POLICY AND PRACTICE
Secteur privé, ressources humaines et franchisage dans le domaine de la santé en Afrique
Sector privado, recursos humanos y licencias para servicios de salud en África
Ndola PrataI,1; Dominic MontaguII; Emma JefferysIII
IBixby Population Program, School of Public Health, University of California, Berkeley, CA 94720-1690, USA
IIInstitute for Global Health, University of California, San Francisco, USA
IIIInstitute for Health Sector Development, London, England
In much of the developing world, private health care providers and pharmacies are the most important sources of medicine and medical care and yet these providers are frequently not considered in planning for public health. This paper presents the available evidence, by socioeconomic status, on which strata of society benefit from publicly provided care and which strata use private health care. Using data from The World Bank's Health Nutrition and Population Poverty Thematic Reports on 22 countries in Africa, an assessment was made of the use of public and private health services, by asset quintile groups, for treatment of diarrhoea and acute respiratory infections, proxies for publicly subsidized services. The evidence and theory on using franchise networks to supplement government programmes in the delivery of public health services was assessed. Examples from health franchises in Africa and Asia are provided to illustrate the potential for franchise systems to leverage private providers and so increase delivery-point availability for public-benefit services. We argue that based on the established demand for private medical services in Africa, these providers should be included in future planning on human resources for public health. Having explored the range of systems that have been tested for working with private providers, from contracting to vouchers to behavioural change and provider education, we conclude that franchising has the greatest potential for integration into large-scale programmes in Africa to address critical illnesses of public health importance.
Keywords: Delivery of health care/methods; Private sector; Contract services/utilization; Quality assurance, Health care; Socioeconomic factors; Africa South of the Sahara (source: MeSH, NLM).
Dans la plus grande partie du monde en développement, les prestataires de soins de santé et les pharmacies privés représentent les sources les plus importantes de médicaments et de soins médicaux. Pourtant, ces prestataires ne sont souvent pas pris en compte dans la planification relative à la santé publique. Les éléments disponibles présentés par l'article font apparaître, par statut socio-économique, les couches de la société bénéficiant de soins publics et celles faisant appel à des soins de santé privés. Les données figurant dans les rapports thématiques sur la santé, la nutrition, la population et la pauvreté, consacrés par la Banque mondiale à 22 pays d'Afrique, ont permis d'évaluer, par groupes de quintile de revenus, le recours aux services de santé publics et privés pour le traitement de la diarrhée et des infections respiratoires aiguës, exemples représentatifs de pathologies donnant accès à des services subventionnés par l'État. On a examiné les faits et la théorie à propos de l'utilisation des réseaux de franchise pour compléter l'action des programmes gouvernementaux dans la délivrance de services de santé publique. L'article cite des exemples de franchises dans le domaine de la santé en Afrique et en Asie pour illustrer la capacité des réseaux de franchise à faire appel aux prestataires privés et à accroître ainsi la disponibilité des services d'intérêt public au point de délivrance. L'article fait valoir que, d'après la demande en services médicaux privés établie pour l'Afrique, ces prestataires devraient être pris en compte dans la planification future des ressources humaines destinées à la santé publique. Après examen de l'ensemble des systèmes testés dans le cadre de la collaboration avec les prestataires privés, du contrat aux bons d'achat, en passant par la modification des comportements et par l'éducation des prestataires, l'article conclut que le franchisage offre de grandes possibilités d'intégration dans les programmes africains à grande échelle visant à combattre des maladies majeures, qui jouent un rôle important en santé publique.
Mots clés: Délivrance soins/méthodes; Secteur privé; Services, Contract/utilisation; Garantie qualité soins; Facteur socioéconomique; Afrique subsaharienne (source: MeSH, INSERM).
