The Delivery of Primary Care
The definition of primary care (Chapter 2) is a normative definition; that is, it defines what the committee believes the function of primary care should be. Whether the elements of this definition can be achieved and whether primary care can assume its proper role in the delivery of health care will be determined in a world of health care that is being reshaped by the forces described in Chapter 1. Although some of those forces are favorable to aspects of primary care, the committee is not convinced that the current health care market, by itself, will shape primary care to match all aspects of the definition. Further actions will need to be taken to provide the financial incentives and infrastructure that will help the health care system overcome barriers. This chapter includes recommendations for such actions. In addition to barriers that are specific to primary care, the committee notes that the lack of universal entitlement to health care benefits will continue to raise special problems for the uninsured and underinsured in obtaining access to primary care.
The committee is under no illusion that it can, or should, prescribe a single path for delivering primary care in an environment that is so diverse and changing so rapidly. Nevertheless, the definition presents clear guideposts for actions by the many actors in health care: health professions, health plans and organizations, payers for group coverage who set many of the standards within which health care is organized, and government regulators. Diversity in the means of achieving the committee's primary care objectives may be desirable, but the key elements of the definition should be the criteria by which actions to advance primary care are judged.
This chapter is presented in two sections. The first section outlines the
committee's observations about the current trends and characteristics of U.S. health care that form the current context for the delivery of primary care. The second section contains the committee's conclusions and recommendations about changes needed to improve the delivery of primary care in order to realize more fully the potential of primary care to improve the health and satisfaction of patients.
Current Pathways For Primary Care
The rapid pace of change and the diversity of local circumstances are striking characteristics of current health care. Descriptive evidence about current directions of health care, augmented by the committee's five site visits, confirms the magnitude and rapidity of those changes. Ours is a health care system going through a major transition. From an era of growth in expensive services supported by open-ended financing, wide choice of clinicians and hospitals, and almost complete freedom for clinical judgment, the U.S. health system is moving quickly into an era of limits on resources, cost-based competition among health plans and providers, financial risk-sharing by providers, and constraints on patient choice of clinician. No one can predict accurately where the health care system will be in 5 years, let alone 10 or 20 years. Simple generalizations informed by past studies, even studies only a few years old, are limited in their ability to describe or explain current directions in health care. Yet we believe that broad pathways for that change can be identified and need to be taken into consideration.
Some studies have identified stages of the health care market that imply a progression toward "mature" markets (University Hospital Consortium, 1993)—essentially those dominated by a handful of large, fiercely competitive health plans. The committee is wary, however, of any interpretation that such a progression is an orderly one. In visiting several areas of the country that are usually considered more mature health care markets (e.g., Minnesota and southern California), committee members observed that the pace of change continues to be rapid. Wherever these markets are going, they are not there yet.
With these cautions and caveats, we do see broad themes, both in what is happening and in what has not happened.
Spread of Managed Care
The term managed care has come to have many meanings. This committee uses managed care to refer to health plans that have a selective list of providers, both health professionals and hospitals, and that include mechanisms for influencing the nature, quantity, and site of services delivered. Many of these plans have focused initially on using their market power to obtain discounts from physicians, hospitals, and other providers in an oversupplied market. They are
evolving, however, toward more organized arrangements that include some form of involvement of the providers in the risk assumed by the plan. That risk derives from the plan's agreement to deliver a defined package of services for a fixed amount per capita for an enrolled population, such as with the various forms of health maintenance organizations (HMOs). The involvement of providers in the success of the plan is intended to offer incentives for containing costs while maintaining patient satisfaction with the care received.
Bailit (1995) estimates that in 1994, of a total of 180 million insured by private plans, enrollment in managed care totaled about 115 million persons. This estimate uses a definition of managed care that includes HMOs; "point of service" plans that combine HMO enrollment with an option to use providers outside the plan for an additional cost; and PPOs (preferred provider organizations), which offer a less structured arrangement that presents the enrolled person with a financial incentive to choose providers from a preferred list. He estimates that the number of individuals enrolled in managed care in the private market increased about 10 percent from 1993 to 1994.
Enrollment in managed care in the public programs in 1994 was much lower than in private plans, at about 8 percent of Medicare beneficiaries and about 25 percent of those eligible for Medicaid. The rate of increase is greater, however, especially in the Medicaid program. Forty-two states are implementing some form of managed care in their Medicaid programs. Arizona (100 percent), Tennessee (74.9 percent), and Oregon (21.9 percent) lead the way in the percentage of Medicaid dollars spent through managed care arrangements, but many other states are moving aggressively in this direction (Lewin-VHI, 1995). Current congressional deliberations on the future course of the Medicare program may result in further encouragement of enrollment of Medicare beneficiaries in capitated managed care plans.
Of particular significance for this study is that one major objective of most managed care plans is to reduce the use of specialists and to increase the use of primary care clinicians. The path to specialized care in most plans is through the primary care physician or other primary care clinician. Managed care, therefore, enhances the power of the primary care clinician to determine the services provided and by whom. The increasing future opportunities for primary care clinicians and the contrasting decline in the need for specialists have been described by Weiner and others in projecting future physician requirements (Weiner, 1993; COGME, 1995; PPRC, 1995).
The growth of managed care, although substantial, is taking place predominantly in large and medium-sized markets. Those providing services in rural areas are anticipating the move of managed care into their communities, but managed care was not yet evident in the rural areas visited by the committee.
The development of managed care varies widely by region. The most significant market penetration has been in the West, the upper Middle West, and the Northeast. The Southeast and South Central regions have less managed care at
this time (GHAA, 1995). In all areas, managed care on the basis of an enrolled, capitated population is not available to the uninsured and many of the underinsured, a growing proportion of the U.S. population (EBRI, 1995).
The growth of managed care plans is blurring the traditional boundaries between the insuring or financing function, with its strong concern for managing risk, and the provision of services and clinical decisions regarding those services. Managing risk is still important; no plan, regardless how efficient, wants to have a disproportionate share of sicker patients unless that risk can be shared. Most managed care plans, however, are also interested in how to improve the efficiency of services and how to maintain or increase patient satisfaction. Sophisticated buyers, such as the business community in the Twin Cities area, are developing performance standards for health plans that include clinical measures (Institute for Clinical Systems Integration, no date). Older staff and group model HMOs, such as Kaiser Permanente and Group Health Cooperative of Puget Sound, have long combined the insurance and patient care functions under a single organizational umbrella.
Development of Integrated Health Care Delivery Systems
A related and overlapping trend is the development of vertically integrated delivery systems that combine physicians and other health professionals, hospitals, rehabilitation units, social services, chronic care capabilities, mental health and substance abuse programs, and health promotion and disease prevention programs into an organized whole that can provide and coordinate a comprehensive array of services. Some of the motivation behind the development of these systems is to increase and protect market share in areas where there is surplus capacity. It is difficult to quantify the extent of systems change because so much is happening so rapidly. Many examples exist, mostly in larger cities but some in more rural areas, often built on preexisting multispecialty groups such as those of the Mayo, Marshfield, and Geisinger clinics.
Based on examples seen in the site visits, these systems at their best provide opportunities for innovations in arrangements for services, in part by breaking down institutional and professional barriers to delivering services more efficiently. They also provide the critical mass and capital needed for the development of infrastructure, such as information and clinical decision systems, telephone triage programs, and training. In the best of these organizations, the functions of primary care move well beyond the gatekeeper function toward a fuller application of the committee's definition.
These systems are not a new phenomenon; some of the older staff and group model HMOs have had many of these characteristics for some time. What may be new is an environment that encourages change rather than one that regards innovations as a questionable deviation from the norm. The pressure for continuing improvement in the cost-effective provision of services is present in older
integrated systems with long track records of success as well as in newer systems that have been built by combining previously independent providers.
For our purposes, the important point is that all of these systems are built on, or are building on, foundations of primary care clinicians, often by purchasing existing primary care practices. This primary care base is seen as necessary for both building and protecting market share and for creating a mechanism to control access to specialized services. In a capitated system, specialized services are seen as cost centers, rather than as revenue centers, and the organization has strong incentives to control such costs.
Consolidation of Health Plans and Systems
Both health plans and integrated systems are consolidating into larger organizations. They are driven to do so by several factors, including the need for capital, advantages in marketing, and potential economies of scale in developing and using infrastructure such as clinical information systems. Site visits to urban markets (the Twin Cities, southern California, and Boston) provided multiple examples in each site of major consolidations of health plans and provider organizations.
In communities where this consolidation is far along, characterization of health care as a very local and personalized service—a cottage industry as it has often been called—no longer holds. Becoming part of a larger organization is causing considerable stress for clinicians who value highly the autonomy of their practice and personal relationships with their patients. Some patients are also disturbed if they believe that their relationship with a primary care clinician who is committed to their interests is being compromised by a large, impersonal, and perhaps distant organization.
Growth in For-Profit Health Plans and Delivery Systems
Along with consolidation, health plans and integrated systems are increasingly under for-profit ownership. In addition to the growth of existing for-profit plans and their acquisition of not-for-profit plans, some not-for-profit plans are converting to for-profit forms of ownership. The need to raise capital for expansion is often given as the reason for the growth of for-profit ownership. The long range effects of this trend are not clear, but it raises the possibility of conflict between the desires of the stockholders to maximize profit and the objectives of primary care to ensure adequate care for patients. It also underlines the need to have measures of performance that include the interests of patients, not just the financial interests of group purchasers and stockholders, and that are available to guide patients' health care choices (for a fuller discussion of these issues, which is beyond the scope of this report, see IOM, 1986a, and Gray, 1991).
Diversity Among and Within Markets
As noted, markets vary widely in the extent to which services have moved along the pathways described above. Most rural areas have not yet joined these trends, and some urban areas have much lower rates of managed care penetration. Within markets observed on the site visits, some health plans are developing service innovations that improve the efficiency of care; others are focusing on utilization management, sales efforts to increase market share, and risk-sharing with providers as their means to compete successfully in their markets. Some groups of clinicians are tightly organized, and some are looser confederations of clinicians who remain essentially independent contractors with ownership and control of their own practices.
The clinicians involved in primary care services vary from plan to plan and setting to setting. In some plans extensive use is made of nurse practitioners and physician assistants; in others, much less. Other practices continue to emphasize the traditional role of the physician. Diversity is also seen in the type of primary care physicians involved. For example, in rural areas family physicians are prevalent, whereas in urban areas pediatricians and internists play a more prominent role.
Coordination of Primary Care with Other Services
The focus of most of the large delivery systems remains on more traditional medical services—acute and chronic care and preventive services provided by clinicians. The extent to which plans with enrolled populations are dealing with population-based health issues is not clear, although many examples of health education and behavior change programs exist. Cooperation with the public health agencies also seems weak.
Coordination regarding mental health and substance abuse services may be harder because of the trend toward "carve-outs" for these services into separate benefit packages that are independently managed. This new trend is in addition to the continuing patterns of delivery of many of these services by separate organizations and of limitations on these services in benefit packages.
Financial barriers to long-term care remain a significant problem. Few private plans include long-term care benefits. In the public sector, the Department of Veterans Affairs (VA) is a notable exception. All in all, concern about the lack of involvement of primary care clinicians in the medical care of patients in long-term care settings remains high (IOM, 1986b; 1995).
