|Home | About | Journals | Submit | Contact Us | Français|
The doctor–patient relationship has been and remains a keystone of care: the medium in which data are gathered, diagnoses and plans are made, compliance is accomplished, and healing, patient activation, and support are provided.1 To managed care organizations, its importance rests also on market savvy: satisfaction with the doctor–patient relationship is a critical factor in people's decisions to join and stay with a specific organization.2–5
The rapid penetration of managed care into the health care market raises concern for many patients, practitioners, and scholars about the effects that different financial and organizational features might have on the doctor–patient relationship.6–10 Some such concerns represent a blatant backlash on the part of providers against the perceived or feared deleterious effects of the corporatization of health care practices. But objective and theoretical bases for genuine concern remain. This article examines the foundations and features of the doctor–patient relationship, and how it may be affected by managed care.
The relationship between doctors and their patients has received philosophical, sociological, and literary attention since Hippocrates, and is the subject of some 8,000 articles, monographs, chapters, and books in the modern medical literature. A robust science of the doctor–patient encounter and relationship can guide decision making in health care plans. We know much about the average doctor's skills and knowledge in this area, and how to teach doctors to relate more effectively and efficiently.11, 12 We will first review data about the importance of the doctor–patient relationship and the medical encounter, then discuss moral features. We describe problems that exist and are said to exist, we promulgate principles for safeguarding what is good and improving that which requires remediation, and we finish with a brief discussion of practical ways that the doctor–patient relationship can be enhanced in managed care.
The medical interview is the major medium of health care. Most of the medical encounter is spent in discussion between practitioner and patient. The interview has three functions and 14 structural elements (Table 1).13 The three functions are gathering information, developing and maintaining a therapeutic relationship, and communicating information.14 These three functions inextricably interact. For example, a patient who does not trust or like the practitioner will not disclose complete information efficiently. A patient who is anxious will not comprehend information clearly. The relationship therefore directly determines the quality and completeness of information elicited and understood. It is the major influence on practitioner and patient satisfaction and thereby contributes to practice maintenance and prevention of practitioner burnout and turnover, and is the major determinant of compliance.15 Increasing data suggest that patients activated in the medical encounter to ask questions and to participate in their care do better biologically, in quality of life, and have higher satisfaction.16
Effective use of the structural elements of the interview also affect the therapeutic relationship and important outcomes such as biological and psychosocial quality of life, compliance, and satisfaction. Effective use gives patients a sense that they have been heard and allowed to express their major concerns,17 as well as respect,18 caring,19 empathy, self-disclosure, positive regard, congruence, and understanding,20 and allows patients to express and reflect their feelings21 and relate their stories in their own words.22 Interestingly, actual time spent together is less critical than the perception by patients that they are the focus of the time and that they are accurately heard. Other aspects important to the relationship include eliciting patients' own explanations of their illness,23, 24 giving patients information,25, 26 and involving patients in developing a treatment plan.27 (For an overview of this area of research, see Putnam and Lipkin, 1995.28)
A series of organizational or system factors also affect the doctor–patient relationship. The accessibility of personnel, both administrative and clinical, and their courtesy level, provide a sense that patients are important and respected, as do reasonable waiting times and attention to personal comfort. The availability of covering nurses and doctors contributes to a sense of security. Reminders and user-friendly educational materials create an atmosphere of caring and concern. Organizations can promote a patient-centered culture,29 or one that is profit- or physician-centered, with consequences for individual doctor–patient relationships. Organizations (as well as whole health care systems) can promote continuity in clinical relationships, which in turn affects the strength of in those relationships. For instance, a market-based system with health insurance linked to employers' whims, with competitive provider networks and frequent mergers and acquisitions, thwarts long-term relationships. A health plan that includes the spectrum of outpatient and inpatient, acute and chronic services has an opportunity to promote continuity across care settings.
The competition to enroll patients is often characterized by a combination of exaggerated promises and efforts to deliver less. Patients may arrive at the doctor's office expecting all their needs to be met in the way they themselves expect and define. They discover instead that the employer's negotiator defines their needs and the managed care company has communicated them in very fine or incomprehensible print. Primary care doctors thus become the bearers of the bad news, and are seen as closing gates to the patient's wishes and needs. When this happens, an immediate and enduring barrier to a trust-based patient-doctor relationship is created.
