The Value of Primary Care
In setting out its view of the value of primary care, the committee makes two critical assumptions. First, primary care is the logical basis of an effective health care system. Second, primary care is essential to reaching the objectives that constitute value in health care: high quality care (including achieving desired outcomes), good patient satisfaction, and efficient use of resources. If the health care system is to move in the directions identified in this report, the value of primary care must be clear to the American public, policymakers, communities, educators, individual health professionals, and students. All people—adults as well as children, middle-class as well as poor, the healthy as well as the ill—must be seen to benefit.
This chapter addresses the value of primary care from two main perspectives. The first section provides some illustrative examples of the value of primary care for individuals; they are organized by the key elements of primary care in the definition from Chapter 2 and are oriented to primary care as it should be. The second section focuses on the benefits of primary care for populations and for the broader society. In reviewing evidence that primary care improves the quality and efficiency of care as well as access to care for populations, the section focuses, of necessity, on primary care as it is now provided. Because much of the current provision of primary care does not match all of the attributes set out in the definition, the value of primary care may, in our judgment, be understated compared with its potential benefits.
The Value of Primary Care for Individuals
Primary care is valuable to individuals in at least the five ways listed below:
- It provides a place to which patients can bring a wide range of health problems for appropriate attention—a place in which patients can expect, in most instances, that their problems will be resolved without referral.
- It guides patients through the health system, including appropriate referrals for services from other health professionals.
- It facilitates an ongoing relationship between patients and clinicians and fosters participation by patients in decisionmaking about their health and their own care.
- It provides opportunities for disease prevention and health promotion as well as early detection of problems.
- It helps build bridges between personal health care services and patients' families and communities that can assist in meeting the health needs of the patient.
These key components of high quality and efficient health care for individuals are illustrated in vignettes throughout this section of the chapter.1 Reflecting the nature of primary care, the vignettes include situations in which a variety of seemingly routine or simple problems may be embedded in the possibility of a patient's having conditions that could have serious consequences for his or her health. They illustrate the need for excellent primary care training that underlies clinicians' ability to distinguish among simple, serious, and complex conditions and to provide care for all.
Addressing Most Problems That Patients Bring
Most of the problems that people bring to the health care system are appropriately resolved at the level of primary care. Having the capacity to address ''a large majority of personal health care needs" also means that primary care offers patients a sensible and convenient route to appropriate care, which may involve referrals or coordination of services by others; patients do not need to guess for themselves what is causing a symptom or concern to be able to enter the health care system at the right place.
Jan Anderson, a 28-year-old woman, visits her doctor because her lower back has been hurting for a week. She has been a patient of Dr. Bloch, a family physician, since she was 10 years old, for a variety of problems. Dr. Bloch has been involved in treating her scoliosis (when she was a young girl) and in managing, over the years, a recurrent kidney infection, irritable bowel syndrome, and, before she used contraceptive pills, painful menstrual cramps, all possible sources of her pain. Dr. Bloch also knows that Ms. Anderson is an avid exercise enthusiast. Dr. Bloch evaluates the low back pain to determine if it is related to one of the earlier problems or to exercise. After he has diagnosed her problem, he treats her and makes arrangements for follow-up care.
Helping patients sort out and resolve such symptoms and dilemmas is an essential feature of primary care. Sometimes evaluation may reveal that, in addition to the patient's stated reason for a visit, an even more important problem or concern lies unspoken and perhaps unacknowledged or unrecognized by the patient.
Caroline Clark is a 40-year-old married woman who manages her own business. She visits her primary care team and sees the nurse practitioner, Donna Washington, complaining of insomnia. Ms. Washington knows that in the past year, Mrs. Clark has had a severe allergic reaction to a bee sting and lithotripsy for a kidney stone; she also knows that Mrs. Clark's 10-year-old son is being treated for leukemia in a nearby medical center, which causes many trips to the hospital, repeated difficult laboratory tests, and frequent school absences; she is aware that Mr. Clark's profession requires frequent and long trips away from home. Ms. Washington prescribes Mrs. Clark a mild sleeping pill, renews her prescription for adrenalin in case she suffers any bee sting in the future, and advises Mrs. Clark about what she may expect in the future regarding kidney stones. Ms. Washington also provides support in coping with these personal and family stresses that may affect her current and future health, including information about how she can, if she wishes, arrange an appointment with a clinical psychologist who is part of her health plan.
These vignettes illustrate that in addressing the "large majority of health needs" the primary care clinician and the patient benefit from the characteristics of primary care, including integration, the development of a sustained partnership, and attention to the context of family and community.
Guiding Patients in Using the Health Care System
A major element of good primary care is the ability of primary care clinicians to diagnose and manage their patients' health care problems. In many cases, this may require considerable understanding of the local health care scene and how best it might be navigated. When patients (or family) are new to an area or otherwise lacking in knowledge of the full range of resources open to them, the
primary care team can play a significant role in ensuring that those individuals move through the system efficiently and comfortably. For example, for patients with frightening symptoms such as acute dizzy spells upon awakening, a major question is where to go for help. Is this a problem that requires the services of a specialist, such as a neurologist? Is the problem related to other health problems for which the patient is being treated by a medical subspecialist and a psychiatrist, both of whom have prescribed prescription drugs? A primary care clinician can evaluate the problem and either manage the problem or arrange the appropriate referrals.
More generally, with its complex array of personnel, facilities, technologies, and other components, the health care system can confuse and intimidate patients. Primary care clinicians who know how the health system operates and have the expertise to evaluate information can provide instructions that patients can understand and help patients and families to make appropriate decisions and use the health care system to best advantage. In pediatrics, this concept is known as a "medical home";2 it is an appealing concept for all ages.
The tasks involved include: coordinating referrals to other specialists and sorting out the sometimes conflicting advice these clinicians give; arranging and overseeing care provided in different settings (e.g., hospitals or nursing homes); and finding and helping to secure ancillary resources such as physical therapy. To be sure, some patients can coordinate much of their own care; all too often, however, this responsibility falls to patients or their families who lack full knowledge of available health care resources. When patients are frail, lack family support, are faced with several difficult options, or have a problem that is complex and not well understood, they need help. Primary care teams can carry out these formidable tasks of coordination on behalf of the patient, drawing on their knowledge of the range of the patient's health problems, family or other social supports, and living arrangements.
Mary Ellerbee, herself elderly, lives with her 83-year-old husband who is being treated for hypertension, diabetes, and poor eyesight. She has moderate dementia and frequently wanders. The couple is cared for by a primary care team; in particular, Robert Griffith, a general internist, and Linda Fuentes, a nurse practitioner, who alternate in seeing the Ellerbees. All of the office staff
are familiar with the couple and their health problems, so that calls can be referred appropriately to Ms. Fuentes and Dr. Griffith. Ms. Fuentes has arranged for a home health aide to assist the couple and has located respite services, and at Mr. Ellerbee's request, Dr. Griffith has been in touch with their only son to give him updates on his parents' health.