Vision care and pharmacy services are collocated in some group model plans, and many plans include a dental care benefit. Dental services as an integral part of the primary care delivery system, however, are seen mostly in programs organized for the poor and by the Indian Health Service.
Judgments may differ as to the likely results of these fissures in services for
common health problems. Nonetheless, the lack of explicit arrangements for coordinating primary care with other services that are needed on a routine and recurring basis by many patients is striking, especially as integration of other aspects of acute services moves ahead rapidly.
Current and Evolving Professional Roles
There is evidence of the increasing demand for primary care physicians as their incomes are rising relative to those of specialists in many areas (Mitka, 1994a,b; 1995). Further evidence of the rising status of primary care among physicians is the desire of many specialists to be designated as primary care physicians. California has given the primary care label to obstetrics and gynecology through state law, and other specialist groups have staked out a claim to the domain of primary care. This desire to be designated as primary care clinicians is the result of managed care plans' requiring that enrollees choose a primary care clinician, usually a family practitioner, general internist, or pediatrician, who will control access to specialized services.
There is also evidence of increasing demand for the use of nurse practitioners and physician assistants in primary care. Training programs for these professionals are expanding rapidly (see Chapters 6 and 7). The committee saw many examples of the involvement of these professions in primary care during its site visits, nearly always as part of a team that included physicians in a key role. Within integrated systems, the use of teams and delegation of primary care functions is proceeding rapidly (see Appendix E). In some locales, supply constraints, in particular, shortages of nurse practitioners, are impeding their greater use.
During site visits, committee members saw examples of further delegation of clinical functions to registered nurses, licensed practical nurses, and desk clerks. Such delegation was the result of a deliberate decision process that examined how specific clinical problems could be managed more efficiently. In some of these settings, primary care physicians were focusing on more complex clinical problems and taking on managerial roles, thus moving the clinical boundaries between primary care physicians and specialists toward more specialized care.
How widespread these changes are is difficult to document because doing so requires knowledge of the details of how particular functions are carried out, and these are only partially reflected in aggregate data on the numbers and types of professionals. This effort on the part of some of the more advanced integrated systems to redefine professional roles within a team concept may prove very important, however, as a future pathway for improving the efficiency and effectiveness of primary care. It may in turn have significant implications for training programs and for workforce policy. The care delivered by other first contact professionals such as dentists, eye care clinicians, and pharmacists is generally less coordinated with the broader functions of primary care.
Primary Care in Rural Settings
Observations made during site visits to rural areas are consistent with the extensive literature on rural care in noting what one host called the ''fragility" of many programs providing primary care to rural populations. Rural care is often dependent on some form of subsidy, as well as on a distant infrastructure that can provide technical assistance and professional backup. The reasons are several: the higher proportion of the uninsured and underinsured in many rural areas; higher costs of transportation; and lower volume of services. Primary care in the rural setting also includes a stronger emphasis on emergency care and the stabilization and transportation of patients with medical emergencies and trauma. Managed care has not yet penetrated most rural settings. The committee observed successful models of rural care, but none that did not have some form of subsidy or assistance (or both). It also observed impressive examples of the importance of community commitment to the maintenance of a primary care capacity in isolated rural areas.
Care of the Urban Poor
Care for low-income or disadvantaged populations, concentrated in the inner cities, is complicated by the lack of universal insurance coverage, the health care needs of illegal immigrants, and the low payments for providers in many states. These problems have often been alleviated by internal cross-subsidies and federal program formulas that favor institutions and care settings that serve a disproportionate share of the poor. The combination of competitive cost pressures and limits on public financing is likely to become much more acute in the near future and to make existing arrangements unstable. In some areas and states, such as Arizona, evidence suggests that managed care may be able to take on an increased share of these populations, but it is not clear how much such an approach can succeed without some form of subsidy that recognizes the extra costs now being incurred to serve the primary care needs of these populations.
Role of Academic Health Centers
In site visits, the committee heard numerous complaints from community programs about the lack of appropriate involvement of academic health centers (AHCs) in primary care and about the resulting lack of fit between the products of their training programs and the needs of managed care and community-based programs. The problems that AHCs face in surviving in the current health care market have been well documented elsewhere (Blumenthal and Meyer, 1993; Fox and Wasserman, 1993; Epstein, 1995; Josiah Macy, Jr. Foundation, 1995). The extra costs of training, the dependence on patient care income from referrals for tertiary services, the higher proportion of the poor in their service area, and
governance processes that make difficult a quick response to market changes are all handicaps for these institutions in a highly competitive health care market. For many of these institutions these factors constitute barriers to greater focus on primary care. Despite these barriers, there are also examples of effective involvement of AHCs in strengthening primary care. In one state, an AHC's mission statement was explicit about its commitment to primary care, and this mission was reflected in the curriculum and in assistance to communities in providing primary care.
Moving Toward Delivery Of Primary Care As Defined
Some aspects of the current health care scene favor further emphasis on primary care as the foundation for the health care system. Despite these favorable forces, many obstacles remain to be overcome in reorienting a large and complex health care system. As a sector of the economy that consumes about one-seventh of this society's resources and that is still growing faster than the rest of the economy, many powerful interest groups have a financial and professional stake in the status quo. Market forces seem to have the strength to require significant alterations in that status quo, but as stated earlier the committee remains skeptical that the market, by itself, will achieve a primary care system that meets its definition and that is widely available to the American public.
Because the benefits of primary care are important for meeting the health care goals of this society, the committee believes that a specific objective for the availability of primary care service should be established, focusing on the central relationship of the clinician and the patient.
Recommendation 5.1 Availability of Primary Care for All Americans
The committee recommends development of primary care delivery systems that will make the services of a primary care clinician available to all Americans.
In order to achieve this goal, steps need to be taken to create conditions favorable to primary care. Some steps involve public policies and the commitment of public resources by federal and state governments (even in a time of stringency for public budgets). Other steps entail voluntary actions to shape existing forces for change so that they more nearly match the committee's definition of primary care. Many of the desired changes will not be achieved rapidly. The results may vary widely in their particulars and still constitute movement in the right directions.
Specific actions in isolation from other needed actions are not likely to be successful. In this sense, bringing about the needed changes in primary care is a systems problem in which many elements interrelate. For example, shifts in the
education of primary care clinicians to encourage the function of a primary care team, as described in Chapter 7, are unlikely to have the desired result if the practice environment does not also support those changes. The rest of this chapter identifies some of the specific areas where action is needed to shape the course of the delivery of primary care toward the objectives that were identified in the definition.
Financing of Primary Care Services
The failure of comprehensive health care reform at the national level (which aimed at providing universal health insurance coverage) and the retreat from reforms at the state level (such as in Washington and Oregon) mean that many Americans remain without health insurance coverage. Furthermore, cost-competitive market forces are likely to exacerbate some of the problems of providing care to the uninsured. The proportions of the population that are either underinsured or uninsured are rising (EBRI, 1995; Short and Banthin, 1995). As long as significant financial barriers to access continue to exist for many millions of people, the objectives and implementing reforms recommended in this report, even if instituted fully, will not make the benefits of primary care available to many of those without health insurance. Addressing specific ways that health care coverage could be extended to everyone is beyond the scope of this report, but we note in the strongest terms that the primary care agenda for the nation will remain incomplete until this extension takes place.
Recommendation 5.2 Health Care Coverage for All Americans
To assure that the benefits of primary care are more uniformly available, the committee recommends that the federal government and the states develop strategies to provide health care coverage for all Americans.
The importance of this recommendation is accentuated by the effects of market forces in reducing the internal cross-subsidies and other forms of implicit subsidies that have helped to cover the health care needs of the uninsured. Likely reductions in the growth of public financing of health care in coming years will make these subsidies even harder to sustain. Therefore, the current situation of financial barriers to primary care for some of the population is almost sure to worsen without some form of public action.
The committee is aware of the controversies that may be engendered about who should be included under the rubric of "all Americans." It is beyond the scope of this committee to address these complicated issues in detail, especially the issues of coverage of undocumented aliens. If universal coverage is realized,
however, the coverage should at least extend to all those who are legal residents, whether or not they are citizens.
In addition to the lack of universal coverage for medical care, this nation seems nowhere near a policy that addresses the need to cover the costs of long-term care for the elderly and the chronically ill. This gap will continue to make more difficult the appropriate coordination of primary care services with long-term care services.
Delivery systems for primary care services need to assure the actual availability of services to all. Universal coverage may by itself encourage availability for some individuals and populations for whom primary care is currently unavailable or very inconvenient. But removing financial barriers to primary care through universal coverage is unlikely by itself to achieve the goal of availability of primary care services set out in Recommendation 5.1, and specific efforts will be required for some populations.
Later sections of this chapter address the need for special efforts to reach some populations with primary care services. Individuals in otherwise well-served areas may also face problems of availability, and these should also be addressed. Arrangements for the financing and monitoring of care would need to include achievement of this goal.
For the large sector of the population that does have health insurance, some methods of paying for services seem more likely than others to encourage primary care. As implemented in the United States, fee-for-service payments have favored procedural services and specialized care. In contrast, financing methods involving a single payment that covers specified services for an enrolled population over a period of time have provided incentives for primary care. Such global capitation payments have been used for many years by HMOs as various forms of managed care have spread and capitation has become more frequent. One study of the development of integrated delivery systems demonstrated that capitated payment mechanisms covering the continuum of care are most likely to promote clinical integration, preventive care, and treatment of patients in the most appropriate setting. As a result there is an incentive to place primary care rather than acute inpatient services at the center of the health care system (Shortell et al., 1994).
By providing an overall cap on resources, however, capitation may also reward health care plans for not providing services, and services necessary for good care could be neglected. Performance monitoring and public dissemination of quality-of-care information, as well as the opportunity for enrollees to change plans at regular intervals, are intended to motivate plans to provide quality services or risk losing their market share in the competitive environment in which most plans function. If these mechanisms to provide good information about plans work, health plans that provide good care efficiently should succeed. Such monitoring mechanisms, however, are still not fully developed in most markets.
Methods for translating capitation into reimbursement for specific primary
care services are numerous and evolving rapidly. For example, some group and staff model HMOs pay salaries to primary care clinicians and may include incentives that are tied to the overall performance of the plan. Many network model HMOs reimburse primary care services through a capitation payment to the primary care clinician, sometimes with specific incentives for desired practice patterns and sometimes placing the primary care clinician at risk for the use of specialty services. Other plans have used a mix of fee-for-service and capitation. Still others pay for primary care on a fee-for-service basis coupled with a financial incentive to discourage high utilization of services.
Capitation payments to the individual clinician may provide incentives not to make referrals that would be in the patient's interests or to skimp on the provision of primary care services by spending too little time with the patient. In particular, deGruy (Appendix D) notes that such incentives may affect the ability of the primary care clinician to deal with the mental health problems presented in the primary care setting, if adequate time for interacting with the patient is not provided. The committee did not have the opportunity to explore in detail the specific methods of paying primary care clinicians that would encourage good primary care. It did note during some site visits that innovations in the patterns of primary care and in the use of teams were associated with salary payment mechanisms. The salary approach also reduces incentives to withhold necessary services. The committee also agrees with the observation by Shortell and his colleagues (1994) that global capitation payments have been associated with an emphasis on primary care services within the overall service mix regardless of the specific method of paying the primary care clinician.
Recommendation 5.3 Payment Methods Favorable to Primary Care
The committee recommends that payment methods favorable to the support of primary care be more widely adopted.
