64. General Primary Care

Scaling Up

Adequate delivery of services (and health care more broadly) at the primary care level is, we believe, fundamental to effective functioning of health systems. However, for the most part, primary care systems in low- and middle-income countries have yet to receive the sustained attention and resources that their importance warrants. Early efforts at primary care expansion in the late 1970s and early 1980s were overtaken in many parts of the developing world by economic crisis, sharp reductions in public spending, political instability, and emerging disease. Although essential packages based on cost-effectiveness criteria have been criticized for their largely disease-oriented and vertical approach, in most poor countries even these limited versions of general primary care remain incompletely applied and largely unaffordable in relation to current per capita health care expenditure. At the same time, renewed awareness of the centrality of the primary level in responding to the consequences of the HIV/AIDS epidemic or to rapidly rising cardiovascular risk means that increasing demands will be placed on primary care services. This section examines critical elements of any strategy to scale up primary care efforts; a prerequisite, however, is an adequate understanding on the part of policy makers and planners of the position and role of primary care in the national health system (Travis and others 2004).

Committing More Financial Resources

In the mid 1980s, Drummond and Mills (1987) found the best estimate of the cost of effective primary health care (including the recurrent and capital costs of basic and village-level health services but not of water and sanitation) to be 2 percent of the annual per capita gross national product (GNP). This amount, they noted, is considerable, given that many governments in developing countries do not spend even 2 percent of annual per capita GNP on their entire health sector.

More recently, the Commission on Macroeconomics and Health (CMH) estimated that an additional US$40 billion to US$52 billion annual expenditure would be required by 2015 to scale up 49 priority health interventions—not all at the primary care level—to reach high levels of coverage in 83 deserving countries (that is, countries with a GNP per capita below US$1,200, plus all countries of Sub-Saharan Africa) (Jha and Mills 2002). Apart from the recurrent and capital costs of the interventions themselves, this estimate included management costs generated at levels above "close-to-client" services, expenditure to improve absorptive capacity, expenditure on improvements in the quality of care, and 100 percent increases in staff salaries to address the problems of staff recruitment and retention. The inclusion of these costs accounts largely for the greatly increased per capita estimates of the CMH package relative to earlier estimates by the World Bank (see table 64.3), and probably provides a better estimate of what is needed, given the enormous challenges facing primary care service delivery.

The CMH has placed great emphasis on donor funding of services to adequate levels (see chapter 12 for a more extensive discussion of sources of financing). Other avenues of funding include reprioritizing government budgets or recovering costs through health insurance schemes and user fees, although these all remain difficult options within low-income settings. In particular the experience of user fee schemes, which proliferated in the 1990s, suggests that such schemes have negative impacts on equity, especially at the primary care level, and should be applied with great caution in poor communities (Gilson 1998). Yet the fact remains that an injection of additional resources is clearly one prerequisite for the successful scaling up of general primary care in the 21st century, backed up by political commitment to the centrality of general primary care (and primary health care more broadly) as a fundamental strategy for tackling the highest-burden diseases and their causes.

Developing Human Resources

Although increased financial resources are imperative, Kurowski and others (2003) emphasize that "human resource availability is likely to determine the capacity to absorb additional financial resources and thus the pace of scaling up." These authors warn that human resource availability is likely to be grossly insufficient to meet the scaling-up needs envisioned by the CMH.

The skills and competencies necessary to deliver and support effective primary care are in some respects similar to those required at other levels of the health system (see chapter 71), but certain competencies warrant special emphasis at the primary care level (see box 64.2). Above all, if local services are to meet community health needs, leaders at the primary care level will have to be freed from the constraints of stifling, rule-bound bureaucracies and encouraged to develop innovative and at times unorthodox responses to the demanding challenges they face. As expressed in the World Health Report 2000 (WHO 2000, 64), "a key challenge in health service delivery is to balance the need for broad policy oversight with sufficient flexibility so that managers and providers can innovate and adapt polices to local needs and contexts in a dynamic way."

The creation of dynamic health teams at the primary level is one of the greatest requirements for scaling up effective primary care. The role of community health workers in such teams remains unresolved and bears further investigation. At the same time, one of the most challenging constraints is to overcome the loss of motivation and sense of resignation of the great body of primary care workers who work in under-staffed settings; who lack consistent, quality support; and who have grown accustomed to a norm of inadequate service delivery (Narasimhan and others 2004). As Hongoro and Normand assert in chapter 71, the extent to which countries can improve access to good quality primary care will depend in large part on a "better matches of skills to needsand clearer understanding of how improved structures and incentives will work."

Harnessing Private Sector Resources

Private sector health care provision is widespread and growing. It extends from local supply of drugs and equipment to fee- for-service and insurance-based medical care to the many forms of traditional practice. Although general primary care in most African, Asian, and Latin American settings is a major feature of publicly financed services that are provided by the public sector, private providers clearly play a significant role in many low- and middle-income countries with respect to the provision of primary care services (Berman and Rose 1996; Palmer and others 2003). Governments have thus viewed private sector providers as contributing additional human and related resources that can be deployed in the service of at least a portion of the population (usually those with means, including employees with access to reasonable health insurance coverage). By alleviating the workloads faced by public providers, private sector providers have allowed the public services to focus more directly on poorer communities and patients without means. Out-of-pocket payments, health insurance, and donations (as opposed to government contracts) that fund private sector services thus result in additional financing for the health sector.