Mr. Ellerbee experiences chest pain; in response to a neighbor's call to 9-1-1, he is hospitalized under Dr. Griffith's care, who arranges a consultation by a cardiologist. Ms. Fuentes helps the son find temporary care for his mother and, at the time of Mr. Ellerbee's discharge, Dr. Griffith assesses both Mr. Ellerbee's needs and those of his wife. Other health professionals are involved in their care as needed and to ensure that the couple can remain at home and that both receive appropriate medical care.
Providing an Ongoing Relationship Between Patient and Clinician
An important feature of primary care is the continuity that results from an ongoing relationship with clinicians who know their patients and their patients' health histories. Such relationships open opportunities for patients to disclose sensitive problems and for clinicians to discover favorable moments to provide counsel and advice.
Some problems are clearly related one to another; some are not clearly related but, when concurrent, may influence each other. Over the years, primary care clinicians will see patients through waves of episodes of care—some spells of illness and treatment stop, others begin, and others overlap. Periods of wellness are interspersed with problems that are chronic, acute, or intermittent. Some patients have only occasional acute problems that can be treated in isolation—a cough, a sprained ankle—and may seek only assured and rapid access to care. Other patients have problems that are recognized only because of patterns of illness that occur over months and years as opposed to hours or days—such as work-related asthma, or depression that results in many physical complaints. All in all, viewing health care as a continuum of interrelated episodes presents a very different picture of health and health care from one in which illness and disease are considered in isolation.
John Williams is an overweight 48-year-old bank executive who comes to his health plan because of his wife's complaints about his snoring. He wonders whether he has a serious condition called sleep apnea and has heard about multiple options for the ailment—diagnostic tests in a sleep disorders laboratory or neurologic testing, and therapies in the form of nasal surgery by an ear, nose, and throat specialist or laser surgery on the uvula by a plastic surgeon. Dr. Xanthos, his general internist, reviews his medical history, probes especially into Mr. Williams's current lifestyle and responsibilities, and orders a preliminary set of tests. His aim is to understand to what extent the condition is
serious and may require surgery, is caused or influenced by Mr. Williams' obesity, his heavy business and social schedule that often involves drinking alcohol in the evening, his sleeping position, or some other problem. Upon the return of the laboratory tests, Dr. Xanthos and Mr. Williams will discuss whether any further evaluations are needed, but in the meantime Mr. Williams is counseled about healthier lifestyle choices he might make in the area of exercise, diet, and alcohol intake.
Time—in effect an element of continuity—is an excellent diagnostic tool. Because of ongoing relationships with patients, primary care clinicians can better evaluate the importance of a patient's symptoms than can practitioners who do not know the patient. This may in turn mean that extensive diagnostic testing for ill-defined symptoms or complaints can be postponed or avoided altogether, because the patient will be assured of follow-up care in person and by telephone.
Another aspect of continuity is having relevant, up-to-date information about a patient available when it is needed. Although this information can reside in paper records and the memories of physicians or others with long-standing relationships to patients, ideally the record will reside on a computer-based patient record that can, with proper authority and passwords and due attention to privacy and confidentiality, be accessed by all clinical members of the primary care team (IOM, 1991; IOM, 1994).
Larry Jones calls his health plan during the weekend because he believes he is experiencing a side effect from his new heart medicine. Although the physician who is on call is not his usual primary care clinician (Dr. Kelly), a list of the patient's current medications, problems, and allergies is available on-line and can be accessed by the on-call physician. Knowing Mr. Jones' diagnosis, the type of medication he has begun and its dosage, the physician changes the dose, assures Mr. Jones that Dr. Kelly will be notified of the change, and advises him to call back if he has further problems.
Accountability reflects the degree to which clinicians or health plans take responsibility for the care they provide and the extent to which they are legally and morally answerable for important attributes of that care, such as quality, patient satisfaction, and efficient use of services. Accountability also involves continuing oversight of the patient's condition and placing occasional acute events in the context of a patient's problems. Being accountable also implies some obligation to maintain adequate, accurate, and retrievable records on patients. As implied above with respect to computer-based patient records, primary care can serve as the hub of an integrated health information system; clinicians can increase the accountability of the system by maintaining patient records of, among other things, medications, allergies, laboratory test results, and family medical histories. Such information serves a variety of purposes. For example, by reviewing
test results, tracking abnormal findings, and making sure patients know about these results, members of a primary care team can ensure that the health system is accountable for follow-up when it is needed. All such information can help patients avoid the problems that sometimes result when they see many different health practitioners, no one of whom has all the relevant information.
Myron Laramie, now 70 years of age, had triple bypass surgery 10 years ago. His current medical problems include diabetes, which he controls with diet; benign prostate enlargement, which his general internist, Dr. Mishalani, checks periodically; and recurrent depression, for which he sometimes takes medications. Recently, he had a cataract operation in his left eye. In all, Mr. Laramie takes six different medications—some several times a day, some daily, and some only when needed—each of which, taken one by one, is effective for a given condition. Mr. Laramie finds it hard to keep track of doses and to recognize side effects and interactions among these medications. Dr. Mishalani knows that overuse of medication in older patients is an important clinical issue and that, if possible, it is preferable to reduce the number of medications a patient must take. Dr. Mishalani cautiously starts a program of carefully withdrawing medications that may not be absolutely necessary. He monitors these medications and helps Mr. Laramie know how to recognize and manage potential side effects and interactions that may arise in the future. The doctor also tracks changes in Mr. Laramie's overall health status and his ability to function independently and confers with specialists who are also treating Mr. Laramie.
Preventing Illness and Detecting Diseases Early
At all ages, patients benefit from the proactive stance that primary care physicians, nurse practitioners, and physician assistants can take to listen, ask questions, and provide information. Indeed, primary care is often considered the front line for many aspects of health promotion and disease prevention.
Annette Nilsson, now 15 years old, has been followed by Dr. O'Brien, a pediatrician, since childhood; she now comes for a visit for treatment of mild acne. This visit presents an opportunity for Dr. O'Brien, who otherwise rarely sees this healthy young woman, to discuss Ms. Nilsson's understandable concerns about changes in her body and to offer appropriate personal guidance concerning smoking, alcohol, sexual activity, and other risk-taking behaviors.
Primary care fosters early detection of various disorders (including those that begin insidiously). The benefits include earlier and less onerous health care interventions, better and less hurried decisionmaking between the primary care clinicians and patients and their families, and likely lower costs of an episode of care.
In a primary care clinic, Dr. Renfroe sees a new patient, Betty Simms, for a sore throat. He also identifies high blood pressure and obesity and learns from Mrs. Simms that she has a two-pack-a-day history of cigarette smoking. Dr.
