Public versus private health care delivery

Beyond the slogans

Von Bart Criel, Daniele Giusti & Xavier de Béthune

In most settings, a „public“ health service refers to a service which belongs to the state. The term „private“ is used when health care is delivered by individuals and/or institutions not administered by the state. In this paper it is argued that such a distinction, which is based on the institutional or administrative identity of the health care provider, is not adequate because it takes for granted that the nature of this identity automatically determines the nature of the service delivered to the population. A different frame of classification between public and private health services is proposed: one which is based on the purpose the health service pursues and on the outputs it yields. A set of five operational criteria to distinguish between health services guided by a public or private purpose is presented. This alternative classification is discussed in relation to a variety of existing situations in sub-Saharan Africa (Mali, Uganda, Zimbabwe). It is hoped that it can be used as a tool in the hands of the health planner in order to bring more rationality in the current altercation between the public and the private health care sector.

There is a growing interest for more and better co-operation between respectively public and private sector in the field of health care delivery, particularly in the developing world. A range of different explanations for this boost in interest can be readily identified. There are, in the first place, the facts that the already scarce resources for health care are dwindling yet further and that linkages with the private sector may raise additional resources. There also is the gradual acknowledgement of the need to develop a systemic approach to health care delivery. The private sector is an important actor in this system, and can, under certain circumstances, substantially contribute to a consistent development of health systems. Our field experience in the delivery of health care in sub-Saharan Africa confronted us with the rigidity, and even the strong emotions, that often tend to colour this debate on co-operation between public and private sector. The relative lack of rationality and objectivity in these discussions has contributed to a state of affairs where the concerned interlocutors clutch at their respective positions. For many of the arguments used in the debate, a consistent explanation can be found in each country's social and political history. However, even if they may have been most valid at one point in time, or in a given social and political environment, they sometimes tend to stiffen and to lead a life on their own. Challenging or questioning the validity of these arguments in changing times is sometimes seen as a disloyal act vis-à-vis the ideological brotherhood. It is common, even natural, to notice a certain diffidence among civil servants and public health managers towards matters that escape their control. In many countries the private health sector has often grown independently from the public health sector and is rarely taken into account in health planning scenarios. This has been the case in Uganda where the non governmental sector, which generally has been in the forefront of the development of primary health care initiatives and which accounts for about 65% of the current primary health care delivery in the country, is rarely taken into account by the District Health Team in their planning exercises. On the other hand, there often is in the private sector an excessive jealousy for own independence, with disregard to policy guidelines and hostility towards regulative measures.

One of the results is the relative scarcity of innovative thinking and experimenting in the field of co-operation between public and private sector; notwithstanding the huge need to gain expertise and design models on the pattern and features such a co-operation could possibly take. The important lipservice and rhetoric policy-makers and health managers of both public and private sector devote to the issue has brought too little changes as yet. But it is increasingly evident that such cooperation is a must in a systemic view of health service provision and for the purpose of avoiding expensive and useless duplications. In this perspective it becomes important move towards an ever progressive integration into the health system of all elements accepting a “public” rationale of operation. But the definition of “public” is exactly, at present, somewhat hazy and needs focusing.

The confusion: what are the real meanings of public and private?

In our view, one of the major stumbling-blocks in this process is the lack of consistent use and interpretation of the terminology public and private, be it conscious or not. The mere attempt to answer the questions "what is a public health service?" or "what is a private health service?" would reveal the heterogeneity of views on the matter. The purpose of this paper is precisely to present some thoughts on the way how these very words public and private are used (and sometimes abused) and to attempt to clarify to what content they should refer. We think that the development of a more coherent vocabulary is a necessary step in this broader process of co-operation between public and private sector, be it in the field of health care or in the field of any other social sector. Whatever the grounds for these differences in perceptions, it is obvious that a dialogue on that matter will benefit from clarity and intellectual honesty.

In the majority of the situations, the definition - both implicit and explicit - of a public health service as found in the literature and in official policy documents refers to health care institutions belonging to the state. In both the developing and industrialised world, health care delivery is often being supplied by private individuals and/or institutions whose ownership and/or administrative guardianship or guardianship is not the state. In that case, the term private is used. It is generally understood that the public health sector should be supported with public money and protected by a series of privileges regulated by law, while the private health sector should operate on private funding, obtained through fees, donations or other means in the arena of a marked oriented provision of service and of competition: this understanding is based on the assumption that the private sector is homogeneous and financially self-sustainable. In reality, a remarkable heterogeneity exits in the private/non government sector (DeJong 1991; Green 1992; Zwarenstein & Price 1990; Smith 1989).

