POLICY AND PRACTICE

The benefits of setting the ground rules and regulating contracting practices

Bénéfices de la fixation de règles de base et de pratiques réglementées pour la passation de contrats

Beneficios derivados del establecimiento de normas fundamentales y de la regulación de las prácticas de contratación

Abatcha Kadaï^I; Farba Lamine Sall^II; Guy Andriantsara^III; Jean Perrot^IV,1

^IDepartment of Planning, Ministry of Public Health, Chad
^IIMinistry of Health and Preventive Medicine, Senegal
^IIIWHO Representative Office, Madagascar
^IVDepartment of Health Systems Financing, World Health Organization, 1211 Geneva 27, Switzerland

ABSTRACT

In recent years, health systems have increasingly made use of contracting practices; despite results that are often promising, there have also been failures and occasionally harsh criticism of such practices. This has made it even more necessary to regulate contracting practices. As part of its stewardship function, in other words its responsibility to protect the public interest, the ministry of health has the responsibility of introducing the tools needed for such regulation. Several tools are available to help it do this. Some of them, such as standard contracts or framework contracts, useful as they may be, are nevertheless specific and ad hoc. Contracting policies, when carefully linked to overall health policies, are undoubtedly the most comprehensive of these tools, since they enable contracting to be accommodated within the management of the health system as a whole and thus take into account its potential contribution to improving health system performance. However, the requirements for success are not present automatically and it has to be ensured that there are mechanisms for vitalizing these regulatory mechanisms and that the key actors make proper use of the framework laid down by the ministry of health. The first three authors of this article have participated in the preparation and implementation of national policies on contracting in their own countries, viz. Chad, Madagascar and Senegal.

RÉSUMÉ

Au cours des dernières années, les systèmes de santé ont recouru de plus en plus à l'externalisation. Malgré des résultats souvent prometteurs, ces pratiques se sont aussi heurtées parfois à des échecs et à des critiques acerbes. Cette situation a rendu encore plus nécessaire une réglementation des pratiques de passation de marché. Dans le cadre de ses fonctions, s'agissant notamment de protéger l'intérêt public, le Ministère de la santé a la charge de mettre en place le dispositif nécessaire pour réglementer l'externalisation. Il existe plusieurs outils pouvant l'aider dans cette voie. Certains d'entre eux, comme les contrats types ou les contrats cadres, aussi utiles qu'ils puissent être, sont néanmoins spécifiques et conçus pour une situation donnée. Lorsqu'elles sont soigneusement élaborées en fonction des politiques sanitaires générales, les politiques de passation de contrats constituent indéniablement les outils les plus complets, car elles permettent à cette opération de s'intégrer dans la gestion globale du système de santé, ce qui permet de prendre en compte leur contribution potentielle à l'amélioration des performances du système de santé. Cependant, les conditions pour que l'externalisation réussisse ne sont pas systématiquement présentes et il faut s'assurer qu'il existe des dispositifs pour dynamiser ces mécanismes réglementaires et que les principaux acteurs font bon usage du cadre établi par le Ministère de la santé. Les trois premiers auteurs de cet article ont participé à la préparation et à la mise en œuvre de politiques nationales de passation de contrats dans leur propre pays, à savoir Madagascar, le Sénégal et le Tchad.

RESUMEN

En los últimos años los sistemas de salud han recurrido cada vez más a prácticas de contratación; aunque se han obtenido unos resultados a menudo prometedores, también se han registrado fracasos, y en ocasiones esas prácticas han recibido duras críticas. Ello ha hecho aún más necesaria si cabe la regulación de las prácticas de contratación. Como parte de su función de rectoría, esto es, de su responsabilidad de proteger el interés público, el ministerio de Salud debe implementar los instrumentos necesarios para llevar a cabo esa regulación. Existen varios instrumentos al efecto de los que puede servirse, algunos de los cuales, como los contratos ordinarios o los contratos marco, aunque útiles, son sin embargo específicos y se conciertan ad hoc. Las políticas de contratación, cuando se vinculan circunstanciadamente a las políticas de salud generales, representan sin duda la más completa de esas herramientas, pues permiten acomodar la contratación en el marco de la gestión del conjunto del sistema sanitario, y tener así en cuenta su contribución potencial a la mejora del desempeño de ese sistema. Sin embargo, los requisitos para el éxito de esas iniciativas no se dan de manera automática, y es preciso asegurarse de que haya mecanismos para robustecer esas normas de regulación y de que los agentes más importantes usen adecuadamente el marco establecido por el ministerio de Salud. Los tres primeros autores de este artículo han participado en la preparación y aplicación de políticas nacionales de contratación en sus respectivos países: Chad, Madagascar y Senegal.