These payment methods should include global capitation that covers all defined services for an enrolled population coupled with methods of paying the primary care clinician and the primary care team that support the characteristics of good primary care as described in this report. Among the methods that seem to be consistent with this objective are (a) salary arrangements and (b) forms of capitation or partial capitation payments (in combination with some form of fee-for-service reimbursement) to the individual provider that are structured to reward good primary care.
As capitation is translated into specific methods of payment for primary care clinicians, clinicians need to be given the flexibility to spend the amount of time with patients that is necessary for good primary care. For example, using rigid productivity guidelines regarding the number of patients to be seen per time period is not consistent with good primary care. The translation of capitation into
specific payment mechanisms should also support collaborative practice. This implies that productivity or improvement of health status ought to be measured for the entire primary care unit with such measures adjusted appropriately for the socioeconomic and health-related characteristics of the patient panel.
An aspect of indemnity insurance that discourages primary care is the use of deductibles and coinsurance that raise the marginal costs of the use of primary care services while expensive, specialized services are often provided at no further cost to the patient. Substantial financial disincentives to the use of routine and recurring care tend to encourage episodic, acute care; they work to the disadvantage of continuous care, care of the chronically ill, and advice about and coordination of other services. Such copayments and deductibles are also sometimes used within capitated systems. This disincentive for some primary care services is accentuated when certain areas, such as preventive services, are excluded entirely from benefit packages.
Although the methods for paying primary care clinicians are likely to continue to evolve and to include salary and capitation arrangements, fee-for-service reimbursement is likely to remain a method of payment for primary care for the foreseeable future. Such reimbursement may come as direct payment from a fee-for-service health insurance plan, as with indemnity insurance plans and the regular Medicare Part B program, and it may be used as the method of payment for individual clinicians under a capitated health plan.
Fee-for-service payments in the U.S. have not typically favored primary care services because they provide higher payment levels for specialized diagnostic and treatment procedures. Traditional patterns of fee-for-service payment are even less likely to support many of the aspects of primary care that are emphasized in the committee's definition, such as coordination of primary care with community-based services, which take clinician time and infrastructure support. Substantial efforts have been made to develop fee schedules that are more favorable to the primary care functions; the most notable is the Resource-Based Relative Value Scale (RBRVS) being implemented by the Medicare program and some private plans. Implementation of RBRVS in the Medicare program up to this time, however, has been disappointing in terms of encouraging primary care (PPRC, 1994). Because the various forms of fee-for-service methods are likely to continue to be used for reimbursing many clinicians in the foreseeable future, it is important for this payment method to provide better incentives for primary care, and the committee makes the following recommendation.
Recommendation 5.4 Payment for Primary Care Services
The committee recommends that when fee-for-service is used to reimburse clinicians for patient care, payments for primary care be up-graded to reflect better the value of these services.
The committee believes that greater emphasis on primary care clinicians may be more than offset by the savings that come from decreased use of specialty care. This issue of encouraging financing mechanisms that support primary care may also arise in proposals to establish "medical savings accounts" (MSAs). These individual accounts, which are created through tax-exempt contributions from employers or individuals (or both), can be used to pay for all types of medical expenses; typically the individuals benefit financially from any residual in their account. This proposal is usually coupled with insurance protection against catastrophic acute health care expenditures (American Academy of Actuaries, 1995; Joint Committee on Taxation, 1995). As of spring 1996, 15 states have adopted some form of MSA (Alpha Center, 1996).
A major aim of such proposals is to provide incentives to patients to use medical care efficiently by giving them a greater role in paying directly for services and a direct stake in the level of expenditure for the services. Such an approach offers greater economic incentives to limit the use of primary care relative to the use of expensive specialized services for two reasons: (1) most of the costs of the latter would be covered by the catastrophic insurance component of the plan and (2) the costs for primary care come out of MSAs. Another factor might be patients' tendencies to forgo preventive services with long-range benefits or other aspects of primary care, if the full benefits are not apparent to them or lie well in the future.
The concept of consumer sovereignty that underlies proposals for MSAs implies that consumers or patients have adequate knowledge to guide their own medical care decisions, but this is probably not true for many consumers. Consumers may postpone care until a major acute episode takes place, and this may be especially true for lower-income persons for whom the economic incentive to postpone care may loom large.
The committee is concerned that the values of primary care as discussed in Chapter 3 may be undermined by this approach to financing. If the funds are used for the purchase of comprehensive benefits that include good primary care coverage, such as an HMO plan, the effects may not be negative for primary care; but if the funds are used in a way that downgrades the function of primary care, the long-range effect on health outcomes and on aggregate health expenditures may be negative. This issue illustrates the limitations of the pure insurance model when paying for health care if it does not include appropriate primary care incentives (see the IOM report [1993b] on employment and health benefits for discussion of some aspects of this issue).
Another aspect of current approaches to financing that causes concern to the committee is the disruption of continuity that may occur when employers change health plans or when patients, motivated by small savings in health plan costs, switch plans. During the committee site visits, committee members heard from primary care clinicians and from patients that frequent changes in health plans offered by employers had forced patients to change physicians and that they
believe this had adverse effects on continuity and access. The committee notes, however, that in some markets employers are beginning to write contracts with health plans that extend for several years rather than just one year. The committee would like to encourage this trend. Three- or even five-year contracts would reduce the possibility that shifts in health plans would force patients to change primary care clinicians. The problem of patient-initiated changes probably needs to be addressed by better education of patients about the benefits to them of continuity in primary care.
Organizing Primary Care Services
The emergence of large integrated delivery systems has emphasized primary care. These organizational arrangements appear to have some economies of scale for the infrastructure of primary care, such as implementing information systems, disseminating clinical decision criteria, developing and evaluating innovative deployment of health personnel and mechanisms of coordinating services, and developing and using patient education materials. Coupled with enrolled populations, these systems offer other potential benefits: improved continuity of care, reduced barriers to movement between different elements of the care system, and pursuit of population-based approaches to disease prevention and health promotion. Whether these goals are realized depends in the longer term on documenting their advantages for patients and for the purchasers of group health benefits. Criteria for success must move beyond crude measures of cost saving to broader measures of systems performance (see discussion of performance measures below).
The committee also has concerns about this trend toward large integrated systems. Its definition emphasizes the importance of the personal relationship between the patient and the clinician or the team of clinicians. Can a large organization nurture and sustain such relationships in the midst of competitive market forces that are sometimes translated into limits on time spent with the patient? Can triage systems be implemented in a manner that appropriately supports regular contact with a clinician who is knowledgeable about the patient and the patient's history? In concept, and in reality, personal relationships can be fostered if the system makes it a high priority reflected in the organization's leadership, procedures, internal incentives, and patient education program.
Another issue is whether integrated systems address effectively the needs of rural populations, the inner city poor, and culturally diverse populations. The record to date is highly variable, and the inclusion of these special populations in large integrated systems has been limited. The trend toward use by states of managed care approaches for the Medicaid population has provided a possible linkage, and the nature and amount of the public funding for these programs will help to determine if integrated systems can meet the needs of these populations for primary care services.
Variation in successful organizational models is great; the committee does not recommend a specific organizational mode as best for primary care. The committee does believe, however, that the potential of integrated systems to provide primary care is substantial and should be encouraged; it also holds that performance measures used for internal and external evaluation of such systems should encompass the desired characteristics of primary care. The use of these systems to meet the needs of vulnerable and underserved populations also needs to be encouraged and measured, although their success in reaching out to these populations will be limited to the extent that these groups continue to lack health insurance coverage.
Understanding Professional Roles in Primary Care
The roles of the various health professions and how those professions should interrelate are both contentious issues. Discussions of professional roles in primary care are influenced by many past tensions: the sometimes strained relationships between nurses and physicians, the struggle of primary care physicians for appropriate status in a medical environment dominated by specialists and subspecialists, and the arguments between such first-contact health professionals as optometrists and some of the medical profession. The tensions have been exacerbated in recent years by the growth of managed care arrangements that make primary care clinicians the path by which patients gain access to specialized medical services.
This shift in power and responsibility for determining the use of medical resources has significant economic implications for most of the health professions. These implications are magnified when many medical specialties are likely to be in surplus and when the hospital is diminishing as a locus of employment for nurses. (See Chapter 6 for an overview of supply and demand issues for physicians, nurse practitioners, and physician assistants.)
For all of these reasons the debate over the label of primary care clinician is intense. At a public hearing organized by this committee and in written statements received by the committee, a wide range of professional groups have expressed the view that their professions provide primary care. The professional and economic stakes are substantial if the function of primary care receives more emphasis at the same time that efforts to hold down expenditures for health care continue. Subsequent discussion of these issues at a workshop on professional roles convened by the study committee provided opportunity to explore further the many dimensions of the roles issue.
From the beginning of its deliberations, the committee has believed that primary care should not be defined solely or primarily by who does it. The definition is a functional definition that provides a basis for determining whether
a particular professional is a primary care clinician. That function is the overriding guide to this discussion of roles. Starting with the functional definition makes more problematic an a priori determination of who is, or is not, a primary care clinician, as was noted in Chapter 2.
For clinicians whose training is explicitly targeted on primary care, their role as primary care clinician is clear. There is little argument that among physicians (both allopathic and osteopathic), family physicians, general internists, and general pediatricians are primary care clinicians. Many nurse practitioners and physician assistants are also trained for primary care and participate in the primary care function. The issues for these professions are mainly how they can work together in the interest of patients. However, the involvement of other physician specialists in primary care is a growing issue, especially for obstetricians and gynecologists and the medical subspecialties. Also at issue is the role of health professionals who independently provide basic services for some health care needs on a first-contact basis—for example, dentists, optometrists, pharmacists, and some physical therapists. Such services form a significant part of health care, but the relationship to primary care is inadequately defined. Each of these issues is addressed in turn below.
The Primary Care Team
In discussing the definition in Chapter 2, the committee indicated that primary care consists of a set of tasks that can often be best carried out by a team rather than by an individual clinician. The team may be organized to achieve a number of purposes: to increase access, to subdivide tasks so that several different kinds of expertise can be brought to bear on the patient's needs through collaborative activity, and to permit the delegation of some tasks by broadening the range of professionals involved in primary care.
Some of these purposes are quite straight forward—accessibility at any hour or on any day is more easily provided by a team than by a single clinician. The achievement of other benefits of collaboration by the team is more complex. Realizing the benefits to patients of truly collaborative practice that draws on the broader expertise of a team of professionals—for instance, health supervision of the child, treatment of recurring infections, palliative care of seriously ill patients, patient education related to a chronic condition such as diabetes, or coordination of community services—will likely require modification of attitudes and beliefs and changes in training and organization. Maintaining a sense of personal relationship between patients and at least some members of the team calls for an organizational emphasis that is sensitive to patients' preferences and needs.
The specific composition of the team will vary with the care setting and the specific needs of the patients being served. The needs of children for routine regular health maintenance will require different knowledge and skills, such as those provided by a pediatric nurse practitioner, than the care of an elderly person
with multiple chronic problems and functional limitations that raise problems for their living environment, which may require the continuing involvement of a social worker. Scheffler (Appendix E) offers a conceptual framework for the variety of team functions in primary care.