The doctor–patient relationship is critical for vulnerable patients as they experience a heightened reliance on the physician's competence, skills, and good will. The relationship need not involve a difference in power but usually does,30 especially to the degree the patient is vulnerable or the physician is autocratic. United States law considers the relationship fiduciary; i.e., physicians are expected and required to act in their patient's interests, even when those interests may conflict with their own.9 In addition, the doctor–patient relationship is remarkable for its centrality during life-altering and meaningful times in persons' lives, times of birth, death, severe illness, and healing. Thus, providing health care, and being a doctor, is a moral enterprise. An incompetent doctor is judged not merely to be a poor businessperson, but also morally blameworthy, as having not lived up to the expectations of patients, and having violated the trust that is an essential and moral feature of the doctor–patient relationship.31 Trust is a fragile state. Deception or other, even minor, betrayals are given weight disproportional to their occurrence, probably because of the vulnerability of the trusting party (R.L. Jackson, unpublished manuscript).
A managed care organization serves a defined population with limited resources in an integrated system of care. Thus, a single organization may both provide and pay for care. Organizations as providers have duties such as competence, skill, and fidelity to sick members. Organizations as payers have duties of stewardship and justice that can conflict with provider duties. Managed care organizations thus have conflicting roles and conflicting accountability.
An organization's accountability to its member population and to individual members has a series of inherent conflicts. Is the organization's primary accountability to its owners, to employer purchasers, to its population of members, or to individual, sick members? If these constituents somehow share the accountability, how are conflicting interests resolved or balanced? For example, the use of the primary care clinician to coordinate or restrain access to other services involves the primary care clinician in accountability for resource use as well as for care of individual patients. Although unrestricted advocacy for all patients is never really achievable, the proper balance and the principles of balancing between accountability to individual patients, a population of patients, or an organization need to be made explicit and to be negotiated in new ways.32–34
Does paying physicians by salary, capitation, risk withholds, or bonuses, with a variety of incentives to withhold (more or less) needed care from patients, represent a conflict of interest for physicians and violate the fiduciary nature of the relationship? All mechanisms for paying physicians, including fee-for-service reimbursement, create financial incentives to practice medicine in certain ways. We still lack a calculus to minimize or even describe in fine detail how such conflicts affect our ability to justify trusting relationships. Even-handed social attention seems appropriate to all the different mechanisms of payment. Balanced assessment of how the details of remuneration systems influence doctor's willingness to act on behalf of patients will best protect both the health of the public and the health of doctor–patient relationships. This is a priority for a new form of empirical, ethical research.
“Whose doctor is it anyway?” expresses one of the most critical problems inherent in managed care for the doctor–patient relationship. Patients correctly wonder if doctors are caring for them, the plan, or their own jobs or incomes (the latter is equally problematic in fee-for-service care). This ambiguity erodes trust, promotes adversarial relationships, and inhibits patient–centered care. The recent controversy over gag rules has only confirmed this set of fears in the mind of the public which is now seeking regulation of the managed care industry through the political process. As illustrated in Figure 1, the interests of patients, plans, and doctors can overlap to a greater or lesser extent. Professional ethics dictate that physicians attempt, as individuals and as a profession, to ensure that their interests and those of their patients are congruent in clinical practice. Plan interests, however, can pull physicians away from this goal, as the organization's values and their implementation inevitably influence attitudes, behavior, and experiences. Alternatively, plans could promote patient-centered care by trying to maximize the extent to which patient, doctor, and plan interests overlap. For example, promoting continuity, communication, and prevention can further all three interests so long as value (and not cost alone) is seen as the plan's product. Similarly, resource stewardship can be honestly promoted as a way to ensure that quality care is available for future patients.
Another feature of managed care organizations is their emphasis, in principle, on primary care. They often rely on primary care clinicians to manage, coordinate, or restrain access to other services. Members are required to choose or are assigned a primary care physician. With the primary care emphasis comes an opportunity for the development of strong relationships between primary care doctors and their patients. In addition, new relationships with patients who in the past never sought care and seldom entered into a doctor–patient relationship may be more likely in a system that emphasizes wellness and primary care, although this may be more apparent than real. It is unclear at present how a “relationship” between a primary care physician and a member of the physician's panel, who have never met, should be characterized, or what responsibilities are associated with it. It is not yet demonstrated that an emphasis, in principle, on primary care leads to stronger relationships, and to what extent countervailing forces such as lack of continuity counter this.