Introduction

In recent years, there has been considerable development of contracting almost everywhere in the health sector. Although a large body of experience on contracting has been documented, the results have often been promising but also occasionally limited, not so much because of the contracting approach itself, but on account of its inappropriate or untimely use.

Contracting is often seen as a tool that the various health actors use in an ad hoc manner to solve specific problems, without always considering how it fits into the overall working of the health system.

In this article, we provide an overview of the context in which contracting has developed before considering it a tool whose use is regulated. Several means of carrying out such a regulation are then described. Subsequently, we examine the requirements for these tools to be successful, restricting ourselves to analyses of experiences that have been published and to our own empirical observations.

Context in which contracting has developed

Increased use of contracting has been part of the evolution of health systems and of the relationships between the various actors in such systems. As a result, the position and role of these actors have evolved considerably, and the changes outlined below may be identified.

Role of the state

The definition of the role of the state that was called into question by analyses which concluded that privatization was the remedy for the state's inability to manage is increasingly giving way to a vision in which the state's role is to steer public interest rather than to provide and finance services.¹ The health sector is no exception to this widespread trend, which proposes that the state should focus on its stewardship function and, as is suggested by the World health report 2000,² "row less and steer more." There is nothing new about this ongoing examination of the state's stewardship function, which is defined as a "function of a government responsible for the welfare of the population, and concerned about the trust and legitimacy with which its activities are viewed by the citizenry"; it was addressed by Jean-Jacques Rousseau in the 18th century, then by Max Weber at the beginning of the 20th century, before being taken up by the Public Choice School in the United States.³

This vision reflects substantial challenges to the practices and techniques of public management. In the view of some analysts of government,^4,5 we now stand at the watershed dividing two periods: that of the "commanding government," which is coming to an end, and the dawn of the period of "government by partners ship." The commanding government, which is constituted and acts by virtue of the impersonal and coercive general rule of law, seems to be increasingly less suited to the environment of modern societies, with their inherent complexities. The current crisis of "governability" has exposed the inefficacy of the state and of its conventional legal regulatory mechanisms. No doubt, laws, decrees and regulations and their application by an authoritarian bureaucratic organization were suited to a certain period in history. Nowadays, however, the results produced by this form of governance are less satisfactory; and it is more through necessity and pragmatism rather than ideology that new forms of public management have developed. "Government by partnership" is characteristic of a state that no longer commands from the top down, but which negotiates with its societal environment. Consequently, modern law should assign an increasingly important place to "law by regulation" or "negotiated law"⁶ (flexible, reflexive, reactive), and no longer aspire to regulate everything but rather provide frameworks for negotiation. The new style of government is that of government by delegation and through the coordination of interlinked networks.^7,8 A modern administration thus becomes a cooperative one which generalizes the practice of negotiating as a day-to-day form of action. The law increasingly resorts to contracts as a means of ensuring it is applied and less to enabling legislation that imposes it unilaterally. There are instances in which a framework law, in addition to setting out the fields for negotiation, requires actors to conclude contractual arrangements by a fixed deadline (for example, in France, Ordinance No. 96-346 of 24 April 1996 requires hospitals to reach agreements with regional hospital agencies).