Renfroe evaluates her sore throat, determines that it is not bacterial in origin, and suggests some useful remedies; in addition, he counsels her about the dangers of her smoking addiction. With Mrs. Simms's agreement, he arranges to follow up her high blood pressure. Finally, he enlists the assistance of the clinic's receptionist to arrange for Mrs. Simms to have a nutrition consultation concerning her weight problem, noting the interactions between her smoking and eating habits and her hypertension.
Bridging Personal Health Care, Family, and Community
Primary care clinicians can establish links with communities and their resources, including those that patients on their own may not be aware of or be able to gain access to. In this way, they can create valuable bridges between what is done to and for patients and their families within the personal health care system and the preventive health or social services that may be available in the area in which patients reside. Knowledge of the family and community may also help the primary care clinician understand better the health problems and health risks faced by the patient. In addition, personnel in primary care teams and settings may often be able to act on behalf of their patients in settings and circumstances outside the traditional health care environment.
Primary Care and the School
Schools are among the settings most amenable to certain types of primary care, at least for persons from school age through late adolescence or early adulthood.3 Obviously, schools are environments in which acute illness, emotional stress, and violence all can occur. They are also windows onto health-related problems whose etiology may not, in the first instance, be obvious to school personnel. Primary care outreach may, therefore, be a useful tool for identifying and managing health-related problems before they irreparably damage a person's educational experience and accomplishments.
Johnny Torres, who is eight years old, has been a good student. Midway through the third grade, however, his teacher reports that he is having difficulty
in reading and is "hyperactive" and disruptive in class; she wonders if Johnny has an attention deficit disorder and would benefit from drug therapy. At the teacher's urging, the family calls Dr. Ursini, their pediatrician, who has cared for Johnny since birth. Dr. Ursini knows that a new baby sister is occupying much of Mrs. Torres's time and that the family is under economic stress because Mr. Torres's firm is anticipating layoffs following a corporate merger. At a quickly arranged visit, Dr. Ursini confirms that Johnny is reacting to stress; during a call to the school shortly thereafter, the pediatrician explains to Johnny's teacher and the school principal that no testing or medication is indicated right now but that Johnny would benefit from extra support and attention. Arrangements are made for a follow-up conference in six weeks.
Primary Care and the Elderly
Consistent aspects of caring for elderly patients include how the aging process affects health problems, the provision of care across different institutional settings, the need to involve family and community, and the benefit of working in a sustained relationship. Families of these patients need to be involved in planning for transportation, for direct care, for managing emergencies, and for issues of advance directives.
Assistance with buying groceries, cooking, managing finances, and personal care can be critical, because mobilizing these home or community services may enable elderly persons to stay in their homes or independent living situations rather than be moved to a nursing home. A primary care team can work closely with older persons and their families (or close friends or members of other social support systems, such as churches) to sustain these connections between personal health care services and long-term-care and social services.
Anthony Villarreal, now 88, has been discharged from the hospital following treatment for an episode of severe congestive heart failure, kidney failure, and paralysis caused by a recent stroke; his prognosis is not encouraging. Mr. Villarreal, his family, and his primary care physician, Dr. Young, have agreed that Mr. Villarreal will end his days at home without rehospitalization. His care is then orchestrated by Dr. Young with the help of Susan Zall (a visiting nurse), social services, and the office receptionist. Mrs. Zall obtains needed laboratory tests on a routine basis and sees that oxygen therapy and other services are arranged through social services; the office receptionist directs phone calls from home. Overall assessment of the patient's condition is based on reports from the family and occasional home visits by Dr. Young and Mrs. Zall, who also have advised the family about procedures they will need to take at the time Mr. Villarreal dies, which happens several months later at home.
Primary Care and Public Health
Links between primary, community, and public health functions are important
parts of primary care. Although primary care clinicians typically cannot intervene to solve public health problems that require community action, their awareness of infections, risks, and sources of morbidity in communities—in the environment, workplaces, homes, neighborhoods, and schools—can prompt important cooperative relationships with those who can intervene at a community level more effectively.
Following a discussion with a teacher, a school-based nurse practitioner, Sarah Aaronson, wonders what might explain the irritable behavior of five-year-old Melissa Edelman. Suspecting lead poisoning, Ms. Aaronson refers Melissa to a community health center. There Jerry Ikle, the center's physician assistant, takes a medical history and does a physical examination and orders appropriate laboratory tests. The lead test result shows elevated levels, and Mr. Ikle notifies his supervising family physician, begins an appropriate protocol-driven treatment, prescribes a follow-up visit with the physician, and alerts the city's medical social worker, Sharon Tang. Mrs. Tang visits Melissa's home, because she has begun to see a pattern of lead poisoning in that part of town, and suspects that it may be caused by old plumbing in many of the houses there. After comparing records, Mr. Ikle and Mrs. Tang alert the local housing authority, the school, and the public health department to both Melissa's case and the broader threat to the community who reside in that area.
The primary care clinician can also be an effective advocate in the community for needed public health actions. Good examples are the successes pediatricians have had in advancing child health through community actions, e.g., lead abatement, child safety seats, safety caps on medicines, community awareness of child abuse, support of poison control centers, and immunization campaigns.
These vignettes describe primary care that can be, and is, provided by well-trained, skillful, and dedicated individuals and teams. It would be naive, however, to conclude that the sorts of coordination, continuity, sustained personal relationships, and linkage of services within the health system, the family, and community that have been described here are either easy or inexpensive. Similarly, it would be naive not to recognize the inherent tensions between a drive for efficiency, as reflected in the private-sector reforms in health care delivery, and a desire to maintain strong patient-clinician relationships.
As to the first, the need to assemble, access, and make sense of huge amounts of information is growing. Also rising in intensity is the drive toward more efficient care in a managed care environment, which can sometimes lead to fragmented care that involves many health professionals who work under increasingly stringent time constraints. In the committee's view, however, achieving efficiency by delivering discrete services in this way—care for a sprain by
one clinician, management of a serious infection by another, adjustment of chronic medication by a third—cannot be the most important goal of health care delivery; indeed, a "division of labor" approach may not necessarily be the most efficient health care delivery.
Rather, a subtlety exists in patient-clinician relationships that so-called efficient systems cannot replace. An integrative function must be nurtured, and it almost certainly requires sensitivity and judgment on the part of a single, specific individual who knows the patient and the patient's circumstances. A group of clinicians—despite the best of intentions and the best-run team management—cannot replace this function. Moreover, medical records or a computerized summary cannot substitute for verbal and nonverbal communication that is based on an ongoing relationship between patient and clinician.
This tension between organized arrangements that facilitate care and efficient practice, on the one hand, and the intimate and personal relationships that are at the core of health care, on the other, is a central challenge for health care delivery systems. In posing these illustrative scenarios, the committee wishes to draw attention not only to the promise of primary care in pulling these threads together on behalf of the patient but also to the obstacles that opposing trends in health care delivery can place in the path of realizing that promise.
