Strengthening Health Systems in Poor Countries: A Code of Conduct for Nongovernmental Organizations

Abstract

The challenges facing efforts in Africa to increase access to antiretroviral HIV treatment underscore the urgent need to strengthen national health systems across the continent. However, donor aid to developing countries continues to be disproportionately channeled to international nongovernmental organizations (NGOs) rather than to ministries of health. The rapid proliferation of NGOs has provoked “brain drain” from the public sector by luring workers away with higher salaries, fragmentation of services, and increased management burdens for local authorities in many countries. Projects by NGOs sometimes can undermine the strengthening of public primary health care systems. We argue for a return to a public focus for donor aid, and for NGOs to adopt a code of conduct that establishes standards and best practices for NGO relationships with public sector health systems.

An estimated 20 million people worldwide could still benefit from antiretroviral therapy (ART) for HIV/AIDS,¹ yet the World Health Organization's “three by five” goal of placing 3 million HIV-infected people on treatment by 2005 has still not been met. Widespread deficiencies in public sector health infrastructure and workforce are at the root of failures to achieve treatment goals.^2,3 According to one analysis, many new HIV treatment projects “are being implemented without adequate investment in strengthening the weak, and in some cases collapsing, health systems in sub-Saharan Africa.”^4(p18) Treatment expansion fueled by new large-scale funding from the President's Emergency Program for AIDS Relief; the Global Fund to Fight AIDS, TB, and Malaria; and the World Bank has been slowed by insufficient health infrastructure.⁵ There is an emerging consensus among donors and local governments recognizing the urgency of strengthening the public sector through workforce expansion, infrastructure investment, and management capacity building.^6–8 With sufficient support, national public sector systems can coordinate large-scale programs and bring integrated, quality services, such as HIV/AIDS treatment and care, to the greatest number of people most equitably.

However, much of the new HIV/AIDS funding is still channeled to international nongovernmental organizations (NGOs), whereas funding for public sector health systems remains constrained. Observers in developing countries with high HIV burdens have become convinced that the practices of NGOs may be causing permanent harm to public systems of care by fragmenting services, promoting internal “brain drain” from the public sector, and creating an excessive management burden for local health managers, who must monitor multiple projects.⁹ Paul Farmer, who pioneered the provision of ART in resource-poor settings in Haiti with the NGO Partners in Health, highlighted this problem in his 2006 American Public Health Association address:

The NGOs, which fight for the right to health care by serving the African poor directly, often do so at the expense of the public sector, creating a local brain drain by luring nurses, doctors, and other professionals from the public hospitals … to NGO-land where salaries are better and the tools of our trade more plentiful.^10(p6)

Recently, notable exceptions including Partners in Health and Médecins Sans Frontières in the Khayelitsha township clinic in South Africa have recognized these challenges and worked to strengthen public services. Through our own experience in Mozambique with Health Alliance International, which supports increased ART through public sector strengthening, we have witnessed the pitfalls associated with the international NGO model. We believe that an international NGO code of conduct could help strengthen health systems by promoting a more constructive role for NGOs at this vital moment in the AIDS crisis.

THE GLOBAL EXPANSION OF NONGOVERNMENTAL ORGANIZATIONS

The struggle to integrate NGOs into the health sector is part of a broader trend. Driven in some measure by donor preferences, the number of NGOs worldwide ballooned during the 1990s from 6000 to 26 000.¹¹ The number of international NGOs supported by the US Agency for International Development increased from 18 in 1970 to 195 in 2000.¹² By 1996 the United States was directing almost one third of its African development assistance through NGOs, and overall aid funding to nonstate organizations from major donors such as the World Bank and European bilateral agencies grew 350% between 1990 and 1999.¹³ Civil society involvement (mostly NGOs) in World Bank operations rose from 21.5% of the total number of projects in 1990 to nearly 72% in 2003.¹²

These statistics, of course, depend on a clear consensus definition of NGO. This has often proven elusive. Influential typologies include the generational model of David Korten and the network analysis of Green and Matthias.^14,15 Such questions are beyond the scope of this article. Rather than linger on definitional issues, we use the term pragmatically, in reference to any international, nonstate organization funded by external agencies to provide health services or technical assistance in developing countries. In addition to groups conventionally identified as NGOs, such as Save the Children, Cooperative for Assistance and Relief Everywhere, or Doctors Without Borders, our definition includes many faith-based organizations, foreign universities that register as NGOs in local settings, for-profit public health agencies, and some donors that occasionally act as service providers.

