Chapter 7: Health Systems: principled integrated care
The global health workforce crisis
The most critical issue facing health care systems is the shortage of the people who make them work. Although this crisis is greatest in developing countries, particularly in sub-Saharan Africa, it affects all nations. It severely constrains the response to the AIDS treatment emergency and the development of health systems driven by primary health care, even as AIDS reduces the available workforce. Botswana's commitment to provide free antiretroviral therapy to all eligible citizens is frustrated, not by financing, but by the severe lack of health personnel (12).
Unfortunately, workforce issues are still considered to be relatively unimportant by both national governments and international agencies. Rapid and substantial strengthening of the workforce is urgently required to capitalize on the funds and pharmaceuticals that are now available.
The health workforce crisis has to be confronted in an economic and policy environment very different from that of 25 years ago. Traditional models in which the government directly recruits, trains, hires and deploys health professionals no longer reflect the reality of most developing countries. Most countries have undergone decentralization and reforms of the civil service and the health sector. There has been a great expansion in the health care roles of nongovernmental organizations and private providers. Furthermore, all countries are now part of the global marketplace for health professionals, and the effects of the demand--supply imbalance will only increase as trade in health services increases (13). Accordingly, new models for health workforce strengthening must be developed and evaluated (14).
Size, composition and distribution of the health workforce
The number of health workers in a country is a key indicator of its capacity to scale up delivery of interventions. This crisis is nowhere greater than in sub-Saharan Africa, where limitations on staffing are now recognized as a major constraint to achieving national health goals and the MDGs (15). In Chad and the United Republic of Tanzania, for example, the current workforce is grossly insufficient for the extensive delivery of priority interventions (16). Countries facing such extreme personnel shortages urgently need a rapid increase in the numbers of health workers to perform key tasks, particularly the delivery of services at community level in underserved areas.
The number of health workers in a country is not the only determinant of access to primary health care. Figure 7.1 shows that the number of births at which skilled attendants are present is only partially related to the number of health professionals in a country.1 Guinea, Indonesia and Paraguay have similar workforce numbers but wide differences in the level of coverage. This is caused by several factors, including the skill mix of health workers, their geographical and functional distribution, and their productivity. These data indicate the importance of using the existing workforce more effectively.

Figure 7.1
Gender discrimination in the health professions has many serious implications for the long-term strength of the health care system and especially for the delivery of services to poor and disadvantaged populations. A specific issue concerns the under-representation of women among those who manage and direct services, even though most health care workers are women. For example, in South Africa, men represent only 29% of personnel in administration overall, but they occupy 65% of all senior management posts (17). Without proper representation at the managerial and leadership levels, women's needs as employees within the health system will continue to be neglected. More generally, workforce policy and planning must consider gender and life-cycle issues, not only out of concern for equity, but also to enable efficient and effective development of a health care system that responds to and meets the particular needs of women.
Workforce training
Meeting urgent health challenges while laying stronger foundations for health systems requires that health planners consider the composition of the health workforce in terms of training levels and skill categories. In developed countries, experimentation with new categories of health worker is a response to cost-containment and quality of care concerns. In developing countries such experimentation is a direct response to limited supply.
To achieve the goals associated with health care systems driven by primary health care, new options for the education and in-service training of health care workers are required so as to ensure a workforce more closely attuned to country needs. Training of students from developing countries at high-prestige institutions in developed countries is useful only when there is no local or regional alternative. Although there are about twice as many nursing schools as medical schools worldwide, in the African Region there are 38 nursing schools and 64 medical schools. This suggests that too many expensive health workers are produced in places that might have a greater need for new types of providers with an education more focused on primary health care. The public health workforce also needs strengthening, based on a new approach to in-country or regional training that emphasizes the management of health problems at the district level (18).
The workforce of doctors is often complemented by training nurse practitioners, "assistant medical officers" and mid-level professionals. These categories are health professionals who can assume many of the responsibilities previously reserved for those with a full medical degree (see Box 7.4). For example, many studies in developed countries show that nurse practitioners can reduce the costs of care without harming, and sometimes actually improving, health outcomes (21,22). In the Pacific Islands, mid-level practitioners, with various titles such as medex, health assistant, or health officer, play an important role in meeting curative and preventive needs, especially in remote or rural areas (23). In other countries, community health workers are trained in very specific and high-priority activities, making it possible to serve populations that are out of the reach of formal health care services.
Box 7.4 Training assistant medical officers: the técnicos of Mozambique
In 1984, a three-year programme was initiated to create assistant medical officers (técnicos de cirurgia) to perform fairly advanced surgical procedures in remote areas where consultants were not available (19). The programme trains middle-level health workers in skills required for three broad priority areas: pregnancy-related complications, trauma-related complications and emergency inflammatory conditions. Two years of lectures and practical sessions in the Maputo Central hospital are followed by a one-year internship at a provincial hospital, under the direct supervision of a surgeon.