The team concept used by the committee means a relatively small group that interacts on a regular basis around the primary care of a defined group of patients. The term ''health care team" is sometimes used in a looser way to mean all those who are involved in patient care. In our usage, referrals to specialists or other independent professionals, or the independent involvement of other professionals on a recurring, first-contact basis (such as dentistry), do not make these other professionals part of the primary care team, although they are providing essential health services.
In the committee's view, and in the many examples of teams observed on the site visits, the team nearly always will include a primary care physician. This often is the person on the team who deals with more complex decisions and usually plays some role in coordinating the efforts of the team. The health care organizations visited provided care in a wide variety of circumstances: the open spaces of the rural West, multicultural urban poverty in south-central Los Angeles, middle-class areas of the Twin Cities. For nearly all, the experience is that most patients want to have access to a physician as an important part of their primary care. Nevertheless, a variety of team arrangements can meet the needs of patients and still have another team member carrying out principal contact with patients for important aspects of their care.
To be efficient, larger and more complex teams that interact face-to-face on a regular basis require a substantial panel of patients. When providing services to an isolated rural population, such teams can be geographically dispersed if they take advantage of modern communication technologies as a substitute for face-to-face contact.
All in all, the committee believes that teams offer the best means to bring to bear the wide range of talents and knowledge needed for primary care. Teams provide a way to achieve efficiencies in the delivery of primary care and to improve access to services on a timely basis while maintaining appropriate personal knowledge of the patient.
Recommendation 5.5 Practice by Interdisciplinary Teams
The committee believes that the quality, efficiency, and responsiveness of primary care are enhanced by the use of interdisciplinary teams and recommends the adoption of the team concept of primary care, wherever feasible.
Role of Specialists in Primary Care
Physician specialists have long had a role in the delivery of primary care. A major study based on a national sample of physicians during 1973–1976, using a definition of primary care that is more limited than the one adopted by this committee, indicated that approximately 20 percent of Americans received continuing care for the majority of their health problems from specialists (Aiken et al., 1979). Those data are now old, but anecdotal evidence and statements presented to this committee by several specialty groups indicate that some specialists still provide substantial portions of their patients' care, although this care may or may not meet the committee's definition.
Two conflicting trends influence specialist delivery of primary care. First is the continued increase in the number of specialists in the past two decades, both in absolute numbers and relative to the number of primary care physicians. The second trend is the growth of managed care plans, which emphasize primary care and control the use of care provided by specialists. The combined effect of these trends is what many analyses have concluded to be a substantial surplus of physicians in many of the specialties, a surplus that is likely to increase in coming years (COGME, 1994; Weiner, 1994; Pew Health Professions Commission, 1995; IOM, 1996). The lack of opportunities to practice their specialty may provide strong incentives for some specialists to increase their involvement in primary care. This involvement can take several forms.
Mixed practices. One form is a mixed practice in which the physician carries out a specialty referral role for some patients and acts as primary care physician for others. The American Society of Internal Medicine (ASIM) has argued for the acceptability of mixed practices. A recent ASIM survey (ASIM, 1995) found that 55 percent of 53 HMOs that responded allow a physician to designate themselves as both a primary care physician carrying out the "gatekeeper" role and a consulting specialist within the same plan; 43 percent make the physician choose one role or another. One plan allowed the specialist to choose only the consulting role. Most (83 percent) permitted the specialist to act as the primary care physician for any patient and not just for those patients with diseases that fall within the physician's specialty (ASIM, 1995). These results suggest that self-designation as primary care physician could become a popular option for the internal medicine subspecialties.
Mixed practice is also reported to be common in the specialty of obstetrics-gynecology (OB/GYN). Women frequently seek general medical care from their OB-GYNs (Horton et al., 1994). One survey reported that about 20 percent of women would choose to receive their primary care from an OB-GYN if asked to make a choice of primary care physician (ACOG, 1993). Nearly all physicians in this specialty also do surgery and provide other specialized care in addition to obstetrics and gynecological care.
Principal physicians. A second pattern of involvement of specialists in primary care is as principal physician. In this role, specialists care for patients whose principal health problems fall within their specialty, e.g., cancer, pulmonary conditions, and advanced heart disease. While providing care for this problem, which often dominates the patients' involvement with health care, specialists can also provide general care for most of the rest of their patients' health care needs; they can make referrals to other specialists as necessary and in some instances refer the patient to a primary care clinician. For these patients, specialists are acting as the principal physician for both specialized and general care, though the extent of preventive care and screening is unknown.
The committee heard many examples of this role for specialists, but it was not able to quantify the extent of this pattern of practice. The data from the 1970s survey mentioned above (Aiken et al., 1979) suggest that, in those years, this pattern may have been common for selected patients. Today most managed care plans control access to specialty care through a designated primary care clinician, so this pattern of specialist practice is likely to be less common for patients in managed care arrangements.
For the committee, the issue is whether these patterns of specialist provision of primary care—mixed practice and the principal physician role—provide primary care as the committee has defined it. The training, experience, and practice patterns of many specialists are not likely to prepare them to engage in the full range of primary care. Two special cases, however, deserve comment.
The first is the case of the many physicians who received training as generalists before going on to specialty training and practice. Physicians in the specialty areas of internal medicine and pediatrics have typically received three years of training in general internal medicine or general pediatrics. Many of their colleagues in these same training programs have gone on to primary care practice. Whether, however, the first three years of training as provided in past years, with its heavy emphasis on hospital-based care of very ill patients, is appropriate for primary care is addressed in Chapter 7. Many of these specialists, especially in the internal medicine subspecialties, have continued to provide services of a primary care nature along with care in their role as specialist consultants.
The second special case is the specialty of OB/GYN. These physicians provide a considerable amount of general care for women, particularly in the childbearing years. Based on this pattern of practice, the American College of Obstetricians and Gynecologists (ACOG) has advocated that the specialty should be recognized as a primary care specialty and has sought to formalize this recognition in state and federal legislation (ACOG testimony to the IOM Committee on the Future of Primary Care, 1994). Reacting to the concern that obstetricians and gynecologists may not be appropriately trained for primary care, ACOG had also sought to strengthen the training for primary care in the OB-GYN residency.
In both of these special cases, the evidence suggests that considerable numbers of patients receive whatever primary care they receive from these specialists
and that many members of the specialties perform primary care in the mixed model described above. In both cases, some basis exists for the claim that residency training will prepare these physicians for the primary care role, at least for more recent and future trainees.
No basis exists at the current time either for the complete exclusion of specialists from the provision of primary care or for their automatic inclusion. Both roles exist now, and the salient issues are: (a) Are the functions of primary care, as defined, being fulfilled? (b) Does the specialist physician have the appropriate knowledge and training to carry out the functions well? The committee believes that primary care requires appropriate training just as specialty care does. Chapter 7 addresses the question of retraining. As indicated in Chapter 4 and in the discussion of education for primary care in Chapter 7, primary care has its own characteristics, knowledge base, and decision criteria. Primary care is more than a junior level of specialty care or a triage function for specialty care. To enable the primary care clinician to carry out the primary care function at a level of excellence that best meets patients' needs requires appropriate training, experience, and support systems. The function of primary care is complex and demanding (see Chapter 4); it involves many activities that extend beyond a reductionist focus on the diagnosis and treatment of a specific disease. The committee questions whether this function can be performed adequately by someone whose orientation and time is substantially committed to the different challenges of the specialist role and whose focus is on a particular disease process or organ system and whether someone can keep abreast of the burgeoning literature in both fields.
The committee does not believe that such questions should be answered by legislative fiat. Nor does it believe that primary care is a residual function to which specialists can return solely through self-designation on the basis of their earlier general training. The current trend is toward a more distinct identification of a group of physicians whose specific role is that of primary care physician, and that trend should be encouraged.
Roles of Other First-Contact Health Professionals
Several health professions provide first-contact care for basic health services that are needed by most or all of the population. Principal among these professions are dentistry, optometry, and pharmacy. Each of these professions has a unique history in the American context, and the evolution of each has been largely independent of the development of the medical profession and the development of the other health professions with closer relationships to medicine, such as nursing and many of the allied health professions. The resulting patterns of basic and continuing services being provided by independent service settings are more a product of history than of logic, but there is no mandate from either health professionals or patients to change these historic patterns.
These services can be considered part of primary care. Representatives from dentistry (American Association of Dental Schools) and optometry (American Optometric Association) have advocated (Appendix B) that these professions be included in any classification of primary care clinicians or that these professions be considered part of the primary care team. The roles of these professions in primary care were also discussed at the committee's workshop on professional roles in primary care.
The committee definition clearly describes functions that extend far beyond the services provided by dentists and optometrists. The independence that characterizes their typical practice does not seem to be consistent with their inclusion in the primary care team as it has been described earlier in this chapter. Yet it would seem logical that good health care for the whole person, certainly a focus of primary care, should include good oral health and vision care.
The historical pattern of separation of these professions and the services that they provide is reflected in such practical matters as the design of health benefit packages, the exclusion of these services from the "gatekeeper" requirements of most managed care plans, and the fact that access to these services is typically through direct contact by the patient. The practical issue would seem to be how to strengthen the relationship to the rest of primary care rather than incorporating these professions into primary care which would involve changes in the historic patterns of professional independence that would probably be resisted by many in these professions. Some have argued that the committee should make explicit that it is not dealing with these professions, but only with primary medical care. This approach, however, would seem to neglect the opportunity to build desirable relationships among the professions that could lead to a more integrated approach in the future.
The committee, therefore, would encourage strengthening the two-way relationship between the primary care clinician or team and the provision of dental care, routine eye care, and pharmacy services. For example, the primary care clinician could determine whether patients are receiving preventive and restorative dental services and encourage them to obtain such routine care. Some screening for oral health problems can be carried out in the primary care setting and lead to appropriate referral for dental services (specific screening instructions appear in Greene and Greene, 1995). Conversely, the dentist can screen for medical problems to be brought to the attention of the primary care clinician. Screening for oral cancer is already common; screening for diabetes and hypertension would take advantage of the sometimes routine contact between the dentist and patient. Reference to a common, computer-based patient record would facilitate such interaction.
Similar interaction could take place between primary care clinicians or teams and those providing routine vision care, whether optometrists or ophthalmologists. Pharmacists already serve as a frequent source of medical advice for alleviation of common problems, and they often provide patient information
regarding the use of pharmaceuticals. Computer systems in the pharmacy can be used to identify possible drug interactions and dosage problems. Again, this interaction could also be strengthened by access to a common computer-based medical record.
In some sites visited by the committee, the interaction of these professions and the primary care team was encouraged and facilitated by a common site of services, particularly in programs that serve the urban and rural poor, such as services provided in community health centers and Indian Health Service clinics. These settings could provide the basis for more systematic study of the benefits of closer integration between these services and primary care, in terms of patient convenience, access, and health outcomes. The extension of a more integrated model of primary care to include a closer relationship would be a logical development in integrated delivery systems, especially because the desirable infrastructure, such as clinical data systems, is already being put in place.
Another set of health professionals, including physical therapists and podiatrists, may also provide first-contact services. Access to these services is often through regular referral mechanisms, and managed care plans frequently require referral by the primary care clinician. Thus, considering these services as referral services seems to be more sensible. In some primary care practices, however, these professions might be a direct part of the primary care team.
