacity-building initiatives, authority functions and shared responsibilities at the local level (Crook and Manor 1998; Heller 2001).

Negative CLR Evidence on the Impact of ELTLG

Here in three bullet points, the reader has some negative examples summarised on Empowerment of Lower Tiers of Local Governments (ELTLG) in the last 20 years:

  • Decentralization has increased dependencies and inequities in authority amongst lower government tiers at the subnational level (Pérez et al. 1995; Bach et al. 2009).

  • Central political agendas have undermined local authorities, staff motivation and social accountability (Okuonzi and Macrae 1995; De Looper and Lafortune 2009).

  • Organisational gaps have become standard issues, which decentralization not always can overcome (Smithson 1995; Bossuyt and Gould 2000).

Increasing the Adoption of New Ways/Approaches of Health Service Provision

The Increase in the Adoption of New Ways/Approaches of Health Service Provision (IANHP) has been achieved by adapting and adopting health services to local situations and conditions, increasing the autonomy of local governments and institutions (e.g. delegation or devolution) through decentralization policies. Also, authority responsibilities have been aligned and coordinated to serve population catchment areas, when they have been within one health district and two separate local government authorities, or vice versa (Davies et al. 1999; Macintyre et al. 2001; Chew and Osborne 2008).

Positive CLR Evidence on the Impact of IANHP

Here in three bullet points, the reader has some positive examples summarised on the Increased Adoption of New Ways/Approaches of Health Service Provision (IANHP) in the last 20 years:

  • The enhancement of health allied and non-allied professionals roles and responsibilities has been accomplished by more local participative monitoring and evaluation feedback (Cutler 2002; Chapman et al. 2006).

  • The quality and extension of delegation processes have been key components for successful decentralization approaches (Richards et al. 2000; Davies et al. 1999).

  • Long-term infrastructure and human resource initiatives have benefited economically, socially, epidemiologically and environmentally to local communities (Fulop et al. 2003; Chew and Osborne 2008).

Negative CLR Evidence on the Impact of IANHP

Here in three bullet points, the reader has some negative examples summarised on the Increased Adoption of New Ways/Approaches of Health Service Provision (IANHP) in the last 20 years:

  • Devolution has sometimes lead health providers to respond to local preferences for curative rather than preventive healthcare services (Belshaw 2000; Goetz and Gaventa 2001; Burau and Blank 2006).

  • Local authorities have invested and prioritised other service sectors; hence, health services have starved of resources (PAHO 2004; Scott-Samuel et al. 2006).

  • The private sector often participated more into goal curative-oriented approaches, usually without regulations (Meuwissen 2002; Standing 2004).

Innovating the Health Service Delivery

Innovation the Health Service Delivery (IHSD) has been achieved by creating new conceptual and operative frameworks to integrate health system networks and deliver better services by improving and integrating health services, information systems, staff performance and training and institutional and social accountability mechanisms related to decisions made/taken. This multi-sectoral integration has usually been provided with requirements, sanctions and penalties, supported by regulations and laws (Cutler 2002; De Looper and Lafortune 2009; Ramsay et al. 2009).

Positive CLR Evidence on the Impact of IHSD

Here in three bullet points, the reader has some positive examples summarised on the Innovation of the Health Service Delivery (IHSD) in the last 20 years:

  • Decentralization corrective measures have reinforced strict budget constraints, strong management, fiscal discipline and rigorous accountability in health sectors (Mills 1996; De Looper and Lafortune 2009).

  • The IFIs have established monitoring and evaluation frameworks for collective management systems to appraise the outcomes of fiscal policies over social services (Loewenson 2000; Ramsay et al. 2009).

  • Redistribution of resources through equity tax policies has been resolved through checks and balances mechanisms placed by the international aid community (Mills 1994; De Looper and Lafortune 2009).

Negative CLR Evidence on the Impact of IHSD

Here in three bullet points, the reader has some negative examples summarised on the Innovation of the Health Service Delivery (IHSD) in the last 20 years:

  • Decentralization has transferred spending and revenue-raising authority, but the lack of administrative capacity has lead to wasting resources (Docteur and Oxley 2003; Erlandsen 2007).

  • Tension between vertical and horizontal integration has been associated with centralised vertical programmes and hierarchical donor funding (World 2001, 2004; Hofmarcher et al. 2007).

  • District health committees have been established; however, their participatory effectiveness is mixed, because of clientelism, corruption or disinterest in involvement (Cutler 2002; Wendt et al. 2009; de Leon 2010).