Integrated systems, characteristic of most managed care plans, introduce opportunities for improvement in continuity across the spectrum of care. For example, opportunities arise for case management or for coordinating care between doctors' offices, hospitals, nursing homes, and home care so that individuals do not fall through the cracks of a fragmented system. With integration come new responsibilities for doctors and other health care practitioners for communication, teamwork, and a more longitudinal approach to patient care. This continuity may be thwarted, however, by turnover in staff or members.
Standardization of practice, sometimes relying on “evidence–based medicine,” is often used by managed care to minimize costs or maximize or ensure quality of care. Standardization is often touted as promoting fairness by treating like individuals in like manner. Both standardization and the application of evidence-based principles in choosing care standards, however, rely on value judgments about what counts as good evidence and how that evidence should be interpreted and applied. The danger to the doctor–patient relationship in these movements is that individual patients with their individual needs and preferences may be considered secondary to following practice guidelines, adherence to which may form part of an evaluation measure of physician's performance. Using practice guidelines and the “standard of care” to determine which benefits are covered, and for whom, ignores the incredible variation in patient preferences and characteristics. This approach treats the disease without reference to the illness.35 Rather than treating individuals with similar illnesses in like manner, the result is that individuals who merely have the same disease are treated in like manner. Fairness is sacrificed to uniformity.36 Reliance on “data” may discount the patient's own story, thus discounting specific evidence about personal aspects of disease and its meaning and value. Obviously, discounting the person depreciates the relationship.
Continuous quality improvement and total quality management are industrial strategies37 lately applied in the health care arena. Although quality improvement efforts are by no means unique to managed care organizations (MCOs) in the health care industry, a few individual MCOs and the American Association of Health Plans have been leaders in promoting quality initiatives and include them in the accreditation process. Implementing continuous quality improvement may work for the doctor–patient relationship by enhancing competence and the perception of competence, or it may work against the doctor–patient relationship if it diminishes practitioner flexibility or accountability, or if it is perceived by practitioners as a manifestation of distrust by the organization.
The effort to cut costs to increase competitiveness or profit means having doctors be more “productive” by seeing patients faster. The first thing dropped as visit length shortens is psychosocial discussion.38 So far, the average length of visits in the United States does not seem to have dropped significantly, probably because of inherent inefficiencies in scheduling and doctors' abilities to finagle time to fit the needs of patients.39 Yet both patients and doctors feel a heightened sense of time pressure, and patients worry about being on a conveyor belt with a production-line-oriented doctor. As companies attempt to increase providers' efficiency, these fears will be realized unless thwarted by consumers, professionals, or more visionary organizations. Less time, otherwise, will mean less relating time and damage to care: less-accurate and incomplete data; difficulty in identifying the real problems; less efficiency in test and treatment choices based on knowledge of the individual patient; less trust; less healing; more errors and more waste.39 A penny of good communication time may avert a pound of unnecessary or even harmful spending used to reassure an anxious patient or substitute for a sketchy history.
We believe that in the long run the trust of the public that the physician is doing the absolute best for the patient must be maintained so that the doctor–patient relationship preserves its healing functions. At the moment, the momentum of control is such that industry and corporate leaders have the upper hand and care is or will suffer as a result. Only if consumers and the medical profession stand together and insist on standards that protect the doctor–patient relationship will it endure the acid raining against its delicate face.
Table 2 lists several principles physicians can follow to retain professional standards and nurture and sustain the public's trust in doctor–patient relationships. The first priority is to enhance knowledge, skills, and attitudes of doctors, patients, and plans in the doctor–patient relationship. Currently, neither doctors and patients, nor plans have adequate skills in the doctor–patient relationship. Most doctors currently practicing have never been critically observed interviewing a patient, breaking bad news, or denying a patient's request for an unnecessary test. Doctors need no longer suffer from a lack of this skill—it is learnable and quickly taught. Physicians should each ensure their own competence in this vital area.