In this study we have not dealt with the role of a large group of other health care personnel who provide services on a first-contact basis that might overlap significantly with services provided in a primary care setting. These include chiropractors, traditional folk healers, and other providers outside the dominant medical model. Some of these services are already included in the primary care offered in some settings or are considered a covered health insurance benefit; others are not. To the extent that such services can be established as effective, the committee would welcome a path of greater convergence between these services and primary care. The National Institutes of Health is implementing a program of research in alternative medicine, which may clarify this issue in future years.
Ensuring Primary Care for Underserved Populations
Earlier in this chapter, the special problems in providing primary care services to underserved populations—particularly rural populations and the urban poor—were underscored. These populations often have special health problems related to low income and social circumstances: for example, trauma related to family and community violence, substance abuse, disease such as diabetes aggravated by poor diet, a higher incidence of infectious diseases such as acquired immunodeficiency syndrome (AIDS) and tuberculosis (TB), and health problems caused by occupational hazards such as injuries and exposure to toxic chemicals
among agricultural workers. The primary care clinicians in these areas also face special barriers to making services available. These include lack of health insurance coverage; low reimbursement under many state Medicaid programs; geographic isolation of the population to be served; lack of transportation; problems of recruiting and retaining health care personnel, especially physicians; language and cultural differences that often complicate communication; and special challenges in coordinating primary care with other health and social services.
For many years, federal and state programs and private foundations have directed specific resources toward providing primary care for these populations. These efforts take the form of community health center grants, rural health clinics, the National Health Service Corps, the Maternal and Child Health Block Grant program, the Indian Health Service, direct state and local government provision of services through public health clinics, and many others. These targeted programs have supplemented the large subsidy for medical care for the underserved provided through the federal-state Medicaid program. As valuable as these programs have been in helping to provide these populations with primary care, many gaps remain and the future is uncertain.
Primary care for these special populations is embedded in the social and economic circumstances of the communities and individuals, and primary care, by itself, is not likely to alter these fundamental circumstances. Problems of increasing disparities in incomes, social disintegration, and continued declines in rural populations and the infrastructure of rural communities will make the delivery of primary care more difficult. At the same time, health insurance for the working poor seems likely to continue its decline in the absence of comprehensive health care reform (EBRI, 1995; Short and Banthin, 1995), and the programs and policies that have helped make primary care available for many of these populations are now facing budget cuts and policy changes of historic proportions.
Managed Care and Underserved Populations
Managed care arrangements, integrated health delivery systems, and capitated financing of primary care services—which are likely to be the principal arrangements for organizing and financing primary care in future years—have been slow to include many underserved populations, especially those in rural areas. On site visits to rural areas, for example, the committee saw little evidence that managed care had penetrated these particular rural markets. This situation may change significantly in the near future, however, as states move to implement managed care arrangements for their Medicaid programs. According to a recent study (Lewin-VHI, 1995), as of June 1994 all but eight states had implemented one or more models of managed care for their Medicaid programs and total enrollment in these programs had doubled since June 1993.
Furthermore, the fastest growing form of Medicaid managed care has been
full-risk capitation programs. While the growth is rapid, these full-risk capitation programs have been implemented in only a few states, usually states that had substantial market penetration by managed care plans for the population as a whole. Overall, however, many other states are implementing or considering moves to more aggressive managed care arrangements for their Medicaid programs. All of the Medicaid managed care models, including partial capitation and primary care case management, emphasize primary care relative to specialty care.
The trend toward managed care in the Medicaid program does not solve the problems of health insurance coverage for those who are ineligible for Medicaid or who lack private health insurance coverage. State contracts with managed care plans do offer some opportunity to assure the provision of primary care for underserved populations that qualify for Medicaid. This objective could be served both by including the committee's definition of primary care in the criteria by which states select managed care plans and by setting up performance monitoring that includes measures of access, quality, and patient satisfaction that relate to the elements of the definition. Among the factors that ought to be included in managed care contracts are integration of services that makes the process of care more seamless for patients; accessibility; development of sustained partnerships between clinicians and patients; and efforts to relate the health care needs of patients to their families and community. This would help ensure that criteria for awards and subsequent renewals of state contracts would extend well beyond the lowest cost package. Many existing community and rural centers that have sought to provide comprehensive primary care that is responsive to the health care needs of their populations might be considered an asset to managed care plans under these criteria.
A higher proportion of the Medicare population may also be moving into managed care, based on trends in markets that have a high proportion of the under-65 population in managed care. This trend might be advanced further by federal policy changes. As Medicare patients move to managed care settings, the Health Care Financing Administration (HCFA) might (a) include the implementation of the IOM's definition of primary care in its criteria for awarding those contracts; and (b) require improved access for underserved Medicare populations. In sum, conditions for success in acquiring Medicaid and Medicare managed care contracts could constitute a powerful incentive to shape the nature of and access to care provided to underserved populations.
Recommendation 5.6 The Underserved and Those with Special Needs
The committee recommends that public or private programs designed to cover underserved populations and those with special needs include the provision of primary care services as defined in this report. It further recommends that the agencies or organizations funding these programs
carefully monitor them to ensure that such primary care is provided.
Other Approaches for Underserved Populations
Regardless of the method of paying for primary care services for underserved populations, the issues remain about whether meeting their primary care needs requires special services and additional expenses. An additional question is how the costs for patients without any form of health insurance will be met. For hospital care, the disproportionate share provision in the federal financing programs is based on the presumption that a hospital's costs are higher if its patients include a large proportion of Medicare and Medicaid beneficiaries and other low-income patients. For primary care programs supported in part by federal grants, the grant covers costs that are above and beyond reimbursement received through Medicaid, Medicare, private health insurance, and self-pay. One fear of these community clinics—urban or rural, public or private—expressed to the committee during its site visits was that as managed care plans spread to these communities, they would fail to recognize the higher costs associated with meeting the health needs of these populations or that such plans would seek to serve only those patients for whom adequate payment was available. This could well jeopardize current arrangements and leave patients who have no insurance even more vulnerable to a loss of local clinics.
The special problems of serving isolated populations and meeting their primary care needs have been well documented (OTA, 1990). Rural health clinics typically combine the primary care function with some functions of emergency medicine, including ambulance service, and transportation and outreach for patients living in remote locations are other important functions. Support for the professional staff, such as locum tenens programs, is sometimes used to retain staff in isolated areas. Low volume of some services may raise unit costs. The economy of the area being served may be in decline, which raises the incidence of problems such as depression and family violence. Coordination with other needed services, such as social services and home care, may be more difficult because of distance, so the primary care unit may need to be more self-sufficient.
For reasons such as these, the committee believes that some form of subsidy and infrastructure support, in addition to third-party reimbursements, will be needed to make these programs viable. Moreover, if the managed care arrangements extend to these communities, then those groups will need to take these extra costs into consideration.
In some circumstances, integrated health care systems providing managed care may be able to provide the capital, support systems, and personnel to help make rural programs viable. However, specific subsidy of the health system may be required to provide compensation for the extra costs, especially if the system is competing in other markets with managed care plans that do not serve isolated
rural populations. Some form of internal cross-subsidy may be required; such subsidies were used in public utility regulation where service to rural communities was required for a license to serve affluent, urban populations.
Serving the urban poor with appropriate primary care services also requires attention to the special problems associated with these populations. Primary care services for these populations have often been concentrated in public clinics, community health centers, or the emergency rooms of public hospitals. The extent to which these populations can be included in the mainstream of primary care services provided through integrated health care systems is complicated by the presence of substantial numbers of those with no source of payment, including the illegal immigrant. Administrators of some integrated systems argue that they can provide primary care for populations with Medicaid eligibility and cover the costs of any needed additional services by their greater efficiency. This is also the general presumption of Medicaid managed care plans. Careful monitoring of the results of managed care plans serving these populations, as recommended, will yield useful information about the extent to which primary care, as defined, can be extended to these populations without compromising care through gaps in needed services, subtle barriers to access, or avoidance of high-risk patients.
Coordinating Primary Care with Other Services
Coordination of services by the primary care clinician is necessary to meet the full range of health needs of the patient and to integrate those services so that the care process can be coherent from the perspective of the patient. Many primary care clinicians and integrated health systems devote considerable attention to coordinating services for the patient within the constraints of current organizational and financial arrangements.
Many of those organizational and financial arrangements, however, are not conducive to coordination of an increasingly complex array of services. For example, the coordination function is often not adequately recognized in the payment for primary care services. Separate administrative structures, funding streams, and organizational and professional cultures may impede coordination. Current knowledge of the wide array of services available for the patient may be difficult for the primary care clinician or team to maintain without some organizational assistance. Specific attention to the means of effective coordination across the array of services needs to be part of the explicit mission of the primary care clinicians and the organizational arrangements within which the primary care function is carried out. The complexities of the coordinating function are another argument for the development of integrated health care systems that can provide appropriate resources for effective coordination.
A familiar aspect of coordination involves the role of the primary care clinician in integrating the diagnostic and treatment services provided by medical
specialists. The primary care clinician has an active role in coordinating specialty services on the patient's behalf. This role is much more than serving as a triage point or gateway for those services. As medicine becomes more complex, an active coordinating role for the primary care clinician is essential to assure the effective use of specialized resources. The decision to refer should be accompanied by an exchange of information. In this exchange, ideally, the primary care clinician's knowledge of the patient's history, other health problems, and family and community circumstances is provided to the specialist; the specialist reciprocates with information that is relevant to the comprehensive care of the patient over time, including prevention of disease, maintenance of function, and appropriate treatment of the patient's other health problems.
How to achieve this active interaction should be part of the training of both the primary care clinician and the specialist. Also essential is the organization and financing of services to support the coordination of services. Coordination is also necessary within the primary care team so that the functioning of the team appears to be seamless and coherent to the patient. If the trend is toward larger, integrated health care systems, special attention will need to be given to ensuring that the team approach enhances rather than displaces personal attention to the needs of the patient. The idea of shared responsibility can mean that no one individual feels fully accountable.
Another aspect of coordination—coordination of primary care with other service systems—is the focus of the rest of this section. Primary care teams need to deal with many service systems, including school systems and workplaces. The committee illustrates the importance and nature of this coordinating function by describing desirable interactions with three other health activities where close relationships with primary care are often of great importance to the patient. These are public health, mental health, and long-term care.
Public Health and Primary Care
Basic linkages. In Chapter 2, the committee acknowledges that population-based public health activities aimed at health promotion and disease prevention have a larger impact on improving health status of populations than personal health services. The committee's definition has focused on primary care as a personal health service and has not incorporated the population-based activities that are the heart of public health; it differs from the World Health Organization (WHO) definition of primary care, which includes the population-based activities of public health under the rubric of primary care. The committee recognizes that effective population-based public health services are essential to the health of the public and acknowledges that rising expenditures for personal health services have often competed in public and private budgets with adequate funding for population-based public health activities (IOM, 1988). The committee holds, however, that the population-based functions of public health and the primary
care services delivered to individuals are complementary functions, and strengthening the relationship should be the focus of action in both arenas. Incorporating public health in its totality into primary care would obscure rather than enhance the importance of public health, at least in the American context.
Furthermore, the agenda for primary care is already very challenging without adding responsibility for the full range of population-based public health activities to the primary care function; in particular, the committee would not accept the idea that primary care should include the enforcement responsibilities that are an essential part of the public health function and are legally based on the police power of the state. Rather than competing for attention and funds, the committee believes that both primary care and public health would gain if these functions are viewed as natural allies. The issue then becomes: How can the relationship between primary care and public health be strengthened so that each function will enhance the other?