Many observers link this shift to NGOs, to structural adjustment programs promoted by the World Bank and International Monetary Fund in developing countries to limit public sector spending and privatize services to address the mounting foreign debt crises experienced by many poor countries since the early 1980s.^16–18 Debt-burdened countries were persuaded to impose public sector salary and hiring caps, cut construction, and reduce funding for training institutions to rectify balance of payment problems.¹⁹ As public services were cut back, some argued that NGOs had a “comparative advantage” in service delivery because they could presumably reach poor communities more effectively, efficiently, and compassionately.^20–31 In this view, state services are plagued by inefficiency, corruption, poor service quality, and unmotivated staff; by contrast, NGOs attract those eager to work with the poor and subsequently provide higher-quality services. Many donors celebrated the presumed virtues of the private sector in meeting market demand and allocating resources more efficiently; the US Agency for International Development often refers to NGOs as Private Voluntary Organizations.³²

However, many observers in Africa are now questioning this model as the proliferation of NGOs has led to management burden on local health managers, fragmentation of services, “brain drain” from public sector services to NGOs, and myriad projects that collapse when NGO grant funding ends.^17,30,33,34 Driven by donor demands, NGOs often focus narrowly on vertical programs that serve limited populations in confined geographical settings for single health problems.³⁵ As a result, NGOs frequently create showcase projects with questionable sustainability and perfunctory linkages to local health services; for instance, NGO-led HIV/AIDS testing and treatment projects are often not integrated into existing primary health care services. Some argue that NGOs can put more people on ART faster than could public sector systems in such projects by circumventing slow government bureaucracies—a claim that attracts donors eager to show dramatic results quickly.³⁶ However, as stated by Loewenson and McCoy,

Vertical programs established to achieve rapid delivery against unrealistic targets can divert scarce resources from strained public health services and bring undesirable opportunity costs and inefficiencies through the creation of parallel management and administrative systems.^37(p242)

COMPARATIVE ADVANTAGE OF THE PUBLIC SECTOR

With the challenges now confronting efforts to expand HIV/AIDS treatment, many believe the comparative advantage actually lies with adequately funded national health systems. As implied in the Universal Declaration of Human Rights and clarified in subsequent covenants, governments must guarantee the right to quality health care that is available, accessible, and acceptable.^38–40 They can only meet this obligation through strong national health systems; nonstate actors and international agencies must commit to supporting this role to meet their own human rights obligations.

African governments can establish standards of care, ensure equity in service delivery, harmonize information systems, achieve geographic coverage, and carry out long-term planning based on local health priorities. Aid channeled into well-planned national health system strategies can prevent fragmentation of service delivery while efficiencies are gained by reducing the transaction costs from working across multiple vertical projects. Through economies of scale and coordination of funds, national health systems can reach the most people more efficiently and cost-effectively than could private sector commercial providers, charities, or NGOs. Their influence over all sectors of the health delivery system ensures decisions made at one level can intersect with those at another. Public sector health systems are also lasting institutions in which complex health programs can be sustained and improved over time. And, unlike NGOs, national health systems should, in principle, be accountable to those they serve because they are controlled through local mechanisms of governance. Although those mechanisms are frequently inadequate, public sectors are premised on their responsibility to serve their constituents.