The private sector

In most health systems, the private sector is developing, becoming more diversified and playing or aspiring to play an increasingly important role. As far as provision of health services is concerned, private providers, both for-profit and not-for-profit, are sometimes more important than public service providers. For example, the private sector provides one-third or more of health services in most African countries.⁹ This sector has become diversified, and private non-profit providers, who used to be mainly from religious orders, now include nondenominational organizations. The private for-profit sector, which used to be present mainly in towns and which essentially provided curative services for the well off, is now spreading to more diverse locations and addressing less privileged sectors of the population.¹⁰

Working in isolation is no longer desirable; gradually, the actors involved are realizing the need to build relationships that are more formal and less on an ad hoc basis. As in the business world, the actors involved consider contracting as a tool that enables them to solve their own problems, without regard to the public interest.¹¹ We thus witness the juxtaposition of specific contractual arrangements that suit everyone, including the health authorities, even though contracting as a concept remains limited and relatively unknown. However, there are drawbacks to such arrangements, as discussed below.

• Within the health sector, contracting is being used opportunistically, rather than as part of a strategy that has been clearly laid down by the ministry of health. In Benin, for example, the decision to delegate management of the Ménontin Health Centre in 1992 is an isolated one. Similarly, a study covering 10 countries in the Eastern Mediterranean Region¹² that use sub-contracting reveals considerable differences between them that cannot be accounted for by any explicit strategy.

• Occasionally, use of contracting entails excesses and abuses that are unacceptable. For example, in some hospitals in Morocco that had subcontracting cleaning services, the subcontractors were found to be paying their workers less than the minimum wage. Elsewhere, sub-contracting has led to the dismissal of public employees and consequently the opposition of trade unions;¹³

• There is often a lack of professionalism in the decision to use contracting; for example, some of the actors involved, stimulated by a particular positive experience they have had, plunge in heedlessly, and their amateurism often results in failure.¹⁴ Handling a contractual process and drafting a contract that sets out, in both word and spirit, the understanding reached between the players involved, is far from being as easy as it seems. For example, in Senegal, the contracts signed between the ministry of health and several nongovernmental organizations (NGOs) as part of the Integrated Health Development Programme were badly worded, largely accounting for the poor results achieved.

When used badly, contracting may also prove dangerous. For this reason, it is accused of being a vehicle for privatization or of indicating the withdrawal of the state, with as a corollary, its perceived loss of concern for the public interest.¹⁵ The realization that contracting might be seen as a factor favouring the development of privatization and the withdrawal of the state has been one of the main arguments in favour of the development of a policy on contracting in Chad, Madagascar and Senegal.

Regulatory tools

In recent years, many countries have felt the need to introduce regulation of contractual practices. The need has been expressed not only by ministries of health, but also by other players. For example, Medicus Mundi International would like the use of contracting to be circumscribed by a clearly defined framework between the state and NGOs in countries where it provides support.¹⁶

To understand the concept of regulation, we need to first define it: "Regulation occurs when government controls or deliberatively tries to influence the activities of individuals or actors through manipulation of target variables such as price, quantity and quality."¹⁷ This broad vision of regulation encompasses both the texts and the tools used to control and supervise — the regulation in a narrow meaning — as well as the incentives, the trends, the strategies and policy. The aim is at the same time to stimulate initiatives by all actors involved, to encourage them to contribute towards the health of populations, and to provide a framework for contracting so as to avert its potentially negative effects. Establishing normative documents is a first stage of the regulation process: laws and rules may be sufficient to avoid abuses and corrupt practices. However, countries are increasingly introducing other measures to regulate the contractual practices, as discussed below.

Harmonization of contractual documents

Many countries harmonize contractual documents by drawing up model or standard contracts, in the same way as health-insurance companies use stands dard contracts to govern their relations with their members. This is illustrated by the following examples.

In France, Act No. 99-477 of 9 June 1999, guaranteeing access to palliative care, stipulates that "those associations that organize voluntary work in public or private establishments, welfare and medical and welfare establishments must sign with the establishments concerned a standard agreement defined by a decree of the Council of State".