Public health and managed care. An important dimension of this issue is how the growth of managed care and integrated health care systems with enrolled populations should affect the interaction between primary care and public health. The California Medicaid program (MediCal) illustrates this redefining of the roles and relationships of primary care and public health. As MediCal moves to a managed care model, health plans that are chosen to enroll its beneficiaries will be required to work out agreements with county health departments that will specify responsibilities for various aspects of public health programs (James Haughton, Los Angeles County Health Department, personal communication, 1995). For some functions, the county may contract with the plans to carry out activities such as maternal and child health services and some forms of screening. For other services for which the health department has special expertise and experience, such as TB treatment and control, the health department could be identified as the place to which cases of TB are referred.
For the rest of the population enrolled in managed care plans, there are numerous examples of how these plans can play an important role in health promotion and disease prevention functions that involve services to and interaction with individuals. Plans with capitated funding typically include a wide range of preventive services in their benefit package, in contrast to the exclusion of many or most preventive services in traditional health indemnity plans. These services may include immunizations, periodic screening for disease, health education and behavior change programs, and even discounts at health clubs. The combination of an enrolled population and appropriate data systems makes possible the notification of the patient when a preventive service is due.
Granting that this emphasis on health promotion and disease prevention may be used as a marketing strategy to assure enrollment of healthy persons with a strong interest in maintaining their health, the opportunities to use the plans as instruments of a public health agenda are significant. The primary care team and
clinicians can, and often do, fulfill important roles as health educators for individuals and as advocates and activists in community health education programs. There are many examples of this natural alliance between primary care and the public health functions, such as the work of the pediatricians in developing a variety of community-based programs that enhance child safety, ranging from encouraging the use of child car seats to child-proof safety caps on medications.
Enhancing the relationship of primary care and public health. Because of the importance of this relationship to the health of the populations being served by both primary care and public health entities, the committee commissioned a paper by Welton and his colleagues on enhancing the relationship (Appendix F). The authors note the many barriers to effective coordination of the spheres of public health, with its population focus, and primary care, with its focus on clinical preventive services and education and behavior change for the individual patient. Commenting that ''we must view both public health and primary care as two interacting and mutually supportive components of an increasingly complex integrated system having the single common goal of improving the health of a community and its diverse population," they outline a systems approach for bringing about this integration. This approach involves developing a means by which to identify the functions of the public health agencies for population-based health activities and those of integrated health systems for personal health services (including preventive and health promotion services for individuals). They also describe the role of the public agency in monitoring the health of the population, including inputs from primary care services, through publicly accountable community health information networks.
Welton and his coauthors also identify the many barriers to accomplishing this degree of integration, including the conceptual, educational, and experiential gaps between public health and primary care professionals. The roles and methods of primary care and public health have often been defined independently of each other. Public health agencies are often organized in a compartmentalized way that makes it more difficult to define the functional relationships to primary care. On the health care side, HMOs and other health care organizations differ substantially in the degree of interest in the long-term health of the population being served and their commitment to provide the professional time and infrastructure necessary to coordinate the primary care and public health functions.
The many barriers and obstacles to relating primary care to the health needs of the community identified by the IOM report on community-oriented primary care (COPC) (IOM, 1984) continue at this time (see Chapter 2), and the COPC model has not expanded its practice base in this country to any great extent, despite some excellent models (see Appendix F).
Welton et al. lay out an ambitious plan for bringing about fuller integration of public health and primary care that would create, on the primary care side, fully developed COPC practices in the context of organized health care systems.
The committee endorses efforts in the organization and financing of primary care services that would move in this direction. The path toward a fully integrated approach that is focused on improving the health of populations will be long and arduous, and the particulars and pace of development will vary from place to place. Moving primary care services toward a more population-based approach will also require changes in the education of primary care clinicians (see Chapter 7) that can build on many activities already under way, such as those supported by private foundations under the "Health of the Public" projects of The Robert Wood Johnson Foundation and The Pew Charitable Trusts and the community-oriented health education programs supported by the W.K. Kellogg Foundation.
The committee encourages the many local efforts that are trying to better integrate primary care and public health activities supported by governments at all levels and by private foundations, many of which are focused on underserved populations. The committee would also like to encourage managed health plans to move toward the "natural alliance" between primary care and public health to which we have referred, based on their mutual interest in improving and maintaining the health of the populations they served. A logical starting place is the ongoing effort to encourage (or require) beneficiaries of federal and state programs (Medicare and Medicaid in whatever form they take in the future) to enroll in capitated managed care plans.
Recommendation 5.7 Primary Care and Public Health
The committee recommends that health care plans and public health agencies develop specific written agreements regarding their respective roles and relationships in (a) maintaining and improving the health of the communities they serve and (b) ensuring coordination of preventive services and health promotion activities related to primary care.
Agreements with public program beneficiaries could serve as a model for agreements covering managed care enrollees who are privately insured. Because most managed care plans will enroll both public and private beneficiaries, this extension of the agreements for coordination should be a logical development.
Stimulated by these agreements the committee encourages the development of community- (population-) based information systems that will serve the joint purposes of public health agencies and the managed care plans by providing better data on the health problems of communities. Such agreements should also address joint development of health promotion strategies; these would combine the individualized approaches of the health plan with population-based approaches of the health department and other voluntary health agencies, such as those focused on specific diseases. Although the committee has advocated specific agreements in order to move beyond a rhetorical commitment to a common agenda, such agreements should not replace the dynamics of ongoing and voluntary
cooperation. Agreements should be modified frequently to reflect new opportunities and should encourage ongoing dialogue among primary care managers, public health officials, and the communities they jointly serve.
The committee recognizes that managed care plans do not now cover the entire population; as long as health insurance is not universally available to the population, this will be true. This means that public health agencies will continue to be responsible for reaching the entire population with disease prevention and health promotion services. In the short term, the public health sector will also have to continue in many places as the primary care provider of last resort.
Mental Health and Primary Care
Another critical interface that requires attention and coordination is the relationship between primary care and mental health. Part of this relationship is encompassed in the referral of patients with mental health problems to specialized mental health providers; at times this follows the usual pattern of referral to individual specialists and at other times involves referral to separate mental health services delivery systems (sometimes called behavioral health care plans) with their own organization and financing. Describing the evolution of these separate delivery systems in both the public and the private sectors is beyond the scope of this report, but the separateness continues in new forms, the latest of which is the growing use of "carve-outs" for mental health and substance abuse services in private health benefit plans.
The existence of this separate, parallel mental health services system implies that the function of primary care regarding mental illness is initial diagnosis and referral for treatment. Yet the reality is that the primary care clinician not only identifies but treats a large portion of mental disorders. Furthermore, as described by deGruy (Appendix D), many aspects of primary care and mental health are indivisible. To quote deGruy:
A major portion of mental health care is rendered in the primary care setting and always will be, sometimes despite strong disincentives; … a sensible vision of primary health care must have mental health woven into its fabric; … the primary setting is well suited to the provision of most mental health services; … despite suboptimal recognition and management of mental disorders and attention to mental health, the structure and operation of primary care can be modified so as to greatly augment the provision of these services; and … current efforts under way in the U.S. to reform the health care system offer an opportunity to find the most effective of these modifications and to discover fruitful collaborative structures both within the primary care setting and between primary care clinicians and mental health professionals.
The issue of referrals is further complicated by the way that mental health services are financed. Many health plans put limits on the number of visits for
mental health services, or they set higher co-payments for such services (or they do both). If a managed care plan capitates its primary care clinicians, it may create an inappropriate incentive for patients to be referred to the "carve out" plan. Referral to mental health providers may compromise the ability of primary care clinicians to maintain continuity of care and to focus on patients' related health problems.
A further issue is the underrecognition and undertreatment of mental disorders in the primary care setting. Most studies of these problems have focused on the diagnosis and treatment of depression, and they indicate that from one-half to two-thirds of patients meeting the criteria for mental disorders are not diagnosed in the primary care setting. For depression, evidence suggests that treatments known to be effective are underutilized by primary care clinicians.
An additional complicating issue is the existence of a "primary" mental health system parallel to the primary care system. Entry to this parallel system—essentially comprising community mental health centers and individual mental health professionals—is at the initiative of patients who identify their problem as primarily mental. For patients who have serious mental disorders or who have mental health problems but no other significant health problems, the diagnosis and treatment by the specialized mental health provider is appropriate. However, for patients with significant overlapping health conditions, this self-referral may raise problems for the adequacy of the treatment of other health conditions.
Given these complexities, the problems of effective coordination of services are not simple to solve. The nature of the relationships between primary care and mental health services, as indicated by deGruy's analysis of the literature and current directions in the health care system, indicate some clear directions for more effective coordination.
First, the important role of primary care in the diagnosis and treatment of mental disorders needs to be recognized and strengthened through appropriate training and organization and financing of primary care services. This role includes dealing with the extensive interrelationships of mental and physical illness.
Second, models of assistance to primary care clinicians by mental health professionals need to be further developed, implemented, and evaluated.
Third, financial and organizational disincentives for a strengthened primary care role need to be reduced so that the primary care clinician can and will provide needed and effective services for those mental health problems that will inevitably present in the primary care setting and that are often imbedded in other health problems. Arrangements such as carve-outs for mental health services and special payment and service limits triggered by a diagnosis of mental problems need to be carefully examined, so that disincentives for appropriate roles of either primary care in mental health or specialized referral services are reduced.
Fourth, collaborative service models that integrate rather than separate specialized mental health services and primary care services need to be encouraged.
This will enable patients in either setting to benefit from coordinated treatment plans dealing with the full range of their health problems, and it will improve diagnosis and treatment of mental health problems, including those requiring specialized services. In many care settings, this means increased consultation and involvement of primary care clinicians and mental health professionals with each other's service domains. Integrated health care systems would seem to be the logical home for such collaborative approaches. Finally, the primary care research program discussed in Chapter 8 should include a significant focus on the primary care role in mental health, including study and evaluation of care models and natural experiments.
Recommendation 5.8 Primary Care and Mental Health Services
The committee recommends the reduction of financial and organizational disincentives for the expanded role of primary care in the provision of mental health services. It further recommends the development and evaluation of collaborative care models that integrate primary care and mental health services more effectively. These models should involve both primary care clinicians and mental health professionals.
Long-Term Care and Primary Care
The importance of long-term care is growing as the number of the elderly, especially the very old, increases. These services raise difficult issues for coordinating care. Long-term care extends well beyond the provision of personal health services to encompass issues of housing, nutrition, assistance in the activities of daily living, social services, transportation, and the roles of voluntary caregivers. Looming large over the breadth and content of these services is the lack of a coherent set of social policies concerning funding for long-term care services.
The roles of primary care in the provision of long-term care services are intertwined with these issues. Nearly all persons who receive long-term care services, either formally organized or provided by family and friends, are high users of medical services, including primary care. Because the elderly, or the seriously disabled of any age, typically have multiple medical problems, including a high incidence of mental health problems, the problems of coordination by primary care clinicians or teams are compounded. Furthermore, many of these patients are in declining health, and this calls for a different mind-set than does the provision of acute services with the intent of providing cure or significant alleviation of symptoms. Markers of effective performance by clinicians in terms of desired patient outcomes are different at least in degree if not in kind for this population. Maintenance of function and emotional support, rather than treatment of a physiological condition, become even more important objectives for the primary care clinician.