Numerous empirical examples of this advantage in HIV care are now emerging. Brazil was among the first countries to implement a universal ART program and by 2003 was providing free medication to approximately 125 000 people. According to the 2004 World Health Report, Brazil's success derived from aggressive drug price negotiation, a drugs logistic system, laboratory capacity, and drug delivery through a network of more than 1000 public care and testing services.⁴¹ In Malawi, the public sector, with NGO help, has been able to deliver treatment to a large number of patients relatively quickly with good outcomes; the number of patients on ART rose from about 4000 in early 2004 to nearly 38 000 by late 2005, nearly 47% of the national target.⁴² In Mozambique, NGOs provide critical support, but the national system created the framework through which treatment has been successfully expanded to nearly 80 000 people in 3 years (more than 90% of its goal for the period),⁴³ a feat impossible to achieve through an NGO-centered approach.

Donors are beginning to notice this advantage. The Paris Declaration on Aid Effectiveness, with more than 100 signatory countries, affirms donor commitment to “increasing alignment of aid with partner countries' priorities, systems and procedures and helping to strengthen their capacities.”^44(p1) Curiously, however, this recognition has yet to translate into major shifts in funding. An analysis of 2004–2005 President's Emergency Program for AIDS Relief central awards grants (those managed by the US Agency for International Development and the Centers for Disease Control and Prevention directly out of the Washington, DC, headquarters) shows only 16% of funds were paid to host governments and the remainder was paid to NGOs, universities, the World Health Organization, and private companies.⁴⁵ In Mozambique, in 2008, only 26% of health sector foreign aid will be managed independently by the ministry of health, 26% will be jointly managed by the ministry of health with donors, and 48% will be channeled to NGOs.⁴⁶ Recognizing that NGOs will continue to play key roles in many developing countries, especially for increasing ART, we believe the time is ideal to discuss a code of conduct for NGOs to strengthen public sector health systems.

ELEMENTS OF A CODE OF CONDUCT

Existing codes of conduct have typically been written by country-specific NGO associations and national governments to address local concerns. As local circumstances vary widely, it may be challenging to develop a useful unifying international code. Nongovernmental organizations do not operate in isolation and any such code should acknowledge the varied possible relationships among international NGOs; local community, civic, and faith-based organizations; national and local governments; international institutions; and public and private donors. Although the challenges are daunting, we look to the success of the Code of Conduct for the International Red Cross and Red Crescent Movement and Non-Governmental Organizations in Disaster Relief as a precedent for this movement.⁴⁷ The International Committee of the Red Cross code, launched just after the Rwanda genocide of April 1994 with more than 300 signatories, has since been usefully invoked in humanitarian crises in Afghanistan, Iraq, Gujarat, and other areas.⁴⁸ Other useful lessons are offered by the Code of Good Practice for NGOs Responding to HIV/AIDS, launched in 10 countries, with the endorsement of more than 160 NGOs.⁴⁹ The specific urgent need now for national health system support leads us to propose an NGO Code of Conduct for Health System Strengthening to address 3 major concerns about NGO activity not included in other codes (Table 1 and the box on page 2137).

TABLE 1.

Nongovernmental Organization (NGO) Impact on National Health Systems

Area	Negative Impact	Positive Impact
Management	Burden	Support
	Multiple projects to oversee	Support for management capacity building
	Divergent financial and program reporting requirements	Support for financial coordination and harmonized program reporting
	Diversion of planning to meet NGO needs	Support for integrated planning
Operations	Fragmentation of services, vertical technical assistance	Technical assistance, innovation, pilot projects
	Showcase projects with limited sustainability	New, innovative programs to meet MOH priorities
	Imbalances in geographic and programmatic resource allocation	Contribution of resources to MOH technical assistance priorities
	Vertical programs that undermine service integration	Innovative methods to channel vertical funds into integrated services
	Concentration of scarce MOH human resources within NGO-related projects	Allocation of human resources to MOH for innovative projects
Human resources	Shortages	Capacity building
	“Brain drain” to NGOs	On-the-job training for MOH staff
	Lack of sustainability for new programs	Funding for additional MOH workforce for new program needs
	Lower morale among health workers	Advocacy to improve work conditions, capacity, and workloads
	Weakened management through loss of skilled staff	Provision of management training and funding for new management tools

Note. MOH = ministry of health.