Primary Care And Costs, Access, And Quality
Empirical research, though sometimes indirect, indicates that primary care reduces costs, increases access to appropriate medical services for the population being served, and does not reduce the quality of care, thereby advancing the broader social interests in health care. This section reviews a portion of the literature comparing resource utilization, quality, and access to care among generalists and specialists. Some of this literature is reviewed in greater detail by Bowman (1989), Starfield (1992), Moore (1992), and Blumenthal and Mort (1992) and includes international comparisons of health status and costs as related to a country's primary care orientation, retrospective review of care given to patients in different settings, and randomized studies that assign patients to primary care and non-primary-care arms of a study. Investigators measure and compare the use of resources and the processes and outcomes of care to understand better whether the frequent claims that primary care reduces costs of care and improves quality and access to appropriate care can be justified. As noted at the start of this chapter, such empirical research is based on primary care as it has been or is now delivered in a number of settings. Demonstration of the full benefits of primary care as this committee has defined it will require prospective studies.
As noted by a preeminent researcher in this area, the effectiveness of primary care can be assessed by measuring each attribute of primary care and determining the impact on outcomes such as health status, satisfaction, use of services, and
costs of care (Starfield, 1992). Some comparisons are based on structural features that permit or facilitate the provision of primary care and on the performance of that system. Many studies, however, assume that if the structural features of primary care are present (for example, "a usual source of care"), then primary care is being provided. Other studies use the provision of care by health maintenance organizations (HMOs) as a proxy for primary care without estimating the extent to which primary care is actually provided. Readers need to consider these issues when reviewing assertions in the literature about quality and costs of primary care. They must also keep in mind that primary care is a moving target, evolving in response to social, economic, and professional factors as it is being studied.
Costs of Care
The primary care model is widely believed to be less expensive than specialty medicine, in part because payments to primary care clinicians are lower and in part because primary care clinicians tend to use fewer resources than other specialists.
Several studies suggest that primary care physicians tend to deliver less intensive care than specialists, particularly in hospital settings. Manu and Schwartz (1983) studied procedures ordered in a medical service ward of a teaching hospital. When the ward team was headed by a subspecialist, substantially more procedures were ordered, including more colonoscopies, bone marrow biopsies, and exercise treadmill tests. Since then Cherkin et al. (1987) found that recent graduates of internal medicine programs, which included many individuals headed for careers in subspecialties, were twice as likely as recently graduated family physicians to order blood tests, blood counts, chest x-rays, and electrocardiograms for their patients.
The Medical Outcomes Study (MOS) is a major observational study of more than 20,000 patients conducted in the late 1980s and into the 1990s.4 Greenfield et al. (1992) compared use of resources in specialty practice (cardiology and endocrinology) and generalist practice (family practice and general internal medicine) in five different systems of care that included both fee-for-service and prepaid practice. Adjusting for patient mix and comparing hospital admission rates, annual office visits, prescription drugs, and common tests and procedures, the authors concluded that specialty training as well as payment system and practice organization had independent effects on resource use.
In particular, cardiologists and endocrinologists had higher rates of hospitalization than did family practitioners and general internists. With respect to office visits, endocrinologists had significantly higher rates than the other physician groups. For prescription drugs, the rates for family practice and general internal medicine were "considerably lower" than the rates for the subspecialties, and the proportion of patients having tests and procedures and the mean number of tests and procedures per visit and per year were generally lower for the generalists than for the subspecialists. Overall differences across the four specialties were highly significant statistically.
Another study compared expenditures for Colorado Medicaid patients who were and were not enrolled in a primary care physician program using as outcome variables the use of emergency department and inpatient services (Fryer, 1991). Fryer found a slight increase in expenditures for physician services, but this was more than offset by decreases in inpatient and emergency department expenditures. Overall, there was a 15 percent decrease in costs for the group enrolled in the primary care physician program as compared to usual costs in the Medicaid program in which patients did not have access to a usual primary care physician.
Evidence demonstrating that primary care providers are more efficient in their use of resources has led managed care organizations to use "networks" of primary care physicians. Premiums for managed care plans have been about 7 percent lower than they are for more traditional indemnity insurance plans (Barents Group, 1995). Whether these cost savings can be attributed to better management of care, economies of scale realized through administrative efficiencies, selection bias, or all three, remains unresolved. What cannot be disputed is the rapid growth of managed care based on primary care as a principal way to moderate increases in health care costs.
The supply of primary care physicians in a geographic area also appears to be associated with the level of costs. Dor and Holahan (1990) reported that Medicare physician expenditures were lower in areas with higher numbers of general practitioners (GPs) and family physicians (FPs). Total Medicare expenditures per beneficiary—adjusted for the prevailing charge index—decreased by 1 percent for every 10 percent increase in the supply of GPs and FPs. Similarly, according to Welch et al. (1993), expenditures for the delivery of physicians' services to Medicare beneficiaries are higher in areas of the country with a lower proportion of primary care practitioners. A recent study (Mark et al., 1995) found that U.S. urban counties with higher population densities of family practice and general internal medicine physicians have lower total Medicare Part B reimbursements per beneficiary.
Although evidence indicates that organizational models that emphasize primary care are less expensive than organizational models emphasizing specialty medicine, skeptics may still ask whether such savings come at the expense of good quality care.
The early work of the Ambulatory Sentinel Practice Network (ASPN, 1988)
(see Chapter 8), a practice-based research network, included a study that suggested that excellent results can be attained in primary care with less intensive use of services than are indicated by specialty-based practice standards. In a study of usual care of miscarriage as managed in primary care practices, for about half of patients, management departed from textbook recommendations (in which all patients should be hospitalized), but results at follow-up were no different among patients treated according to standard teaching and those who were not. The primary care physicians were evidently able to discriminate on the basis of clinical presentations those women who would do well with less intensive treatment than recommended.
The same network later reported a series of investigations concerning the evaluation of headache and the detection of intracranial tumors, subarachnoid hemorrhages, and subdural hematomas in primary care patients (Becker et al., 1993a, 1993b). Becker and his colleagues found that primary care clinicians used computed tomography (CT) scanning very selectively and that more extensive use of CT scans would be a weak strategy to improve detection of these serious disorders because increased use would lead to higher health care costs and to unintended adverse effects, but they provide little if any benefit. Although these studies are not conclusive, they suggest that policies directed toward the use of low-cost providers will not necessarily lead to a deterioration in the quality of care.
Access to Care
According to an IOM report on access (IOM 1993a), access is the timely receipt of appropriate care. The concept is relevant to primary, specialty, and even exotic or experimental care, but in all cases, access to appropriate care is influenced by the number and distribution of primary care clinicians.
To cite cases in point, when individuals do not have a usual source of primary care because of geographic, financial, or other barriers, the care they receive through emergency departments may be both costly and inefficient (Shea et al., 1992). Lack of health insurance or gaps in insurance can mean loss of a source of primary care (Berman, 1995; Kogan et al., 1995). Having a ''regular source of care" is sometimes used as a proxy for availability of primary care and of continuity. The Rand Health Insurance Study demonstrated the benefit of access to primary care services, in particular for the poor, that resulted in improved vision, more complete immunization, better blood pressure control, enhanced dental status, and reduction in estimated mortality in comparison to low-income individuals and their children who had financial barriers to access consisting of cost sharing (Lohr et al., 1986; Goldberg and Newhouse, 1987; Newhouse and the Health Insurance Group, 1993).