Physicians should focus on continuity: in their relationships with individual patients, between their patients and other clinicians (including specialists and nurses), and with the organization as a whole. Trust is most realistic when a relationship has a history of reliability, advocacy, beneficence, and good will (R.L. Jackson, unpublished manuscript). Continuity encourages trust, provides an opportunity for patients and providers to know each other as persons and provides a foundation for making decisions with a particular individual. It allows physicians to be better advocates for their patients and allows patients some power by virtue of the personal relationship they have with this physician. Patients value continuity in and of itself, apart from its effect on health outcomes,40, 41 although its current value seems to be about $15 per month in added premium. Industry estimates are that an average patient will change plans and doctors if continuity costs more than $180 per year.42 Rapid changes between plans, mergers, acquisitions, closings, changing panels of providers within plans, and physician non-competition clauses all detract from the continuity of patient care. Physicians should advocate for continuity as an important goal for themselves in their individual practices, as members of a group practice, as a profession, and within their organizations.
Practitioners should work to protect the interests and the preferences of individuals. Utilization management, standardization, guidelines, and other cost–containment efforts are morally neutral. They may be necessary to ensure that resources needed to care for those who are not yet sick are available when the time comes. Whereas administrators and managers must responsibly steward the pooled resources of health insurance premiums, each physician in a managed care organization should primarily be an advocate for individual patients. This is not to say that physicians should ignore the cost implications of their decisions, or that they should be unconcerned with resource stewardship, merely that their primary responsibility as practitioners should be for the care of their patients.
Health care administrators, whose primary responsibility is stewardship, should not ignore the need for competence, compassion, and individualization of care. Physicians' roles as patient advocates mean they must attend to the needs of individual patients who may be exceptions to the rules or otherwise have special needs. As patient advocates, physicians must ensure that policies and procedures put in place that threaten the ability to individualize care do not go unchecked. Since this power may be beyond the capacity of individual physicians, it may require organization at the level of the whole profession.
Practitioners should contribute to quality improvement efforts. For efforts to be focused on improving the quality of care and not solely on restraining resource use, the role of physicians is indispensable. Physicians know when access is too tightly restrained and their patients' care is suffering, when restrictions on the use of particular drugs or equipment constitute unacceptable impingements on the quality of care, or in what circumstances a procedure is probably unnecessary. Physicians can, and should, serve as “quality police” by noticing, remarking, and, ideally, working for change when they see a feature that is detrimental to patient care. In addition, they should be proactive in spearheading and making clinically and humanly relevant quality improvement efforts in their organization.
Practitioners can practice prudence. Physicians should be prudent in their use of resources, and at a minimum should not waste resources by providing services of no benefit to patients. Physicians often complain that patients come in asking for x-rays, blood tests, and other services when physicians are skeptical of any benefit. Conversely, many patients have noted physician's overuse of “tests.” The role of insurers in the health care system means that a service rarely has direct costs for an individual patient, though it may be costly. Indeed, our culture seems to rely on technology to answer questions with a greater certainty than the technology can deliver. Physicians themselves have contributed to a culture of medical practice in which objective test results are given more credence and are felt to be more reliable than the subjective story of the patient or assessment of the physicians. In truth more than 80% of diagnoses are made by history alone.43 Physicians need to control their own reliance on objective but noncontributing data. By fostering a system of care in which concern for cost is acceptable and unnecessary services are not provided, physicians can be perceived as being socially responsible and perhaps restore some credibility in this area to the profession.
Because it is a matter of integrity not to waste resources on tests or other services, physicians must talk to patients, find out why they are requesting certain services, and meet those needs in other ways. We must educate patients about the limited ability of medical technology and the potential for harm in any treatment. This, again, involves skills that many physicians need to learn in order to understand the patient's underlying concerns, cultural background, and life history.
Physicians need to pay close attention to financial and nonfinancial incentives that might provide a strong conflict of interest when making decisions for individual patients. Physicians must look at how they are paid, realize how it might influence the care of their patients, and take steps to ensure that such concerns do not intrude unduly into decisions at the individual patient level. Remuneration schemes must be scrutinized for this possibility by paying attention to the number of patients the scheme affects, the ability to spread risks over a large population of patients in the case of capitated payment schemes, the implicit and explicit goals of remunerative strategies (including cost containment, but also potentially quality, patient satisfaction, continuity, and other worthy goals), and the extent to which the arrangements are public or, at least, open and understandable to patients. It is important to recognize that large fee–for–service payments and salaries without productivity standards or quality standards are equally likely to influence the care of individual patients and should be scrutinized with equal seriousness. Similarly, things like the size of a physician's panel of patients, its cultural variety, or morbidity can affect relationships because of their influence on time available per patient visit.