En gran parte del mundo en desarrollo, los proveedores y las farmacias del sector privado son las fuentes más importantes de atención médica y de medicamentos, pese a lo cual esos proveedores son ignorados con frecuencia en la planificación de la salud pública. En este artículo se presenta la evidencia disponible para identificar, teniendo en cuenta la situación socioeconómica, los estratos de la sociedad que se benefician de la atención pública y los que recurren a la atención privada. Utilizando datos extraídos de los informes temáticos sobre salud, nutrición y pobreza de la población del Banco Mundial para 22 países de África, se hizo una evaluación del uso de los servicios de salud públicos y privados, por quintiles de bienes, para el tratamiento de casos de diarrea y de infecciones respiratorias agudas, como indicadores aproximados de los servicios subvencionados con fondos públicos. Se evaluaron la evidencia y la teoría sobre el uso de redes de licencias como complemento de los programas gubernamentales de prestación de servicios de salud pública. Se aportan ejemplos de licencias para servicios de salud en África y Asia a fin de ilustrar las posibilidades que brindan los sistemas de concesión de licencias para multiplicar los proveedores privados y ampliar así la oferta de puntos de prestación de servicios públicos. Sostenemos que, considerando la demanda establecida de servicios médicos privados en África, en el futuro se debería tener en cuenta a estos proveedores a la hora de planificar los recursos humanos destinados a la salud pública. Habiendo estudiado todos los sistemas que se han ensayado para colaborar con proveedores privados, desde la contratación, pasando por los comprobantes, hasta los cambios de comportamiento y la educación de los proveedores, hemos llegado a la conclusión de que la concesión de licencias brinda las máximas posibilidades para la integración en programas a gran escala en África en lo que atañe a combatir enfermedades de crucial importancia para la salud pública.
Palabras clave: Prestación de atención de salud/métodos; Sector privado; Servicios contratados/utilización; Garantía de la calidad de atención de salud; Factores socioeconómicos; África del Sur del Sahara (fuente: DeCS, BIREME).
Poverty and health are inextricably interrelated, and the poor are especially vulnerable to further impoverishment if faced with the high costs of illness or the death of a family member (1, 2). Health inequality has been studied using a number of wellness measures. These include health status, health service spending or financing and health service use. The evidence available from multi-country surveys shows large differences between the rich and the poor (3). Public spending on curative health care in Africa does not seem to be reaching most of the poor. On average, government subsidies for curative health care are imperfectly targeted and primarily benefit the wealthy (2). Although poverty is closely associated with rural areas, the majority of government health-care facilities in Africa are located in urban areas (4). The constraints arising from distance from care providers, combined with the uncertainty of receiving the necessary drugs or treatment from public services, too often leave the poor with only two options: locally available private health care providers, or doing without health-care services altogether.
Even where public facilities do exist, equivalent privately delivered services are often perceived by the users to be of higher quality, irrespective of the empirical evidence that often suggests the opposite (5). In developing countries, private-sector delivery of primary health care is usually poorly regulated and prices are usually scaled according to the ability-to-pay of the client (68). As a result, when the poor seek treatment from private providers they are likely to spend a greater proportion of their income on health care, leading to an increased financial burden. This quandary is a form of market failure: the source of health care is often private, but the private sector is not structured to assure quality or affordability of health care. Strategies to improve access to health-care services and products in developing countries need to take into account the health-seeking behaviour of the various socioeconomic groups so that the poor can be protected and served appropriately. In theory, regulation and enforcement can improve private-sector care, but such measures have a limited track record in areas where government presence is already weak.
The limited human resources available to governments necessitate a strategy of greater involvement with the private sector. An increase in government services, when and if it comes, will not be sufficient to increase diagnosis and entry into treatment to reach the rates set by the Millennium Development Goals (911). The opportunities to intervene with private providers to improve access and quality, ensure equity of prices, and empower clients have been studied both in theory and in practice.
Health franchising is an application of commercial franchising systems to socially motivated health programmes (12, 13). Individual franchisees operate for-profit outlets or clinics, in accordance with clear and strictly defined clinical and quality guidelines set out in a contractual relationship with the franchiser. As a method of organizing an unstructured private sector, franchising is attractive because it incorporates into one system all of the interventions that have been shown to have some effect individually (training, oversight, performance-based incentives, accreditation and certification, vouchers or other external payment schemes, ongoing support relationships and monitoring). Often called social franchising programmes, these programmes have been used successfully for nearly 10 years by family planning clinics in Asia and Africa, and for essential drug provision and programmes for voluntary counselling and testing for human immunodeficiency virus (HIV) in Africa.
The purpose of the present study was to examine the scale of the private health sector in Africa, its importance for mobilization towards meeting public health goals, and the potential of health franchising to function as a primary mechanism for this mobilization. This was done by determining which strata of society benefit from publicly or privately provided services; by shedding light on the size and composition of health-care providers in sub-Saharan Africa; and by providing evidence that health franchising in Africa has the potential to improve service availability to the poor through large-scale programmes.