Some patients identify the emergency department as their regular source of care (Baker et al., 1994), but this cannot be considered primary care, and as a
regular source of care it may not be appropriate to their needs. Hurley et al. (1988) randomized Medicaid patients into two groups: those with a primary care physician and those without. Patients who were assigned to a primary care physician had substantially fewer emergency department visits without an accompanying increase in office visits to a primary care physician.
With respect to hospital admissions, Parchman and Culler (1994) showed that, even after controlling for per capita income, preventable hospitalization rates among adults and children were significantly lower where the ratio of family and general practice physicians to population was greater. Bindman et al. (1995a) and Starfield (1995) report evidence suggesting that preventable hospitalizations are associated with a lack of primary care. Communities in which residents reported lower access to medical care (meaning principally primary care) had higher rates of preventable admissions for chronic medical conditions such as asthma, hypertension, congestive heart failure, chronic obstructive pulmonary disease, and diabetes.
Lurie et al. (1984, 1986) studied the effects of termination of Medi-Cal benefits for California's 270,000 medically indigent adults. They found that access to care and health status of those who lost their health coverage worsened. One year after their benefits were ended, only half of these low-income adults could identify a regular doctor, indicating a lack of access to primary care, and only two in five thought they could obtain care when they needed it. Sixty-eight percent of the group reported a specific episode in which they had not obtained care that they believed they needed; of those patients, 78 percent listed cost as a reason for not obtaining care. The percentage of medically indigent adults satisfied with their care decreased from 97 percent before termination to 40 percent one year after termination of benefits. While these findings extend beyond primary care, the finding of loss of a regular doctor for many would indicate that access to primary care was an important casualty of the loss of health benefits.
Quality of Care
How might we know if primary care produces equivalent or better outcomes and increases patient satisfaction compared to other health care delivery arrangements for serving populations with similar needs? Measures can include the classic triad of structure, process, and outcome described by Donabedian (1966, 1980, 1982, 1985). One can quantify underuse and overuse of services as well as technical and interpersonal quality in primary care settings, offices of medical subspecialists, and other settings such as emergency departments. To make such comparisons, one needs, first, to know how to measure primary care and, second, to determine which settings are providing primary care.
Measurement of primary care is made more difficult because the committee's
definition emphasizes characteristics of primary care that extend well beyond the competence with which a specific medical service is performed. Both process and outcome data 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, and knowledge of relationships to family and community relevant to the provision of primary care.
An important task in comparing the quality of care in primary care and non-primary-care settings is the need to control for variation in the kinds of patients seen in each setting—that is, to account accurately for demographic characteristics and severity and type of illness or injury. As Bindman (1994) notes, the extensive literature comparing generalist and specialist practice is difficult to interpret because of such differences. Studies that have avoided selection bias or adjusted for differences in patient populations have found that primary care physicians use fewer technologies and admit patients to the hospital less frequently.
A primary care orientation has been an important variable in improving health status. It enables individuals to obtain services for illnesses before they become severe (Gonnella et al., 1977). It can improve health by controlling chronic conditions and thereby reducing preventable hospitalizations and what is usually thought to be inefficient utilization of nonemergency services provided by emergency departments. For example, Shea et al. (1992) determined that patients who had uncontrolled hypertension and did not have any primary care physician were more likely to seek emergency department care or to be admitted to a hospital than those with primary care physicians, even after controlling for patients' insurance status and compliance with medical therapy.
Higher levels of primary care in a geographic area are associated with lower mortality rates. Shi (1992) showed a consistent relationship between the availability of primary care physicians and positive health status in 50 states and the District of Columbia, as assessed by age-adjusted and standardized overall mortality, mortality associated with cancer and heart disease, neonatal mortality, and life expectancy; the association held even after controlling for the effect of urban-rural differences, poverty rates, education, and lifestyle factors. His results confirmed an earlier study by Farmer et al. (1991), which found that the ratio of primary care physicians to population was the only consistent predictor of age-specific mortality rates, even when considering such other characteristics as rurality, percentage of female-headed households, education levels, minority status, and poverty rates.
Information from countries with strong systems of primary care is illuminating as well. For example, in the 11 European, Scandinavian, and North American countries studied by Starfield (1994), countries whose health systems are more
oriented toward primary care generally realize better population outcomes. They achieve better health status (based on 14 indicators such as low birthweight ratio, neonatal mortality, age-adjusted life expectancy, and years of potential life lost), higher satisfaction with health services among their populations, lower expenditures per capita, and lower medication use.
The only study to compare outcomes of care in general and subspecialist practice is the Medical Outcomes Study. Outcomes of care for primary care and specialty care for patients with hypertension and diabetes mellitus were compared in an observational study with follow-up at three periods. Measured outcomes included mortality, disease-specific physiological markers, and measures of physical and emotional health. The authors found that clinicians in medical subspecialties (cardiology and endocrinology) used more services than did clinicians in family medicine and general internal medicine for patients with cardiac disease and diabetes mellitus, even after controlling for patient mix (patients' sociodemographic characteristics and severity of their illness). The research team also determined that the number of office visits, percentage of patients tested per visit, and the percentage of patients admitted to the hospital were higher for medical subspecialists than for clinicians in family medicine or general internal medicine (Greenfield et al., 1995).
In terms of outcomes, no meaningful differences were found in the mean health outcomes (including 7-year mortality) for moderately-ill patients with hypertension or non-insulin-dependent diabetes mellitus. Without further research, MOS conclusions based on the care of patients with diabetes mellitus cannot be extrapolated to the management of other conditions, but the evidence from this study indicates that care for these conditions by specialists does not result in better outcomes than care provided by generalists.
Attributes of Primary Care
Comprehensiveness. Certain attributes of primary care, including comprehensiveness, continuity, and coordination of care, are associated with better health outcomes and patient satisfaction. Hochheiser et al. (1971) compared the number of emergency department visits by children in 1967, before and after the opening of a neighborhood health center in Rochester, New York. They reported a 38 percent decrease in emergency department visits by center-area children from 1967 to 1970. For routine care of these children, the primary care setting can be presumed to be more appropriate and less expensive than alternative settings, such as emergency departments or hospitals.
Alpert et al. (1976) compared the effectiveness of comprehensive family-focused pediatric care with the episodic pediatric care provided at hospitals and public clinics. The patients with comprehensive care had fewer hospitalizations, operations, illness visits, and "no-show" appointments; they had more health
supervision visits and used more preventive services; and they reported higher patient satisfaction.
Continuity. Continuity, in the primary care context, has several meanings. It refers to care over time by a single individual or team of health professionals, but it can also refer to continuity of information about the patient.