The aspect of the definition that speaks to the context of family and community becomes especially important in these circumstances, and the need for coordination of services is great. Effective treatment of medical problems requires that primary care clinicians be aware of patients' living circumstances, personal capabilities, and other persons involved in their care. Coordination of treatment plans with others involved in provision of long-term care services is often essential. Simply involving primary clinicians often becomes a problem because of the inability of patients to go to a clinician's office; the resulting special demands on clinicians' time, for which there may be little financial or emotional compensation, pose yet further obstacles. Even today, there are complaints that primary care clinicians do not visit the home- or institution-bound patient and do not take an active role in their care or care plans (IOM, 1986b).
Many aspects of improving long-term care have not been adequately addressed by society as a whole. Among these issues are the preoccupation with holding down acute care costs for the elderly served by Medicare; the reluctance to extend entitlement any further; the possibility of new and strong incentives for states to reduce their exposure to long-term care costs through caps on Medicaid expenditure; and the steady increase in the numbers of the very old. Taken together, these factors almost guarantee that coordination of long-term care and primary care will remain beset with problems and frustrations for both clinicians and patients.
Some avenues for improvement and some care models show promise of better integration of services. Demonstration programs such as the Social Health Maintenance Organization (S/HMO) programs, the Program for the All-Inclusive Care of the Elderly (PACE program), and others pool Medicare, Medicaid, and private funding sources to provide a coordinated approach to care that includes medical services (IOM, 1995). Coordination seems more likely in integrated health care systems that are built on a base of primary care and that have an extended primary care team, because these approaches can include nurse practitioners and social workers who are well informed about the care of the dependent elderly and about community resources that can help. These members of the primary care team can also maintain personal contact with patients in the home or long-term care setting and monitor their medical condition and treatment plans. The primary care team members in turn should participate in the joint planning with those providing long-term care services to develop plans that include attention to the patients' needs for primary care and for the coordination of other medical services. A primary care team member can serve as case manager in coordinating an array of services for the individual with long-term care needs or can work with a case manager from outside the team. Finally, the primary care clinicians and team members can help provide emotional support and counseling for patients whose medical and living circumstances interact to accentuate fear and anxiety.
Recommendation 5.9 Primary Care and Long-Term Care
To improve the continuity and effectiveness of services for those requiring long-term care, the committee recommends that third-party payers (including Medicare and Medicaid), health care organizations, and health professionals promote the integration of primary care and long-term care by coordinating or pooling financing and removing regulatory or other barriers to such coordination.
Performance Monitoring for Primary Care
In an era when resource constraints for health care will be a continuing reality, monitoring the performance of the health care system in terms of quality and patient outcomes will become increasingly important. Costs are quantifiable and a source of intense concern to large payers for health care, so one can safely assume that comparative cost data will become more widely available. The debate over future expenditures for the Medicare and Medicaid programs and the close attention to health plan premiums by employers and, where they bear part of the premium cost, by individuals assure that costs will remain in the forefront as one marker of performance.
Other measures of performance, including technical quality of care, health status, and patient satisfaction, are also increasingly available. Examples include HEDIS (Health Plan Employer Data and Information Set), a system to measure HMO performance pioneered by the National Committee for Quality Assurance, and the requirements set out by such private groups as the employers in the Twin Cities area and CALPERS (the California Public Employees Retirement System), which provide fringe benefits for public employees in that state. A governmental equivalent is the competitive contracting process for state Medicaid programs that selects managed care plans to serve the Medicaid population, as in Arizona, Tennessee, and a number of other states.
Performance monitoring systems should also include measures of access, which would require population-based data on such indicators as those recommended by the IOM in 1993 (IOM, 1993a). Such data cannot be gathered entirely by the health plans themselves, at least as long as a growing number of Americans are excluded from any health plan. Regardless of the prevailing interest among elected officials or the public in questions of access, the committee believes that levels of access should be considered an important indicator of overall performance of the health system, including primary care. This view is consistent with the committee's recommendation that access to primary care for everyone should remain an objective for American society (Recommendation 5.1).
Potential users of information about health care performance include employers,
governments on behalf of their beneficiaries and employees, individuals choosing among competing health plans, and health plans themselves.
The developing performance monitoring models are aimed at total health system performance, yet most managed care plans make the operating assumption that increasing primary care as a proportion of the total health care activity will make the totality of care less costly without compromising quality or patient satisfaction. (At a minimum plans may assume that outcomes will be sustained at a level that will not cause the plan to lose enrollees or contracts with employers or government agencies.) Therefore, information on how well the primary care component of the plan is performing is likely to be very important to plan managers, purchasers, regulators, and patients.
Until very recent years most of the resources to develop programs to assess quality have focused on inpatient services. In the current health care environment, however, it is imperative that substantial effort be put into further development of approaches to monitor the performance of primary care, particularly on dimensions of health care outcomes, patient health status, and patient satisfaction. The market in health care, even in those locations that have proceeded quite far down the road of competition among managed care plans, seems too compromised by lack of informed choice for the ultimate consumer, the patient, to be the sole arbiter of health system performance. Whichever mix of regulation—choice by large payers on behalf of consumers or direct choice by patients—emerges as the means of shaping desired performance by the health care providers and plans, better information will be the key. In the area of primary care, where the tradition of measurement is less and where the technical challenges of developing and implementing are formidable, an increased level of effort in developing those systems should have a high priority.
As discussed elsewhere, the objective of accountability in primary care requires performance measurement. Other aspects of the definition make this task more rather than less difficult, because they emphasize characteristics of primary care that extend well beyond the competence with which a specific medical encounter is performed. Both process and outcome data will need to relate to the objectives of integration (continuity, comprehensiveness, and coordination), accessibility of services, sustained partnership with patients, the scope of services and the pattern of referrals (already tracked by most managed care plans), and knowledge of relationships to family and community relevant to the provision of primary care. The technical problems of case mix, instability of enrollments, and the multiple factors affecting outcomes, among others, will complicate the measurement task. The unit of review—health care organization or individual primary care clinician or practice—is yet another issue.
Fortunately, this effort can build on work already done; the need to balance information on utilization and cost with information about the other measures of care that are necessary to measure performance and value should provide the motivation to proceed. The issue is not new, however. According to Kerr White,
an important early figure in identifying the need for more emphasis on primary care in the U.S. health care system, "Performance and results are the criteria that society is using with increasing sophistication to assess the medical profession and its efforts; activity and costs are no longer adequate measures" (White, 1967, p. 848).
The issue of who should be responsible for developing the measures and how they should be implemented is also complex, given Americans' general skepticism of the role of government. In recent years, the health professions have also become wary of the motivations of health plans competing in a market that is very sensitive to cost. A governmental model is illustrated by the classic state role in licensure of health professionals and institutions, by the quality assurance and improvement efforts for the Medicare program, and by the regulation of nursing homes under a federal-state relationship related mostly to the Medicaid program. Nongovernmental models are illustrated by the decades-long accreditation programs of the Joint Commission on the Accreditation of Healthcare Organizations and NCQA's accreditation of HMOs and the HEDIS effort cited earlier.
A public-private collaborative model might be appropriate for efforts to develop performance measures, especially if it could continue over time to advance the state of the art of performance monitoring. The users could be both governmental agencies and private sector plans, with the public-private entity assuming a data audit function to certify the quality of the data. The committee has no firm view about which model is best, but history would suggest that a public-private consortium would match the distributed nature of health care responsibility in the United States.
Recommendation 5.10 Quality of Primary Care
The committee recommends the development and adoption of uniform methods and measures to monitor the performance of health care systems and individual clinicians in delivering primary care as defined in this report. Performance measures should include cost, quality, access, and patient and clinician satisfaction. The results should be made available to public and private purchasers of care, provider organizations, clinicians, and the general public.
Infrastructure Development for Primary Care
Primary care practices in the future are likely to require an infrastructure that extends beyond the usual capital requirements of facilities, land, and equipment. This factor in turn will call for investments that are beyond the capabilities of the individual primary care unit (i.e., a small group or team). These infrastructure
needs constitute a lengthy and very complex list of systems and information sources, such as:
- systems for
- — recording and maintaining clinical data,
- — providing assists to clinical decisionmaking (e.g., clinical practice guidelines, clinical algorithms),
- — monitoring quality of care, and
- — overall practice management;
- patient education materials relating to healthy behaviors and as background information that patients can use in participating in clinical decisionmaking about their care;
- information on the community and the population being served, including disease and injury patterns, environmental and workplace hazards, social and economic characteristics of the locale;
- information about community services available in the community, including health and social services, transportation services for patients without transport options of their own, and for rural areas, emergency medical services capabilities and transport systems, telecommunication links, and locum tenens support; and
- continuing education support for primary care staff.
Extensive as this inventory of infrastructure needs is, it is not all-inclusive. (For example, not mentioned here is support for professional education and research, which are discussed in Chapters 7 and 8.) Furthermore, although the needs have been recognized for many years, a decade ago an IOM committee identified the lack of appropriate infrastructure support as one factor that inhibited the development of COPC practices in the United States (IOM, 1984).
Some of these activities, such as clinical information systems, are generic to all of medical care; The Computer-Based Patient Record: An Essential Technology for Health Care (IOM, 1991) highlights this point. Even generic infrastructure needs, however, have aspects that are particularly related to primary care. For example, the chair of the IOM committee on the computer-based patient record has argued that such systems, although often developed in institutional settings, are even more pertinent to primary care because of the need to deal with patient data covering many problems and to follow the patient over substantial time (Detmer and Finney, 1992).
It is not the intent of this report to deal with infrastructure needs separately but rather to address the questions of how, collectively, they might be met. Several basic approaches to infrastructure development and support might be considered:
- Methods of payment for primary care services should recognize the costs
- of infrastructure, thereby creating a market for these infrastructure services that will then be purchased by the primary care practices. This market would encourage others to bear development and marketing costs, as, for example, vendors of clinical information systems.
- Aggregation of primary care practices into larger integrated systems results in economies of scale that allow use of internal capital to develop infrastructure.
- Direct subsidy of infrastructure development either by public subsidy or voluntary contributions from organizations.
The first two approaches by themselves have drawbacks. For example, low reimbursement rates or inadequate plan incomes in cost-competitive markets may mean that individual plans cannot finance infrastructure purchases. In addition, concentration of market power in a few large entities may give them competitive advantages (in part through well-capitalized infrastructure) that in turn will inhibit market entry by smaller health organizations. The health care market in many locales is likely to remain a mix of small and large primary care organizations in the near and medium term, yet the large organizations, especially for-profit enterprises, will have significant advantages in raising capital. As long as the health care market is skewed by such factors, the third approach may be desirable, at least for underwriting those infrastructure needs that require extensive initial capital for technical development (such as clinical information and decision systems), especially if those technologies are to focus specifically on the requirements of primary care. The development of infrastructure for primary care and assuring its wide dissemination could be advanced by creation of a new organization devoted to this purpose as well as other related functions, including relevant applied research. These are long-term strategic issues, and the committee returns to some of them in its final chapter on implementation.
Role of Academic Health Centers in Delivery Of Care
The academic health center (AHC) has as its principal missions the education of health professionals, patient care, and the conduct of research to advance health. These institutions have been and remain major providers of health care, primarily through their affiliated teaching hospitals and clinics. The patient care function has historically been seen as supportive of the education and research roles. It has been predominantly hospital based and focused on advanced, tertiary care. While most of the AHCs have provided some primary care, the primary care activities have remained a small part of the institutions' service role.