Data and methods
Data on health-care service utilization by socioeconomic status for 22 countries in sub-Saharan Africa were published in the Health Nutrition and Population Poverty Thematic Reports of The World Bank (3). Health service use for the treatment of two very common childhood diseases, diarrhoea and acute respiratory infection (ARI), was examined. Treatments for these two diseases can be considered to be good proxies for public services that are either free-of-charge or highly subsidized. For each disease, service use was assessed by socioeconomic status and type of provider (i.e. public or private) for both rural and urban populations. Gwatkin et al. defined socioeconomic status in terms of asset wealth quintiles, gathered through the Demographic and Health Surveys (DHS) household questionnaire (14). Provider type was categorized as either public or private. Public facilities included government hospitals, health centres and dispensaries. Private providers included private physicians, mission hospitals and clinics, other private hospitals and clinics, and pharmacies.
Data from an Institute for Health Sector Development (IHSD) study were used to estimate the density of private health-care providers in urban and rural areas. The IHSD study was conducted by local experts, using secondary-source analysis and direct interviews with policy-makers and data managers at the ministerial level in Burkina Faso, Cameroon, Ethiopia, Malawi, Mozambique, Nigeria, Rwanda, Uganda and the United Republic of Tanzania. Further details on the methodology have been published elsewhere (15).
The experience of health franchising in Africa is limited and few survey data are available, but a number of established programmes in Asia have conducted studies and produced public and internal data. Data from public and grey-literature sources were collected through referrals obtained during interviews with public health researchers, health franchise managers and donors supporting franchise programmes. Documents were verified and supported by face-to-face interviews with health practitioners and programmes managers directly involved in the direction of all health franchises reviewed for the present study.
Service utilization by type of provider
Table 1 shows the distribution of children aged under 5 years who were ill, and the use of health services for treatment of diarrhoeal disease and ARI, by socioeconomic status, in 22 sub-Saharan African countries. Children from the poorest quintile were more likely to have had a recent illness than the children from the richest quintile. In the 22 countries studied, the poorest families were least likely to seek medical care when a child was ill. The poorest children were also those most likely to live in areas that were underserved by public health-care providers.
In the majority of the sub-Saharan African countries for which DHS data were available, of those children seen by a medical practitioner, the use of public services by the rich was not significantly different from that by the poor. On average, of those children seen by a medical practitioner, most of those from the poorest quintile sought care from private providers for both diarrhoeal disease and ARI (Table 1). The DHS surveys from which these data were taken were nationally representative samples. The reported source of care (i.e. public or private) was therefore inferred from the results to reflect the combined effects of availability of services and choice of provider. The use of private services did not differ significantly between socioeconomic groups, the differences were mainly between those receiving and those not receiving services, with a higher proportion of the poor not receiving any medical services.
In only three of the countries (Namibia, the United Republic of Tanzania and Zambia), did half or more than half of the poorest children receive treatment for ARI from the public health care sector. To further explore the role of the private sector in service provision to poor people, we took the country examples, Mozambique and Uganda, and chose Namibia as being representative of the three outliers.
In Mozambique, of those children who were ill with diarrhoea, the percentage seen at private sector health-care facilities did not differ significantly between socioeconomic groups. Among the children who were ill, even in rural areas, the difference between use of private services by the poor (18.8%) and the rich (21.9%) was not statistically significant (P-value = 0.782). The largest differences seen between rich and poor were in whether or not they received any treatment at all; the richest were most likely to receive care. Mozambique is an example of a country in which the rich made more use of the public services than the poor (45.8% versus 6.4%; P-value < 0.0001 for rural populations).
The use of services for the treatment of acute respiratory diseases in Uganda, exemplifies another group of countries where most of those who sought medical care at any socioeconomic level did so through the private sector. Moreover, of those who were ill, the difference in use of private services between the rural poor (37.9%) and the rural rich (48%) was not statistically significant (P-value = 0.063). In Uganda, the public sector facilities for treating children with respiratory infections were used by only a small fraction across all socioeconomic groups. Nevertheless, even the rural poor made significantly less use (P-value < 0.0001) of the public services (10.6%) than the rich (23.8%).