In Canada (Province of Quebec), a standard contract has been drawn up to harmonize contracts between network clinics and Health and Social Services Centres (CSSS); the purpose of these contracts is to coordinate the services offered by the network clinics and the CSSS in order to offer clients access, via a clearly identified portal, to a continuos ous range of services required by their state of health.

In the United Kingdom, the Department of Health has agreed the content of the General Medical Services Standard Contract with the General Practitioners' Committee (GPC) and the National Health Service (NHS) Confederation. Thus, by offering a formal framework for specific contractual relations, the standard contract technique makes it possible to harmonize practices even if it remains focused on the well-defined aspects of contractual relations.

Guidance documents

Certain countries use guidance documents relating to specific areas, which resort to contracting. These include Canada (Province of Quebec), which inspired by the United Kingdom, is developing, through Act 61 of 2004, public-private partnerships to renovate public infrastructure and improve the quality of services provided to citizens.

In France, the "Hospitals" ordinance of 4 September 2003 has authorized the use of "long-term (emphyteutic) hospital leases" (BEH), a particular type of partnership contract, and established the National Mission to Support Investment in Hospitals (MAINH). The latter is a form of contractual arrangement covering the funding, design, construction, maintenance and operation of the building and in some cases the overall provision of services associated with it. These official documents precisely determine the use of this type of contracting, albeit within a clearly determined area.

Framework agreements

The use of framework agreements reflects different objectives, as discussed below.

Certain framework agreements define contractual terms with which actors may or may not wish to comply. In France, for example, this applies to the agreement between health insurance funds and general practitioners' professional organizations: practitioners may simply send a letter stating that they wish to adhere to the general agreement regulating relations between the funds and general practitioners; this means that there is no specific contract binding a physician to a health insurance fund.

Other framework agreements are designed more as documents setting out the major contractual guidelines, leaving it for the actors to define, within their framework, their specific contractual relations. For example, the major national NGOs and religious bodies that own and manage numerous health facilities in low-income countries are keen to have framework agreements drawn up, to which they may then refer in negotiating specific contractual agreements. For example, in Burundi¹⁸ and Benin,¹⁹ churches find that specific contractual arrangements require often arduous case-by-case negotiations. Moreover, each contractual arrangement is considered an exception, as the contractual strategy is not part of the national health policy. To make up for these shortcomings, the churches have requested the introduction of a frame of reference for their negotiations to provide strength and credibility to any contractual arrangements into which they may subsequently enter.

In some cases, framework agreements take the form of a memorandum of understanding; for example, to implement the DOTS strategy to treat tuberculosis, WHO's Stop TB Department has recommended that ministries of health sign a memorandum of understanding with private practitioners, setting out the terms of collaboration between the public and private sectors.²⁰ Private practitioners are then free to sign contracts to implement the DOTS strategy.

In comparison with a laissez-faire approach, the above-mentioned agreements provide a framework for contractual practices. From the practical angle, benchmarks and limits are laid down for certain types of contract, in respect of which the rules need to be spelt out. However, this does not mean that contracting is part of a systemic approach, i.e. a framework within which it is considered to be a tool to improve health-system performance.

Integration of contracting into health policy

If contracting is to be seen within the broader framework of health-sector reform, it is in countries' interest to integrate it within their health policy. To do so, they have two options.

• Draw up "national health policy" documents, which address the evolution of the health sector together with the reforms required. Countries that wish to adopt contracting may define the terms and strategies for its implementation within this overall policy framework.

• Draw up a specific "contracting policy" document. This is recommended by the resolution adopted by WHO on contracting,²¹ which stipulates that "The Fifty-sixth World Health Assembly ... urges Member States: ... to frame contractual policies that maximize impact on the performance of health systems and harmonize the practices of all parties in a transparent way, to avoid adverse effects". The purpose of a policy on contracting will thus be to define relations between actors; it will determine the place of the contract in relations between actors operating in the field, lay down the principles and objectives of contractual relations, determine priorities and which actions are subject to contract and lay down certain ground rules (such as requirements for registering contracts).