When taking on responsibility for a panel of patients, physicians could be said to join a relationship in theory that does not yet exist in reality. Physicians, working with their plan, should spearhead efforts to reach out to such members if only to ensure they are educated about preventive medicine issues and encourage them to follow healthy lifestyles. Although patients and doctors alike will not find frequent visits necessary when someone remains healthy, still the relationship between patient and physician may become important later, should the patient become seriously ill. Something as simple as an annual “Health Care Maintenance Reminder” postcard (with the doctor's name) may help members feel their faceless doctor is nonetheless caring for them. Developing relationships with all enrolled members is also a way for physicians and plans to become more accountable for the care of those who are not seen in clinical practice.
A number of strategies that MCOs can use to strengthen doctor–patient relationships are listed in Table 2. Often, plans do not know how to detect and remediate problems in doctor–patient relationships, how to train their practitioners and their staff to relate effectively and efficiently, or how to train their enrollees to be effective in their own care. As we now know how to do all of these things, there is no longer justification for poor performance in the encounters between providers and patients. Doctors need training in dealing with difficult patients, about common aspects of life adjustment such as reaction to illness, in recognizing the underlying psychological problems that remain a leading cause of seeking medical care, in negotiating, and in handling tough situations like breaking bad news. Courses such as those of the American Academy on Physician and Patient (AAPP) can provide such skill. Patients need to be taught to organize their approach to care, to ask questions, to negotiate, and to discuss feelings. The AAPP, the Northwest Institute, the Bager Institute, and others can provide such training.
Plans can promote a culture that is patient- and member-centered. This variation on “put the customer first” acknowledges the vulnerability of patients as ill persons needing care, compassion, and special attention. It also implicitly and explicitly makes care, not profit, the center of attention for those doing the daily work of providing health care. Physicians and other clinicians are encouraged to put their patients' good first, ahead of profit (their own or the organization's), politics (e.g., reluctance to whistleblow or disclose mistakes), or personnel (e.g., the convenience of the other staff). Conserving resources for future patients or to expand services becomes an important part of serving the member population. Although creating a culture that is patient-centered is not a quick or easy task, there are resources available.44
It is useful for plans to separate patient care from administrative rules communication. Too often, the practitioner is the person who has the difficult task of saying “no” to a patient.45 Plans can be purposefully deceptive or vague in communicating what they will not do for a member, when they are trying to enroll new members.46 It would ease the situation between doctor and patient if the patient clearly understood when the doctor said no that (when applicable) this is not the doctor's decision but the plan's. This approach is likely to require regulatory change.
Plans can structure contracts with employers that encourage accountability to the membership rather than the employer. It is hard to balance the competing interests of sick and well members, those who need resources now and those who may need them later, staff and the community. Employers' standing in decisions that affect primarily their employee members adds more complexity, and is fraught with conflict. The illusion remains that employers pay for health insurance. Actually their not paying the premiums would increase real wages for their employees, drop the cost of living, increase profits, or increase income due to greater competitiveness. This illusion, however, affects how health insurers view their accountability. Managed care plans do what it takes to please employers, because employees are their customers. The member, sick or well, has little voice. One way to alleviate this situation is to ensure that members have a voice, either through their employer or union, or in the health plan itself, for example, through representation on guideline development initiatives or benefits committees. If policies can be said to be self-imposed by the membership, physicians making judgments about resource use are acting for their patients, current and future, and not for employers.47, 48 Another strategy is to require management to use the same plans their employees do.
Plans must eliminate intrusive incentives in contracting with physicians. Intrusive incentives are those that combine strength (i.e., are large either in absolute or relative terms) with a tight linkage to individual patient care decisions. If a single decision about a single patient (including the decision to accept a chronically ill person into one's practice) is likely to result in a significant financial loss to the physician, then the relevant incentive is too intrusive. The intrusiveness of incentives is a product of the incentive's size (e.g., how much money is at stake) and its link to individual care decisions. For instance, if referring a patient to a specialist “costs” a physician a loss out of the physician's pool, it is tightly linked. If, however, a prepaid arrangement covers several thousand patients, the relative size (or impact) of the incentive is small. Incentives need not be only financial; peer pressure, leisure time, the threat of deselection, or a sense of fulfillment from work may also influence patient care decisions and thus also should be subject to scrutiny.