To carry out the education functions discussed in Chapter 7, however, the service role of AHCs must develop a much stronger base in primary care. This requires creative new strategies that may involve affiliation with other health care organizations and primary care practice sites. These sites for primary care will be
the equivalent of the historic role of the teaching hospital; they will need to provide high-quality service while fulfilling the teaching and mentoring responsibilities for new professionals and helping to advance the state of the art of primary care through applied research.
The need to strengthen the primary care role of AHCs comes at a time when the financing of these institutions is under great strain; competing health care organizations draw clinical activities and revenues away from the AHCs and opportunities for internal cross-subsidies are limited. Many of these institutions also care for a larger share of the uninsured than do their competitors. As noted in Chapter 7, therefore, these functions will require some direct subsidy, just as the teaching hospital function has been subsidized for decades. Part of the challenge will be to provide primary care experiences that will prepare students for practice in a health care environment that is concerned about efficient use of resources. Therefore, the subsidy for education should not be used to shield the primary care teaching setting from the need to focus on efficiency and value.
This strengthened role in primary care will call for an explicit modification of the mission of these institutions. As noted throughout the committee's site visits, many other health care organizations are skeptical about the commitment of the AHCs to primary care. Actions will be needed on their part to back up statements about the importance of primary care. It is also reasonable, however, for these institutions to expect that the extra costs of the educational function and of research and demonstrations in primary care will be covered by funding sources.
Recommendation 5.11 Primary Care in Academic Health Centers
The committee recommends that academic health centers explicitly accept primary care as one of their core missions and provide leadership in the development of primary care teaching, research, and service delivery programs.
This chapter has outlined several features of the U.S. health care scene that will influence the extent to which primary care evolves in this country. These include the spread of managed care, the expansion of integrated delivery systems, the consolidation of health plans and systems, growth in for-profit ownership of health plans and integrated delivery systems, the diversity between and within health care markets, the special challenges of primary care in rural areas and for the urban poor, the need for primary care to coordinate with other types of services, current and evolving roles for health care professionals, and the role of academic health centers in primary care delivery.
Having reviewed these topics, the committee considered what conclusions and recommendations it would make to overcome the barriers, or exploit the advantages, that these factors pose, or offer, to full implementation of the committee's vision of primary care in the future. In all, the committee advances 11 separate recommendations in the several different arenas. First, the committee recommends establishing as a goal the availability of the services of a primary care clinician for all Americans. Second, the committee makes several recommendations to assure that mechanisms for financing primary care services provide appropriate incentives for sustaining a strong primary care function. In this context the committee makes a strong statement about the need to have universal health care coverage to make possible universal access to primary care. Another recommendation concerns the organization of primary care and emphasizes the importance of the primary care team. With respect to underserved populations, the committee returns to its earlier themes to underscore the importance of primary care for populations who have special health care needs or who are traditionally underserved. Another major thesis of this chapter is the need for primary care to develop strong relationships with three other types of health activities—public health, mental health, and long-term care—and the committee offers three specific recommendations intended to reinforce the coordination and collaboration efforts in these areas. A tenth recommendation calls for specific steps to develop tools and approaches for monitoring and improving the quality of primary care and to make performance information available to a wide audience. The final recommendation calls on AHCs to make primary care a core element of their mission and to provide leadership in education, research, and service delivery related to primary care.
ACOG (American College of Obstetricians and Gynecologists). Findings from a study conducted for ACOG by the Gallup Organization to examine women's attitudes and experiences with OB/GYNs as their primary care physicians. Washington, D.C.: ACOG, 1993.
Aiken, L.H., Lewis, C.E., Craig, J., et al. The Contribution of Specialists to the Delivery of Primary Care: A New Perspective. New England Journal of Medicine 300:1363-1370, 1979.
Alpha Center. MSAs: Issues for States. State Initiatives in Health Care Reform, 16(Jan./Feb.):7, 1996.
American Academy of Actuaries. Public Policy Monograph. Medical Savings Accounts: An Analysis of the Family Medical Savings and Investment Act of 1995. Washington, D.C.: American Academy of Actuaries, 1995.
ASIM (American Society of Internal Medicine). Results of a survey of IPA model HMOs. Conveyed in letter dated May 3, 1995, from J.P. DuMoulin, Director, Managed Care and Regulatory Affairs. ASIM Survey of HMOs. Washington, D.C.: ASIM, 1995.
Bailit, H.L. Market Strategies and the Growth of Managed Care. Pp. 3–13 in Academic Health Centers in the Managed Care Environment. D. Korn, C.J. McLaughlin, and M. Osterweis, eds. Washington, D.C.: Association of Academic Health Centers, 1995.
Blumenthal, D., and Meyer, G.S. The Future of the Academic Medical Center Under Health Reform. New England Journal of Medicine 329:1812–1814, 1993.
COGME (Council on Graduate Medical Education). Recommendations to Improve Access to Health Care Through Physician Workforce Reform. Fourth Report to Congress and the Department of Health and Human Services Secretary. Rockville, Md.: Health Resources and Services Administration, Department of Health and Human Services, 1994.
COGME. COGME 1995 Physician Workforce Funding Recommendations for Department of Health and Human Services' Programs. Seventh Report to Congress and the Department of Health and Human Services. Rockville, Md.: Health Resources and Services Administration, Department of Health and Human Services, 1995.
Detmer, D.E., and Finney, M.D. The Catalyst of Technology: How Will Advances in Information Technology Change the Role of the Primary Care Practitioner? Pp. 167–182 in Proceedings of the National Primary Care Conference, Vol. 2. Washington, D.C.: U.S. Department of Health and Human Services, Public Health Service, Health Resources and Services Administration, 1992.
EBRI (Employee Benefit Research Institute). Sources of Health Insurance and Characteristics of the Uninsured: Analysis of the March 1994 Current Population Survey. EBRI Special Report SR-28 and Issue Brief Number 158. Washington, D.C.: EBRI, 1995.
Epstein, A.M. U.S. Teaching Hospitals in the Evolving Health Care System. Journal of the American Medical Association 273:1203-1207, 1995.
Fox, P.D., and Wasserman, J. Academic Medical Centers and Managed Care: Uneasy Partners. Health Affairs 12(1):85-93, 1993.
GHAA (Group Health Association of America). Patterns in HMO Enrollment . 4th ed. Washington, D.C.: GHAA, 1995.
Gray, B.H. The Profit Motive in Patient Care. Cambridge, Mass.: Harvard University Press, 1991.
Greene, J.C., and Greene, A.R. Chapter 15: Oral Health. Pp. 315–334 in Health Promotion and Disease Prevention in Clinical Practice. S.H. Woolf, S. Jonas, and R.S. Lawrence, eds. Baltimore: Williams and Wilkins, 1995.
Horton, J.A., Murphy, P., and Hale, R.W. Obstetrician-Gynecologists as Primary Care Providers: A National Survey of Women. Primary Care Update for OB/GYNS 1:212-215, 1994.
Institute for Clinical Systems Integration. 1993 Annual Report. Minneapolis, Minn. : Institute for Clinical Systems Integration, no date.
IOM. (Institute of Medicine) Community-Oriented Primary Care: A Practical Assessment. Volume I. The Committee Report. Washington, D.C.: National Academy Press, 1984.
IOM. For-Profit Enterprise in Health Care. B.H. Gray, ed. Washington, D.C.: National Academy Press, 1986a.
IOM. Improving the Quality of Care in Nursing Homes. Washington, D.C.: National Academy Press, 1986b.
IOM. The Future of Public Health. Washington, D.C.: National Academy Press, 1988.
IOM. The Computer-Based Patient Record: An Essential Technology for Health Care. Richard S. Dick and Elaine B. Steen, eds. Washington, D.C.: National Academy Press, 1991.
IOM. Access to Health Care in America. M. Millman, ed. Washington, D.C.: National Academy Press, 1993a.
IOM. Employment and Health Benefits: A Connection at Risk. M.J. Field and H.T. Shapiro, eds. Washington, D.C.: National Academy Press, 1993b.
IOM. Real People, Real Problems: An Evaluation of the Long-term Care Ombudsman Programs of the Older Americans Act. J. Harris-Wehling, J.C. Feasley, and C.L. Estes, eds. Washington, D.C.: National Academy Press, 1995.
IOM. The Nation's Physician Workforce: Options for Balancing Supply and Requirements. K.N. Lohr, N.A. Vanselow, and D.E. Detmer, eds. Washington, D.C.: National Academy Press, 1996.
Joint Committee on Taxation. Description and Analysis of H.R. 1818 (The ''Family Medical Savings and Investment Act of 1995"). Doc. No. JCX-28-95. Washington, D.C.: Joint Committee on Taxation, U.S. Congress, 1995.
Josiah Macy, Jr. Foundation. The Financing of Medical Schools in an Era of Health Care Reform. Proceedings of a Conference Chaired by Eli Ginzberg, M.D. Conference proceedings edited by R.H. Ebert and case studies edited by M. Ostrow. New York: Josiah Macy, Jr. Foundation, 1995.
Lewin-VHI. The States and Private Sector: Leading Health Care Reform. States as Payers: Managed Care for Medicaid Populations. Washington, D.C.: The National Institute for Health Care Management, 1995.
Mitka, M. Higher Pay for Primary Care. American Medical News 37:3,7, October 3, 1994a.
Mitka, M. Physician Income up 5.4%. American Medical News 37:3,12, December 12, 1994b.
Mitka, M. Academic Pay Slowing, Primary Care Growing. American Medical News 38:5, July 24, 1995.
OTA (Office of Technology Assessment). Health Care in Rural America . Washington, D.C.: U.S. Government Printing Office, 1990.
Pew Health Professions Commission. Critical Challenges: Revitalizing the Health Professions for the Twenty-First Century. San Francisco: Pew Health Professions Commission, 1995.
PPRC (Physician Payment Review Commission). Annual Report to Congress, 1994. Washington, D.C.: PPRC, 1994.
PPRC. Annual Report to Congress. 1995. Washington, D.C.: PPRC, 1995.
Short, P.F., and Banthin, J.S. Caring for the Uninsured and Underinsured: New Estimates of the Underinsured Younger than 65 Years. Journal of the American Medical Association 274:1302–1306, 1995.
Shortell, S.M., Gillies, R.R., and Anderson, D.A. The New World of Managed Care: Creating Organized Delivery Systems. Health Affairs 13;(Winter):46–64, 1994.
University Hospital Consortium. Competing in the Maturing Health Care Marketplace: Strategies for Academic Medical Centers. Prepared in consultation with APM Management Consultants. Oakbrook, Ill.: University Hospital Consortium, 1993.
Weiner, J.P. Assessing the Impact of Managed Care on the U.S. Physician Workforce. Policy paper commissioned by the Bureau of Health Professions of the U.S. Department of Health and Human Services on behalf of the Council on Graduate Medical Education. November 1993.
Weiner, J.P. Forecasting the Effects of Health Reform on U.S. Physician Workforce Requirement: Evidence from HMO Staffing Patterns. Journal of the American Medical Association 272:222–230, 1994.
White, K.L. Primary Medical Care for Families—Organization and Evaluation. New England Journal of Medicine 277:847–852, 1967.