Few countries have actually formally drafted such "national contracting policy" documents: Chad,²² Madagascar,²³ Senegal²⁴. However, there are several that intend to do so: Benin, Burkina Faso, Burundi, Cameroon, Mali and Morocco.

Clearly, these two strategies are not mutually exclusive; ideally, they are complementary. Depending on when they are drafted, the latter will complete the former, or the former will incorporate the latter.

Of the tools described here, contracting policies are perhaps the most comprehensive and innovative; they make it possible to integrate contracting within a systemic approach, along the lines of policies adopted for other areas, such as those on drugs, human resources and health financing. Because in many cases contracting is a new tool which may occasionally give rise to criticism from certain actors, it is in the interest of the ministry of health to define carefully its orientations and strategy.

Requirements for success

All the regulatory tools described here will have an impact only if they are properly defined and implemented. There is no failsafe solution, although it is possible to identify several requirements, as discussed below.

National policy on contracting

A national policy on contracting cannot be drafted without sound political support from the ministry of health, if not the government itself. Realization of the need to lay down a framework for contractual practices may be a technical matter; however, without any political commitment, technical arguments may be insufficient. Additionally, the different actors will more closely comply with a contractual policy if there is real commitment from politicians. For example, Morocco has drafted a strategy for introducing internal contracting between the central ministerial level and health regions. Although the regional directors did not initiate this strategy, they are now committed to the policy because they have understood the political determination to implement the reform. It will be easier to secure such political commitment if there is support from the ministry of health's partners. Within the framework laid down by the resolution, WHO supports countries that wish to develop a policy on contracting; likewise, the World Bank, via operations funded by it in countries as well as through the training programmes it offers, provides similar support to countries wishing to receive it; and lastly, before they provide technical and financial support, several NGOs, such as Medicus Mundi International, suggest that countries introduce rules for contracting.

Mutual understanding between the actors involved

Regardless of its nature, there is little likelihood that a regulatory tool will produce satisfactory results if it is imposed by those responsible for drafting it; this is particularly true of contracting, which relies on mutual understanding between actors. Thus, in Benin, the ministry of health realized that some trade unions were vehemently opposed to certain forms of contracting (lease contract for public hospitals and some forms of sub-contracting or externalization). Rather than impose its views in the document on national contracting policy, which it intends to draw up, the ministry of health has preferred to persuade the unions first of all to discuss the strategy. Three countries — Chad, Madagascar and Senegal — which worked out their policy of contractualization have, throughout the development process, sought the consensus of all actors of health. The search for a consensus when drawing up regulatory tools will often prove to be a prerequisite for effective implementation of the regulations.

Drafting regulatory tools

The drafting of regulatory tools is a prerequisite that will prove fully effective if it is linked to follow-up mechanisms, which may take a variety of forms, as outlined below. One or more monitoring bodies need to be set up. If regulatory instruments or policy documents are to be implemented and complied with, there need to be agencies within the ministry of health that are charged with oversight. The task may be entrusted to existing agencies, or to an ad hoc body, which is given the necessary resources to perform its tasks. Disregard for actual operating conditions means that sound measures, strategies and policies frequently remain defunct. In Madagascar and Senegal, provision for such monitoring bodies was made in the actual contracting policy document, although they face many difficulties in carrying out their tasks. The ministry of health has preferred to persuade the unions to first of all discuss the strategy. An authority for follow-up of the national policy of contractualization should not, however, be confused with a technical support structure for contractualization which helps the actors who need it.

Implementing contracting policies

If contracting policies are to be implemented, they often need to be backed up by legal texts (laws and regulations). For this reason, health actors often give up the idea of establishing contractual relations because of the difficulties posed by a meddlesome administration. A legal framework will determine the "decision space"²⁵ allowed to each actor. This framework varies considerably from one country to another, depending on the institutional reforms they have carried out. In Colombia, for example, the law allows newly autonomous hospitals to sign contracts with private health insurances systems, whereas this is not possible under a similar law in Chile. The three documents on contracting policy to which we have most frequently referred in this article,^22-24 provide for a system for registering contractual documents; so far, not one of the three countries has taken any action in this respect.