Generally, when the service is rendered without lucrative purpose the specification "nonprofit" is added. The term "non governmental" is used to indicate organisations offering services without lucrative purposes, and whose ownership and/or administrative guardianship is not the state. We think that a distinction between public and private based on the institutional or administrative identity or guardianship is not adequate in dealing properly with the variety of existing situations. In the logic of this classification, the qualification public is reserved to institutions belonging to the government, while the term private is meant to cover all other configurations. The fact that the category private needed further subdivision into "for-profit" and "non-profit", illustrates the insufficiency of this approach, even if, in practice, these terms have already been consecrated by use.

The limits of this classification can be exemplified by the mushrooming number of non governmental organisations operating for outright or hidden "profit" goals. At the same time, there are "public" services which operate, to varying extents, on a lucrative for profit basis even if the intensity and the sometimes radical character of this shift in rationale from non-profit to profit within public facilities has not necessarily been the result of a positive and planned choice of policy makers. Examples of such a shift is the situation of some government (and thus "public") hospitals in Zimbabwe and Uganda. In both countries, medical officers are allowed to develop private practice next to their responsibilities and tasks in the hospital. In the case of Zimbabwe, this measure is part of a broader effort aiming to attract (or to keep) national medical officers in the public sector in a context of massive brain-drain to neighbouring countries or to the private sector in Zimbabwe. In the case of Uganda, it grew out of the legitimate concern to increase the revenue of national doctors beyond the extremely poor level of government salaries. In both countries, government medical officers (i.e. belonging to the „public“ sector) are allowed to use the hospital infrastructure and hospital resources for treatment of private patients who pay them a fee, but without retribution of the hospital.

The gloomy prospect is governments ending up in subsidising - with tax-payer money - a private-for-profit sector where basic measures of quality control are lacking and with a poor accessibility for lower income population groups. A „two speed“ health care system becomes a real threat... The same Government would instead deny subsidies to private institutions striving (and finding it progressively difficult) to offer financially accessible services, often at lower costs (in absolute terms) than those observable in the public (state) institutions.

The core of the matter really is that the adjectives private and public refer to the institutional identity and type of administrative guardianship of a given health service, taking for granted that the nature of this administrative identity automatically determines the nature of the service that is actually offered to people. In an epoch of reform of most health systems, with decentralisation as a key element, we think that this assumption is no longer justified. If a distinction between public and private needs to be made, we think it cannot be based exclusively on the institutional set-up of a given service, but rather on the performance and output of that very service.

Maintaining a distinction between public and private on the ground of the administrative guardianship will only perpetuate confusion, prejudices and discrimination (positive or negative but, in either case, inadequate to the changing scenario). In Uganda for instance, the non government sector (mainly Church-related non-profit organisations) has been able to develop acceptable levels of health care delivery in sometimes very remote and insecure areas of the country and in environments characterised by important social and political unrest with a de facto absence of the State. The private status of these institutions has been a clear-cut hindrance to the development of consistent and long term government policies of support to the health care sector: for example, the posting of national doctors to these institutions has become very difficult because of uncertain career and training perspectives for those who choose to work in them; nurses trained in NGO training schools, which are formally recognised by the national Nursing Council and the final examinations of which are supervised by Government officials, can make their way to the Government service only with great difficulty; no or very little government subsidies are being allocated to NGO facilities which are considered by District Health Teams as falling outside their scope of responsibility, even when their importance for the system is openly recognised etc.

Such a distinction will hinder the dialogue between the different components of the health system in an epoch when each one's contribution and cooperation is necessary. Indeed, in the light of decentralisation policies being implemented in many developing countries, the institutional set-up of many decentralised "public" health services is far less clear-cut. In the past all public health services, with few exceptions, belonged to and were financed by the State, represented by the Ministry of Health. Today, there is a trend towards decentralised ownership and management by local communities, co-operatives, districts etc.