Elements of a Nongovernmental Organization (NGO) Code of Conduct for Health System Strengthening

(1) Hiring practices that ensure long-term health system sustainability

Limit hiring out of public systems

Obtain consent from the local MOH for any hiring from public systems

(2) Compensation practices that strengthen the public sector

Limit pay inequity between the public and private sectors, and compensate community health workers

Support pay incentives for rural service

Grant similar privileges to expatriate and national employees

(3) Human resources support for local health systems

Support increases in the number and capacity of health professionals

Support training to build management and service capacity in MOHs

(4) NGO management support for MOHs

Commit to joint planning

Follow MOH geographic, administrative, and personnel norms

Advocate for flexible donor funding to mitigate effects of vertical funding

(5) Health system community support

Support communities' linkages to health systems while promoting government accountability

Help protect oppressed populations

(6) Advocacy to eliminate wage bill caps and limitations on health system investment promoted by IFIs

Notes. MOH = ministry of health; IFIs = international financial institutions.

Management Burden

International NGOs often promote pet projects with idiosyncratic accounting systems, individual reporting systems, and objectives distinct from those of ministries of health. These create enormous management burdens for local health officials.⁵⁰ Disruptive turf wars sometimes erupt between competing NGOs as they vie for access to specific geographic or health domains, requiring mediation by local authorities. Many ministry of health officials find it impossible to refuse desperately needed resources, even when they are channeled to NGO projects and away from national priorities.^51,52

International NGOs should instead match their resources and projects to existing ministry of health priorities and management capabilities. To do this, they should engage in joint planning and implementation, support the strengthening of existing administrative and managerial structures, and strengthen management capacity of local and national governments. The NGOs should also share budgetary and financial information.

Fragmentation of the Health Sector

NGOs are normally pressured by donors to produce short-term gains quickly (within 1 to 2 years) in a limited population, creating conflict with longer-term system strengthening. Showcase projects by NGOs are frequently designed as vertical programs with no plans for expansion or sustainability, and little integration with local health systems. The result is fragmented and inequitable health care delivery, where, for example, viral load measurement may be available, but cesarean sections are not; where one district has a state-of-the-art hospital and the next district has a building serving as a makeshift health post.^30,53

Nongovernmental organizations can minimize this fragmentation and help build a strong primary health care base by creatively integrating vertical donor-funded projects into the existing public sector health system. Donors should allow flexibility to NGOs in tailoring programs to existing conditions and systems. The code of conduct should include a commitment to help build local systems and use funding in ways that will most benefit comprehensive primary health care.

Brain Drain

Nongovernmental organizations often contribute to the human resources “brain drain” crisis in Africa when they lure government health workers away into highly paid NGO positions.^10,30,33 In our experience in Mozambique, this internal “brain drain” has had a more severe impact on the local health system than has the more widely recognized international migration of health workers. The NGO salaries may be 5- to 20-times higher than are public-sector salaries while providing more comfortable working environments and benefits.^{27,30,33,54–56} Structural adjustment program–related public sector salary and hiring caps restrict the ability of governments to compete with NGO offers, or train sufficient numbers of new health workers. When NGOs provide a few lucky health workers with high salaries, they contribute to morale and management problems among those left behind. Instead, NGOs should strengthen local human resource capacity by working within existing salary structures and complementing local training capacity. Rather than hiring workers out of the public system to work in a parallel program, NGOs can integrate projects into local systems and fund additional workers in the public system in accordance with local pay structures. Nongovernmental organizations can also support other incentives to retain staff, such as payment for overtime or after-hours service expansion, or stipends for extra training and additional job responsibilities.