Incentives

A ministry of health may introduce incentives to back up the implementation of its texts and policies. In this way, the ministry may influence the decisions of actors by persuading them to re-examine their interest in entering into contractual relations. For example, it may decide to tie the award of a bonus, subsidy or tax exemption to the signing of contractual arrangements.

In France, for example, the ministry of health will award hospitals (both public and private) certain subsidies only if they have signed a contract setting out objectives and means with the regional hospital agency on which they depend. Incentives will be particularly relevant for all contractual policies whose aim is to persuade the private sector to work hand in hand with the public sector. Such a strategy has been adopted by certain Asian countries to introduce the DOTS strategy: ministries of health offer incentives to encourage those who sign contracts with them.²⁶ Moreover, incentives lie at the heart of the current trend for "performance-related contracts"; these are contracts in which assignment of resources (whether human or financial) by one actor depends on another, who is responsible for providing health services and attaining results that are defined in the contract. The Malian Ministry of Health is currently drawing up performance-related contracts with different actors, such as autonomous hospitals, NGOs and local authorities and this strategy will be set forth in the document on contracting policy which is being drafted. Similarly, in Haiti, the United States Agency for International Development (USAID), which finances health activities through NGOs which have signed contracts with it, has introduced incentive mechanisms in the form of bonuses for NGOs that achieve the levels of performance defined therein.²⁷

Evaluation

Use of contracting must be subject to a process of evaluation, to which all those involved should contribute. The ministry of health is responsible for evaluating experiences with contracting to draw lessons: its relevance as a tool, the effects produced by contracting and tracking the contractual process. The evaluation should also highlight good practices, which may then be applied to improve the proposed regulatory mechanisms. Three years after adopting its contractual policy, the Ministry of Public Health in Chad carried out an evaluation of its implementation. The main finding was that although the policy did make it possible to increase significantly the number of contracts signed, in particular with NGOs, it was also necessary to give it a new lease of life; for example, many of those responsible for drawing up the policy are no longer with the health system.

Conclusion

Within a health system, large-scale use of contracting almost automatically involves forms of regulation. A laissez-faire approach, which may be allowed as long as contracting is used only exceptionally, is no longer appropriate when contracting is used by a variety of actors in varied settings. The need to lay down a framework is self-evident, and the need for it is frequently felt and expressed by the actors themselves. However, it may prove to be a perilous exercise, as the framework should not be too restrictive, so as not to dampen the inherent vitality of the contractual approach. We believe that by drawing up national policies on contracting it will be possible to address all its facets and to determine sound guidelines and strategies, which will be taken up by all those involved in health care.

Competing interests: none declared.

References

1. Figueras J, Robinson R, Jakubowski E, editors. Purchasing to improve health systems performance. European Observatory on Health Systems and Policies Series. Berkshire (England): Open University Press; 2005.        

2. The world health report 2000. Health systems: improving performance. Geneva: WHO; 2000.        

3. Saltman RB, Ferroussier-Davis O. The concept of stewardship in health policy. Bull World Health Organ 2000;78:732-9.        

4. Mayntz R. Governing failures and the problem of governability: some comments on a theoretical paradigm. In: J. Kooiman, editor. Modern governance, new government-society Interactions, London: Sage; 1993. p. 9-20.        

5. Papadopoulos Y. Complexité sociale et politiques publiques. Paris: Montchrestien (Clefs/politique); 1995.        

6. Senant Ph, Kirsh-Woik Th. Evolution de l'approche contractuelle dans la province de Mahajanga (Madagascar). Communication présentée au Colloque international "Financement des systèmes de santé dans les pays à faible revenu d'Afrique et d'Asie, France, Clermont-Ferrand, 30 novembre -1 décembre 2000.        