Forty-six assistant medical officers were trained between 1984 and 1999, and the evaluation of their influence on quality of care is promising. For example, a comparison of 1000 consecutive caesarean sections conducted by técnicos de cirurgia with the same number conducted by obstetricians or gynaecologists indicated that there were no differences in the outcomes of this type of delivery or in the associated surgical interventions (20). Many countries have now started or are considering similar programmes, based on their claimed cost-effectiveness. The potential impact of this type of health worker on both quality and efficiency of health care must continue to be evaluated.
In the past, primary health care strategies based on community health workers or other alternative health care providers have been difficult to sustain (24). However, evidence suggests that such strategies can be effective, given appropriate training (25,26). To be successful, the creation of new types of health worker requires that they be valued for their distinctive contribution, rather than treated as second-class providers. This means offering them career development prospects, rotation to and from rural and underserved areas, good working conditions, the chance to work as a team with other professionals, and an adequate salary. New cadres can be seen not only as a pragmatic response to current shortages, but as a cohort whose skills can be continually upgraded through in-service training, leading in the longer term to their incorporation in the more highly qualified professional categories. Evidence is growing that community members can carry out a wide range of health care tasks, including treatment of more complex conditions (10, 25--28).
Migration of health workers
Policy-makers in all countries are concerned about "brain drain" of the health workforce within and between countries, although relevant research is still in its infancy (29). The movement of health professionals closely follows the migration pattern of all professionals. While doctors and nurses make up only a small proportion of professional migrants, their loss weakens health systems.
The internal movement of the workforce to urban areas is common to all countries. Within a region, there is also movement from poorer to richer countries, for example from Zambia to South Africa or from the United Republic of Tanzania to Botswana. The most controversial "brain drain" is international professional migration from poorer to wealthier countries (30). While the departure of doctors receives the most attention, it is the departure of nurses and other health professionals that can easily cripple a health system. Nurses are in high demand in developed countries, partly because of population ageing. Some efforts have been made to promote ethical practices in international recruitment, but results have yet to be assessed (31).
Workforce mobility creates additional imbalances just when increased financial resources are beginning to flow to some developing countries. This requires better workforce planning in developed countries, attention to issues of pay and other rewards in developing countries, and improved management of the workforce in all countries.
Paying more and paying differently
Raising wages may increase the number of health workers and their productivity and may succeed in countries where health workers are paid less than comparable professions. It may be less successful, however, in countries where health sector wages are higher than those of comparable professions. Wages take the single largest share of health expenditure, so increases have to be carefully evaluated for their impact both on the availability and productivity of health workers and on aggregate budgets. The role of public sector unions in negotiating with governments for pay increases is an additional complexity.
As well as raising salaries, other strategies to improve productivity include non-financial benefits such as housing, electricity and telephones, on-the-job training with professional supervision, and opportunities for rotation and promotion. In rural Senegal, providing nurses with motorcycles not only made it possible to increase immunization coverage but also improved their access to technical support and reduced their isolation (32).
Both financial and non-financial incentives can also reduce geographical imbalances in the distribution of health workers. For example, in Indonesia, a bonus of as much as 100% of the normal salary attracted medical graduates from Jakarta to the outer islands (33). Recruitment and training of people from remote areas, who are committed to their region of origin, have also been proposed.
Finally, nongovernmental organizations concerned with health and private providers are a large and increasing presence in most countries. Governments could consider partnerships in which the public sector provides financial support and the nongovernmental organizations manage and provide the direct services. Often, private health workers are available in places that the public sector finds difficult to reach. In such situations, establishing formal programmes either to contract private providers or to reimburse the services they provide may be the most pragmatic response. In the mid-1990s, the Government of Guatemala was obliged to expand health care services to unserved populations as part of the negotiated peace agreements that took place at the time. It contracted more than 100 nongovernmental organizations to provide basic health care services to some 3 million of the country's citizens, predominantly indigenous and rural people, who previously had no access to services (34). Recently, Cambodia has successfully experimented with contracting nongovernmental organizations and private providers to deliver basic services to underserved groups(35).
Improving quality: workers and systems
All health systems need to create an environment for effective team learning for quality improvement. While the quality of care depends to some extent on the individual characteristics of health care workers, levels of performance are determined to a far greater extent by the organization of the health care system in which they work. Many health care organizations are moving from a practice of blaming individual health care workers for deficiencies to a culture of team learning and shared quality improvement.
Recognizing that the quality of health care is fundamentally a system issue is the first step towards making improvements in processes and outcomes of care. System-level quality improvements require a clear definition of optimal care and a framework for changing the system. Essential components of optimal care include decision support tools for health care workers, with written care guidelines and diagnostic and treatment algorithms; and necessary supplies, medical equipment, laboratory access and medications. Also needed are clinical information systems (which can be computerized or paper-based), guidelines for planned visits and active follow-up, and systematic support to patients for self-management of their conditions and referral to community resources. Evidence suggests that health systems with strong, integrated primary care are associated with better outcomes, probably because they provide for more comprehensive, longitudinal and coordinated care (36).