Patients often prefer the private sector for a number of reasons. These reasons include geographic accessibility, convenient opening hours, and more favorable staff attitudes, as well as perceived better quality in terms of shorter waiting times, greater privacy, higher standards of diagnosis, better (perceived) treatment, and counseling (Doherty and Govender 2004). Although private providers are generally thought of in relation to curative care, interest is growing in the role they could play in meeting public health objectives, especially with respect to the scaling up of primary care services (Palmer and others 2003).

It is important that the potential contribution of private sector resources be optimized through appropriate use of public-private partnership mechanisms, public sector contracts, and government regulation. These mechanisms are generally easier for not-for-profit providers to contemplate, because they have often been instrumental in bringing primary care to poor communities. Some nongovernmental organizations (NGOs) are able to offer services that can fill notable gaps, home-based care to HIV and AIDS sufferers being but one example.

The potential of for-profit providers to contribute to the care of the poor is less obvious, especially given the incentives to overservice that are inherent in the fee-for-service reimbursement system. Mills and others (2002) find that consumers of private sector primary care are often unable to assess the technical quality of services, tending to place more weight on aspects of perceived quality, such as interpersonal skills of providers and the comfort of the environment in which treatment occurs, than on technical competence. Mills and others (2002,) argue that the effectiveness of private services is by and large rather low: "poor treatment practices have been reported for diseases such as tuberculosis and sexually transmitted infections, with implications not only for the individuals treated but also for disease transmission and the development of drug resistance."

Palmer and others (2003, 292) have reviewed a "new model of private primary care provision" that has emerged in South Africa. This innovation involves commercial companies providing "standardized primary care services at relatively low cost" that are targeted at the low-income employed rather than the very poor (Palmer and others 2003,). Regarding the growing popularity of these private clinics, the authors find that they maintain excellent standards with respect to the quality of services. The clinics also run at a cost per visit that is comparable with public sector primary care clinics, demonstrating that the acceptability of services to users and low-cost service delivery are not incompatible objectives.

Palmer and others (2003, 295) suggest that the increasing popularity of these (affordable) private clinics may provide an opportunity to encourage employed but low-income workers (who historically have used public sector health services at little or no charge) to make use of these clinic networks, which would enable the public sector to better tend to its "role as regulator and providing services to the poorest." Potentially, this redirection of care could remove some of the burden on the public sector, and the task of regulation might be made easier by the strong hierarchical control exercised within these clinic chains. To some extent, this shift has been the experience in Sri Lanka, where government services have been designed with the explicit assumption that certain forms of care will be provided through the private sector (Rannan-Eliya 2001).

However, Palmer and others (2003) point out that the model has potential drawbacks. The comprehensiveness and continuity of services provided by these private clinics fall short of that available in the public sector. Furthermore, how the behavior of private clinics would change under a system of contractual arrangements with the public sector is not clear. Whereas contracting with the not-for-profit sector tends to accommodate government objectives fairly easily (Gilson and others 1997), the experience of contracting with the for-profit sector has had mixed results. These and other concerns imply that, although the for-profit sector is an important resource, arrangements for the delivery of care through this sector should be developed with caution. It also bears mention that, where public sector systems are weak, private sector services gain ground to the extent of unbalancing the public-private mix, with potentially serious consequences for costs and continuity in patient care and for coverage and equity more generally.

Setting Population Health and Clinical Care Priorities

Along with securing additional resources for primary care delivery, country capacity to generate the information necessary for setting and reviewing public health and clinical care priorities must be strengthened as a fundamental measure (Commission on Health Research for Development 1990). This principle lies at the heart of influential pilot work—at times referred to as community-oriented primary care—that emerged in the first half of the 20th century and now underpins the Tanzanian Essential Health Interventions Program (TEHIP), 1997-2004. TEHIP, through a research and development arm tasked with devising practical tools for decentralized health planning, has tested "how and to what extent evidence can guide planning of the health sector at district level[in order to] improve technical and allocative efficiency with regard to local choices for resource allocation and services offered" (de Savigny and others 2002,).² A dynamic process of using high-quality local information, coupled with local problem solving, planning, and ownership, was central to appropriate decision making and consequent implementation.

Because local data on intervention costs and coverage are generally not available to district planners and managers, local cost-effectiveness analysis is difficult to incorporate into decentralized priority setting. With TEHIP, priority setting was driven more by the shares of the burden of disease that known cost-effective interventions could address. New analytic tools were devised that would help focus resource allocations on the major "intervention-addressable" disease burdens; targeted sets of cost-effective interventions were then applied—in place of embarking on a disease-by-disease or detailed cost-effectiveness approach (D. de Savigny, personal communication,). Available understanding on cost-effectiveness was used to eliminate interventions known to be grossly cost-ineffective; it was not used to prioritize or rank interventions generally considered to be highly cost-effective.