7. Williamson OE. Transaction-cost economics: the governance of contractual relations. J Law Econ 1979;22:233-61.        

8. Kettl DF. Sharing power: public governance and private markets. Washington (DC): Brookings Institution; 1993.        

9. Marek T, O'Farrell C, Yamamoto C, Zable I. Trends and opportunities in public-private partnerships to improve health service delivery in Africa. Washington (DC): The World Bank. Sara Project; 2005. p. 45.        

10. Desplats D, Koné Y, Razakarison C. Pour une médecine générale communautaire en première ligne. Médecine Tropicale 2004;64:539-44.        

11. McPake B, Mills A. What can we learn from international comparisons of health systems and health system reform? Bull World Health Organ 2000;78:811-20.        

12. Siddiqi S, Masud TI, Sabri B. Contracting but not without caution: experience with outsourcing of health services in countries of the Eastern Mediterranean. Bull World Health Organ 2006;84: 867-75.        

13. Rigoli F, Dussault G. The interface between health sector reform and human resources in health. Hum Resour Health 2003;1:9.        

14. Bennett S, Mills A. Government capacity to contract: health sector experience and lessons. Public Adm Dev 1998;18:307-26.        

15. Ridde V. Performance-based partnership agreements for the reconstruction of health system in Afghanistan. Dev Pract 2005;15.        

16. Medicus Mundi International. Contracting NGOs for health. Available from: http://www.medicusmundi.org/Files/contract.doc.        

17. Kumaranayake L. Economic aspects of health sector regulation: strategic choices for low and middle income countries. Public Health and Policy, Department Paper No. 29. London: London School of Hygiene and Tropical Medicine;1998.        

18. Bigirimana F. Collaboration entre l'Etat et l'Eglise Catholique du Burundi dans le domaine de la santé. Communication présentée à la réunion organisée par l'Organisation mondiale de la Santé. L'approche contractuelle dans les services de santé décentralisés en Afrique. Sénégal, Dakar, 19-22 juin 2000.        

19. Association des Œuvres médicales privées confessionnelles et sociales au Bénin (AMCES). Accord-cadre portant détermination des rapports de partenariat entre l'association. Bénin: AMCES et le Ministère de la Santé Publique; 2002.        

20. World Health Organization. Public-private mix for DOTS-Practical tools to help implementation. Geneva: Stop TB Department; 2003. WHO document WHO/CDS/TB/2003.325.        

21. World Health Organization. The role of contractual arrangements in improving health systems' performance. World Health Assembly; 2003, WHA56.25.2003.        

22. Ministère de la Santé Publique. Politique contractuelle dans le secteur de la santé au Tchad. Tchad: Ministère de la Santé Publique; 2001. Available from http://www.who.int/contracting/countries/en/politique_contractuelle_tchad.pdf.        

23. Ministère de la Santé et du Planning Familial. Politique nationale de contractualisation dans le secteur de la santé à Madagascar; 2004. Available from http://www.sante.gov.mg/mambots/editors/tinymce/docs/
politique_contractualisation.pdf.

24. Ministère de la Santé et de la Prévention Médicale. Politique de contractualisation dans le secteur de la santé au Sénégal. Sénégal: Ministère de la Santé et de la Prévention Médicale; 2004." Available from http://www.sante.gouv.sn/structures/document_final.pdf#search=%22politique%20
contractualisation%20s%C3%A9n%C3%A9gal%22.

25. Bossert TJ, Beauvais JC. Decentralization of health systems in Ghana, Zambia, Uganda and the Philippines: a comparative analysis of decision space. Health Policy Plan 2002;17:14-31.        

26. Zafar Ullah AN, Newell JN, Ahmed JU, Hyder MK, Islam A. Government - NGO collaboration: the case of tuberculosis control in Bangladesh. Health Policy Plan 2006;21:143-55.        

27. Eichler R, Auxila P, Pollack J. Performance based reimbursement to improve impact: evidence from Haïti. Washington (DC): Latin America and Caribbean Health Sector Reform Initiative; 2000 (No. 44).        

(Submitted: 6 April 2006 - Final revised version received: 30 June 2006 - Accepted: 9 August 2006)

1 Correspondence to this author (email: perrotj@who.int).