Starfield and others have reviewed research evidence that continuity of care is "associated with more indicated preventive care, better identification of patients' psychosocial problems, fewer emergency hospitalizations, fewer hospitalizations in general, shorter lengths of stay, better compliance with appointments and taking of medications, and more timely care for problems" (Starfield, 1986, p. 194). Research has linked continuity of care to improved health outcomes. For example, Shear et al. (1983) used pregnancy as a tracer condition to analyze the association between clinician continuity and the quality of ambulatory care. Utilizing a retrospective cohort study design, they examined two groups of pregnant women—one cared for in family practice centers and the other in obstetric clinics. The newborn infants of women in the family practice group, who had much higher clinician continuity, were of higher birthweight, even after controlling for race, income, education, and parity of their mothers.
Using a double-blind randomized trial of elderly men assigned to either a "provider-continuity group" or a "provider-discontinuity group," Wasson et al. (1984) found that patients in the continuity group had fewer emergency admissions and shorter hospital lengths of stay. These patients also viewed their providers as more knowledgeable, thorough, and interested in patient education. Billings and Teicholz (1990) reported that patients with a single individual whom they considered to be in charge of their care experienced much lower rates of preventable hospital admissions.
Coordination. When patient care is well coordinated, it reflects an appropriate range of services that are orchestrated in a rational, cost-effective manner. Coordinated care can lower the risk of harmful complications of unnecessary tests and procedures (Franks et al., 1992). Furthermore, because coordination of care can often reduce the numbers of tests and procedures performed, it can lower the overall costs of care. Although several authors have expressed concern about the risks that undertreatment might pose for patient outcomes (Hillman, 1987; Reagan, 1987; Stephens, 1989), little if any evidence indicates that coordination of care might be associated with unfavorable outcomes, once confounding factors are taken into account (Franks et al., 1992).
Computer-based information systems lie in the future for primary care and are an important element of both continuity and coordination. For example, in one randomized controlled trial, Rogers and Haring (1979) found that computerized feedback of certain types of information enhanced patient care by facilitating coordination. Summaries with information about patients—including a problem
list, medications, results of laboratory tests, and suggested courses of action for care—were given to some physicians before their patient visits. Those patients whose doctors received such summaries had more completed procedures and referrals, more designated diets, and more discovery of new problems. These patients also spent, on average, fewer days in the hospital.
Management of Referrals
An important tenet of primary care is that self-referral defeats coordination of care, risks picking the wrong type of clinician and receiving less than optimum care, may result in additional and sometimes inappropriate referrals by specialists to other specialists, and increases the cost of medical care. Most managed care plans insist that the primary care clinician be the pathway to specialty care. Some empirical work supports this principle in terms of its effect on quality. For example, Roos (1979) found that the appropriateness and outcomes of tonsillectomy and adenoidectomy were better when patients had been referred to specialists by primary care physicians than when they were self-referred.
Specialists tend to refer to other specialists less appropriately than generalists (Rothert et al., 1984; Flood et al., 1993). Self-referral may be associated with other quality-of-care problems. For instance, although specialists seem to achieve better results than primary care physicians when treating patients with problems within their specialty, they do less well outside their specialty area (Rhee et al., 1981).
When referrals by primary care physicians are required before visits to specialists, use of specialty services and emergency room visits drops. Martin et al. (1989) randomized patients into two groups; one required a referral for specialty services and the other did not. The patients in the plan with the referral requirement had an average of 0.3 fewer visits to a specialist over a one-year period. These findings have obvious implications for costs of services as well as for the appropriateness of care, illustrating how cost and quality considerations are often intertwined.
Studies show conflicting evidence about the comparative levels of preventive services provided by generalists and specialists, though Dietrich and Goldberg (1984) found that both generalists and specialists were well below preventive services guidelines in providing these services to their patients. More recently, a telephone survey in urban California found that having a regular source of primary care has several positive features (Bindman et al., 1995b); compared to individuals who did not have a regular source of care, those who did (after controlling for differences in health insurance status) received more preventive care services.
The Limits Of Primary Care In Improving Population Health
What are the limits of medical care—in particular, of primary care—in improving health status? Chapter 2 defined primary care and differentiated it from primary health care as defined by the World Health Organization (1978, p. 3). As noted there, primary health care includes population-oriented services such as sanitation and safe drinking water. By contrast, primary care as defined by this committee includes personal health services but not population-based, public health services.
The distinction between public and personal health services is not the only boundary of interest, however. Those who emphasize community-oriented primary care (COPC) view COPC as a strategy for focusing attention on community determinants of health, especially socioeconomic determinants (Abramson and Kark, 1983; IOM, 1984). COPC proponents and others recognize that health care by itself, whether primary or specialty-based services, will have a limited impact on health status until or unless these determinants of public and social health are addressed.
The aggregate benefits in health status to be gained from increasing income or education greatly outweigh the gains from medical intervention. For example, health status has been demonstrated repeatedly to have a direct, positive relationship to per capita income and to level of education. Similarly, preventing injuries from violence, child neglect, or motor vehicle crashes and deterring the adverse health effects of teenage pregnancy, substance abuse, and sexually transmitted diseases are critical to the health of the community.
Despite diligent efforts by individual clinicians to assist individual patients, these broader influences on health may outweigh the contributions of traditional personal health services. During the committee's site visit to one rural area, a primary care clinician described the community's poverty, illiteracy, lack of transportation, and lack of knowledge about self-care, all of which made caring for acutely ill children and the elderly with common chronic problems particularly difficult and discouraging. She depicted her primary care services as "a cup bobbing on a sea of social problems."
High levels of teenage pregnancy, prenatal mortality, substance abuse, or occupational illness all signal factors far beyond the capacity of individual health care or even health promotion and disease prevention programs to cope with successfully. Primary care clinicians do, however, form an important bridge between the health and public health realms—that is, between personal and population health services. They have knowledge of community and environmental conditions and understand how their particular patients may be affected by those conditions. Clearly, primary care clinicians are not "responsible" for the lack of prenatal care, substance abuse, outbreaks of infectious disease, or malnutrition, and they cannot alone shoulder the burdens of social dysfunction. They can and
do, however, promote collaborative working relationships that include community resources, employer- or school-based initiatives, lay workers, and volunteer support groups. As discussed in Chapter 5, primary care is an arrangement well suited to forming these relationships.
Finally, most primary care interventions are undertaken at the level of personal health services. Nonetheless, the committee believes that such interventions—whether counseling, referral, or active listening—are made more effective by a sustained and personal relationship with patients' families and knowledge of their communities. In this way, an important conceptual and practical link between personal and population health services is both maintained and enhanced.