TEHIP indicates that gross technical and allocative efficiencies are relatively easy to address when incremental funding is available. As described by de Savigny and others (2002), the net effect of decentralized funding, together with a mutually reinforcing series of planning, management, and capacity-development inputs, was a proportional and absolute increase in resources for more efficient delivery of prioritized, cost-effective interventions addressing the largest shares of the preventable local burden of disease; an increase in the use of government health services; and a decrease in mortality in infants, children under five, adolescents, and adults. This effect was achieved with relatively limited resources.

TEHIP and related experience make clear that delivery of effective primary care requires a greatly stepped-up capacity to provide an evidence base that is founded on current and evolving local disease and risk factor burdens, the performance of local health services, client use of public as well as private and traditional services, and (where appropriate) the costs of providing care. Effective use of such information can profoundly enhance the ability of the health system to deliver on its core service functions, target high-risk and vulnerable groups, assess coverage in service provision, and gauge health effects. Moreover, such information is vital to establishing the dimensions of the local disease burden that should be managed at the primary care level (Kahn and others 1999). As cogently stated by the Bellagio Study Group on Child Survival (2003, 324),"the capacity of countries to obtain and use information to support child-health programmes will be a determining factor in reducing child mortality."

Developing a District Health System

Drawing from theory and experience in other branches of the public sector (Mills and others 1990)—and often as part of wider public sector developments—the health sector introduced decentralization widely in low- and middle-income countries throughout the 1980s and 1990s (WHO 2000). Positive justification for this method of delivering health care and primary care in particular lay, first, in its intended benefits—for patients and communities—through the provision of context-appropriate services of steadily improving quality. This service delivery was rightly seen as also conferring substantial financial benefit on households. Second, decentralization was expected to lead to the strengthening of local responsibility and accountability, with growing authority of district management teams over local cost centers. Third, it was presumed that the more central management levels would invest in enhanced support systems, including management support, further training, financial management and administration, laboratory services, and drug supply systems (World Bank 1994).

In developing settings, few health systems did not decentralize in some form or another over this period, and most based services development on a so-called district (or subdistrict) health system model. Considerable effort was devoted to achieving a balance between primary care service delivery and referral to the first-level (or district) hospital. Incentives as well as penalties were invoked to encourage first use of primary care facilities.

Notwithstanding the theoretical appeal of health system decentralization, numerous difficulties in implementation were encountered, with the consequence that the performance of decentralized, primary care-oriented systems and national-level support to these systems have fallen way below expectations (Bellagio Study Group on Child Survival 2003). Although various factors can account for this outcome—and although these factors will differ according to local and regional circumstances—common difficulties include inadequate or insufficient primary care skills and competencies, which result in poor-quality care; breakdown in referral systems for emergencies and more complex cases (McCord and Chowdhury 2003; Snow and others 1994); delegation of responsibilities without the concomitant delegation of authority, especially in relation to budgeting; authoritarian or strictly hierarchical managerial styles that are not conducive to local health services support and development; and weak or absent measures to develop workable cost-management systems appropriate to different service levels.

These problems in achieving successful delivery need to be addressed if decentralization is to achieve its intended benefits. Again, greater appreciation of the role of decentralized systems in the broader health care architecture, the support needed to ensure their effectiveness, and the time required to build the necessary capability are all necessary. As Bryce and others (2003, 160) put it, "although research on interventions is plentiful, little is known about the characteristics of delivery strategies capable of achieving and maintaining high coverage for specific interventions in various epidemiological, health system, and cultural contexts." From this perspective, a too-narrow preoccupation with the cost-effectiveness of interventions cannot but have shortcomings: "whatever package of services is delivered, the resulting effectiveness and equity will almost certainly depend on how the services are delivered, [in other words] the strategy for organizing the care" (B. Starfield, personal communication,).

Primary care is delivered through a system of facilities, equipment, and personnel; tackling inefficiencies in the system may have major positive benefits for quality of care, program coverage, and cost-effectiveness. In many settings a real opportunity exists to increase the efficiency of general primary care teams by giving attention to working conditions, ensuring functional equipment, and maintaining a stable drug supply. Meaningful step-ups in care, workable referral and communication systems, gatekeeper functions where indicated, and effectively aligned management and support are all needed. Achieving such efficiencies should result in many more patients being assessed and managed properly. Significant cost savings may accrue to the health service (through patients being managed at the primary care level rather than the first or specialist referral level) and to patients, families, and households (through care being delivered more rapidly and nearer to home).

Demanding Services: Relationships with Local Communities

Among poor and vulnerable communities, the need for care is demonstrably high, and the effectiveness of primary care services is likely to substantially influence demand on the public sector. In relation to infectious as well as noncommunicable disease, outreach services have a major role to play in promoting positive health and health-seeking behaviors and in supporting community-level preventive and promotive efforts. More generally, renewed efforts to enhance community relationships with primary care workers and the health system as a whole—and to ensure that community voices actively and appropriately bear on local service development and decision making—can help bring clients and communities into constructive public health care partnerships.