In limited cases where NGOs are faced with hiring workers from a ministry of health, local health authorities should approve and coordinate the process. We propose that in these rare situations, NGOs commit to replacing (via support for preservice training position and salary) each health worker they hire out of the public system. Planners for NGOs should also commit to limiting pay inequities between NGO and ministry of health workers. Although market forces pressure NGOs to raise salaries, a collective effort within the NGO community could place ceilings on pay to keep pay ratios more reasonable.

NEXT STEPS

International NGOs have unique standing and opportunity to influence donors, governments, and multilateral organizations to strengthen national health systems.⁵⁷ Policies that restrict investment in public sector health systems, such as structural adjustment programs, should be exposed and decried by NGOs who have firsthand experience of their destructive effects. Indeed, reversing these policies and freeing governments to invest in their beleaguered health care systems would provide more benefit than a legion of geographically constrained and time-limited NGO programs. The code should therefore explicitly compel signatories to advocate with donors, governments, and international financial institutions to remove constraints and increase aid directed to health systems strengthening.

We have already launched a series of discussions among a small number of concerned actors to develop an international NGO Code of Conduct for Health System Strengthening, building on earlier efforts at codes designed to influence NGO ethical behavior and accountability. Nongovernmental organization forums have developed country-specific codes of conduct in Mozambique, Botswana, Uganda, Afghanistan, Pakistan, India, Philippines, South Africa, and the Democratic Republic of the Congo, among others.^47,58–66 The recent Mozambique code of conduct outlines principles of partnership and coordination in which NGOs are to abide by the Ministry of Health strategic plan, adhere to a code of conduct for bilateral and multilateral partners, avoid duplication of efforts, and promote public sector capacity building.⁶⁷ In addition, it states that projects involving more than US $1 million should include construction of a health center, rehabilitation of equipment, or scholarships for training courses.

In the development of the new code suggested here, there are a number of other contentious issues to resolve; for example, should international NGOs pay taxes in countries where they operate? There are few suitable global arenas to discuss and debate such a code, and NGOs range across the ideological spectrum with varying attitudes toward private versus public sector involvement in service delivery. Public sector support may be controversial in fragile states, conflict settings, or in disaster relief programs. However, even in these precarious situations, NGO actions can either contribute to or hinder the state's ability to serve local needs.

Code compliance will be a challenge, and experience with existing codes in specific countries has been mixed. In Mozambique, code enforcement is constrained by lack of designated processes and personnel to follow-up and monitor activities. However, compliance with a broader international code could be encouraged with the participation of major donors in collaboration with local governments. Key donors could tailor their awarding mechanisms to promote adherence to code guidelines. The Joint United Nations Programme on HIV/AIDS or the World Health Organization could promote the code through their country missions. The code would also provide local ministries of health with a tool and shared set of expectations to use in negotiation, management, and coordination of NGO activities.

As an important follow-up to the Paris Declaration, the International Health Partnership was launched in September 2007 as a pact among donor and recipient countries, international health agencies, and foundations that aims to help developing countries strengthen national health systems as a whole, ensure better coordination, and provide long-term, predictable financing to countries.⁶⁸ It is also encouraging that the Global Fund to Fight AIDS, TB, and Malaria has recognized the importance of health system strengthening, in addition to providing support for disease-specific programs, and has also committed to the Paris Principles of Harmonization and Alignment. The Global Fund to Fight AIDS, TB, and Malaria also mandates that recipient countries establish country coordinating mechanisms that seek to align state and nonstate activities. In practice, country coordinating mechanisms have had mixed results; nevertheless, they provide potential institutionalized donor support for code compliance.

In spite of formidable challenges, the existence of a well-known and widely disseminated code could shift expectations across the aid industry and, thus, provide a kind of professional peer pressure to adhere to certain standards of conduct or risk ostracism and damage to organizational reputation with eventual impact on funding. There is increasingly broad agreement that without strengthening health systems, a long-term solution to the AIDS crisis and the high burden of disease faced by the poorest countries will not be possible. With the help of an international code of conduct, a reorientation of NGO practice toward system strengthening could add significant momentum to the major new efforts now underway to bring HIV/AIDS treatment and health care to all.