Enhancing the Mechanisms of Accountability in Health Systems

The Enhancement of the Mechanisms of Accountability in Health Systems (EMAHS) have been achieved by establishing and improving the accountability strategies and mechanisms when transforming health systems by including governmentwide anti-corruption campaigns, health sector reform programmes, fiscal strategic approaches, transparency networks, etc. Accountability is still in constant construction and has been a permanent struggle for the grassroots movements. It has also been redefined and restructured to be more open, participative, horizontal, pluralistic and transparent (Cornwall et al. 2000; Heller 2001; Brinkerhoff 2004).

Positive CLR Evidence on the Impact of EMAHS

Here in three bullet points, the reader has some positive examples summarised on the Enhancement of the Mechanisms of Accountability in the Health System (EMAHS) in the last 20 years:

  • Financial accountability has been developed to track down and report the allocation, disbursement and utilisation of financial resources (Brinkerhoff 2004; WHO 2009).

  • Performance accountability has been implemented to assess the healthcare services, outputs and outcomes (Mills 1994).

  • Political accountability has been created to ensure that elected governments deliver on electoral promises (Eaton 1998; Cornwall et al. 2000).

Negative CLR Evidence on the Impact of EMAHS

Here in three bullet points, the reader has some negative examples summarised on the Enhancement of the Mechanisms of Accountability in the Health System (EMAHS) in the last 20 years:

  • Decentralization policies have not been properly analysed to ensure that authorities truly address the changing needs of people (Cole 1999; Brinkerhoff 2004).

  • Local political and organisational capacity has not been strategically supported by central government authorities (Cornwall et al. 2000; Opwora et al. 2010).

  • Some decentralization objectives have not been met yet in some countries due to structural constrains and political environments (Pérez et al. 1995; Brinkerhoff 2004).

Raising the Quality of Health Services

The Raise in the Quality of Health Services (RQHS) has been achieved by designing and enhancing the infrastructure and the provision of health services; by improving the quality and availability of physical, economic and human resources; by integrating the health services, the information systems and qualified staff/personnel; and, also in some countries, by utilising decentralization policies to open and extend the access to healthcare services for vulnerable groups, where the public health sector has tended to go towards privatisation under the banner of more efficient and effective health service provision (Goetz and Gaventa 2001; Gragnolati and Marini 2003; Gilson and Erasmus 2006).

Positive CLR Evidence on the Impact of RQHS

Here in three bullet points, the reader has some positive examples summarised on the Raise of the Quality of Health Services (RQHS) in the last 20 years:

  • Analytical frameworks in developing countries have positively generalised those efficiency factors behind ‘successful and unsuccessful’ decentralization experiences (Braveman and Gruskin 2003).

  • Managerial efficiency improvements have been critical to prepare staff for their new roles due to decentralization (Carmel and Harlock 2008).

  • Responsiveness to local demands has been a very important determinant for any decentralization initiative to be successful (Bossert 1998; Nsibambi 1998).

Negative CLR Evidence on the Impact of RQHS

Here in three bullet points, the reader has some negative examples summarised on the Raise of the Quality of Health Services (RQHS) in the last 20 years:

  • Confusion amongst decentralization objectives have been detected when applied in shifts in service mix, away from locally prioritised services and health needs (Gilson and Erasmus 2006).

  • Devolution has lead health providers responding to local preferences, offering curative rather than preventive services (Bossert et. al. 2002; Heller 2001).

  • Conflicts of interest have appeared when a health district works with two separate local authorities or vice versa, creating competition and health inequalities (Nsibambi 1998; Bossert et. al. 2002).

Increasing the Equity and Accessibility of Health Services

The Increase in the Equity and Accessibility of Health Services (IEAHS) has been achieved by allocating resources according to local needs, regarding the assessment of endemic profiles through local, national and international public health programmes. Also, by enabling local organisations to better meet the needs of particular groups; by providing equitable distribution of resources towards marginalised regions and groups, through cross-subsidy balancing mechanisms; and by developing alternative strategies and resources to tackle the health problems and better ways of organising institutional and non-institutional resources, amongst other initiatives (Bloom 2000; WHO 2000; Cutler 2002; Braveman and Gruskin 2003; Docteur and Oxley 2003; Schou 2003; Jacobs et al. 2007; De Looper and Lafortune 2009).