Plans can standardize “with heart.” Moderating the variation in clinical practice has often been touted as a way to save money without compromising quality of care. Yet some variation is necessary and inevitable. An organization that does not allow clinicians to open the gate for the justifiable exception to the rule, or is overly skeptical of clinical judgment about those with rare or poorly characterized conditions, ignores to its peril the rich variety of the human condition.
The openness and honesty of a system or organization can contribute to a climate of trustworthiness. For instance, discrepancies between marketing messages (“we provide everything”) and the availability of medications, equipment, or specialty care (“that's not covered in your plan”) create entitlement and convert it to disenchantment, resulting in an atmosphere of distrust that inevitably includes the doctor–patient relationship. Health care organizations may not relish the idea of promoting honest talk about limited resources and their consequences, but should at a minimum not try to raise expectations of unlimited access to unlimited services.
Plans should promote patient privacy and confidentiality. The expectation of privacy is one of the most important aspects of the doctor–patient relationship and influences the disposition to trust, but confidentiality is no longer solely in the doctor's control. Organizational personnel have access to patient information and must be required to keep it private, taught how to keep it private, and monitored to be sure they do.
Time is another prerequisite for trust. Plans should determine a reasonable minimum average time for doctor visits. They should pay attention when doctors or patients complain they do not have enough time together. Because the time of visit varies by type of visit, type of doctor, and complexity of the patient, patient complaints about visit time may be a useful patient-centered indicator of potential trouble in doctor–patient relationships.
Plans can encourage consideration of psychosocial issues in all forms of patient care. An organization can use continuing education, promotional materials, patient-directed education, and quality improvement efforts to promote this aspect of patient care. In doing so, discussions about these areas between doctors and patients will be enabled, patient satisfaction will increase, and unnecessary visits, such as to the emergency department for panic attacks, may even go down. Organizational change may be a more efficient way to promote caring than changing either medical education or the process by which medical students are selected.49
Plans should avoid business decisions that interrupt continuity between doctors and patients. Mergers and acquisitions, adding and deleting physician groups, agreeing to short-term contracts with employers, expanding or selling out, all are decisions with profound implications for one-on-one relationships between doctors and patients. To minimize harm when these decisions are unavoidable, exceptions can be made for those with important, established relationships. The “old doctor” may accept the standard fee, or the patient may be willing to contribute to some degree. If necessary, the patient's care can be gradually (as opposed to abruptly) established with a new physician “in the plan.” The latter strategy enables patients to take control over their choice of doctors and gives them time to find one acceptable to them in the network.
As Chairman Mao said, the first step in solving a problem is calling it by its right name.50 Only then can it be discussed and its particular features in a given site identified. The second step is agreeing on its high priority. The third step is obtaining appropriate consultation and choosing solutions. The solution will often be training practitioners and staff. To everyone's regret, there is no quick fix here although major improvements can be initiated in as short as a daylong course.51 Such interventions need to be part of an ongoing commitment to this area, steady work through a continuous quality improvement-type process, and regular training and renewal of skills. Groups like the AAPP can provide such long-range training efforts. Many plans already monitor practitioner skills in these areas through patient satisfaction surveys, and these may effectively identify those needing extra help. Attention to the training of patients is another critical part of creating effective partners for care. So also is employers' education as to the importance of this area, as their decisions may be critical in directing resource allocation. Finally, we believe the medical profession needs to provide data-based standards and establish principles physicians will not violate and to which plans must adhere. Otherwise, this will be done in a haphazard way by corporate interests.
We have outlined briefly the fundamentals of the doctor–patient relationship, some features of the health care system found particularly in managed care settings that affect it, and approaches for protecting and sustaining the doctor–patient relationship in these settings. These are aimed at physicians and plans, but should be of interest to policy makers, other health care administrators, and consumer groups. In change there is opportunity. Our current opportunity is to examine the doctor–patient relationship, the context in which that relationship operates, and in particular, the influence of changes in the financing and organization of health care. The doctor–patient relationship deserves our serious attention and protection during these dangerous times.
Dr. Goold's contribution to this work was supported by the Picker Commonwealth Scholars Program (National Program Office, Association of Health Services Research, 1130 Connecticut Ave., N.W., Suite 700, Washington, DC 20036) and the Department of Veterans Affairs.