Such a trend can be exemplified by the case of the network of community health centres („centres de santé communautaires“) gradually put in place in Bamako (Mali) from 1989 on. Former rural community based experiences in the public sector served as an inspiration basis for young medical doctors who could not be hired by the government and who remained, jobless, in the capital of the country. With some initial external help, three or four health centres were organised so as to offer basic curative, preventive and promotional services. The owners of the facilities were the members of community associations created for that purpose and the aim of these health centres was to provide health care to the members through a system of cost-recovery. Later, a "second" generation of centres was put in place with virtually no other external help than small in kind loans by existing centres. These new centres built up their revolving drug fund through the initial voluntary work of their employees. Several of them later acquired grants from different donors.

The government played a promotional and regulatory role by considering these centres as active partners in its health development efforts. The existing centres constituted the starting point for geographical health coverage maps drawn up by the urban district teams. They also received small subsidies in kind especially for immunisations and family planning services. Their revenue was tax exempted and they were granted a special licence to sell generic essential drugs provided no profit was made on them.

The debate on the status of these institutions is still ongoing. Legal texts have defined both the governments and the health centres responsibilities. But the way the centres were put in place and the pressure from unemployed health workers in Bamako explain that some of the attention has been diverted from the equitable provision of health care to the raising of revenue mainly to hire additional staff.

An alternative classification?

What really matters to the health planner and to the public, are the contents, the quality and the costs of the package of services offered. For planning and evaluation purposes, and for the allocation of the meagre resources available, it is important that a clear, and explicit, declaration of intent, or mission statement, of the health care institution exists; so that the output and accessibility of these services can be evaluated against such declaration. (should the notion of contract here be introduced - also a formalised contract???? In an era of rapid change, it is also necessary to evaluate in a systematic way the faithfulness and/or parting of each health care institution from the mission statement. Hence, we propose a different frame for classification of health services based on the objectives the health service declares to pursue and on the output it yields. From thereon, a dichotomous classification in health services with respectively a public or private finality can be proposed. More specifically, we propose some five criteria for classification of a health institution in the category of public.

1. A social perspective: a concern to enhance people's well-being and autonomy in a perspective of human promotion. In the case of health services, it more specifically means contributing to people's realisation of a socially productive life, in a climate of dialogue and in harmony with the prevailing overall socio-economic development. • non discrimination: a concern to offer people accessible and quality health care without discrimination whatsoever with regard to race, sex, religion, political affiliation, social status, income level etc.

2. Positive discrimination: for specified groups of people, deemed to be in particular need of a specific health service (e.g. women, handicapped, blind etc.)

3. Population-based: a concern to take responsibility for, and to be accountable to, a well-defined population for its health care delivery. This accountability could be based on a contract to be established with the population, specifying the mission statement of the service or institution

4. Government policy guided: a concern to comply with government health policies for the level of care provided and to fit in a broader masterplan. Should any different views arise with regards to official policy, then it is necessary that they be argued, discussed and, when possible, formalised in official agreement between the health institution and the national health authorities.

5. Non profit goal: a concern not to reduce the purpose of the service to profit making. This does of course not mean that good working and living conditions would not be a right for staff, nor that the service must be run at a loss. On the contrary, it is desirable that any service be self-sustained (this is not always possible; it is even virtually impossible in the case of hospitals). Anyhow, in order to preserve the public finality of the service, profits made should be reinvested in the same service or in other socially oriented activities.

These five criteria, which have been identified as suitable in an African context at District level, do not exhaust the variety of possible criteria identifiable in different contexts. Nonetheless, they provide an instrumental frame for the assessment of the finality or purpose of health services with a strong focus on the very nature of the service offered to people rather then on the administrative/institutional set-up of the institution. Both perspectives can be represented in a simple two by two table:

Administrative guardianship and/or institutional identity of the health service Public Private Finality or purpose of the health service Public a b Private c d

The different configurations of each of the four cells of this table can be exemplified as follows: a corresponds to National Health Service (NHS) hospitals in the United Kingdom (although the current reforms of the NHS represent a gradual shift from a to b); b corresponds to most church-related hospitals in Uganda; a shift from a to c is taking place in many government hospitals in Uganda and in some government hospitals in Zimbabwe; and d corresponds to the situation of many hospitals in the United States of America. The relative strength of the actors involved with the Bamako health centre network will determine whether these centres end up in categories b or d, or remain somewhere in between.