Positive CLR Evidence on the Impact of IEAHS

Here in three bullet points, the reader has some positive examples summarised on the Increase in Equity and Accessibility of Health Services (IEAHS) in the last 20 years:

  • ‘Horizontal inequity’ has partially been solved by implementing subnational redistribution policies (WHO 2000; Braveman and Gruskin 2003).

  • Results have also been mixed with programmes trying to reduce tax-base disparities amongst local governments (Bloom 2000; Cutler 2002).

  • Some countries in transition have instituted programmes to reduce disparities in the per capita revenues of subnational governments (Jacobs et al. 2007; De Looper and Lafortune 2009).

Negative CLR Evidence on the Impact of IEAHS

Here in three bullet points, the reader has some negative examples summarised on the Increase in Equity and Accessibility of Health Services (IEAHS) in the last 20 years:

  • Decentralization planning and implementation policies have not properly been identified or determined with regard to health priorities, according to health realities (Docteur and Oxley 2003; Schou 2003).

  • The health services have expanded within a general culture of uncertainty, abuse of authority, opportunism, clientelism and patrimony (Bloom 2000; Braveman and Gruskin 2003; Jacobs et al. 2007).

  • There has been a brain-drain problem or high inequalities in human health resources, which decentralization efforts have not be able to resolve (Jacobs et al. 2007; De Looper and Lafortune 2009).

Conclusion

From the 48 examples using CLR (24 positive and 24 negative) and the 8 examples showed in Tables 8.1–8.8 using EBR, the conclusions are the following: sectoral (heath) decentralization after more than 20 years since it became the flag of structural changes for the existing health systems and their administration and provision of social services worldwide still is ambiguous, not clearly determined and understood due to its complex holistic nature. On the one hand, from some of the positive examples on the decentralization impact on health systems and health and populations described in this chapter, it is possible to interpret some of the benefits brought to some countries and their health systems worldwide. On the other hand, from the negative examples, it is possible to interpret that this is a continuous process in the short, middle and long terms, in which negative outcomes have to be corrected in different ways as well as adapt and adopt more creative and proper ways to carry out the transformation based on the previous experiences of such decentralising processes to obtain better results overall. Thus, the impact of decentralization in health systems has to be constantly appraised, and conclusions and corrective approaches, strategies, actions and measures have to be taken to avoid as much as possible its negative outcomes. Nevertheless, all this rationale unfortunately will heavily depend on the political environment of the country, the political history, political decision-making processes, the prevalent institutional and organisational culture, socio-economic and ethnic problems and inequities, corruption and so on.

Further Analytical and Reflective Discussion/Exercise with Questions

  1. 1.

    What are your own conclusions over the negative/positive evidence on the impact of health decentralization upon health systems worldwide after reading this chapter? Provide your opinion with 3 positive and 3 negative examples.

  2. 2.

    Do you think that is very important to continue implementing health decentralization worldwide? If you think affirmatively, explain why? If you think negatively, explain why?

  3. 3.

    Assume you are a health planner and you need to further improve (more efficiently and effectively) the impact of the eight main objectives of health decentralization. What would you do? (Write a couple of pages mapping your ideas based on the information provided by this chapter, the recommended readings below and a chosen country by yourself, which will provide you with a setting or context.)

Recommended Reading

The first reading below is a study which examines whether decentralization has improved health system performance and presents an example from a middle-income country at the subnational level.

  1. 1.

    Atkinson, S., & Haran, D. (2004, November). Back to basics: Does decentralization improve health system performance? Evidence from Ceará in north-east Brazil. Bulletin of the World Health Organization, 82(11). http://www.scielosp.org/pdf/bwho/ v82n11/v82n11a06.pdf

The second reading below is a study which comparatively analyses the ‘decision space’ in two very different countries. There are few studies focused on this key topic to understand the dynamics and changes through a decentralization process.

  1. 2.

    Bossert, T., Bowser, D., & Amenyah, J. (2007). Is decentralization good for logistics systems? Evidence on essential medicine logistics in Ghana and Guatemala. Health Policy and Planning, 22, 73–82. http://heapol.oxfordjournals.org/content/22/2/73.full.pdf+html

The third reading below, presents the new tendencies in some European countries, on rethinking their decentralization process and adjusting their health systems to the elasticity and contraction of the regional and global market.

  1. 3.

    Saltman, R. B. (2008). Decentralization, re-centralization and future European health policy. The European Journal of Public Health, 18, 104–106 http://eurpub.oxfordjournals.org/content/18/2/104.full.pdf+html