There is also a requirement for a method that will enable health care workers to make improvements (37). Health care teams should be able to develop and test changes in their local settings, which will enable them systematically to make improvements in processes of health care. One such method is the model for improvement (38), which enables rapid testing and evaluation by health care teams of potential improvements in their work. Working together in groups of teams helps communication and spreads innovations to larger groups (39,40).
Responding to the workforce crisis
Taking action to meet the workforce crisis is not easy and requires paying attention to all areas of workforce needs, from training to morale, and from local to global determinants. WHO has a major advocacy role in building and sustaining awareness of the extent of the crisis.
Some actions can be taken immediately, others require more time and planning (see Box 7.5). In the most urgent circumstances, for example scaling up HIV/AIDS treatment, countries can mobilize community resources, volunteers and traditional healers to collaborate in expanding access to primary health care (see Box 3.2 in Chapter 3). Health care workers' productivity can be improved through better supervision, support and morale-building. In the short term, governments can initiate programmes that mobilize nongovernmental and private sector resources by contracting the delivery of services, upgrading staff skills, and making sure that workers have the drugs and medical supplies they need to do their job.
Box 7.5 Creating a skilled workforce for tomorrow
The process of building a motivated workforce with the relevant skills can begin immediately, using existing budgets and staff. These activities also require continuous investments of time and leadership, and the involvement of health workers and communities in planning and managing their own futures.
Immediate actions that should be taken include: mobilizing communities and community workers; engaging traditional healers and enlisting volunteers; raising productivity among current health workers through improved supervision and support; and assessing and obtaining feedback on quality of practice.
Reorienting managerial staff to new functions takes time and planning. This process includes:
- developing and implementing on-the-job training to upgrade skills;
- contracting with the private sector and nongovernmental organizations;
- introducing flexible new contract opportunities for part-time work; improving working conditions with better drug distribution and supply of other essential medical supplies;
- strengthening collaboration among health workers, traditional healers, volunteers and community members.
Preparing for changes to institutions, policies and legislation requires undertaking studies and analyses that need to be started immediately. Design, approval and implementation of the changes, however, need time and will have an impact in the medium term. Some actions in this category include:
- developing pay and non-pay incentive packages to improve staff recruitment and retention;
- developing a plan to improve training capacity and management practices;
- coordinating donor contributions to workforce development; designing and implementing safety guidelines, clinical protocols, and anti-discrimination policies to improve working conditions;
- building extensive partnerships with civil society.
Urgent problems require urgent action. Governments must not lose sight, however, of the long-term requirements of the health system. While tackling more urgent activities, governments can lay the foundations for effective workforce policies, by:
- analysing the labour market, relative wages and supply trends so as to be able to design appropriate recruitment, retention, and wage policies;
- developing long-term plans for achieving appropriate mixes of skills and geographical distribution of health care professionals;
- expanding opportunities for management training and for the improvement of management practices;
- developing strategies for strengthening the relationship between public and private providers and financing.
In the medium term, governments can bring in changes requiring more planning, reorientation of administrative staff, and changes in budgeting such as pay policies, non-pay incentives, and expanded training capacities. New guidelines and policies can be adopted. Collaboration with communities and local governments can be deepened. Important lessons can be learnt from country experiences, whether very successful or less so (14). Policy-makers may suggest that any proposed new project or policy include a formal "human resources impact assessment" during its preparation; international agencies and donors could also be brought into this process (41).
Finally, governments should keep sight of the conditions necessary to motivate and sustain good health service delivery. This means taking the dynamic nature of labour markets seriously, and recognizing the long-term limits and expectations of health care workers. It also means establishing more clearly the expected roles of public and private providers in a future system of universal coverage. WHO is actively working with countries to develop long-term and short-term solutions. An example from the Region of the Americas is provided in Box 7.6.
Box 7.6 The Observatory of Human Resources in Health Sector Reforms
The Observatory is a cooperative initiative promoted by the Pan American Health Organization/WHO Regional Office for the Americas. Its goals are to produce and share knowledge among the countries of the region to support human resources policy decisions and improve workforce development in the health services sector.
The initiative supports the creation of national inter-institutional groups (for example, ministries of health, universities, and professional associations) to collect information on the stock of human resources for health and to analyse imbalances and trends, to prioritize an agenda of issues to be tackled, and to advise on long-term and medium-term policy development. Nineteen countries participate in the initiative, with different emphases and priorities, according to national concerns. The networking efforts are geared to making the country experiences applicable in other contexts, through construction of databases and dissemination of lessons learnt. The main lessons to date are as follows:
- The Observatory is a way to improve the stewardship role of ministries of health in human resources.
- The inter-institutional Observatory groups can help to maintain the human resources agenda during the transition between administrations.
- There is a need to integrate key stakeholders: universities, ministries of health, and professional associations, even though there may be substantial conflicts between them.
- Evidence needs to be developed from more reliable and stable statistical sources (for example, the International Labour Organization and household surveys).
- New approaches should be found to use the information in shaping policies (for example, to improve geographical distribution and to correct public--private imbalances).
General information about the network, its meetings and useful links can be found at http://observatorio_rh.tripod.com/.