In Namibia (the outlier) the majority of those seeking medical treatment for diarrhoeal disease did so through the public sector. Even in rural areas, the comparison of private facility use across socioeconomic groups was not statistically significant (P-values > 0.05). However, the use of private sector health-care for childhood diseases, although much lower than the public sector use, did not differ significantly between socioeconomic groups (comparison of proportions public versus private with P-values > 0.05). The private sector reached poor and rich equally.
Distribution of private sector health-care in sub- Saharan Africa
According to a recent IHSD study on private sector health care in sub-Saharan Africa (15), the size of the independent private sector in sub-Saharan Africa varied enormously between countries. On average, doctors and pharmacists were the most likely to operate privately, and they were mostly concentrated in urban centres. The largest numbers of nurses operating privately were in Cameroon, Malawi, Nigeria, Uganda and the United Republic of Tanzania (12%, 10%, 25%, 8% and 13%, respectively, of the nurses operating in each country). In all of the countries included in the study, some staff employed by the public sector also worked in the private sector, although the estimated proportion varied between countries. Some of the countries had no specific laws or regulations either authorizing or prohibiting the practice of working in both the public and private sectors (Ethiopia, Malawi, Mozambique and Nigeria), although in many countries such a practice was informally recognized, as long as it occurred outside the working hours of the main employment.
Our study demonstrated that large numbers of the poor received care from private providers. Governments in sub-Saharan Africa increasingly regard publicprivate partnerships as a necessary step towards expanding access to basic health-care services (16, 17). As a result, there is a need for systems that can organize existing private providers to ensure the availability of diagnostics and treatment for public health priority illnesses. The components of availability must include provider competence, diagnostic capacity, existing and assured supply of treatment medications, and pricing or payment exemptions that ensure the affordability of these services.
A number of schemes involving private providers have had an effect on some of these components. Voucher systems have increased the affordability of specific health-care treatments, but are prohibitively expensive to manage through large numbers of service-delivery points (18, 19). Postgraduate education for private providers has led to improvements in the quality of clinical care, but the benefits declined over time in the absence of systems for continued regular engagement with public and private systems (2022). Regulation can be effective, but has a poor track record of continued enforcement in Africa (6, 7). Accreditation and certification systems have worked well for hospitals in wealthy and middle-income countries, but have had little success in poor developing countries (23). There are no accreditation and certification programmes that have had proven impacts on private solo practitioners. Publicprivate partnerships, where governments take the initiative to contact and encourage referrals from private health-care providers such as pulmonary specialists, sometimes including supply to the private provider of free treatment medication, have demonstrated successes for tuberculosis control, but no studies on integration with other diseases have yet been documented (24).
Health franchising is an attractive innovation for integrating private providers into public health programmes because it combines critical aspects of all of the initiatives above. Health franchising is based on contractual agreements with medical providers, in which the providers sell services (often subsidized so that there is a lower cost to the end-user) and receive member-specific benefits. Such benefits include the right to use the franchise brand; training; access to certain drugs; business loans; prestige from name-association; and advertising. Thanks to these benefits, the experience has been that franchisees usually enjoy a profitable business and increased clientele, and that client satisfaction is higher in users of franchised clinics than in those who use equivalent non-franchised clinics (25).
Client satisfaction is a result of the contract: for the franchisee, membership benefits are conditional upon the delivery of quality care. Quality regulations for example, on clinic cleanliness, patient interaction, and application of appropriate clinical protocols are monitored by the franchiser through client exit interviews, tracking of drug sales, and in some cases trained actors may pose as clients ("mystery clients"). If the franchisee fails to follow the regulations set out in the initial contract, the franchise is revoked. As long as the value of the opportunity is greater than the value of breaking the rules and there is a credible threat of enforcement, franchisees follow the rules and self-regulate, lowering the overall cost of monitoring. This self-regulation makes this particular system of service expansion and quality improvement cost-effective in a way that is only possible because the goals of the provider (selling medicine and treating patients) are aligned with the goals of the franchiser (ensuring availability and appropriate care).
The specific model of franchising adopted varies depending on how critical it is to maintain quality assurance and compliance of providers with standards. Unlike other systems for involving private providers in the pursuit of public health goals, franchising can promise and deliver a high quality of care and it has been proven to work on a large scale. The experiences of the Greenstar programmes in Pakistan, Janani in India, Child and Family Welfare shops in Kenya, and many others have demonstrated the potential for a franchise model to greatly increase service availability through the mobilization of existing private health care human resources.