It is clear that the variable finality does not completely fit the nature of a dichotomous variable; indeed, in reality it covers the range of intermediate situations that exist in the wide spectrum going from public to private finality. The same comment holds for the other variable in this table, i.e. the administrative guardianship . This two by two table is thus an oversimplification of reality; but we nevertheless think that it is useful to illustrate our point. If governments agree and accept the rationale for this classification according to the very finality of the service, then it would allow them to achieve more accuracy in targeting their support to health care institutions and organisations - both government and non government - who serve a public purpose. The case of designated district hospitals in Tanzania or Ghana, illustrates that it is possible to define consistent policies on that matter. Referring back to the case of Uganda, it appears that many (but certainly not all) of the non-governmental and church-related organisations would sufficiently fit the criteria defining a public finality. This classification could also be helpful to distinguish in the present mushrooming of private practice throughout the developing world: it may help to separate the corn from the wheat. A consistent policy would then be to support these organisations, and these individuals, that pursue a public mission and not those that fit a given administrative status.

In operative terms, we see it appropriate, for a public administration aiming at rationalising its health services, to start using these criteria (or others as it may apply in each specific context) in order to identify those elements in health system which need to be secured to the rationale of public oriented health service provision, rather than allowing them to drift towards a market-oriented rationale of operation. It is not impossible to develop, out of the criteria, some easily applicable indicators (quantitative and qualitative). In Uganda, as example, the criteria “population based” and “non-profit”, are progressively being used to identify those element of the health system eligible for integration and, sometimes, partial financial support. Some districts are being “zoned” and each zone is entrusted to a Hospital or to another level of unit, regardless of its being private or government, with the mandate to guarantee the support to the provision of a package of PHC activities. Such mandate is regulated by a formal memorandum of understanding, specifying mutual rights and duties, and the nature of service to be provided.


We have argued that a distinction between private and public based on the institutional set-up of a given service is inadequate in defining the very nature of the service offered, the latter being of paramount importance to the health planner at any level of the health system. For example, many private hospitals and health centres in developing countries operate according to a rationale which could be defined as public; at the same time lucrative goals are being introduced into public health services which, eventually, endanger their adequacy, relevancy and accessibility. An operational definition of what could be considered to be a public health service is still lacking. This is not without consequences in an era where, on the one hand, most governments are (or have become) unable to answer in a satisfactory way to the health needs of people and where, on the other hand, the contribution of the private sector is called upon. This paper attempts to identify some operational criteria which would allow to distinguish between services operating according to a public or private rationale. We hope that these criteria can be used as a tool in the hands of the health planner in order to bring more rationality in the current altercation between public and private, and in order to help the various actors in the debate to dialogue - beyond the slogans.


Health Policy and Planning, Volume 12, Number 3, September 1997, pp. 192-198. Copyright © 1997 by Oxford University Press. Reprinted with kind permission of the authors and the journal. Text version based on:


  • Collins C & Green A 1994. Decentralisation and Primary Health Care: some negative implications in Developing Countries. International Journal of Health Services, vol 24, N°3, pp 459-475.
  • Dejong J 1991. Non governmental Organisations and Health delivery in Sub-Saharan Africa. Working Paper. The World Bank. Population and Human Resources Department.
  • Gilson L, Priti DS, Shirin M & Phare M 1994. The potential of health sector non governmental organisations: policy options. Health Policy and Planning, 9(1): pp 14-24.
  • Green A 1992. Planning for Health. In: An introduction to Health Planning in Developing Countries. Oxford Medical Publications. p 77.
  • Korten D 1987. Third generation NGO strategies: a key to people-centred development. World Development, 15 (suppl): 145-159.
  • Ogoh Alubo S 1990. Doctoring as Business: a study of Entrepreneurial Medicine in Nigeria. Medical Anthropology, vol 12, pp 305-324.
  • Smith K 1989. Non Governmental Organisations in the health field: collaboration, integration and contrasting aims. Social Science and Medicine, 29(3): pp 395-402.
  • Unger JP 1991. Rôle des districts sanitaires et méthodologie de leur développement en Afrique. PhD Thesis. Free University of Brussels: p 7.
  • Walley J, Tefera B & McDonald MA 1991. Integrating Health Services: the experience of NGO’s in Ethiopia. Health Policy and Planning, 6(4): pp 327-335.
  • World Bank 1993. Investing in Health: the World Development Report. The World Bank. Washington.
  • Zwarenstein M & Price MR 1990. The 1983 distribution of hospitals and hospital beds in the Republic of South Africa. South African Medical Journal, 77, pp 448-52.

Passwort vergessen?
Neuer Benutzer?