The value of primary care to individuals is found in all the core elements of the definition of primary care. The vignettes in this chapter illustrate that primary care provides a place to which patients can bring a wide range of health problems for appropriate attention; guides patients through the health system; facilitates an ongoing relationship between patients and clinicians in which patients participate in decisionmaking about their health and their own care; provides opportunities for disease prevention and health promotion as well as early detection of problems; and helps build bridges between clinicians and patients' families and communities. Empirical research also indicates the merits of primary care as a means of improving the overall performance of the health care system, by improving the quality and efficiency of care and expanding access to care. The chapter comments on the relationships between personal health care services (i.e., primary care) and public health services focused on the population. Chapter 4 explores in more detail the nature of primary care.
AAP (American Academy of Pediatrics). Ad Hoc Task Force on Definition of the Medical Home. The Medical Home. In Emergency Medical Services for Children: The Role of the Primary Care Provider. J. Singer and S. Ludwig, eds. Elk Grove Village, Ill.: American Academy of Pediatrics, 1992.
Abramson, J.H., and Kark, S.L. Community-Oriented Primary Care: Meaning and Scope. Pp. 21–59 in: Community-Oriented Primary Care—New Directions for Health Services. Washington, D.C.: National Academy Press, 1983.
Alpert, J.J., Robertson, L.S., Kosa, J., et al. Delivery of Health Care for Children: Report of an Experiment. Pediatrics 57:917-930, 1976.
ASPN. Spontaneous abortion in Primary Care. A Report from ASPN. Journal of the American Board of Family Practice. 1:15–23, 1988.
Baker, D.W., Stevens, C.D., and Brook, R.H. Regular Source of Ambulatory Care and Medical Utilization by Patients Presenting to a Public Hospital Emergency Department. Journal of the American Medical Association 271:1909–1912, 1994.
Barents Group. The Role of Primary Care Physicians in Controlling Health Care Costs: Evidence
and Effects. Prepared for the American Academy of Family Physicians. Washington, D.C.: Barents Group, 1995.
Becker, L.A., Green, L.A, Beaufait, D., et al. Use of CT Scans for the Investigation of Headache: A Report from ASPN, Part 1. Journal of Family Practice 37:135–141, 1993a.
Becker L.A., Green L.A., Beaufait D, et al. Detection of Intracranial Tumors, Subarachnoid Hemorrhages, and Subdural Hematomas in Primary Care Patients: A Report from ASPN, Part 2. Journal of Family Practice 37: 135–141, 1993b.
Berman, S. Uninsured Children. An Unintended Consequence of Health Care System Reform Efforts. Journal of the American Medical Association 274:1472–1473, 1995.
Billings, J., and Teicholz, N. Data Watch. Uninsured Patients in District of Columbia Hospitals. Health Affairs 9(Winter):158-165, 1990.
Bindman, A.B. Primary and Managed Care: Ingredients for Health Care Reform. Western Journal of Medicine 161:78-82, 1994.
Bindman, A.B., Grumbach, K., Osmond, D., et al. Preventable Hospitalizations and Access to Health Care. Journal of the American Medical Association 274:305-311, 1995a.
Bindman, A.B., Grumbach, K., Osmond, D., et al. Primary Care and Receipt of Preventive Services. Unpublished document. San Francisco: Bindman, San Francisco General Hospital and the Institute for Health Policy Studies, University of California, San Francisco, 1995b.
Blumenthal, D., and Mort, E. Primary Care for the Uninsured: A Review of the Literature. Draft of a paper prepared under contract for the Office of Technology Assessment, U.S. Congress. Boston: Massachusetts General Hospital and Harvard Medical School, 1992.
Bowman, M.A. The Quality of Care Provided by Family Physicians. Journal of Family Practice 28:346–355, 1989.
Cherkin, D.C., Rosenblatt, R.A, Hart, L.G., et al. The Use of Medical Resources by Residency-Trained Family Physicians and General Internists. Medical Care 25:455–468, 1987.
Dietrich, A.J., and Goldberg, H. Preventive Content of Adult Primary Care: Do Generalists and Subspecialists Differ? American Journal of Public Health 74:223–227, 1984.
Donabedian, A. Evaluating the Quality of Medical Care. Milbank Memorial Fund Quarterly 44:166–203, July (part 2), 1966.
Donabedian, A. Explorations in Quality Assessment and Monitoring. Vols. 1–3. Ann Arbor, Mich.: Health Administration, 1980, 1982, 1985.
Dor, A., and Holahan, J. Urban-Rural Differences in Medicare Physician Expenditures. Inquiry 27:307–318, 1990.
Farmer, F.L., Stokes, C.S., Fiser, R.H., and Papini, D.P. Poverty, Primary Care and Age-Specific Mortality. Journal of Rural Health 7:153–169, 1991.
Flood, A.B., Fremont, A.M., Bott, D.M., et al. Comparing Disease-Specific Practice Patterns of Generalists and Specialists. Presentation at the annual meeting of the Association for Health Services Research, Washington, D.C., June 1993.
Franks, P., Clancy, C.M., and Nutting, P.A. Gatekeeping Revisited—Protecting Patients from Overtreatment. New England Journal of Medicine 327:424-429, 1992.
Fryer, G.E. Evaluation of the Primary Care Physician Program. Unpublished report to the Colorado Department of Social Services, November, 1991.
Goldberg, G.A., and Newhouse, J.P. Effects of Cost Sharing on Physiological Health, Health Practices, and Worry. Health Services Research 22:279–306, 1987.
Gonnella, J., Cattani, J., Louis, D., et al. Use of Outcome Measures in Ambulatory Care Evaluation. In: G. Giebink, N. White, and E. Short, eds. Ambulatory Medical Care—Quality Assurance 1977. La Jolla, Calif.: La Jolla Health Science Publications, 1977.
Greenfield, S., Nelson, E.C., Zubkoff, M., et al. Variations in Resource Utilization Among Medical Specialties and Systems of Care: Results from the Medical Outcomes Study. Journal of the American Medical Association 267:1624–1630, 1992.
Greenfield, S., Rogers, W., Mangotich, M., et al. Outcomes of Patients With Hypertension and Non-Insulin-Dependent Diabetes Mellitus Treated by Different Systems and Specialties: Results
from the Medical Outcomes Study. Journal of the American Medical Association 274:1436–1444, 1995.
Hillman, A.L. Financial Incentives for Physicians in HMOs: Is There a Conflict of Interest? New England Journal of Medicine 317:1743–1748, 1987.
Hochheiser, L.I., Woodward, K., and Charney, E. Effect of the Neighborhood Health Center on the Use of Pediatric Emergency Departments in Rochester, New York. New England Journal of Medicine 285:148-152, 1971.
Hurley R.E., Freund, D.A., and Taylor, D.E. Emergency Room Use and Primary Care Case Management: Evidence from Four Medicaid Demonstration Programs. American Journal of Public Health 79:843–846, 1988.
IOM (Institute of Medicine). Community-Oriented Primary Care: A Practical Assessment. Volume I. The Committee Report. Washington, D.C.: National Academy Press, 1984.