One attraction of health franchising is that it has been successful in vastly different societies. In India, a health franchise has improved the sexual health of inter-city truck drivers through provision of education, contraceptives and diagnosis and treatment of sexually transmitted infections near motorway rest stops (13). In Nicaragua, Marie Stopes International, a British non-profit organization specializing in reproductive health, runs a similar health franchise for sexual health services. The Well-Family Midwife Clinic franchise in the Philippines provides midwives trained in safe birth practices to attend deliveries through more than 100 outlets.
The franchise system has also proven successful in sub-Saharan Africa: in Ethiopia the Biruh Tesfa (Ray of Hope) programmes increased contraceptive use by 30% among the 10 million people covered by its 92 clinics (25). In Zimbabwe, New Start franchised testing and counselling for HIV; this had increased monthly visits from 230 in 1999 to 4000 in 2003 (26). In Kenya, the Sustainable Healthcare Enterprise Foundation's Child and Family Welfare Shops (SHEF/CFW) programmes have provided affordable generic drugs through franchised community health workers. SHEF/CFW generates income from 80% of its franchisees, despite serving low-income customers in rural areas (27). Survey data from India, Nepal, Pakistan and elsewhere have shown that clients respond positively to franchise brands, and that the volume of branded services (offered by a provider and quality-assured through oversight by the agency) provided by franchisees is higher than that provided by equivalent non-franchised private providers. It is difficult to make quality measures in the private sector, but one unpublished study from Nepal (D Montague, personal communication) found that counselling provided to mystery clients was more complete and more objective when provided by franchise members than when provided by non-franchise members. A multi-country survey of franchises found that patient-to-staff ratios were significantly lower at franchised facilities than at non-franchised facilities across a number of franchise programmes in Africa and Asia (28). The existing evidence remains limited, but indicates that franchising of private providers improves both accessibility and quality of services.
Evidence from DHS data confirm earlier studies that have shown that public sector services disproportionately serve the wealthy in developing countries (2). Our analysis has further clarified the role of the private sector in filling the gap left by the absence of public sector facilities in serving the poor. The use of private health care for the treatment of childhood diseases does not differ significantly by socioeconomic group. In most countries the poor must usually choose between using private services and not using any health services at all.
A recent review of interventions focusing on improving the quality of health care provided for children by the private sector concluded that the experience with the private sector offered considerable promise for improving child health (29). The importance of private providers is especially great today in light of the current challenges of acquired immunodeficiency syndrome (AIDS) and tuberculosis. In Africa, many poor people seek care for tuberculosis and sexually transmitted diseases from private providers because of the stigma these diseases carry (30, 31).
Extrapolating from the context of need and the health delivery systems currently operating in Africa, our study has concluded that it is critical for African governments to actively engage existing private health-care providers in rapidly expanding the availability of health-care services to low-income populations. Despite the challenges in Africa, the experience of programmes around the world leads us to believe that a system to group and improve the quality of existing private providers would be both viable and beneficial to the poor in a number of countries. Franchising provides an attractive addition to the available tools for leveraging existing human resources and offers a system for standardizing the outputs from a heterogeneous group of practitioners. In addition, it can potentially increase human resources: it works with existing private practitioners who are currently outside public health programming and are almost certainly not providing quality priority disease care for the poor due to restricted drug supplies or lack of ability and support. The inclusion of these providers in public health campaigns is likely to result in a net gain for national programmes.
As the governments in Africa are increasingly challenged by the demands of treating AIDS, and pressure from The World Bank and other donors to expand the reach of publicprivate partnerships, there is a need for new ideas to involve the hitherto unutilized human resources available in the private sector. There are few models that can be adopted to effectively motivate private providers to support public health goals. Health franchising has a track record of successes and provides a possible solution to this urgent and challenging problem.
Competing interests: none declared.
1. The World Bank. World Development Report 1990: Poverty. Washington, DC: World Bank/Oxford University Press; 1990.
2. Castro-Leal F, Dayton J, Demery L, Mehra K. Public spending on health care in Africa: do the poor benefit? Bulletin of the World Health Organization 2000;78:66-74.
3. Gwatkin DR. Health inequalities and the health of the poor: what do we know? What can we do? Bulletin of the World Health Organization 2000;78:3-18.
4. Hjortsberg CA, Mwikisa CN. Cost of access to health services in Zambia. Health Policy and Planning 2002;17:71-7.
5. Brugha R, Zwi A. Improving the quality of private sector delivery of public health services: challenges and strategies. Health Policy and Planning 1998;13:107-20.