IOM. The Computer-Based Patient Record: An Essential Technology for Health Care. R.S. Dick and E.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. Emergency Medical Services for Children. J.S. Durch and K.N. Lohr, eds. Washington, D.C.: National Academy Press, 1993b.
IOM. Health Data in the Information Age: Use, Disclosure, and Privacy . M.S. Donaldson and K.N. Lohr, eds. Washington, D.C.: National Academy Press, 1994.
IOM. Defining a Comprehensive School Health Program: An Interim Statement D. Allensworth, J. Wyche, E. Lawson, and L. Nicholson, eds. Washington, D.C.: National Academy Press, 1995.
IOM. Schools and Health: Our Nation's Investment. D. Allensworth, J. Wyche, E. Lawson, and L. Nicholson, eds. Washington, D.C.: National Academy Press, in press.
Kogan, M.D., Alexander, G.R., Teitelbaum, M.A., et al. The Effect of Gaps in Health Insurance on Continuity of a Regular Source of Care Among Preschool-Aged Children in the United States. Journal of the American Medical Association 274:1429–1435, 1995.
Lohr, K.N., Brook, R.H., Kamberg, C.J., et al. Use of Medical Care in the Rand Health Insurance Experiment: Diagnosis- and Service-Specific Analyses in a Randomized Controlled Trial. Medical Care 24:(9 Suppl.):S1–S87, 1986.
Lurie, N., Ward, N.B., Shapiro, M.F., and Brook, R.H. Termination from Medi-Cal—Does It Affect Health? New England Journal of Medicine 311:480–484, 1984.
Lurie, N., Ward, N.B., Shapiro, M.F., et al. Special Report. Termination of Medi-Cal Benefits: A Follow-Up Study One Year Later. New England Journal of Medicine 314:1266-1268, 1986.
Manu, P., and Schwartz, S.E. Patterns of Diagnostic Testing in the Academic Setting: The Influence of Medical Attendings' Subspecialty Training. Social Science and Medicine 17:1339-1342, 1983.
Mark, D.H., Gottleib, M.S., Zellner, B.B., et al. Medicare Costs and the Supply of Primary Care Physicians. Unpublished document. Milwaukee, Wisc.: D.H. Mark, Medical College of Wisconsin, 1995.
Martin, D.P., Diehr, P., Price, K.F., and Richardson, W.C. Effect of a Gatekeeper Plan on Health Services Use and Charges: A Randomized Trial. American Journal of Public Health 79:1628–1632, 1989.
Moore, G.T. The Case of the Disappearing Generalist: Does It Need to Be Solved? The Milbank Quarterly 70:361–379, 1992.
Newhouse, J.P., and the Health Insurance Group. Free for All? Lessons from the RAND Health Insurance Experiment. Cambridge:Harvard University Press, 1993.
Parchman, M.L., and Culler, S. Primary Care Physicians and Avoidable Hospitalization. Journal of Family Practice 39:123–128, 1994.
Reagan, M.D. Sounding Board. Physicians as Gatekeepers: A Complex Challenge . New England Journal of Medicine 317:1731–1734, 1987.
Rhee, S.-O., Luke, R.D., Lyons, T.F., and Payne, B.C. Domain of Practice and the Quality of Physician Performance. Medical Care 19:14–23, 1981.
Rogers, J.L., and Haring, O.M. The Impact of a Computerized Medical Record Summary System on Incidence and Length of Hospitalization. Medical Care 17:618–630, 1979.
Rogers, W.H., Wells, K.B., Meredith, L.S., et al. Outcomes for Adult Outpatients with Depression Under Prepaid or Fee-for-Service Financing. Archives of General Psychiatry 50:517–525, 1993.
Roos, N.P. Who Should Do the Surgery? Tonsillectomy-Adenoidectomy in One Canadian Province. Inquiry 16:73-83, 1979.
Rothert, M.L., Rovner, D.R., Elstein, A.S., et al. Differences in Medical Referral Decisions for Obesity Among Family Practitioners, General Internists, and Gynecologists. Medical Care 22:42–55, 1984.
Rubin, H.R., Gandek, B., Kosinski, M., et al. Patient Ratings of Outpatient Visits in Different Practice Settings. Results from the Medical Outcomes Study. Journal of the American Medical Association 270:835–840, 1993.
Safran, D., Tarlov, A.R., and Rogers, W. Primary Care Performance in Fee-for-Service and Prepaid Health Care Systems. Results from the Medical Outcomes Study. Journal of the American Medical Association 271:1579–1586, 1994.
Shea, S., Misra, D., Ehrlich, M.H., et al. Predisposing Factors for Severe, Uncontrolled Hypertension in an Inner-City Minority Population. New England Journal of Medicine 327:776-781, 1992.
Shear, C.L., Gipe, B.T., Mattheis, J.K., et al. Provider Continuity and Quality of Medical Care: A Retrospective Analysis of Prenatal and Prenatal Outcome. Medical Care 21:1204-1210, 1983.
Shi, L. The Relationship Between Primary Care and Life Chances. Journal of Health Care for the Poor and Underserved 3:321–335, 1992.
Starfield, B. Primary Care in the United States. International Journal of Health Services 16:179–198, 1986.
Starfield, B. Primary Care: Concept, Evaluation, and Policy. New York:Oxford University Press, 1992.
Starfield, B. Is Primary Care Essential? The Lancet 344:1129–1133, 1994.
Starfield, B. Editorial. Access—Perceived or Real, and to What? Journal of the American Medical Association 274:346–347, 1995.
Stephens, G. Can the Family Physician Avoid Conflict of Interest in the Gatekeeper Role? An Opposing View. Journal of Family Practice 28:701–704, 1989.
Stewart, A.L., and Ware, J.E., eds. Measuring Functioning and Well-Being: The Medical Outcomes Study Approach. Raleigh-Durham, N.C.: Duke University Press, 1992.
Tarlov, A.R., Ware, J.E., Greenfield, S., et al. The Medical Outcomes Study: An Application of Methods for Monitoring the Results of Medical Care. Journal of the American Medical Association 262:925–930, 1989.
Wasson, J.H., Sauvigne, A.E., Mogielnicki, P., et al. Continuity of Outpatient Medical Care in Elderly Men: A Randomized Trial. Journal of the American Medical Association 252:2413-2417, 1984.
Welch, W.P., Miller, M.E., Welch, H.G., et al. Geographic Variation in Expenditures for Physicians' Services in the United States. New England Journal of Medicine 328:621-627, 1993.
Wells, K.B., Hays, R.D., Burnam, M.A., et al. Detection of Depressive Disorder for Patients Receiving Prepaid or Fee-for-Service Care: Results from the Medical Outcomes Study. Journal of the American Medical Association 262:3298–3302, 1989.
World Health Organization. Alma-Ata 1978: Primary Health Care. Report of the International Conference on Primary Health Care, Alma-Ata, USSR, 6–12 September 1978. Geneva: World Health Organization, 1978.