6. Hongoro C, Kumaranayake L. Do they work? Regulating for-profit providers in Zimbabwe. Health Policy and Planning 2000;15:368-77.
7. Kumaranayake L, Mujinja P, Hongoro C, Mpembeni R. How do countries regulate the health sector? Evidence from Tanzania and Zimbabwe. Health Policy and Planning 2000;15:357-67.
8. Soderlund N, Tangcharoensathien V. Health sector regulation understanding the range of responses from Government. Health Policy and Planning 2000;15:347-8.
9. Leonard D. Africa's changing markets for health and veterinary services. New York: Saint Martin's Press; 2000.
10. Ngalande-Banda E, Walt G. The private health sector in Malawi: opening Pandora's box? Journal of International Development 1995;7:403-22.
11. Rosen J, Conly S. Getting down to business. Washington, DC: Population Action International; 1999.
12. Montagu D. Franchising of health services in low-income countries. Health Policy and Planning. 2002;17:121-30.
13. Smith E. Social franchising reproductive health services: can it work? London: Marie Stopes International; 2002. Report No. 5.
14. Gwatkin DR, Rustein S, Johnson K, Pande R, Wagstaff A. Socio-economic differences in health, nutrition, and population. Washington, DC: World Bank; 2000.
15. Jefferys E. Evaluating the private sector potential for franchising TB and HIV/AIDS diagnosis and care in Sub-Saharan Africa. London: Institute for Health Sector Development; 2004. Available from: URL: http://www.ihsd.org/docs/JefferysFranchising04.pdf
16. Lambo E, Sambo LG. Health sector reform in sub-Saharan Africa: a synthesis of country experiences. East African Medical Journal 2003;80 (6 Suppl):S1-20.
17. Sekwat A. Health financing reform in sub-Saharan Africa: major constraints, goals, and strategies. Journal of Health Care Finance 2003;29:67-78.
18. Mushi AK, Schellenberg JR, Mponda H, Lengeler C. Targeted subsidy for malaria control with treated nets using a discount voucher system in Tanzania. Health Policy Planning 2003;18:163-71.
19. Gorter A, Sandiford P, Rojas Z, Salvetto M. Competitive voucher schemes for health, background paper. Washington, DC: Central American Health Institute/World Bank; 2003.
20. Ibrahim S, Isani Z. Evaluation of doctors trained at Diarrhoea Training Unit of National Institute of Child Health, Karachi. Journal of the Pakistan Medical Association 1997;47:7-11.
21. Choudhry AJ, Mubasher M. Factors influencing the prescribing patterns in acute watery diarrhoea. Journal of the Pakistan Medical Association 1997;47:32-5.
22. Luby S, Zaidi N, Rehman S, Northrup R. Improving private practitioner sick-child case management in two urban communities in Pakistan. Tropical Medicine and International Health. 2002;7:210-9.
23. Shawn C. External assessment of health care. BMJ 2001;322:851-4.
24. Lonnroth K, Uplekar M, Arora VK, Juvekar S, Lan NT, Mwaniki D, et al. Public-private mix for DOTS implementation: what makes it work? Bulletin of the World Health Organization 2004;82:580-6.
25. Stephenson R, Tsui AO, Sulzbach S, Bardsley P, Bekele G, Giday T, et al. Franchising reproductive health services. Health Services Research 2004;39:2053-80.
26. Population Services International. New Hope with New Start. Vol. 2004. Washington, DC: Population Services International; 2004.
27. Sustainable Healthcare Enterprise Foundation. Facts about CFW. Vol. 2004. Minneapolis, USA: Sustainable Healthcare Enterprise Foundation; 2004.
28. Sulzbach S. Franchising reproductive health services. Minneapolis: Population Association of America; 2002.
29. Waters H, Hatt L, Peters D. Working with the private sector for child health. Health Policy and Planning 2003;18:127-37.
30. Berman P. Organization of ambulatory care provision: a critical determinant of health system performance in developing countries. Bulletin of the World Health Organization 2000;78:791-802.
31. Brugha R. Antiretroviral treatment in developing countries: the peril of neglecting private providers. BMJ 2003;326:1382-4.
(Submitted: 14 July 2004 Final revised version received: 19 December 2004 Accepted: 20 December 2004)